SDVASDFV

u <OU_1 64297 >m
OSMANIA UNIVERSITY LIBRARY

Accession No.
Author
Title
This byok should be returned on or before the date Last marked bdtov
EDUCATIONAL PSYCHOLOGY
THE MACMILLAN COMPANY
MEW YORK BOSTON CHICAGO DALLAS
ATLANTA SAN FRANCISCO
MACMILLAN & CO., LIMITED

LONDON BOMBAY CALCUTTA
MELBOURNE
THE MACMILLAN CO. OF CANADA. LTD.
TORONTO
BY
DANIEL STARCH, PH. Do
UNIVERSITY OK WISCONSIN
fork
THE MACMILLAN COMP^
1924
jt right'
CoPYttlQHT, 1919,
BY THE MACMIL1AN COMPANY.
Set up and elcclrolyped. Published JUH

Reprinted January, April, August, December.
mber. 19*0;
19'
November, 1921; October, 19*2; April, OctoL

19*3; January, 19*4.
'
VTING COMPANY
TO MY WIFE
PREFACE
The preparation of this book has been carried out according to
two fundamental purposes: First, to present that material which
seems to be most useful and relevant to the problems of educa-
tional psychology; and second, to maintain a strictly experimental,
viewpoint in discussing these problems. The result of

scientific
these aims has been a considerable reduction in the amount of
space usually devoted in texts on educational psychology to cer-
tain topics such as, instinct, fatigue, and imagery, and the inclusion
of new topics such as tests of intelligence, studying, transference
of training in school subjects, the assignment of marks, and much
of the material in Part III which has as yet not found a place in
text-books.
The space devoted to the discussion of instinct has been ma-
terially reduced for two reasons: In the first place, while the in-
stincts are fundamental in human life, too much time has usually
been devoted to their consideration for the amount of direct
benefit gained. The actual use in school work that can be made
of a detailed knowledge of instincts, which in our present stage
of information is largely analytical and theoretical, is relatively
small when it comes to dealing face to face with concrete school
problems. In the second place, a great deal of experimental

and statistical material has accumulated in recent years which is
more immediately valuable in solving the problems of the 1 ii
<y
-
chology and pedagogy of learning.

It would have been desirable to include a discussion of the
psychology of more of the high school subjects; but this is impossi-

ble at the present time. The discussion of the school subjects in
Part III has been confined to tangible, scientific investigations.
Obviously there is little or no material of this sort on most of the
high school subjects. The consideration of educational tests in
the chapters of Part III is perhaps brief; but a detailed treatment
of the theoretical and statistical principles underlying their con-
struction belongs rather in special treatises. XII on
Chapter
How to Study is not altogether satisfactory, because of the scar-
city of definite or substantial material in this field. It was, how-

vii
viu PREFACE
ever, included because the topic is exceedingly important in school

work and because it was hoped that its inclusion would stimulate
discussion of it by teachers and prospective teachers.
I take pleasure in expressing obligations to the persons who
my
have assisted me in various ways in the preparation of this book;
namely, to Dr. Helen Hubbert Caldwell and Mr. A. O. Hansen,
who have read the manuscript and offered many helpful sugges-
tions, to Mr. W. R. Ames who has prepared the drawings, and
especially to Dr. C. L. Hull who has critically examined every
portion of the manuscript and has offered many suggestions which
have been incorporated in the book.
DANIEL STARCH.
MADISON, WISCONSIN,
October 5, 1918.
CONTENTS
CHAPTER PAGE
I. Problems and Scope of Educational Psychology i
PART I. THE NATIVE EQUIPMENT OF HUMAN BEINGS

II. The Instinctive Elements of Native Equipment 9
III. Variation in Human Capacities 26
IV. Correlation Among Human Capacities 49
^V. Sex Differences 63
'VI. The Inheritance of Mental Traits 73
VII. The Measurement of Mental Capacities 97
PART II. THE PSYCHOLOGY OF LEARNING: A. IN GENERAL

nil. Analysis of Problems 115
IX. The Reception of Stimuli: A. Sensory Defects 121
X. The Reception of Stimuli: B. Perception and Observation of
Sensory Material 132
XI. The Rate and Progress of Learning 141
XII. How to Study 176
XIII. Transference of Training in Special Mental Functions 191
XIV. Transference of Training in Abilities in School Subjects. . .
.217
PART III. THE PSYCHOLOGY OF LEARNING: B. OF SCHOOL SUBJECTS

XV. The Psychology of Learning School Subjects 259
/
XVI. Reading 261
XVII. Handwriting 207
322
^VIII. Spelling
XIX. Language 349
y XX. Arithmetic 374
XXI. History 416
XXII. Marks as Measures of School Work 426
Bibliography 451
Index 465
iz
CHAPTER I
PROBLEMS AND SCOPE OF EDUCATIONAL PSYCHOLOGY
What is Education? The problems and the scope of educational

psychology are necessarily determined by our conception of what
education is. If we conceive education to be primarily self-devel:
opment, our problems will be of one sort; if we conceive education
to be fundamentally social adaptation, our problems will be of
another sort. In the former case, education would mean the
complete training of the mental and physical capacities irrespective
of environment; in the latter case, education would mean the
training of those capacities which will adapt the individual most
adequately to the social and physical environment in which he is
to live. For our present purpose it is not necessary to define in
complete detail the aim and meaning of education. It will be
terms what education is as a
sufficient to state in the simplest
psychological process.
JnThelBfoadest sense, educationis
thejDroduction of useful
changes in human beings.
1
These~cnanges may be classified into
three" drvlslonsT cMhges in knowledge, in skill^ and in ideals.
Through education the child is to acquire useful knowledge; he is
to acquire skill, both motor and intellectual, in the use of his

muscles and in the manipulation of ideas and concepts; and,
finally, he is to acquire the right ideals of life which will actually
function in his behavior. Probably all changes wrought in human
beings which in any sense are educational, fall under these three
heads. Obviously then, education is the most momentous, as well
as the most essential, business of the human race; for the welfare
of the race depends upon education as it depends upon nothing
else.
1
Thorndike has defined the purpose of education thus: "The aim of education is, as
we have seen, to change uman
1 beings for the better, so that they will have more humane
and useful wants and be inore able to satisfy them." ('12, p. 52.)
2 EDUCATIONAL PSYCHOLOGY
Which changes are useful and which are not is a question that
cannot be answered as easily as it would seem at first glance.
Learning to read or to figure are obviously useful modifications;
but it is not so easy to say whether the study of a given drama, or
the knowledge of certain facts of history, or the understanding of
a certain theory of matter, or ability to read a given foreign lan-
guage, are useful, or sufficiently useful to be included in the com-
mon school, in the high school, or in the university. The term
useful should not be limited narrowly to the things which are
immediately applicable in making a living, but should include
all ^changes which will broaden and enrich, the life of the in-
. . .
dividual.
TEe Problems of Educational Psychology. In accordance with
our definition, the fundamental problems that we must raise
concerning education are as follows:
1. What changes are to be made in human beings?
2. What are the agencies by which the changes may be brought
about?
3. What are the capacities which human beings possess for
acquiring the changes?
4. What are the most economical methods by which these
changes may be brought about?

The first problem is primarily a problem for philosophy and
sociology. What changes are ultimately to be made depends upon
our idgaTs of life and our views of society. The modifications
that have been sought by different nations and different sys-
tems of education have varied from century to century and
from race to race. The ultimate aims of education sought by
the ancient Greeks or by the mediaeval monks were very
different from those sought by the modern Americans or
Europeans.
The second problem deals primarily with the value of school
subjects and exercises in bringing about the changes that are to be
made. To what extent will the study of arithmetic, the study of
grammar, or the study of physics or Latin be able to produce the
training that philosophy and sociology dictate? This problem is
partly sociological and partly psychological. It is sociological in
so far as the determination of educational ager.cies depends upon
the physical and social environment of mankind; it is psychological
in so far as it necessitates a study of mental processes affected or
PROBLEMS AND SCOPE 3
brought about by these agencies. This latter phase of the problem

merges into problems three and four.
Problems three and four are fundamentally psychological and,
together with the second phase of problem two, constitute the
main scope of educational psychology. It is a psychological prob-
lem to determine what capacity and equipment human beings
have for acquiring the changes that are to be made. Likewise, it
is fundamentally a psychological problem to discover the most
economical methods of learning. Accordingly, the field of edu-

cational psychology is divided into two large divisions which we
may designate as:

I. The native equipment of human beings;
II. The psychology of learning.
Psychology and Teaching. Methods of teaching rest funda-
mentally upon the psychology of learning. Since the experimental
analysis of learning processes will have to reveal the principles ac-
cording to which the human mind learns, and learns most economic-
ally, it follows that the methods of teaching will have to be based
upon these discoveries. This may be illustrated in the case of
reading. If the child learns to read most economically by the word
method, it follows that the most economical way of teaching read-

ing would be by the word method. Likewise, if a child learns to
spell homonyms more easily by studying them together, or memo-
rizes prose or poetry more readily by wholes than by parts, it follows
that these exercises should be taught accordingly. Evidently the
fundamental principles of teaching must be based upon the psy-
chological laws and principles of economic learning.
Waste in Education. Exact information concerning the proper
procedure in educational matters is exceedingly rare. Definite,
scientific knowledge of the proper methods of learning and teaching
school subjects and of the efficient administration of our schools
is surprisingly small, and the field of educational psychology in its
broadest sense opens up endless problems for the future to solve.
We know relatively little in a scientific way about the learning of
any single school subject. For example, we do not know with any
definite assurance what is the most economical amount of time to
devote to any one of the school subjects. From such investigations
as have been made, we may infer that there is an enormous waste
in our educational practices which is indicated by such facts as the
following: It has been found by recent tests and measurements

that some schools obtain

just as good results by devoting only
one-half as much time
to writing as other schools do. Similar facts
have been brought out in the case of reading, arithmetic and other
school subjects. Schools which have devoted as much as 100
minutes a week, or 20 minutes a day, have obtained no better re-
sults than other schools devoting 50 minutes a week, or 10 minutes
a day, to the same subject. If these facts actually represent the
real possibilities, it seems quite obvious that there is an enormous
waste in our schools and this waste is far greater than we realize
until we make definite calculations of the possible saving of time.
If by some means it were possible to save one minute a day for
every school day during the eight years of a child's school life,
we would be able to save one entire week of school time. If we
could save four minutes a day for the same length of time, we
would be able to save one month; if we were able to save 18 min-
utes a day, we would be able to save one-half of a school year; and
if by more economical methods of learning and distribution of
time we were able to save 36 minutes a day for eight years, we

would be able to save an entire school year. Such a saving is not
impossible; indeed, by a better use of time and more effective
methods of learning, it is highly probable. Eighteen minutes a
day would mean a reduction of only tf4 minutes in each of four
subjects; 36 minutes a day would mean a reduction of only 9
minutes a day in each of four subjects. This time could be de-
voted with greater advantage to other and possibly more advanced
school subjects and school exercises.
The Specific Topics and Problems. In order that we may be
properly orientated with reference to the problems that will be
discussed under the two large divisions of educational psychology,
the following order of topics will be considered:
I. The native equipment of human beings.

a. What does it consist of?
b. To what extent does it vary?
1. Among individuals.
(a) In single traits.

(b) In combinations and relationships of traits.
2. At different times of life in the same individual.

3. Between the sexes.
c. To what extent is it inherited?
d. How may it be measured?
PROBLEMS AND SCOPE 5
II. The Psychology of Learning.

a. The psychology of learning in general.
1. Observation and perception.
2. The rate and progress of learning.
3. Transference of training.
b. The psychology of learning school subjects in particular.
1. The psychological processes involved in each subject.
2. The measurement of ability and progress in learning
each subject.
3. The most economical methods of learning the material
of each subject.
PARTI
THE NATIVE EQUIPMENT OF HUMAN BEINGS
CHAPTER H
THE INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT
Reflexes, Instincts, and Capacities. The equipment with

which human beings start in life may be divided into three types
of inherited responses and abilities: Reflexes, instincts, and capaci-
ties. The distinction among these three is primarily one of definite-
ness and degree of complexity. AnUnstinct/niay be defined from

the neurological side as an inborn neural connection between sense
organ and muscle. It may be denned from the functional side
as an inborn capacity of responding in definite ways under definite
circumstances. These responses are prior to experience and train-
ing,and need not be learned. To close the eyes when an object
suddenly approaches them, to get food when hungry, to strike
when struck, and to be afraid of thunder and of the dark, are illus-
trations of instinctive responses. They reflexive /md instinctive
responses are inherited in the sense that there is present in the
nervous system, either at the time of birth or later on as a result
of growth, a set of nervous connections already formed for the
carrying out of a particular action in response to a given situation.
If the child closes his eyes when an object suddenly approaches,
it means that the mo tor(impulses) travel from the retina to the
visual center of the brain, from there to the motor center which
controls the movement of the eyelids, and from there out to the
muscles of the eyelids to cause the contraction. In the case of
inherited responses, the connection from the sensory to the motor
centers is already present and ready to operate in carrying out
the action. In the case of acquired responses, such as habits,

these nervous connections must be formed us a result of effort
and on the part of the individual.
trial
The difference between reflexes and instincts is largely a differ-

ence of complexity. Both are inherited types of responses. Re-
flexes are simpler forms of reaction usually involving a limited set
of muscles and occurring in response to precise stimuli. The
contraction or expansion of the iris, the closing of the eyelids, the
knee jerk, are illustrations of reflexes. Instincts are complex re-
actions involving the use of large groups of muscles or, in many

instances, the entire muscular system of the body. They may be
aroused either by extenial stimuli or situations or possibly by
internal stimulation. To make movements in the direction of
getting food when hungry, to seek shelter when cold, to offer re-
sistance when hemmed in, to spit out what tastes bad, and the
like, are instinctive reactions. Capacities are distinguished from
reflexes and instincts in bemjênera^naeiital abilities rather than
specific motor responses and in referring primarily to the native
mental equipment, such as the powers of sensation, perception,
retention, attention, imagination, and all the varied complex
psychic processes. f
Classification of Instinctive Responses. The olderjd^ssifica-
tions of instincts usually divided them intq^three or four large
groups of responses, such as individual, racial, and social, and re-
garded them rather as generartendencies than as specific responses.
The presentconccption of Instincts is to regard them as specific
responses with inherited neuraj. mechanisms which will be set
into action by specific stimuli or situations. On
this basis the
classification consists of an enumeration of as many definite, identi-
unlearned reactions to specific situations as can be observed
fiable,
and as can be recognized in human beings prior to training and
habituation in each particular type of activity. Accordingly,
Thorndike ('14, 1) enumerates forty or more different types of in-
stinctive reactions as follows:
1. Food getting and protective responses.

1. Eating.
2. Reaching, grasping, and putting objects into the
mouth.
3. Acquisition and possession.
4. Hunting.
5. Collecting and hoarding.
6. Avoidance and repulsion.
7. Rivalry and cooperation.
8. Habitation.
9. Response to confinement.
10. Migration and domesticity.
11. Fear.
12. Fighting.
13. Anger.
INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT II
II. Responses to behavior of other human beings.

14. Motherly behavior.
15. Gregariousness.
1 6. Responses of attention to other human beings.
17. Attention getting.
1 8. Responses to approving and to scornful behavior.
19. Responses by approving and scornful behavior.
20. Mastering and submissive behavior.
21. Display.
22. Shyness.
23. Self-conscious behavior.

24. Sex behavior.
25. Secretiveness.
26. Rivalry.
27. Cooperation.
28. Suggestibility and opposition.
29. Envious and jealous behavior.
30. Greed.
31. Ownership.
32. Kindliness.
33. Teasing, tormenting, and bullying.
34. Imitation.
III. Minor bodily movements and cerebral connections.
35. Vocalizations.
36. Visual exploration.
37. Manipulation.
38. Cleanliness.
39. Curiosity.
40. Multiform mental activities.
41. Multiform physical activities.
42. Play.
Relation of Education to Native Endowment. The inherited

equipment of the human being is the foundation upon which educa-
tion must build; it consists of the faculties and capacities which
the child has for reacting to his environment. It is the utilization,
the training or the curbing of these endowments which education
attempts to accomplish. In much of the writing and thinking con-
cerning educational problems, there has been a relative overem-
phasis, in space and time, upon instincts and an underemphasis upon
the mental capacities. Education in the sense of schooling has as
much if not more to do with the latter than with the former. The
direct appeal to, and use of, instinctive reactions in actual concrete
instances in school work are not as frequent and specific as is com-
monly implied. The number of instincts enumerated in the pre-
ceding list which may be directly and concretely appealed to in the
learning of a school subject is relatively small. The best way to be
convinced on this point is to take the various instincts one by one,
and to determine to what extent each one may be appealed to or
used in teaching the various subjects. The number of specific
applications is much smaller than one is likely to anticipate. Two-
thirds or three-fourths of them are probably never immediately
but only indirectly concerned in school exercises, and most of the
remaining ones, such as rivalry, cooperation, collecting and hoard-
ing, are serviceable chiefly as general motives. As such they are,
to be sure, highly important.
Wemust, of course, not minimize .the place and importance of
instinctive reactions in behavior as a whole. They furnish the
general motives and mechanisms for doing and learning, but the
mental capacities are more directly and concretely involved in the
acquisition of knowledge and skill in school subjects.
The instinctive elements in learning any school subject are for
the most part simple reflex actions or undeveloped connections.
Take learning to read as an illustration. The chief instinctive
elements probably are the reflexes in the control of the eyes, the
neural mechanism for receiving and transmitting visual impulses
to the brain, the capacities for attentiveness and retentiveness,
and partial motor control of the speech organs. The process of
learning to read assumes these and uses them; but, what is
more important from the practical side of getting the meaning
of the printed word is the establishment of countless new cbn-
nections.
Perhaps the most important role of instinct in education lies

|
and energizing the learning processes. There can
in motivating
^be
no education except through the activity of the child himself;
and no activity can take place which does not ultimately depend
upon native tendencies. They are the origin of effort, the springs of
action. The skillful teacher plays upon them and appeals to them
in countless ways. The ability to do this is an art which is not
from books; it is acquired rather by patient practice
easily learned
and by sympathetic contact with children.
The energizing poêr of instinct makes itself felt largely through
INSTINCTIVE ELEMENTS OF NATIVE EQUIPMENT 13
itscontrol of the attention processes. Owing to the peculiarities

of our inherited nervous organization, certain impressions have a
potency over others in attracting the attention and interest of the

child. A flash of lightning, a holiday parade, one's name in the
newspaper, or a moving picture make certain instinctive appeals to
the attention of a young girl which the study-lamp, the doing of
errands for mother, the seeing of a stranger's name in the newspaper,
or the reading of the history lesson do not make. The great im-'
portance of attention for the learning process lies in the fact that
associations, analyses, and indeed all mental processes are carried
out much more effectively when they occupy the focus of attention.
Ebbinghaus found that, after inattentively reading over nonsense
syllables until many successive persons had learned them per-
he himself could repeat very few of them. Impressions
fectly,
must occupy the focus of consciousness in order to be retained
effectively.
Considering the three main types of attention, passive, active,
and secondary passive, the most simple and the one most directly
related to the instinctive life is probably the first. Passive atten-
tion is such as one gives spontaneously to any curious or interesting
sight or sound. Active^ attention is such as one gives perhaps to an
inherently distasteful task which requires an effort of will to keep
the mind upon it While such a task itself does not supply the
stimulus for vigorous instinctive reaction, it is in some direct way
connected with one that does. A little girl will apply herself to the
disagreeable task of learning a spelling lesson, not because the
words in themselves have any charm "for her but because she has
the instinctive craving for the approval of her teacher. The third
type of attention, secondary passive or derived attention, is at-
tention which has become passive~onTy after having passed through
an initial active stage. It is illustrated by the common experience
of becoming absorbed in a task which at first required a distinct
effort. In the beginning the motivation lay outside the task, say
in a sense of duty or social obligation; but after a time an adequate
stimulus for activity was encountered in the work itself.
Apparently back of every act of attention lies somewhere a more
or less primitive, innate tendency to action. To focus the atten-
tion of a class upon various associations involved in learning the
multiplication table, a skillful teacher may on one day appeal to

curiosity in the novelty of the combinations; on another day she
may appeal to the native pleasure in rhythm by making the table
into a rhyme or song. Or the tendency to play will be utilized by

making the number combinations into a game. More than likely
the game itself will depend upon instinctive rivalry and emulation.
The love of social approval is appealed to by giving distinctions,
marks, gilt stars, and the like. Future advantage may be used as
an inducement for present application. The instinct of pugnacity
may be utilized in wanting to succeed in a hard task. The teacher
herself, in standing position with face and body in animated
attitudes, may appeal to the fundamental interest in change and
action. Lastly may be mentioned the more doubtful negative
motives of deprivation from coveted privileges and, biologically
perhaps the most fundamental and powerful of all motives, physical
pain. It may be worth noting that animal psychologists have
found pain in some cases a more potent motive to learning than
pleasure. Hoge and Stocking ('12) found that rats when rewarded
by food alone had by no means learned perfectly certain sensory
habits in 610 trials; when they were punished for failures they
learned the habits perfectly in this number of trials, but when
they were both rewarded for successes and punished for failures
they learned the habits perfectly in 530 trials.
Educational Doctrines Based upon Instincts. Some very far-
reaching speculations and theories with regard to the nature and

value of instincts for education have been spun out, some of which
are largely imaginary and questionable, and are based upon analogy
rather than fact. For the most part these educational doctrines
liave centered around three concepts.
The first is that instincts are the great dynamic forces of human
nature which determine the actions, desires, and achievements in an
individual's life. Hence the injunction to the school has been to
work with nature rather than against her or apart from her. We
!
hall call this the dynamic theory of instincts.
The second is that these instincts are highly transitory; that they
burst out at certain times in the growth of the individual with
more or less dramatic force and suddenness, and that if they are
not allowed to manifest themselves, they will disappear never to be
revived again. From this assumption has been derived the peda-
gogical application of the maxim, "strike while the iron is hot."
This is the theory of the transitoriness of instincts.
The is that instincts appear in the
third growth of the child in
the order in which they appeared in the evolution of the race. From
this assumption has been derived the pedagogical maxim, "teach
the child his activities in the order in which the interests for
them appear." This is known as the recapitulation theory of
instincts.
Critique of the Dynamic Theory of Instincts. To work with

nature rather than against her is undoubtedly a sound principle.
The fundamental instincts of man are the driving forces of human
-life that determine ultimately the motives and causes of behavior.
They are so deep-seated in the human psychophysicai organism

that we may almost say that to work apart from, or against, nature
is a futile task. If, through the instinct of multiform activities, a
child manifests a tendency to draw, the school should take ad-

vantage of this original impulse and build upon it. All the original
manifestations of a child's nature should be used in the acquisi-
tion and training of those exercises which education considers
valuable.
This dynamic theory of instincts, however, involves on the one
hand a difficulty and on the other a questionable assumption. The
difficulty is that the principle is general and as such does
nni-
pninfr
ojLitJhejjarticular ways
inwhich the school may cooperate with the
inborn forces of 'child nature. It is easy enough to say "work with
nature," but just how is that to be done in teaching a pupil how to
make the letter "a," or to learn the reading of a printed word, or to
acquire correct speech, or to learn the grammatical rules of a foreign
language? Ultimately, the concrete use of the principle must be
determined experimentally. Our definite knowledge of the tech-
nique of learning in the case of school subjects is appallingly limited.
Only by careful and painstaking experimentation can this principle
be made useful in the concrete work of the school in anything more
than an offhand impressionistic manner.
The questionable assumption is that the instincts are infallible
guides of human life. It may be argued that since instincts are*
such powerful springs to action as to have maintained the individual
and the race for numberless generations, they must necessarily be
dependable in producing action and interest of the right sort. But
the question may fairly be raised: Are the native tendencies always
right so that we should always cooperate with them and never coun-
teract or curb them? The theory of the infallibility of instincts is
based on the belief that for countless ages nature has found by ex-
perimentation and natural selection what is best for the individual.
Whatever the child is inclined to do by virtue of his natural pro-
clivities is right and good for him; or, if apparently not useful, it is a
necessary precursor or a necessary accompaniment of useful tenden-

cies. On the theory of immunization or catharsis, the undesirable
tendencies prepare the ground for the proper development and
growth of the desirable ones. However, the theory of catharsis
is highly questionable and runs counter to the law of use and disuse
which in general operates to the effect of making permanent the
functions exercised. The belief is that if a boy has proclivities
toward thieving or lying or dishonesty, and is allowed to exercise
these unconstrained, he will purge himself of these tendencies and
be the more honest and truthful later on in life. Concrete data,
however, in addition to the general principle of use and disuse, seem
to point in the opposite direction. Experimental and statistical
inquiries show that the relative strength of the various traits
remains fairly constant throughout life; that if a child manifests
certain abilities and tendencies even during childhood, these abili-
ties and tendencies will remain relatively dominant during his adult
life. Early interests and intellectual capacities are very certain, on
the whole, of being prophecies of similar interests and intellectual

abilities later in life. As will be pointed out in a succeeding chapter,
scholastic ability remains fairly constant in any given child all
through his educational career.
Furthermore, we must remember that nature, in the develop-
ment of instincts and in securing adaptation to environment,
works on the whole in a very slow and prodigal manner; and that
conditions of life may change long before the organism through its
behavior appropriately adapts itself to the surroundings by fur-
nishing the necessary native equipment within the organism. On
this account a great deal of our native equipment is out of date and
has adapted man to primitive conditions of uncivilized life. As a
result of this, we manifest many tendencies which are not par-
ticularly useful at the present time. We would be better off if in

place of them we had instinctive capacities for meeting situations
with which we arc to-day confronted in civilized life. As Thorn*
dike has pointed out:
"The imperfections and misleadings of original nature are in fact

many and momentous. The common good requires that each child
learn countlessnew lessons and unlearn a large fraction of his natural
birthright.The main reason for this is that original equipment is archaic,
adapting the human animal for the life that might be led by a family
group of wild men in the woods, amongst the brute forces of land, water,
wind, rain, plants, animals, and other groups of wild men. The life to
which original nature adapts man is probably far more like the life of
the wolf or ape, than like the life that now is, as a result of human art,
habit and reasoning, perpetuating themselves in language, tools, build-
ings, books and customs." ('14, 1, p. 280.)
That these primitive tendencies persist with great strength is

shown by the ready manner in which the veneer of civilization comes
off and by the fact that men and women in strained circumstances
wilt easily revert to their primitive, brutal instincts. The chief ad-
vocate of the theory of infallibility and catharsis of instincts has
been G. Stanley Hall, who has said:
"In education, don't cut off the tadpole's tail."

"Rousseau would leave prepubesccnt years to nature and to these
primal hereditary impulsions and allow the fundamental traits of savagery
their fling till twelve. Biological psychology finds many and cogent
reasons to confirm this view if only a proper environment could be pro-
vided. The child revels in savagery; and if its tribal, predatory, hunting,
fishing, fighting, roving, idle, playing proclivities could be indulged in
the country and under conditions that now, alas! seem hopelessly ideal,
they could conceivably be so organized and directed as to be far more
truly humanistic and liberal than all that the best modern school can
provide. Rudimentary organs of the soul, now suppressed, perverted, or
delayed, to crop out in menacing forms later, would be developed in
their season so that we should be immune to them in maturer years, on
the principle of the Aristotelian catharsis for which I have tried to sug-
gest a far broader application than the Stagirite could see in his day."
(Hall, '08, p. 2.)
"lie should have fought, whipped and been whipped, used language
offensive to the prude and to the prim precisian, been in some scrapes,
had something to do with bad, if more with good, associates, and been
exposed to and already recovering from as many forms of cthicalj|umps
and measles as, by having in mild form now he can be rendered immune
to later when they become far more dangerous, because his moral and
religious as well as his rational nature is normally rudimentary." ('08,
P- 235.)
The Theory of the Transitoriness of Instincts. This concep-

tion of instincts contains two elements: First, the suddenness of the
appearance of instincts, and second, the unrevivabte disappearance
of instincts. The most conspicuous advocate of the former idea in
this country, has been G. Stanley Hall, who states it as follows:
"But with the teens all this begins to be changed and many of these
precepts must be gradually reversed. There is an outburst of growth
i8 EDUCATIONAL PSYCHOLOGY
that needs a large part of the total kinetic energy of the body. There is
;i new interest in adults, a passion to be treated like one's elders, to
170
160
150
'Z 140
^
1 180
120
55 6.5 7.5 8.5 9.5 10.5 11.5 12.5 13.5 14.5 15.5 16.5 17.5 18.5
Age in Years
FIG. i. Height of boys and girls measured in centimeters, based on measure-

ments of 45,151 boys and 43,298 girls. After Boas ('96-97)-
make plans for the future, a new sensitiveness to adult praise or blame.
The large muscles have their innings and there is a new clumsiness of
body and mind. The blood-vessels expand and blushing is increased,
new sensations and feelings arise, the imagination blossoms, love of
140
16 17 18
FIG. 2. Rate of tapping. Number of taps made with right hand in 30

seconds. Based on tests with 395 boys and 495 girls. After Smedley
('oo-'oi).
nature is born, music is felt in a new, more inward way, fatigue come.,
easier and sooner; and if heredity and environment enable the individual
to cross this bridge successfully, there is sometimes almost a break of

continuity, and a new being emerges." ('08, p. 236.)
The correctness of the theory of sudden appearance is primarily

a question of fact. Thus Hall describes the social instincts at the
time of adolescence as follows:
"The social instincts undergo sudden unfoldment and the new life of
love awakens. and of religion, of rapid fluctua-
It is the age of sentiment
tion of mood, and the world seems strange and new. Interest in adult
life and in vocations develops. Youth awakes to a new world and under-
stands neither it nor himself." ('04, Preface p. XV.)
The advocates of this viewpoint maintain, therefore, that there is

a nascent period for motor activity, for memory and habituation,
for reason and logical thinking and the like; that the school should
seize these opportunities to teach those activities which will exer-
cise the particular capacities that occupy the stage of youth at that
period; and that more can be accomplished at those periods in a

given length of time than can be accomplished in several fold as
much time later on.
The facts do not seem to warrant an interpretation of such
marked suddenness but indicate rather a gradual waxing of in-
stincts. There appears to be no special time during which the child
suddenly begins to reason or to reason very much more than he had

done theretofore. The same description seems to be true of
memory, motor and other capaci-
ability, the collecting instinct,
ties, as indicated in the accompanying graphs.
A great deal of the dramatic bursting forth of instincts is chiefly
a dramatic bursting forth of descriptive words. The actual facts
seem to justify more nearly an interpretation of gradual unfold-
ment instead of a sudden bursting forth. Growth in height and
weight proceeds by a very uniform increase even during the adoles-
cent period. Motor capacity grows steadily and uniformly without
particularly sudden leaps or bounds. Memory ability increases
steadily for both rote and logical material up to adulthood, during
which it probably remains fairly constant until senility sets in.
There is no memory period during which the child memorizes very
much more readily than he did before or than he ever will later.
The memory of the average adult for both mechanical and logical
material and for either immediate or permanent retention is superior
to the memory of the average child at any age. Even reasoning
ability, which is usually described as appearing suddenly at the

dawn of adolescence, is a matter of gradual development. To argue
FIG. 3. Memory for digits based upon tests of 937 pupils. After Smedle>
FIG. 4. Development in arithmetical reasoning as measured by the Courtis

test No. A.
8, Series The vertical axis shows the number of problems done cor-
rectly in SAX minutes.
that school exercises which consist mainly of memorizing should be

placed at the "memory age" on the ground that the pupil will
learn them more readily at that time than at a later time in life,
is fallacious. It may be advisable to begin the study of foreign
languages earlier than is customary, but not for any reason of more
rapid memorizing at an earlier age. If rapidity of tapping, Figure 2,
is any indication at all of endurance or of quickness of becoming
fatigued, it does not seem to be true that "fatigue comes easier

and sooner" during the adolescent stage. The graphs for tapping
FIG. 5. Development in arithmetical reasoning as measured by the arith-

metical scale A. After Starch ('16). The vertical axis represents the scale step
or the number of problems done correctly.
do not drop but tend to rise gradually even during the years from
eleven to fifteen.There is practically a level at the age of eleven
but no drop.
The unrevivability of instincts through disuse has been advocated
chiefly by James as follows:
"This leads us to the law of transitorin^ss which is this: Many in-

stincts ripen at a certain age and then fade away. A
consequence of
this law is that if, during the time of such an instinct's vivacity, objects
adequate to arouse it are met with, a habit of acting on them is formed,

which remains when the original instinct has passed away; but that if
no such objects are met with, then no habit will be formed; and, later
on in life, when the animal meets the objects, he will altogether fail to
react, as at the earlier epoch he would instinctively have done." ('90,
ii, P 39 8.)
.
"Leaving lower animals aside, and turning to human instincts, we

see the law of transiency corroborated on the widest scale by the altera-
tion of different interests and passions as human life goes on. With the
child, life is all play and fairy-talcs and learning the external properties
of 'things'; with the youth, it is bodily exercises of a more systematic
sort, novels of the real world, boon-fellowship and song, friendship and
love, nature, travel and adventure, science and philosophy; with the
678 10 11 12 13 14 15 16 17
Ages
FIG. 6. The number of collections made by children at various ages. After
;
C. F. Burk ( oo).
man, ambition and policy, acquisitiveness, responsibility to others, and

the selfish zest of the battle of life. If a boy grows up alone at the age of
games and sports, and learns neither to play ball, nor row, nor sail, nor
ride, nor skate, nor fish, nor shoot, probably he will be sedentary to the
end of his days; and, though the best of opportunities be afforded him
for learning these things later, it is a hundred to one but he will pass
them by and shrink back from the effort of taking those necessary first
steps the prospect of which, at an earlier age, would have filled him with
eager delight. The sexual passion expires after a protracted reign; but
it is well known that its peculiar manifestations in a given individual
depend almost entirely on the habits he may form during the early
period of its activity. Exposure to bad company then makes him a
loose liver all his days; chastity kept at first makes the same easy later
on. In all pedagogy the great thingis to strike the iron while hot, and
to seize the wave

of the pupil's interest in each successive subject before
its ebb has come, so that knowledge may be got and a habit of skill
acquired a headway of interest, in short secured, on which afterward
the individual may float. There is a happy moment for fixing skill in
drawing, for making boys collectors in natural history, and presently
dissectors and botanists; then for initiating them into the harmonics of
mechanics and the wonders of physical and chemical law. Later, intro-
spective psychology and the metaphysical and religious mysteries take
their turn; and, last of all, the drama of human affairs and worldly
wisdom in the widest sense of the term." ('90, II, pp. 400-401.)
Facts seem to support this aspect of the nature of instincts more

than the theory of the sudden appearance. The law of disuse of
functions is necessarily in general support of the theory.
Any
functions will, as a rule, be strengthened through exercise. The
assumption, however, that instincts, if not exercised when they first
manifest themselves, will become dormant beyond the possibility
of reawakening, or that they actually become dormant, is question-
able. James gives isolated instances in favor of his viewpoint.
Experimental and comprehensive observations are missing at the
present time. Isolated illustrations of the opposite viewpoint,
however, also are to be found. Thus it frequently happens that
through the change of circumstances of life, instincts apparently
long dormant or never given opportunity to manifest themselves,
will quickly appear for action. For example, the writer has a friend
who as a boy had never developed the instinctive tendencies in-
volved in fishing. About the age of thirty, through the opportuni-
ties of a new environment, the instinct appeared so strongly that
he will go to great lengths at any time of day or night to follow this
sport. But isolated instances are dangerous bases on which to
generalize, and future inquiries will have to solve the problem.
Many instincts apparently are dormant only because no opportu-
nity of expressing themselves are at hand, or because other more
dominant interests prevail, but may, when appropriate circum-
stances arise, rapidly appear for action.
The Recapitulation Theory of Instincts. The principle of re-
capitulation was formulated by biologists to account for the develop-
ment of animal organisms in the early stages of their growth.
The theory of recapitulation holds that the individual retraces in
its growth the successive stages of development of the entire animal
series. Thus Hall says:
"Holding that the child and the race are each keys to the other, I have
constantly suggested phyletic explanations. ..." ('04, I, p. VIII.)
"The best index and guide to the stated activities of adults in past
ages is found in the instinctive, untaught, and non-imitative plays of
children. ... In play every mood and movement is instinct with
heredity. Thus we rehearse the activities of our ancestors, back we know
not how far, and repeat their life work in summative and adumbrated
ways. It is reminiscent, albeit unconsciously of our line of descent, and
each is the key to the other. . . . Thus stage by stage we enact their
(our ancestors') lives. Once in the phylon many of these activities were
elaborated in the life and death struggle for existence. Now the elements
and combinations oldest in the muscle history of the race are re-presented
earliest in the individual, and those later follow in order." ('04, I,
pp. 202-203.)
And Puffer says: "We are by turns vertebrates, gill-breathing verte-
brates, lung-breathing vertebrates (we make the great change at birth),
little monkeys, little savages, and finally civilized men and women."
('12, p. 77,)
The evidences given for the principle of recapitulation are largely

embryological and structural. Vestigial organs such as the vermi-
form appendix, gill slits, etc., are further cited as evidences of the
remainder of structures once useful. The facts seem to be that
such recapitulation as takes place is very brief and confined almost
wholly to the prenatal period of an individual's development.
Davidson, after a comprehensive review of the biological evidence
for the theory, concludes thus:
"The history of recapitulation is an instructive one. A

principle of
limited application within the field of its origin was elevated to a position
of wide generality, and so gave rise to a conception in the main mislead-
ing. Carried into anew territory without a sufficient examination of
its merits, it was applied broadly as an explanatory principle and thus
distributed its misleading influence beyond its own borders." (' 14, p. 99.)
"A more thorough consideration of the facts has led to a view of de-
velopment essentially contradictory to this recapitulatory one. Ontogeny
represents the ancient life-cycle which as such has been transmitted from
the beginning. The chronological sequence from egg to maturity is not
a rehearsal of a like historical series of events throughout the phylogeny
of species; it is but the recurrence of an order which has been repeated
in the lifetime of each individual from the beginning. In general, the
effect of the modifications induced by germinal mutations and selection
in the successive ontogenies, to make them over, and to destroy the
.resemblance of later ones to their predecessors."
The recapitulation theory with its pedagogical corollate, the

culture epochs theory, has been developed largely as an analogy
with many of the analogues missing. Its usefulness for educational
thinking seems to the writer to be greatly exaggerated. It has
built pedagogical mountains out of biological molehills. It is
primarily an anatomical principle proposed to account for the

embryological development of biological organisms, and has been
brought over into human behavior to explain on the one hand, the
order and dates of appearances of instincts, and to furnish on the
other hand, a basis for the order and dates of teaching subjects in
the school curriculum. The former assumption is more or less
dubious, since most, if not all, of the demonstrable recapitulation
occurs before birth; and the latter assumption is quite certainly
dubious, since the anatomical and probably also the functional
recapitulation has long ceased when the child begins his definite
education.
CHAPTER III
VARIATION IN HUMAN CAPACITIES
How May They be Measured and Represented? Differences

among human beings are quantitative rather than qualitative.
That is, all human beings have the same reflexes, instincts, and
capacities; all have the powers of perception, discrimination, at-
tentiveness, retentiveness, reasoning, and so forth. All persons,
consequently, have the same general qualitative make-up. The

variations from person to person are, therefore, primarily differ-
ences in the strength of the various abilities that each individual
possesses, and in the manner in which amounts of the various
20-
210
F*o- 7. Distribution of memory ability of 173 University students. The

test consisted in dictating ten monosyllabic nouns. The persons then recorded
the words that they remembered. The horizontal axis indicates the number of
words and the vertical axis indicates the number of persons having each memory
ability.
26
60 70 80
Letters per Minute
FIG. 8. Distribution of ability in the A-test. Based on 164 University stu-

dents. The horizontal axis represents the number of A's canceled in one minute;
the vertical axis represents the number of persons of each ability.
85i
30
25
15
10
1284 66 7 8 9 10 11 12 18 14 15
Figures per Minute

Q. Distribution of ability in canceling a certain geometrical figure in *
page of figures. Time allowed, one minute. Based on 164 persons.
traits combine in the same person. The differences are qualitative

'
only in the sense that combinations of varying amounts of diverse

traits occur.
The most convenient manner in which to represent and deter-
mine the amount of variation in a given trait is by means of the dis-
tribution curve, or the surface of frequency. The distribution
curve is a curve designed to represent how frequently each amount
10 20 80 40
Associations per 15 Seconds
FIG. 10. Distribution of ability in giving associations in response to a

stimulus word. The horizontal axis gives the number of words given in 15 sec-
onds. The vertical axis gives the number of persons of each ability. Based on
135 persons.
or strength of a given trait occurs in a given group of persons. The

range of ability from a small amount to a large amount is repre-
sented along the base line, or x axis, from left to right, and the
number of times each particular ability occurs is represented ver-
along the ordinates, or y axis. (See Figures 7, 8,. 9, and 10.)

tically
How Wide are the Differences? The investigation of this
problem in recent years has brought out the fact that the differ-
ences among human beings are very much greater than has com-
monly been thought. If we measure a group of pupils in a given
VARIATION IN HUMAN CAPACITIES 29
class or grade, we find that on the average the best pupil is able to
do from two to twenty-five times as much as the poorest pupil,
or is able to do the same task from two to twenty-five times as well
as the poorest pupil. The accompanying table shows the range of
differences between the highest and the lowest in a series of tests
made upon fifty university students.
TABLE i
Range between the best and the poorest in a series of mental

of differences
tests. Based upon the writer s Experiments in Educational Psychology,
page 8, which may be consulted for the nature of the tests.
What is the Nature of the Variation? From the general ap-

pearance and form of the distribution curves of mental traits, we
note that abilities range without break from the lowest to the
highest. Our common terminology of dividing groups of persons
into various classes as dull, mediocre, and bright, on the assumption
that they may be divided into distinct classes with gaps between
them, is psychologically incorrect. The fact rather is that all
grades of ability, varying by infinitesimally small amounts from
the lowest to the highest, are found in the human species.
The next conspicuous feature about the nature of the distribu-
tion of mental abilities is the general shape of the curve. This
indicates that the large majority of individuals cluster about the
center. In the accompanying illustration it will be noticed that
ifthe entire range of abilities is divided along the base line into
three equal sections so that we may designate the one at the right
as the superior section, the one in the middle as the medium sec-
tion, and the one at the left as the inferior section, we find that
approximately two-thirds, or 66% of all persons fall into the middle

third; one-sixth or 17% fall into the superior one- third, and the
remaining one-sixth, or 17% fall into the inferior one-third of the
range of abilities. In other words, the normal distribution curve

is a symmetrical, bell-shaped figure, having its mode in the center
and dropping at first rather gradually, then very rapidly and finally
"
very slowly. The statement attributed to Lincoln that God must
"
have loved the common people because He made so many of them
is psychologically true. If the middle third of the entire range
of abilities represents the common people, then two- thirds of all
persons are common people.
Fie. ii. Distribution of chest measurements of English soldiers.
The third interesting fact to be noted is that psychological and

biological traits vary universally in the same manner in conformity
with the normal, bell-shaped curve. Note for example the dis-
tribution of such biological traits as chest measurement, height,
girth of head, and so forth, as represented in the accompanying
graphs, Figures n, 12, and 13. The number of men who are ex-
tremely tall or extremely short very small, and the number less
is
tall or less short is larger and larger as the median is being ap-
proached. This uniformity throughout organic nature is an in-
teresting and significant fact. Apparently nowhere are there traits
which are discontinuous so that gaps would exist within the ranges
do we find that, on the whole, traits are distributed
of the traits, nor
68 69 70
FIG. 12. Distribution of the height of 1,052 women.
in a skewed manner, so that the great majority of individuals

would lie either in the upper or lower range of abilities.
250
200
i 160
100
61 62 63 54 66 56 67 58 59 60 61
Head Girths in Centimeters
FIG. 13. Distribution of the head girth of 1,071 boys, 16-19 years of age,
Finally, the variation in both psychological and biological traits

occurs apparently according to the law of chance, that is, according
to the frequency of occurrence of a chance event. Consequently,

on this assumption, the statistical treatment of the distribution of
mental abilities becomes subject to the mathematical properties of
the probability integral. What we mean by the statement that the
variation occurs according to the law of chance may be illustrated
in the following manner: If we toss up ten pennies at one time,
count the number of heads up and keep a record of it, then repeat
the tossing a thousand times and keep a record each time, it will
be found that the number of tunes no heads are up will occur very
240
210
180
160
120
90
60
80
10
FIG. 14. Distribution of the number of heads up in tossing ten pennies 1,000
times. The horizontal axis gives the number of possible heads up in each tossing;
the vertical axis gives the number of times each number of heads was up.
rarely, likewise, thenumber of times all ten heads are up will occur
very rarely, the number of times one head is up or nine heads are
up will occur less rarely, and as you approach from either side
toward four, five, and six, the occurrences will be more and more
frequent. The actual records of a thousand such tossings are
represented in Figure 14. It would seem as though nature, in the
production of her creatures, aimed at a target. The largest number
of trials strikes somewhere near the bull's-eye, a smaller number
strikes within the next circle, and a still smaller number within the
next circle, and so on. The correspondence between the actual
distribution of abilities and the values of the probability integral is
exceedingly useful in permitting statistical treatment of series of
measurements of any trait. Figure 15 gives the mathematical

or theoretical probability curve.
Variation in Abilities in School Subjects. The differences in
FIG. 15. The theoretical probability curve.
abilities in school subjects are fully as wide as in special psycho-

logical capacities. They are probably due primarily to native
FIG. Distribution of pupils (in one school) in grades 2 to 8 in reading

1 6.
measured by the author's tests. The horizontal axis represents speed

ability as
and comprehension combined in terms of speed, i. e., words read per second.
ability rather than to differences in opportunity, training, or en-

vironment. Table 2 shows the range of difference in ability
Grade 8
. IS
Grade 7
Grade
Grade 5
Grade 4
Grade 8
Grade 2
20 30 40 60 60 70 80 00 100 HO 120
FIG. 17.Distribution and overlapping of pupils in writing. Speed and
quality combined into one score as explained in the author's Educational
Measurements. The numbers along the horizontal axis represent speed and
quality in terms of speed, i. e., letters per minute. Quality was measured by the
Thorndike scale.
GradeS
!..yj::
Grade?
Grade6
Grade 5
Grade 4
GradeS
I
Grade 2
102080406060708090100
FIG. 18. Distribution and overlapping of pupils in spelling as measured by
the author's test. The numbers along the horizontal axis are the numbers of
words spelled out of a list of 100.
in various school subjects as found in a class of 36 eighth-grade
pupils. Abilities in reading, arithmetical reasoning, spelling,
grammar, and history were measured by the author's tests ('16).
FIG. 19. Distribution and overlapping of pupils in ability to solve arith-

metical problems as measured by the author's Scale A. The numbers along the
horizontal axis are the steps on the scale.
Quality of writing was

measured by the Ayres scale, the four
fundamental operations in arithmetic by the Courtis tests (iseries
B), and composition by
the Hillegas scale ('12.)
TABLE 2
Ranges of difference between the best and the poorest in a class of 36 eighth-
grade pupils.
In the accompanying diagrams, Figures 16-26, the complete

distribution of the abilities of the pupils in each grade in the sub-
jects of reading, writing, spelling, etc., are shown as determined
by methods of measurement described elsewhere. These graphs
show that the range from the lowest to the highest ability in any
given subject within any given grade, is approximately as great
as that found for special mental functions referred to in a preceding
section. The best pupil in reading or spelling or any school subject
isfrom one and a half to twenty-five times as capable as the poorest
pupil. As a result of this wide range of abilities, there appears an
enormous amount of overlapping of the abilities possessed by the
pupils in other grades in the same school. Thus it will be noted
that the best pupil in arithmetical reasoning in the third grade is
as capable as the poorest pupil in the eighth grade. All pupils
had been tested by the same set of problems. The same statement
applies with practically identical details to any school subject.
Putting the situation in a little different statement, it has been
shown that 60% of the best pupils in any given grade could be
exchanged with the 60% of the poorest pupils in the next higher
grade, with the result that there would be no change in average
ability of the two grades.
The question next arising is this: Granting that the range of
ability in any one subject is as large as the results of the tests show
it to be, may, however, a given pupil not be two or three years
ahead of his grade in arithmetic, two or three years behind his
grade in spelling, up to the average of his class in reading, etc.,

and may he not be placed correctly, after all? The facts, however,
5 10 15
FIG. 20. Distribution and overlapping in addition as measured by the

Courtis test. The numbers along the horliontal ffîs represent the number of
examples done correctly.
seem to be as represented in the accompanying illustration, Figure

27, in which a combined score was obtained for each pupil as fol-
lows: In reading and writing in grade r; in reading, writing, and
spelling in grade 2;. in reading, writing, spelling, and arithmetic
in grades 3 and 4; and in reading, writing, spelling, arithmetic,
language, and composition in grades 5 to 8. Even when the varia-

tions in abilities in different subjects possessed by the same pupil
are counterbalanced and averaged, the range of abilities and the

overlapping is practically as large. It will be noticed, for ex-
ample, that the best pupils in the second and third grades in these
three subjects combined, are almost up to the ability of the poorest
pupils in the eighth grade. The fact is that in every eighth grade
Grade 8
Grade 7
Grade6
Grade5
5 10 15
FIG. 21. Distributionand overlapping in use of correct English as measured

by the author's Grammatical Scale A. The numbers along the baseline are
the steps on that scale.
one pupil in nine is fully equal in ability to the average ability of
the pupils in the second year of high school and could do the work
to be
equally well if he had been allowed to go on rapidly enough
in the second year of high school. Two pupils in every nine are
equal in ability to the average pupil in the first year of high school,
three of the nine pupils are correctly placed in the eighth grade,
two are equal only to the average seventh grader, and one is equal
only to the average sixth grader. Thus by proper promotion or
classification, one pupil in every nine could save two years in eight,
and two pupils in every nine could save one year in eight.
Expressing the same facts in a different form for the school

population as a whole, we may say that:
i pupil in 100 could finish the 8 grades in 4 yrs. or at 10 yrs. of age.

2 pupils 5 II
9 6 12
21
"
7 13
"
33 8 14
"
21 9 15
" 10
4
16
9
"
2 ii 17
i pupil 12 18
10 20 80 40 60 60 80 00 100
FIG. 22. Distribution and overlapping in geography as measured by the

author's geography test. The numbers along the horizontal axis are the scores
in the test. The situation in the case of history is very similar.
The last two groups are composed of pupils so retarded that

they probably never would or could complete the elementary
school. The variation in ability is so great that the children of
any given age are spread out over about nine years of maturity.
For example, children ten years old range in ability all the way
from fourteen-year-olds to six-year-olds or less, and the numbers
of pupils at each age of mentality are approximately those given
above. These facts are further borne out by recent tests of in-
telligence. (See Chapter VII.)
This enormous range of ability and the resulting overlapping of
successive grades, is probably the most important single fact discovered
with reference to education in the last decade. The import of it is so
significant of the situation as it exists in our schools to-day and of
the possibilities in the direction of the proper reclassification or
10 40 50 60 70 80
FIG. 23. Distribution and overlapping in ability to write a composition as

rated by theHiliegas scale. The numbers along the base are values on that
scale.
10 15
FIG. 24. Distribution and overlapping in drawing ability. The numbers

along the horizontal axis are the units of Thorndike's drawing scale.
readjustment of pupils according to ability that we have scarcely

begun to realize how great the differences are or in what manner
the readjustments may be made.
Provisions Made in the School for the Variations in Abilities.
Experimental work has drawn renewed attention to the possibilities
of taking account of the enormous ranges of abilities such as are
Year 4
YcarS
Year 2
Yearl
20 40- 60 100
Scale Values
FIG. 25. Distribution and overlapping of pupils in a high school in ability to

write an English composition. The numbers along the horizontal axis are
values on the Hillegas scale.
found even in an ordinary class of supposedly homogeneous pupils.

To keep an ordinary class of pupils together is no doubt very
wasteful in time both for the gifted as well as for the stupid pupils.
The gifted must listen to questions
and explanations designed
of the dull pupils, but which the bright
chiefly for the benefit
pupils already understand.
The dull pupils, on the other hand,
waste time by being dragged along too rapidly in the endeavor
to keep the bright pupils occupied.
The plans which have been proposed for meeting the varying
two general classes :Jii5l, those which
abilities of pupils fall into
attempt to keep the pupils of a given class together but vary the
manner of instruction for the pupils of different capacities j^sgpond,
those which keep the manner of instruction uniform but promote
or retard pupils according to their achievements.
High School
10
Seniors
High School
Uo Freshmen
,123456789101112181415
Steps -Grammatical Scale A
FIG. 26. Distribution and overlapping of pupils in a high school in ability
in discriminating between correct and incorrect English. The numbers along
the horizontal axis are the steps on the author's Grammatical Scale A.
The principal schemes of the first general method which have

been tried in various schools are known as the individual instruc-
tion oi^Pueblo the monitorial group plan, the extra-work
plan,}
plan, and the supervised study or BatavjajDlan. The individual
instruction plan was employed by Superintendent P. W. Search
in Pueblo, Colorado, and consisted in the abolition of all class
instruction and the substitution of individual teaching according
to the needs of the pupils. The monitorial group plan is carried
out by dividing a class into several groups, usually three, according
VARIATION IN HUMAN CAPACITY 43
to the abilities of the pupils, and by appointing a monitor for each
group from among the members of the class. The extra-work

plan consists in having recitation and class instruction chiefly for
those who need it, and in assigning additional work to the capable
pupils to be done at their desks. The supervised study plan de-
0.5 1.0 1.5 2.0 2.5 3.0 4.0 4.5 50 5.5

Average Scores
FIG. 27. Distribution and overlapping of pupils when their attainments in

different subjects are averaged. Reported in a thesis by Helen Craig in the
library of the University of Wisconsin, 1918.
votes a part of the class period to the usual recitation and instruc-
tional work, and the remainder to study done under the supervision
of the teacher. Sometimes the class period is considerably length-
ened and no home study is done; at other times, the class period
is kept at the usual length with some assignments for home
study.
These plans have been in operation in various schools during
the past thirty years with varying amounts of success or failure.

Most of them have been successful when carried out under the
immediate supervision of the persons who devised them. The
difficulty,however, has usually been that when others have at-
tempted to use them they have not been so satisfactory. Some
of the schemes have been objectionable on other grounds also.
For example, the individual instruction plan is in part unsatisfac-
tory because it removes a large share of the social stimulus and
interaction that is derived from class instruction.
The one type of plan which is being adopted on an extensive
scale and is proving to be generally applicable, is some form of the
supervised, study plan. The methods with which this plan is
carried out diher considerably and great care must be taken to
avoid formality in the division of the time between recitation and
study during the class period and in the order and manner in which
the supervision is carried out. A more detailed discussion will
be given in the chapter on "How to Study" where this subject
properly belongs.
The schemes coming under the second general proposi-
different
namely,
tion, that of keeping the manner of instruction constant
and varying the rate of promotion, have been applied widely,
and many plans designed to produce greater flexi-
different
bility in the rate of promotion have been worked out in various

school systems. As illustrations, two plans will be mentioned
because they have been in successful operation for many years.
In Cambridge, Massachusetts, there has been in operation a plan
for some twenty years, in which the work of grades three to eight
is laid out in three different courses of study. Pursuit of course A
permits the completion of the remaining six grades in six years;
pursuit of course B permits the completion of the work in five
years; and the pursuit of course C makes possible the completion
of the six grades in four years. Transfer from one course to another
may take place at any time.
In the St. Louis schools a method of promotion has been in force
for a great many^years which consists in dividing the school year
into four quarters of ten weeks each. Promotion can be made
at the end of each quarter. Pupils who have made a grade of
excellent maybe promoted to the next higher class at the end of
any ten-week period, and pupils who have made very low grades
or practically failed, must repeat their work beginning with the
various ten-week periods.
The effect of this

plan in shortening the time of a considerable
is shown in a study made
proportion of pupils by W. J. Stevens. 1
This investigation shows the length of time required by each of
1,439 pupils in four elementary schools in St. Louis to complete the
eight grades.
TABLE 3
The average attendance per grade of 1,439 pupils, graduates, required to com-
plete each of the eight grades. Forty weeks is assigned in the course of
btudy for each grade.
NUMBER OF AVERAGE NUMBER OF NUMBER OF AVERAGE NUMBER OF
PUPILS WEEKS TO COMPLETE PUPILS WEEKS TO COMPLETE
EACH GRADE EACH GRADE
i 33
2 18 49 42
I 20 29 43
I 21 27 44
8 22 19 45
8 23 20 46
13 24 15 47
25 9 48
19 26 5 49
25 27 4 50
46 28 4
43 29 2 52
52 30 2 53
83 31 2 54
32 I 55
99 33 2 56
109 34 2 57
92 35 2 58
no 36 I 59
87 37 I 60
104 38 2 62
39 I 63
95
87 40 2 70
Median ..35 weeks
Total average time of attendance. . . 288 weeks
.
To do 320 weeks' work

Double promotions -.17%
Normal promotions -.67%
Repeaters ,.x6%
It will be noticed from this table that thirteen pupils completed

the eight grades in an average of twenty weeks to do forty weeks
1
Reported in a thesis (1914) in the library of the University of Wisconsin.
The study
. C. Elliott.
was carried out under the direction of Professor
of work, that is, amount of time. In other

in half of the prescribed
words, about i% of
the pupils required approximately four years,
6.3% five years, 22.8% six years, 34.6% seven years, 24.9% eight
years, 7.6% nine years, 1.7% ten years, and 1.3% eleven to thirteen
years to complete the eight grades. These results agree quite
closely with the figures suggested on page 39.
Promotion by subjects is a plan adopted in various schools. The
program must be arranged so that all grades recite in the same
subject at the same period in order that a pupil may do his work
with the particular class to which he belongs. For example, a fifth-
grade pupil might recite in spelling with a seventh-grade class, in
reading with a sixth-grade class, in arithmetic with a fourth-grade
class, and so on.
In high school work there is equal need for flexibility in progress.
Plans should be devised whereby a class could be divided into
three sections, a rapid, a normal, and a slow section. For example,
an algebra class, after some early tests, could be divided into three
divisions. Section A could easily do the year's work in two-thirds
of the year and then pass on to geometry or more advanced algebra
or even some other subject. Section B could do the normal work
in the year, and Section C could take a year and a third to do the
normal year's work, or could cover only two- thirds of the ground
in the year and receive only two-thirds credit. Differences in
ability are sufficiently great to make possible as much dif-
ference in progress as is here indicated. The more capable
pupils could easily shorten their high school course by half a year
or a year.
A plan of flexible promotion that can be administered success-
fully has in many respects distinct advantages over any plan which
merely varies the instruction for the reason that it allows the
capable pupil really to gain the advantage of his ability; because
he is able to shorten his elementary school period, which is one of
the aims striven for at the present time. The elementary school
course is considered too long. Any plan which varies the method
of instruction so as to require more work of the capable pupil no
doubt occupies the time of these pupils and gives them the benefit
of the additional work achieved, but it does not give the pupil the
full benefit that he deserves in accordance with his capacities. In
practical life the capable man performs several times as much

work or makes several times as rapid progress in the same period
of time as the incapable man, both having equal opportunities.
Why should not the schools permit progress according to ability

and achievement? Greater flexibility in promotion or retardation
is an advantage both to the more gifted and to the less gifted
pu-
pils. The former will be able to step forward whenever they are
ready and the latter will not need to step back so far whenever a
part of the work must be gone over again. Promotion once a year
works to the disadvantage of both types of pupils. The bright
pupils cannot well jump an entire year and so will not be able to
progress as rapidly as their abilities warrant, while the slower
pupils will have to repeat an entire year when a quarter or half of
a year would be sufficient. School progress is determined too
much by the calendar and not enough by capacity. The most
capable one-third of pupils are advanced too slowly, and the least
capable one-third are advanced too rapidly. A saving of half a
year or a year on the part of a fourth or a third of the pupils would
be of inestimable value to the pupils themselves and to the com-
munity at large, either in getting an earlier start in their life
work, or, preferably, in securing more advanced and thorough

training.
Finally, one of the most important, if not the most important
aspect of the principle of progress according to performance, is its
appeal to the individual to do the best he can. Few incentives are
as strong as the personal impulse of going forward as rapidly as
possible and of putting forth the best that is in one. If a child
knows that, if he can spell as well as the pupils in the grade above
him, he will be put with them, he will be stimulated as he would
be in no other way to reach that degree of attainment. Likewise,
if he knows that he is likely to be put back to recite in spelling
with the pupils of a lower grade if he falls behind, he will put forth
his best efforts to hold his own. Dawdling could hardly be en-
couraged more than it is in many of our schools. Rewards in
adult life are more nearly according to ability and performance.
The same conditions would work to the advantage of school
life.
The schools have given special attention to the backward pupils

by organizing separate classes for them and by giving them extra
help, but they have given little or no attention to the advanced
pupils. Society would be compensated far more for paying at
least equal attention to the gifted pupils since they primarily will
determine the future progress of mankind. The leaders of society
will come from the right end rather than from the left end of the
distribution curve. Wisdom would dictate that we devote at

least as much care and thought to them, that we surround them
with an atmosphere of high aspiration and achievement and stim-
ulate to the full their powers of originality and discovery. This
would make for maximum progress based upon ability and per-
formance, not upon birth or social caste.
CHAPTER IV
CORRELATION AMONG HUMAN CAPACITIES
Problem. Any given single trait varies over an enormously

wide range among the members of the human race as a whole.
The problem, however, before us now is: To what extent is a given
amount of any capacity accompanied in general in the same person
by equal, larger or smaller amounts of any other ability? To what
extent a good memory in the same person accompanied by an
is
equally good capacity for reasoning or attention or perception or

judgment? To what extent is poor or mediocre abiUty in memory
accompanied by poor or mediocre ability in other directions? If
all mental abilities were measured on a scale of o to TO, the con-
crete problem would be: To what extent would a memory ability

of 7 be accompanied in the same person by a perception ability of
7, or a judgment ability of 7? If it is not accompanied by the

same amount of other abilities, by how large or small an amount
of any other ability is it
accompanied?
Educationally the problem is important and takes the following
form: To what extent may we expect pupils, who are excellent,
mediocre, or poor in one subject to be excellent, mediocre or poor in
other subjects? To what extent is a statement such as the following
true in general: "I simply cannot learn languages or mathematics,
although I get along very well in my other studies" ? To what
extent is freedom of electives in studies justifiable on the basis of
variation in the combination of capacities in the same individual?
To what extent are mental and physical traits correlated? To
what extent are abilities similar at different times of life in the
same individual? To what extent is ability in childhood or
youth
a forerunner of ability in adulthood?
Methods of Measuring Combinations of Traits. The extent
to which various amounts of abilities accompany one another is
measured or expressed definitely by the coefficient of correlation.
The value of the coefficient of correlation ranges from i.oo through
o to -i.oo. A coefficient of correlation of i.oo means a complete
agreement. If the coefficient of correlation between ability in
49
Latin and ability in German were i.oo, it would mean that the
best pupil in Latin would be also the best pupil in German, the
second best pupil in Latin would be the second best pupil in Ger-
man, etc., down to the poorest pupil in Latin who would also be
the poorest in German. As the correlation drops farther and
farther below i.oo toward o, the closeness of this agreement be-
comes correspondingly less until o is reached. If the coefficient
of correlation between ability in Latin and ability in German
were i.oo, it would mean that the best pupil in Latin would be
the poorest pupil hi German, the second best pupil in Latin would
be the second poorest in German, etc As the correlation rises
above i.oo toward o the reversal becomes less and less until o is
reached. A coefficient of o means that no relationship exists.
A pupil might have any amount of ability in one subject and any
other amount of ability in the other subject. 1
The Correlation Among Specific Mental Abilities. The early
investigations in this field found surprisingly small correlations
even among apparently very similar or closely related capacities.
Thus it was thought that a person might have a good
memory for
words but not for numbers or faces; he might have a keen percep-
tion of words but not of letters or geometrical figures and the like.
As typical of the earlier results on correlations we may cite a few
from Wissler (*oi) as follows:
Auditory memory of figures and visual memory of figures .29 to .39

11 " " "
auditory "passage .04
" " "
memory of length of line .00
" " " " " " "
"passage" .07
" " " " "
quickness in naming colors .03
" " " reaction time
.12
" " " " " "
figures .17
The significance of these coefficients may be interpreted approxi-

mately as follows: A
coefficient of o means that no correlation
exists, and roughly speaking, a coefficient of .30 or less is small
and practically means very little agreement. Correspondingly, a
coefficient lying between .30 and .50 means a moderate amount of
agreement, a coefficient between .50 and .75 means a considerable

correlation, while a coefficient above .75 indicates a very close
1
For methods of computing the coefficient of correlation, consult the author's Ex-
periments in Educational Psychology, (igir Edition) Chapter IV; Whipple, Manual
tf Mental and Physical Tests, Chapter III; Thorndike, Menial and Social Measurements,
Chapter XI; and Rugg, H. 0., Statistical Methods Applied to Education, Chapter DC,
CORRELATION AMONG HUMAN CAPACITIES 51
relationship and, as it approaches i.oo, indicates practically perfect

agreement. From the table, it appears that many coefficients are
very low and imply little or no agreement. The coefficients so
low as to indicate practically very little correlation are stated to
exist between auditory memory of figures and visual memory of
figures. This would mean that a person might have a very

good memory for figures seen but a poor memory for figures
heard.
The difficulty with these coefficients is that they are based upon
unreliable and incomplete measurements of the traits concerned.
Many of the measurements of the early investigations of correla-

tions were derived from group tests which had been made but
once. Measurements thus obtained have been shown by subse-
quent investigations to be rather uncertain indications of the real
amount of a given trait possessed by the individual. In order to
obtain a fairly accurate measurement of a given capacity, it is
necessary to repeat several times under favorable circumstances
the measurement of the trait in question. You cannot obtain
anything like an accurate measure of any amount or quantity
by a single measurement made under distracting conditions. If
we should wish to measure the memory capacity of a given in-
dividual, we should not consider the result very trustworthy if a
single test were given consisting of eight lines of poetry learned
in three or four minutes. We ought at least to repeat the test
with several similar passages, preferably on different days, and
derive therefrom an average. This is, in fact, the sort of procedure
that has been followed in subsequent researches. Inaccurate
measurements, as indicated by recent analyses of correlations,
tend to reduce very materially the computed coefficients below
the actual amounts of correlation.
Recent researches have shown that among many traits quite
close, and among other traits very close, correlations exist. An
investigation by J. A. Stevenson ('18) showed remarkably close
correlations between various types of sensory discrimination.
The plan of the investigation consisted in making extensive and
repeated measurements of discrimination in lengths of lines, in
intensities of sound, in degrees of brightness, in shades of gray and
in pressures on the end of the first finger, The correlations com-
puted on the basis of these measurements with ten persons were
as follows:
TABLE 4. After Stevenson ('18).
Lines and intensity of sound 90

Brightness and intensity of sound 90
Pressure and intensity of sound 36
Pressure and lines 39
Lines and brightness 92
Pressure and brightness 41
A similar investigation in the field of memory, conducted by

Miss N. F. Bennett ('16), showed on the basis of numerous and
repeated tests with nine subjects fairly close correlations among
the capacities to remember various kinds of material such as
syllables,numbers, nouns, prose, and faces, between visual and
auditory presentations of the material, and between mediate and
immediate learning. Her conclusions are stated thus:
"i. There a high correlation between mediate and immediate re-

is
tention if a number and variety of measurements for each

sufficient
type of memory are taken, and the results amalgamated to determine

ranks.
"
2. There is a high correlation between the memory span, or imme-
diate retention for disconnected materials, and the ability to learn the
same."
Hollingworth made a study to determine the increase in the

coefficients of correlationamong six different capacities with the
increase in the number of measurements made upon each capacity.
His results are set forth in the following table. They indicate a
very marked rise in the coefficients with the increase in the number
of tests.
TABLE 5
The average correlation of each test with all others at various points in the
curve of practice. After Hollingworth ('12).
As an illustration of a series of correlations among special mental

functions based upon measurements repeated several times but
not as frequently as those in the preceding tables, we may cite the
coefficients obtained by Simpson. These coefficients are unusually
high because they are based upon tests performed on two ex-
treme groups of subjects, the one a highly intelligent group and
the other distinctly unintelligent.
TABLE 6
Correlations among certain mental abiHt'cs. After Simpson ('12).
Burt obtained the following correlations (Table 7) from a va-

riety of tests of specialized mental functions made upon forty-three
pupils. The test designated as dotting was regarded as a measure
of voluntary attention; the tests designated as spot pattern, mirror
and memory were designed to measure memorial
and associative
capacities; the tests called alphabetand sorting referred to sensori-
motor capacities; dealing and tapping to motor functions; and
the remainder to sensory discrimination.
TABLE 7
After Burt ('09).
The import of the researches up to the present time seems quite

certainly to prove that the higher mental capacities are on the
whole rather closely correlated. The coefficients lie for the most
part above .50, and some of them reach up to .80 and .90. The
same statement holds approximately for sensory capacities among
themselves and also probably for motor capacities among them-
selves. The cross-correlations among from these three levels
traits
is, so far as we can judge at the present stage of our knowledge,

lower than among the traits within a given level. This seems to
be particularly true of the correlation of motor capacities with
intellectual capacities.
Correlations Among Abilities in School Subjects. The develop-
ment of knowledge concerning this aspect of our problem has had a
history similar to that of the special mental functions. The early
correlations among abilities in school subjects were computed
upon relatively uncertain data. About 1903, coefficients obtained
by various investigators, were as follows:
TABLE 8
Summary of coefficients of correlation between abilities in high school subjects

as reported up to about 1903. (Thorndike '03, pp. 26, 30-31).
B = After Burns, based on nearly 1,000 pupils.

P = After Parker, based on 245 pupils.
Br = After Brinckerhoff, Morris, and Thorndike.
In the case of grammar school subjects, A. G. Smith (Thorn-

dike '03 p. 13), computed the following correlations:
English and Mathematics 39

" "
Geography 43
" "
Drawing 15
Mathematics and Geography 36
" "
Drawing 14
" "
Geography 12
These coefficients for the most part Indicate only a moderate

amount of correlation. Thus Thorndike interpreted them in 1903
by the following statement: "For our purpose the most striking
thing about these figures is their small amount. It is safe
to say
that in a grammar or high school student a deviation from the
average ability in any one subject implies by and large a deviation
in any other not more than half as great. The most talented
scholar in one field will be less than half as talented in any other,
The most hopeless scholar in one field will in another be not so
very far below mediocrity." ('03, pp. 37-38).
The coefficients here quoted were based usually upon marks of a
single teacher in any given subject. Recent studies have called
attention to the unreliability of marks and the differences in stand-
ards of marking employed by See Chapter
different teachers.
XXII. This necessarily produces a considerable reduction in co-
efficients based upon them.
A computation based upon the average mark of each pupil in
each subject in grades five to eight yielded the following coefficients
(Table 9):
TABLE 9
Correlations among abilities in school subjects. After Starch ('13).
Arithmetic and language 85

" "
geography 83
" "
history 73
" "
reading 67
" "
spelling 55
Language and geography 85
" "
history 77
" "
reading 83
" "
spelling 71
Geography and history 81
" "
80
reading
" "
spelling 52
History and reading 67
" "
spelling 37
Reading and spelling 58
These coefficients are almost twice as high as those previously

quoted and represent very close correlations. They would warrant
the interpretation that the pupil who is good, mediocre, or poor
in a given subject, is good, mediocre, or poor to very nearly the
same, but not equal, degree in all other subjects so far as his abili-
ties are concerned. Such lack of agreement as does exist is due

probably to a difference of interest and industry on the part of
the pupil in different subjects at different times and to a real
difference in abilities in the various fields. Thus spelling ability
correlates apparently less closely with ability in other subjects
than abilities in these other subjects correlate among themselves.
The up-shot of the whole problem concerning the variation in the
combination of traits, or the extent to which different amounts of
mental traits accompany one another, may fairly be stated as
follows:
First, no negative correlations exist either among the abilities
involved in school subjects or among the special mental functions
measured by Popular and "short-story" psychology
special tests.
is false in and description of antagonisms of mental
the assumption
traits. They apparently do not exist among desirable and useful
traits. Advice, given to prospective wives, such as "if he is good-
natured, he may be lazy; if he is scholarly, he may be cold; if he is
thrifty, he may be stingy; if he is generous he may be wasteful,"
may produce caution, but it is not true psychology. Good-natured
men are probably on the whole no more lazy than ill-natured men
are, and scholarly men are probably on the whole no more cold-
hearted than stupid men are. In fact the opposite is more likely
"
to be true. And such statements as Johnny is very bright in read-
ing, but he simply cannot get arithmetic" is a soothing salve for the
feelings of parents, but not apt to be sound psychology.
Most of the opinions of students who state that they "simply can-
not get" mathematics or language or history are in part probably
due to a relatively small discrepancy in abilities, that is, to some-
what less ability in mathematics, language or whatever the sub-
ject may be; but to a larger extent they are illusory, because, when
the actual facts are obtained or when more careful measurements
of the abilities in various directions are made, the abilities corre-
late much more closely than the student's statements would lead
one to believe. As a
concrete example the following case of a col-
lege freshman, brought to the author's attention, may be cited: The
student claimed that he had always had great difficulty in learning
foreign languages but that other subjects were easy for him. He
stated that in high school he never was able to obtain a grade in
languages higher than about 75 but that in other subjects his grades
were always high, as high as 95. Since his trouble seemed to be
language it was thought that he might have a defective memory
$8 EDUCATIONAL PSYCHOLOGY
or an abnormal type of imagery. Some memory and imagery

he had normal memory and imagery of
tests revealed the fact that
average ability. This at once led to an inquiry into his actual

high school record to ascertain his grades. The various grades for
any given subject are final grades in different courses as follows:
English, 83, 80, 78, 81; History, 88, 75, 83; mathematics, 80, 87,
77; science, 87, 87; Latin, 77, 79, 75; German, 75, 75. When these
marks are compared there is little or nothing to explain. His
marks in Latin and German were somewhat lower than in other
subjects, which is probably largely explained by his own statement
that he "hated" languages, but they were not much lower on the
whole. The highest grade in any subject was in the first year of
history, 88, but he also had a grade of 75 in the second year of
history and 77 in the 3rd course in mathematics. There was no
grade of 95 in the entire list. This record was corroborated by the
grades which he received at the end of the first eight weeks of his
freshman year in college: Spanish, Fair; Geology, Fair; English,
Fair; Mathematics, Fair; History, Poor. His abilities are pretty
uniformly mediocre in all respects.
Exceptions do occur such as that of a boy seventeen years of age
in the second year of the high school who was able to carry his work
satisfactorily, but was able to read no more fluently, either orally
or silently, than the average pupil can at the end of the first grade.
He was a normally intelligent boy. Such cases occur perhaps once
among one or two hundred pupils, and may be regarded as ab-
normal.
Second, intellectual and scholastic abilities are for the most
part closely correlated. Barring certain exceptions, which are
rarer than is generally supposed, abilities are combined in fairly
similar amounts. Intercorrelations between the different levels,
intellectual, sensory, and motor, seem to be smaller and in some
traits, practically zero. Some of the motor abilities, such as hand-
writing, have practically no correlation with intelligence or general
mental abilities.
The wider bearing of the facts about the combinations of
mental with the distribution of mental traits
capacities, together
according to a continuous, bell-shaped curve discussed in the
preceding chapter, are deeply significant for the problem as to
whether there are distinct mental types. Mankind apparently
cannot be divided into three or four separate types. The ancient
classification of temperaments into sanguine, choleric, melancholic,
and phlegmatic, may be conveniently analogous to the four seasons

of the year, spring, summer, autumn, and winter respectively, but
Ithere are no mental types that correspond to such superficial
characteristics and none that are marked off sharply or even vaguely
from one another. If all members of the human race were to be
exhibited in a distribution curve whose base line represented from
"
left to right different amounts of sanguine-melancholic, or choleric-
phlegmatic" natures, the curves would in all probability not be a
series of four distinct curves separated from one another, nor even
possess four modes with depressions between them, but would very
likelybe single continuous distribution surfaces of the usual normal
form with one mode. The human beings who even remotely ap-
proach any one type are very rare. The rule is that each person
possesses more or less of all different traits, and within certain limits,
roughly similar amounts of the various traits. Persons in whom
the divergences are large are the exceptions rather than the rule.
Correlation between Special Mental Capacities and General
Intelligence. So far as definite data are available on this pointy
the inference be drawn that many special mental functions
may
are correlated anywhere from moderately to very closely with
general intelligence. Men of intelligence have, on the whole, keen'
powers of perception, observation, and attention, remarkable re-
tentiveness, exceptionally rapid and varied association processes,

as well as unusually incisive powers of analysis and soundness of
judgment. Wemay note here in passing, by turning to Chapter
VII, the amounts of correlation of certain capacities with general
estimated intelligence as found by Simpson, Burt, and others.
The usefulness of the facts that many specific mental capacities
are reliable symptoms or essential constituents ft f g*mpml mt Dili-
gence will be particularly important in the future in the develop-
ment of tests and methods of measuring intelligence. The value of
this to mankind, not only in education but in all fields of human
endeavor, can hardly be foretold at the present time. Further con-

sideration will be given to it in a later chapter.
Correlations between Mental and Physical Traits. In the case
of adults, the correlations between mental abilities and such physi-
cal characteristics as height, weight, size of head, lung capacity, or
strength of grip, are either very low or zero. In the case of children^
the situation somewhat different. B. T. Baldwin made an elabo-
is
and 1,063 g^ 8 m tne University of Chicago

rate study of 861 boys
elementary and high school, the F. W. Parker school of Chicago,
and the Horace Mann School of Columbia University. Measure-

ments of various physical characteristics were obtained at yearly
and half-yearly intervals on two groups of pupils. One group was
followed continuously through the ages from six to twelve, and the
other from twelve to eighteen, A parallel comparison between
the physical measurements and the school records of the same
pupils was then made. From these results, Baldwin has derived
the following conclusion:
Taller, heavier children mature physically in advance of the shorter,

lighter ones. Those whose physiological age is accelerated complete
the last grade of the elementary school at 12 years, 9 5/6 months of age
with an average of 84.3%, and those below average or of retarded phys-
iological development, complete the elementary school work at 13 years
7 4/13 months of age, with an average of 81.7%. (Bulletin of Bureau of
Education No. 581, 1914. Page 82.)
Correlations Between Early and Later Mental Abilities. The

problem here is, to what extent will a given pupil maintain his
record of excellence, mediocrity, or stupidity all through his educa-
tional career or all through his life? Will the pupil who has high,
medium, or low ability in the elementary school also have high,
medium, or low ability in high school and in college? The first
extensive study in this field was made by W. F. Dearborn ('09) who
traced through the high school and through the university the
scholastic records of various groups of students, varying in size
from 92 to 472, and coming from eight large and four small high
schools in Wisconsin. He divided the pupils into four quartiles
according to their marks in high school, and then ascertained to
what extent the pupils remained in the same quartiles during their
university course. His records showed that the pupils maintained
the same records with remarkable consistency. He states his con-
clusion in the following words:
We may say then, on the basis of the results secured in this group
(472 pupils) which is sufficiently large to be representative, that if a
pupil has stood in the first quarter of a large class through high school,
the chances are four out of five that he will not fall below the first half
of his class in the university. . . . The chances are but about one in
five that the student who has done poorly in high school who has been
in the lowest quarter of his class will rise above the median or average
of the freshman class at the university, and the chances that he will
prove a superior student at the university are very slim indeed. . . . The
Pearson coefficient of correlation of the standings in the high schools and

in thefreshman year, for this group of 472 pupils is .80. A little . . .
over 80% of those who were found in the lowest or the highest quarter of
the group in high school are found in their respective halves of the
group throughout the university. Three-fourths of the students
. . .
who enter the university from these high schools will maintain through-
out the university approximately the same rank which they held in
high school.
F. 0. Smith made a similar study of 120 students entering the
College of Liberal Arts at the University of Iowa. He traced their
records from high school through the entire university course and
found almost the same situation. Expressed in terms of coefficients
of correlations, the results were as follows:
TABLE 10
Correlations. After Smith. ('12.)
H. S. average and Univ. Freshman Average 48

H. S. Average and Univ. Sophomore Average 39
H. Average and Univ. Junior Average
S. 47
H. Average and Univ. Senior Average
S. 28
ist and 2nd Year High School 77
ist and 3rd Year High School 67
ist and 4th Year High School 66
University Freshman and Sophomore 73
University Freshmen and Junior 6r
University Freshmen and Senior 45
T. L. Kelley compared the marks of 59 pupils as they passed

from grade five up into the first year of the high school. The extent
of the agreement of their records in successive years is shown in the
following coefficients of correlation:
TABLE ii. After Kelley. ('14).
Correlation between marks in the grades and marks in the first high school
year.
First Year of High School and 7th Grade 72

First Year of High School and $th Grade 53
He then states:
"The net conclusion which may be drawn from these coefficients oL

correlation that it is possible to estimate a person's general ability in
is
the first year (H. S.) class from the marks he has received in the last four
years of elementary school with accuracy represented by a coefficient of

correlation of .789, and that individual idiosyncrasies may be estimated,
in the case of mathematics and English, with an accuracy represented
by a coefficient of correlation of .515. . . . Indeed, it seems that an

estimate of a pupil's ability to carry high school work when the pupil is
in the fourth grade may be nearly as accurate as a judgment given when
the pupil is in the seventh grade."
A study of the permanency of interests was made by Thorndike

('12)by comparing the relative strength of interests and abilities
during each of three periods of a person's school career, during the
elementary school, high school, and college. These comparisons
were made by asking one hundred individuals to estimate in ret-
rospect, their relative interests and abilities in mathematics,
history, literature, science, music, drawing, and manual work.

Such data are necessarily subject to the errors of memory and
judgment, but they are practically the only results available so far
ate strength of interests is concerned. Thorndike inferred from these
Estimates that early interests are not passing whims, but rather
prophetic, with a fair degree of certainty, of later interests and
abilities. He concludes that "A correlation of .60 or .70 seems to
be approximately the true degree of resemblance between the
relative degree of an interest in a child of from ten to fourteen and
.the same person at twenty-one." "Interests are shown to be not
'only permanent but also symptomatic to a very great extent, of

present and future capacity or ability. Either because one likes
what he can do well, or because one gives zeal and effort to what
he likes, or because interest and ability are both symptoms of
some fundamental feature of the individual's original nature, or
because of the combined action of all three of these factors, interest
and ability are bound very closely together. The bond is so close
that either may be used as a symptom for the other almost as well
as for itself. The importance of these facts for the whole field of
practice with respect to early diagnosis, vocational guidance, the
work of social secretaries, deans, adviser, and others who direct
students' choices of schools, studies and careers is obvious."
The impression gained from all these investigations is that human
nature is not a medley of capricious capacities which vary from
year to year, but rather a fairly consistent combination of abilities
life.
throughout
CHAPTER V
SEX DIFFERENCES
Educational Significance of Sex Differences. If we may judge

fairly at the presenttime concerning the nature and amounts of
differencesbetween the sexes in mental characteristics, it would
seem that the differences are so small in native intellectual abili-
ties that they are almost wholly negligible in the education of boys
and girls. That boys and girls ought to be educated differ-
ently may very probably be desirable, but for reasons other than
differences in ability. The professional, business, and domestic
lifeof men and women makes it necessary to have different train-
ing for boys and girls. But so far as the native abilities involved
in school work are concerned, boys and girls might as well pursue
the same courses from the first day of school to the last.
Popular vs. Scientific View of Sex Differences. Probably more

fallacious psychology of sex has been spread abroad by novelists
and journalists than has been disseminated on any psychological
question of popular interest. Occasional and extreme differences
in individuals of either sex have been seized upon and exaggerated
by descriptive phraseology and represented as though they were
the normal divergences between men and women. Up to less than
two decades ago, there was practically no scientific knowledge of
the nature of sex differences available, and the statements of popular
beliefs about such differences were hardly exaggerated by the sort of
differences implied in the Sanscrit myth of the creation of woman.
"In the beginning, when Twashtrai came to the creation of woman,

he found that he had exhausted his materials in the making of man, and
that no solid elements were left. In this dilemna, after profound medita-
tion, he did as follows: He took the rotundity of the moon, and the
curves of the creepers, and the clinging of tendrils, and the trembling of
grass, and the slenderness of the reed, and the bloom of flowers, and the
lightness of leaves, and the timidity of the hare, and the vanity of the
peacock, and the clustering of rows of bees, and the joyous gaiety of sun-
beams, and the weeping of clouds, and the fickleness of the winds, and
the softness of the parrot's bosom, and the hardness of adamant, and the
sweetness of honey, and the cruelty of the tiger, and the warm glow of
63
fire,and the coldness of snow, and the chattering of jays, and the cooing
of the kokila, and the hypocrisy of the crane, and the fidelity of the
chakrawaka, and then compounding all these together, he made woman
and gave her to man. But after one week, man came to him and said:
Lord, this creature that you have given me makes my life miserable.
She chatters incessantly and teases me beyond endurance, never leaving
me alone; and she requires incessant attention, and takes all my time up,
and cries about nothing, and is always idle; and so I have come to give
her back again, as I cannot live with her. So Twashtrai said: Very well;
and he took her back. Then after another week, man came again to him
and said: Lord, I find that my life is very lonely since I gave you back
that creature. I remember how she used to dance and sing to me, and
look at me out of the corner of her eye, and play with me, and cling to
me; and her laughter was music, and she was beautiful to look at, and
soft to touch; so give her back to me again. So Twashtrai said: Very
well, and gave her back again. Then after only three days, man came
back to him again and said: Lord, I know not how it is; but after all I
have come to the conclusion that she is more of a trouble than a pleasure
to me; so please take her back again. But Twashtrai said: Out with you,
Be off. I will have no more of this. You must manage how you can.
Then man said: But I cannot live with her. And Twashtrai replied:
Neither could you live without her, and he turned his back on man, and
went on with his work. Then man said: What is to be done? For I
cannot live either with or without her. (Thomas, Source Book of Social
Origins, p. 512.)
Such popular beliefs have been in part justified by the probability

that many obvious differences are due to the work, and the result-
ing variation in experience and environment, of women as con-
trasted with those of men. Thus men know more about business,
politics, current events and machines because their occupations
bring them much more in contact with these things; but it does not
follow that women could not, or would not, know as much about
them if their occupations were as much concerned with them.
Women know more about cooking, social events, and household
utensils because their occupations bring them much more in con-
tact with them; but it does not follow that men could not, or would
not, acquire as much knowledge or skill in these directions if their
occupations required it.
The differences between the sexes are probably quantitative

rather than qualitative. Both men and women have the same re-
flexes, instincts, and capacities with the exception of certain as-
pects of the sex instinct. These are probably similar in the main
SEX DIFFERENCES 65
and manner of expression. The differences

differ chiefly in their
due to sex and the rearing of children, with the consequent
life
differences in occupations and experiences, will account for many

of the superficially observable differences between men and women.
What are the differences that have been scientifically measured
and compared? In order to produce a complete picture of mental
differences between men and women it would be necessary to
measure each trait in a very large number of persons and to com-
pare the measurements with regard to both the averages of the
abilities and the manner of the distribution of each ability. This
has been done in part only with a few traits and only upon small
groups of persons.
Differences in Average Amounts of Mental Abilities. There
are two methods by which abilities of two groups may be compared.
Either we may state the actual average or median of each group,
or we may state how many members of one group reach or exceed
the average or median of the other group. The latter method is
preferable in many respects to the former in that it makes possible
a comparison of groups of various sizes and indicates the relative
differences more nearly true to fact. The two methods may be
illustrated in the case of a memory test consisting of the oral pres-
entation of ten words at the rate of one word per second and of
asking the subjects to reooqi immediately the number of words
remembered. He ma> then state that the number of words
remembered on the average by men was 6.9 and by women 7.2.
Or we may state that 43.6% of men reached or exceeded the me-
dian of the women. The latter method of comparison represents
probably more true to life the amount and kind of difference or
similarity that actually exist. The differences, hastily inferred
from a comparison of averages only, would lead to the conclusion
that in regard to memory women are distinctly superior to men.
The implication would be that all women have a memory superior
to that of men, whereas the fact is that the number of women
having a memory superior to that of men is really small and that,
in these few women, memory is better only by a very small shade.
If 43% of men reach or exceed the median of women, it means
that if the 7% of women having a slightly superior memory were
omitted, the remaining 93% of the women would have a memory
ability identical with that of the men. A
difference of in the 7%
distributions between two groups is represented by the curves in
Figure 28. The difference is so small that the groups could hardly
be distinguished.
By method of amounts of overlapping in the distribution
the
of one group over the other, the following results have been ob-
tained from students in the University of Wisconsin in a series of
tests on memory as just stated, on perception consisting in the
cancellation within one minute of as many of a certain geometrical
figure as possible, on motor ability consisting in tapping with a
pencil upon a card as rapidly as possible for thirty seconds, and on
mental addition as described elsewhere. 1
FIG. 28. Distribution curves representing a difference of 7% between the

medians of the two groups.
TABLE 12
Percentage of men reaching or exceeding the median of women.
In the interpretation of these percentages of overlapping it
must be remembered that if 50% of one group reaches or exceeds

the median of the other, it means of course that the two groups
are identical in ability and distribution. If the percentage of men
reaching or exceeding the median of the women is over 50% it
means that the men are superior by the number exceeding 50%.
Helen Thompson Woolley made a series of tests as indicated in
the following table upon twenty-five men and women at the Uni-
versity of Chicago, on the basis of which Thorndike has computed
the following percentages of men reaching or exceeding the median
for women:
1
Experiments in Educational Psychology, revised edition, chapter 16.
SEX DIFFERENCES 67
TABLE 13
Percentages of men
reaching or exceeding the median of the women. After
Woolley as computed by Thorndike ('14, III, p. 178).
Reaction time 68%
Tapping 81%
Sorting cards, speed 14%
Sorting cards, accuracy 44%
Thrusting at target 60%
Drawing lines 72%
Threshold of pain 46%
Threshold of taste 34% (22)
Threshold of smell 43%
Lifting weights 66%
Two-point discrimination 18% (43)
Memory (syllables and learning) 32% (46)
Ingenuity 63%
In a similar comparison made on the basis of 100 boys and 100

girlsfrom results obtained by Gilbert, the percentage of boys
reaching or exceeding the median of girls was as follows:
TABLE 14
Percentages of boys reaching or exceeding the median of the girls. After

Gilbert ('94) as computed by Thorndike ('14, III, p. 182).
to 14 years 15 to 17 years
Discrimination of weights 48% 58%
"colors 39% 58%
Reaction time 57% y6%
Resistance to size-weight illusion 55% 68%
Rate of tapping 64% 73%
Thorndike ('14, III, p. 183) reports measurements in which the

comparison of the percentages of boys reaching or exceeding the
median of girls for persons 8 to 14 years old, were as follows:
TABLE 15
Associative tests, opposites, addition, multiplication, etc 48%

Perception, A-test, etc 33%
Memory of words 40%
The writer has made comparisons in the case of school subjects

on the measured by means of tests and scales.
basis of abilities
Speed of writing was measured in terms of letters written per

minute. Quality was rated by the Thorndike scale. Attainments
in theremaining subjects were measured by the author's tests in

these fields.The following percentages of boys, reaching or ex-
ceeding the median of the girts, were obtained:
TABLE 16
Speed of handwriting, about noo boys and noo girls ................ 47%
Quality of handwriting,
" noo " " noo " ............... 39%
" " " "
Arithmetical reasoning, 1250 1250 ............... 60%
" " " "
History, 429 526 ............... 72%
" " "
Geography, 447 472 ............... 48%
Figures of a similar sort computed by Thorndike ('14, III, p.

183) on the basis of teachers' marks showed the following percent-
ages of boys reaching or exceeding the median of girls:
TABLE 17
High school pupils
English .................................................. 41%
Mathematics ............................................. 57%
Latin .................................................... 57%
History ................................................. 60%
College students
English .................................................. 35%
Mathematics ............................................. 45%
History and economics ..................................... 56%
Natural sciences .......................................... 50%
Modern languages ..................................... 40%
The difficulty with many of the measurements is that they are

based on too small a number of persons. Comparisons based on
twenty-five persons from either sex may be indicative but not
final. Summarizing, we say that women and girls are supe-
may
rior in sensibility, in memory, in most forms of perception, in
ot handwriting, and linguistic ^npnry Jt is
y
To note in .this connection that in the survey of mental-tesj:
1
result^
the women excel in
pend linguistic Thus fV> P fern a) 1*3 ^yrpl j^

chiefly upon fluency.
speedTof reading, both oraljand siTcmy in amount of information
given jri describing an object or in making a report^ in the pnus-
'species Test, in the number of words thought of and written per
^nrniTCgjr
m
the
part-whole test,
in the opposite test, in memory
span for words- in memory Io7logi(al-verbal^terial,injhe WQK&

1
Given in the various chapters of Whippfe's Manual of Mental and Physical Tuts,
SEX DIFFERENCES 69
building test and in the Ebbinghaus completion test; while the

males excel in the rate of association and in the sentence building
test.Apparently the popular belief in the greater linguistic fluency
of women
is not without foundation. Men and boys are superior
in motor capacities, such as tapping, quickness of reaction, in
arithmetical reasoning, and in resistance to suggestions as indi-
cated by the size-weight illusion and the use of suggestive ques-
tions in testimony. The two sexes seem to be approximately
equal in associative processes and in most school subjects. The
amounts of difference, however, are very small. This is particu-
larly true of all the traits that have been measured in a sufficiently
1.20
Girls_
i.oo
Boys
.80
.eo
.40
.20
.00
8 9 10 11 12 18 14
Years
FIG. 20. Comparison of general intelligence of boys and girls as measured
by the Stanford revision of the Binet-Simon tests. After Terman ('16, p. 72).
The numbers along the vertical axis are intelligence quotients as explained in
Chapter VII.
large number of persons to make the comparisons safe. Any

differences lying between 40% and 60% of the number of either
sex reaching or exceeding the median of the other are practically
negligible. If 60% of one sex reach or exceed the median of the
other, itmeans that 10 persons in a hundred of the one sex, are by
a small amount superior to the other. Differences larger than this
have been established with a fair degree of certainty practically
only in the case of one large field of capacities, namely, that of
motor abilities. Differences in nearly ail other respects in which
comparisons have been made on large numbers of persons are
almost entirely within the limits of 40% to 60%. Terman found
inmeasuring the general intelligence of nearly 1,000 boys and girls
by means of his revision of the Binet-Simon tests that for the ages
of five to fourteen girls tend to be very slightly superior to boys
and that after fourteen they are practically equal. His results are
set forth in Figure 28.
It seems a likely interpretation that motor superiority has been
carried over to include intellectual superiority as well. For centu-
ries women have been considered intellectually inferior to men.
They were thought to be incapable of acquiring anything more
than an elementary education. It has been only since the middle
of thei gth century that co-education and women's colleges have
been generally established. Intellectual inferiority has probably
id co m 10 o t- oo o o^je^ju
o i c!t
' ' ' ' ' ' ^ 2 i '
FIG. 30. Range of ability of men and women in color discrimination. After
Henmon (*io).
been inferred chiefly from motor and muscular inferiority and from
the conditions of a narrower environment and dependency due
to the bearing and rearing of children. The inference and belief
of intellectual inferiority is apparently unfounded. This conclu-
sion may be fairly drawn both from the specific psychological
tests that have been cited and also from the recent successes of
women in the acquisition of higher education.
Difference in the Range of Variations in Abilities. Besides
comparing the average amounts of any given ability in the two
sexes, we may compare also the range of abilities from the lowest
to the highest in the two sexes. Such comparisons have been made
in a few traits and the general inference has been that the range of
abilities is wider among men than among women. The distribution
of the abilities in the geometrical perception test made upon 193
SEX DIFFERENCES 71
men and 200 women mentioned in a preceding paragraph, was

as follows:
Scores: 2-3 4-5 6-7 8-9 10-11 H-^S
12-13
193 Men 4.5% 15.4% 33.9% 21.9% 15.4% 56% 2.8%
200 Women 3.2% 20.8% 38.9% 21.6% 10.7% 3.9% 0.8%
Thus in the extremely high ability of canceling 14 to 15 geo-
metrical figures in one minute, there were 2% more men than
women, and in the lowest ability of canceling only two to three
geometrical figures, there were 1.3% more men than women.
Comparisons of can be made safely only on large numbers
this sort
of individuals, and consequently there is as yet little reliable
material available.
The male variability has been computed by
ratio of female to
Thorndike on the basis of tests of memory, re-
('14, III, p. 194)
action-time, discrimination of length, opposites, and cancellation
made by Gilbert ('94) upon 100 boys and 100 girls of each age
from 6 to 17. The average ratio in all tests for the ages of 9 to 12
was found to be .92, for the ages 13 to 14 1.025, and for the age
of 15, .97. Girls were slightly less variable at all ages except 13
and 14. In a test of color discrimination Henmon ( io) also found J
a slightly larger variability among men than among women as

shown in Figure 30.
The author made a comparison of the range of abilities in history
and geography as measured by his tests in these subjects, and
found the following distributions:
History, 8th Grade

Percentages of boys and girls attaining the various scores
Scores: o-io 11-20 21-30 31-40 41-50 51-60 61-70
Boys 4-2% 9-3% 15.3% 170% 13.2% 12.2% 11.5%
Girls 6.2% 22.7% 22.7% 16.4% 12.8% 9.4% 6.0%
Scores (continued): 71-80 81-90 91-100 loo-iro Total
Boys (continued) 9.3% 6.6% 2.1% 04% 288
Girls (continued) 2.3% 2.6% 0.9% 0.0% 352
Geography, 7th Grade

The variability of boys in the case of history is somewhat larger

than that of the girls, whereas in the case of geography it is sub-
stantially the same.
ThornJike ('14, III, p. 195) has given the range of ages of boys
and girls in the third year of high schools in Chicago, Philadelphia,
New York, Detroit, Fall River, Los Angeles, Lowell, and Worcester
as follows:
Age 13 14 15 16 17 18 19 20 and over Total

Boys 7 92 594 1246 1203 572 193 67 3974
Girls 4 73 562 1351 1289 554 120 34 3987
There are about twice as many boys as girls at either 13 or 20

or over.
In support of the general belief that the range of general abilities
iswider in men than in women may also be cited the fact that in
the history of the world most of the great geniuses have been mor> r
and also the statistical fact that male and criminals at the
idiots
other extreme of the distribution curve considerably outnum-
ber the female. The fact that the great geniuses of the world
have been men rather than women would accordingly be explained,
not on the basis of luck of opportunity, but mainly on the basis
of greater exceptional ability. The theory seems plausible but
has been proposed rather in advance of a convincingly wide range
of experimental data. If it is true, it would mean that according
to the perception test the one or two per cent most gifted individu-
als are men and the i or 2% least gifted individuals arc also men,
that of the next 10 or 12% of most gifted individuals approximately
two-thirds would be men and one-third women, and likewise of
the next 10 or 12% least gifted individuals at the other extreme,
about two-thirds would be men and one-third women. For the
remainder of the distribution the number would be practically
identical. The facts should not be interpreted as implying that
men as a rule are superior to women, but would mean simply that
only the one or two exceptional persons in a hundred would be
superior to the most gifted women. The remaining 96 or 98%
would be largely identical.
CHAPTER VI
THE INHERITANCE OF MENTAL TRAITS
Problem.In a certain obvious sense, the entire native equip-

ment any human being is inherited. The various capacities and
of
the relative amounts of them with which a person starts in life
are derived from the cells from which the individual originates.
The differences among these original cells, even when derived
from the same parent, arc assumed to vary with regard to any
potentiality according to the normal distribution curve about
the cejitral tendency of that particular parent. Stalks of corn
grown from seed taken from the same ear will vary considerably
from one another because the seeds themselves, even from the
same ear, are different, but yet they will vary around the general
type of the parent stalk. It is therefore obvious that childrerLflf
will not foe
absolutely alike but that they will
vary ahoiitjlie. cont.ra.1 tandnncy of their ancestors. The specific
problem is not: Are mental traits inherited? but rather: How much
do children of the same parents or ancestors resemble one another
in the amounts of different traits possessed, and in the manner
in which the various traits combine? To what extent are abilities
in school work inherited? To what extent are the wide ranges of
abilities, noticed in Chapter III, due to native equipment or to
opportunity and environment? To what extent does a person

make of himself what he does by virtue of his opportunities or by
virtue of his inherent make-up? What part of the future adult
individual is really determined by the school as an agency of
his environment and what part is beyond the control of the
school?
Methods of Studying Heredity. Any individual is the resultant
;>fthe interplay between his inherited equipment and the stimuli
'rom his environment. Hence, theoretically, there are two general
methods of studying the problem: First, by keeping the environ-
ment constant and varying the ancestry, so to speak; or second,
Dy keeping the ancestry constant and varying the environment.
That is, according to the former plan we would place children of
entirely different ancestry into the same environment from birth
up to a given point in life, and then measure the amount of simi-

larity or difference; or according to the latter plan, we would place
children of the same ancestry into entirely different environments
from birth to a given point in life, and then measure the amount
of similarity or difference. Such ideally scientific conditions are
practically impossible to obtain. The best we can do is to measure
the resemblances or differences of children of the same ancestry
and compare them with the resemblances or differences of children
of different ancestry, both groups living in approximately the'
same environment.
General Views Concerning Mental Heredity. Two extreme
views concerning heredity are possible according to our conception
of the relative roles played by heredity and environment in the
production of adult individuals. We may assume on the one hand

that what a person becomes is absolutely and entirely determined
by heredity, and that environment makes no difference whatever;
or we may assume on the other hand that what a person becomes is
completely and entirely determined by his environment, and that
heredity plays no part. Neither view has been held by any serious
student of heredity in recent times. Views very closely approach-
ing these extremes, have, however, been held by prominent writers
and thinkers in times past; whereas various views between these
extremes are generally being held at the present time, depending
upon the conception as to whether the larger, smaller, or equal
share is contributed by heredity or by environment. The view
held by most scientific students of the problem to-day gives
weight to both elements with perhaps the major emphasis upon
heredity.
The Similarity of Abilities among Related Eminent Persons.
This particular method of attacking the problem was historically
the first means of approaching the study of the inheritance of
mental traits. Two extensive investigations on this aspect of the
subject have been made. The first was carried out by Sir Francis
Galton and published in 1869. Galton made a study of 977 emi-
nent men, each of whom was the most eminent among 4,000 persons.
He proceeded to determine how many relatives of equal eminence
and of varying degrees of relationship each person possessed.
In this manner he found that these 977 men had the following
relatives of a like degree of eminence: 89 fathers, 114 brothers, 129
sons, 52 grandfathers, 37 grandsons, 53 uncles, and 61 nephews, or
a total of 535. Galton further pointed out that 977 ordinary men
THE INHERITANCE OF MENTAL TRAITS 75
selected by chance from the population at large would have only

four such eminent relatives. He concluded as follows:
"i. That men who are gifted with high abilities even men of class
E easily rise through all the obstacles caused by inferiority of social
rank.
"2. Countries where there are fewer hindrances than in England, to a
poor man rising in life, produce a much larger proportion of persons of
culture, but not of what I call eminent men. (England and America are
taken as illustration.)
"3. Men who are largely aided by social advantages are unable to
achieve eminence, unless they are endowed with high natural gifts.'*
More recently an extensive study was made by Woods ('06)

on mental and moral heredity in royalty. Woods made a com-
parison of 671 members of royal families in Europe by giving each
person a rating on a scale of i to 10 in which 10 signified excep-
tionally high ability or genius, and i represented exceedingly low
ability or imbecility. These ratings were made by the judgment
of Woods himself according to the reports of these persons in
histories and biographies. On the basis of these estimates, a
tabulation was then made of the relationship of persons of various
degrees of ability. He
found that most of the eminent persons
were grouped about four stocks or families out of fifteen, namely,
the families of Frederick the Great, Queen Isabella of Spain,
William the and Gustavus Adolphus. Likewise, he found
Silent,
that most of the persons of lowest ability were grouped around
certain families in Spain and Russia, and the persons of mediocre
ratings, four to seven, centered about some half dozen royal families
including the houses of Hanover, Saxe-Coburg-Gotha, Reuss,
Mecklenburg, Hapsburg in Austria, Holstein, Denmark, Saxony,
Savoy, Orleans and modern Portugal. The ratings, of course,
were not absolutely correct measurements of their abilities, but
they, no doubt, represented greater validity than general impres-
sions would. He further computed coefficients of resemblance in
intellect and morals as follows:
I. In intellect:
Offspring and father 30

" 16
grandfather
" " *
great-grandfather 15
1J. In morals:
Offspring and father 30
" "
grandfather 175
Dr. Woods then attempted to determine whether or not thefact of

accession to the throne by virtue ofbirth gave an individual greater
opportunity for eminence. This he states in the following manner:
"There is one peculiar way in which a little more than half of all
males have had a considerable advantage over the others in gaining
distinction as important historical characters. The eldest sons, or if not
the eldest, those sons to whom the succession has devolved, have un-
doubtedly had greater opportunities to become illustrious than those to
whom the succession did not fall by right to primogeniture. I think
every one must feel that perhaps much of the greatness of Frederick II
of Prussia, Gustavus Adolphus, and William the Silent, was due to their
official positions; but an actual mathematical count is entirely opposed to
this view. The inheritors of the succession arc no more plentiful in the
higher grades than in the lower. The figures show the number in each
grade who came into power by inheriting the throne."
Grades
Total No. in each grade
i
7
23456789
21 41 49 71 70 68 43 18
10
7
Succession inheritors 5 14 26 31 49 38 45 23 12 4
Per cent 71 67 63 64 69 54 67 54 67 57
"It isthus seen that from 54 to 71% inherited the succession in the
different grades. The upper grades are in no way composed of men
whose opportunities were enhanced by virtue of this high position. Thus
we see that a certain very decided difference in outward circumstances
namely, the right of succession can be proved to have no effect on
intellectual distinction, or at least so small as to be unmeasurable without
/much greater data. The younger sons have made neither a poorer nor a
better showing. ('06, pp. 285-286.)"
"The upshot of it all is, that as regards intellectual life, environment is
a totally inadequate explanation. If it explains certain characters in
certain instances, it always fails to explain as many more; while heredity
not only explains all (or at least 90%) of the intellectual side of character
in practically every instance, but does so best when questions of en-
vironment are left out of the discussion. Therefore, it would seem that
we are forced to the conclusion that all these rough differences in in-
which are susceptible of grading on a scale of ten are
tellectual activity
due to predetermined differences in the primary germ-cells." ('06, p. 286.)
While heredity no doubt plays an important part in the produc-
tion of intellect and character the part attributed to it by Woods
that explains "at least go% of the intellectual side of character
it
in every case" is hardly warranted either by the findings of other

investigators or by the results of Woods himself. His correlation
between father and offspring is only .30.
Similarities of Abilities Among Related Defective and Low

Grade Persons. Quite a number of studies have been made in
recent years concerning the frequency with which defective persons
arc either distantly or closely related. One of the first studies
was that of the Jukes reported by R. G. Dugdale in 1877. Max
Juke, born in 1720, was a shiftless truant, who married an equally
worthless woman. Up to 1877 there had been five generations with
approximately 1,200 descendants among whom have been traced
the following types of persons: 310 paupers, 7 murderers, 60 habit-
ual thieves, 50 prostitutes, 130 convicted of crime, 300 died in
infancy, 440 physical wrecks from debauchery, only 20 learned a
trade, and 10 of these learned it in prison. The estimated cost to
the State of New York has been put at approximately $1,000 a
person. In contrast with this lineage, a comparison has been
suggested with the Jonathan Edwards family, which had approxi-
mately 1,400 descendants in the same period of time. Among
them there have been 120 graduates of Yale alone, 14 college
presidents, over 100 professors, 135 books of merit have been
written by various members of the family, and 118 journals have
been edited by them. Aaron Burr was the only black sheep among
them and he can certainly not be classed as an intellectually de-
fective person. (Winship 'oo).
Poellman of Bonn (Guyer
'16, p. 271) made a study of a family
called the Zeros inwhich 800 descendants were traced through six
generations back to a female drunkard. Among them were found
102 professional beggars, 107 illegitimate offspring, 181 prostitutes,
54 inmates of almshouses, 76 convicted of crime, and 7 murderers.
The cost to the state was placed at $1,206,000.
More recently a very interesting study was conducted by Dr.
Goddard of the Training School at Vineland, New Jersey. Dr.
Goddard ('12) traced the ancestry of a young girl who had been
brought to his institution. It was found that the lineage went
back to a man, Martin Kallikak, a soldier in the Revolutionary
War, who was the progenitor of two lines of descendants. (See
Figure 31.) He had an illegitimate son whose mother was feeble-
minded. This was the establishment of line A which had, down
to the time of the study, 480 direct descendants among whom
were found the following: 143 feeble-minded, 292 unknown, 36
illegitimates, 33 prostitutes, 24 alcoholics, 3 epileptics, 82 died in
infancy, 3 criminals, 8 keepers of disreputable houses, and only 46
normal individuals. Apparently human nature does not gather
grapes of thorns or After his return from the war,

figs of thistles.
Martin married a woman normal intelligence and from this
of
lineage B there had come during the same period of time, 406
direct descendants, of whom all were normal individuals with the
THE
LAWFUL WIFE FEEBLE-MINDED GIRL
L
MMRTIN JR. RHOOA ZABETH
TAKE
GOOD 9 YR8. 4 VI
HOME
D. 86 YR8.
AUUHOUSE
FIG. 31. Descendants of the Kallikak Family. Squares males, circles
females, black squares or circles = feebleminded, open squares or circles =s
normal persons. The lineage was traced back from Deborah. After Goddard.
exception of five, one of whom was reported as mentally defective,

two as alcoholics, one as sexually immoral, and one as a case of
religious mania. There were no epileptics or criminals, and only
15 died in infancy. The remainder were good citizens, including
doctors, lawyers, educators, judges, and business men.
One thing seems to stand out very conspicuously from the
numerous facts of family histories that have been unravelled in

recent years, namely, that much defective mentality, degeneracy,
and crime is a matter of ancestry. General opinion among persons
in charge of institutions for defectives is that two-thirds of all
cases are due to heredity and one-third to environmental or un-
known causes. Thus Dr. Alfred Wilmarth, Superintendent of the
Wisconsin Home for Feeble-minded, says:
"
My own observations, and those of others in this country and Europe,
would indicate that at least two-thirds of the feeble-minded have defec-
tive relatives. This is significant. Mental accident may occur in any
family, but it is rarely a second case occurs unless there is a tendency to
nerve degeneracy. (Quoted by Guyer,'r6, p. 245.)
"I present to you the results of compiling the histories on 1,000 appli-
cations, where our information is most thorough; but I am confident
that these do not tell the whole story. In 311 of these any neurotic taint
in the family history is absolutely denied. In 365 cases at least one near
relative suffers from one of the graver forms of nervous or mental trouble;
in 170, two relatives were found; in 73 cases, three relatives, and in 8r
cases four or more. These figures agree very accurately with the results
of other observers in this country and abroad. It is safe to say that less
than one-third of the defective classes are the results of disease or trau-
matism in families capable of transmitting a healthy, well developed
nervous system."
Dr. Goddard of the Training School, Vineland, New Jersey,

states in connection with his tests of 2,000 children:
"But we now know that 65% of these children have inherited the
condition, and that if they grow up and marry they will transmit the
same condition to their offspring. Indeed, we know that this class of
peopleis increasing at an enormous rate in every community and unless
we do something to stop this great stream of bad protoplasm we shall

"
some day be swamped in a sea of degeneracy.
Likewise Dr. A. C. Rogers of the Minnesota School for Feeble-
minded, at Faribault, says:
"We have no survey of mentality in this country except in very small
areas, but probably about of the feeble-minded children that we
65%
know of are feeble-minded from heredity; that is, they come from families
in which there is much feeble-mindedness, usually associated with various
neuroses or psychoses. There are about 35% approximately that are
acquired cases. These cases develop from various things. Full develop-
ment may be prevented during gestation, or early childhood, or early
adolescence, but these acquired cases are entirely distinct from the
hereditary ones." (Guyer '16, p. 246.)
8o EDUCATIONAL PSYCHOLOGY
Likewise, Dr. Martin W. Barr of the Pennsylvania Training

School for Feeble-minded Children states:
"In my individual study of 4,050 cases of imbecility, I find 2,631 or

65.34%, caused by malign heredities; and of these 1,030, or 25.43%, are
due to direct inheritance of idiocy; and 280, or 6.91%, to insanity."
To one who wishes to argue in favor of environment as the chief

determining element in ability and character, such data as have
been presented from family histories and relationships are not en-
tirely convincing. It might be argued that a given family has
so many individuals of high or low intelligence and achievement
because its members were born in circumstances which did or did
not afford opportunities for development and training and for
achieving higher success. It might be said that the descendants of
the Edwards family were born and reared among favorable cir-
cumstances of educational and financial advantages and conse-
quently were fitted for greater tasks and lived in an environment
in which larger opportunities offered themselves, whereas the
members of such a lineage as the Jukes family would have just the
opposite environment of birth, education, and opportunity in life.
In answer to all this, we must remember, however, that ability
very largely determines the sort of environment in which a person
is satisfied to live, that a really capable person is quite likely to
push forward and to find a way out of the environment in which

he may happen to have been born, or to improve it if he cannot
leave it, and finally, we must remember that the persons of low
ability were born in circumstances of a correspondingly low nature
because of the hereditary stock of the families from which they
came. Their parents were content to live under the circumstances
under which they did live because their abilities and desires sought
for nothing better.
Similarities between Brothers and Sisters in Special Mental
Traits. Pearson has shown that the resemblance in physical
characteristics among brothers and sisters is approximately .50.
He gives the following coefficients of correlation for various
physical traits.
Brother and sister

Hair color 55
Cephalic index 49
Height 50
Eye color. 52
TI1K INHERITANCE OF MENTAL TRAITS 8l
What, however, is the degree of resemblance in mental traits?

The general argument is that mental traits are dependent upon
anatomical and neurological structures, and hence, if these are
inherited, mental traits must also be inherited. What is the evi-
dence from experimental and statistical facts?
In a study made by the writer ('17) a series of tests of capacities
directly affected by school work and another series of tests of capac-

ities not directly affected by school work were applied to 18 pairs
of brothers and sisters in the University of Wisconsin. Each test
was given twice on two different occasions in order to obtain a
fairly accurate measurement of the capacities concerned. The pur-
pose of giving the two types of tests was to ascertain whether
brothers and sisters were more alike in the traits affected by train-
ing than in the traits not directly affected by school training. The
following were the tests and the correlations obtained between
pairs of children of the same family:
TABLE 18
Correlations between abilities of brothers and sisters. ACtcr Starch ('17)
Reading speed 51
Reading comprehension 64
Writing speed 72
Writing quality 46
Size of reading vocabulary 07
Spelling 05
Arithmetical reasoning 38
Addition attempts 71
Addition rights 44
Subtraction attempts 43
Subtraction rights 29
Multiplication attempts 37
Multiplication rights 2$
Division attempts 46
Division rights 56
Average 42
Memory 31
A-test 50
Geometrical form test 07
Tapping 6$
Average 3$
Coefficients based on ranks in all tests combined , .74

In order to grasp the full import of these figures, it is necessary
to remember that the coefficient of correlation between mental

abilities of pairs of unrelated individuals selected by chance would
be zero, and that any coefficient above zero between pairs of
brothers and sisters means a corresponding amount of resemblance.
"Fromthe above table several interesting results appear, (i) The

resemblance of siblings is apparently no greater in those mental traits
which arc directly affected by school work than in those which are not so
affected. The average correlation in the former group of tests is .42 and
in the latter .38. This seems to indicate that the mental similarities of
children of thesame parents are due primarily to heredity rather than to
similarity of environment since the resemblance is no greater in those
traits which arc more directly affected by environment.
"
(2) The resemblance of siblings is approximately as great in mental
traits as in physical traits. Pearson found the correlation between
brother and brother in height to be .50 and in cephalic index (ratio of
length to width of head) .49. These correlations for physical traits are a
little larger than the ones found here for mental traits taken separately.
The correlation, however, calculated on the basis of a combined rank for

each person in all mental tests together was found to be .73. This
greater correlation for all tests combined as compared with the correlation
for single traits is due partly to the variation of the correlations among
the single traits and partly due to the imperfections in the separate tests,
which are counterbalanced to some extent in a combined ranking."
(Starch '17, p. 237.)
Pearson ('04) made a comparison of 2,000 brothers and sisters

who were rated by their teachers in such traits as vivacity, self-
assertion, introspection,popularity, conscientiousness, temper,
ability, and handwriting. On the basis of these ratings he found
coefficients of correlation ranging from .43 to .64 with an average
of .52. These results are interesting and indicative of the resem-
blance of more general traits of character, but they are probably
rendered more or less uncertain by the unreliability of one person's
ratings of such elements of character. The likelihood is that the
teachers would be more inclined to estimate alike the children from
the same families, rather than to estimate them more different
than they really were.
Similarities of Brothers and Sisters in Abilities in School Sub-
jects. In a study made several years ago, Earle ('03) found a
correlation of .50 between the spelling abilities of 180 pairs of
brothers and sisters. The writer ('15) made a study of the scho-
lastic records of children from 63 families. The average grade in all

school subjects was obtained for each pupil and used as the measure
of his academic ability. The correlations based upon these averages
were as follows:
First and second child in a family, 63 pairs 58

Second and third child in a family, 24 pairs 64
First and third child in a family, 24 pairs 34
Average 52
Further comparisons were made for abilities in specific school

subjects which yielded the following correlations:
Spelling, 57 pairs of children from the same parents 21

Reading, 57 pairs of children from the same parents 49
Writing (speed) 24 pairs of children from the same parents 18
Writing (quality) 24 pairs of children from the same parents 06
Another study was made of 38 children from n families. All the

marks that each pupil had received in each study in grades three to
eight were averaged. From these averages the following coefficients
of correlation were obtained.
Arithmetic, 54 pairs 32
Spelling, 54 pairs 21
Reading, 54 pairs 31
Language, 54 pairs 24
"No importance, I believe, can be attached to the differences in cor-

relation between the various studies. The correlation for individual
studies is lower than that for scholarship in general based on the average
performance in all studies combined. This is probably due chiefly to the
fact that the inaccuracies of teachers* marks in individual subjects are
partly eliminated in the averages derived from all studies.
"Abilities in special subjects are inherited, apparently, to no greater
extent in one subject than in another. What is probably inherited is
either general scholarship or else more specialized traits than ability in
arithmetic, or ability in language. Each study involves many mental
faculties and nearly all studies involve the same faculties with varying
emphasis.
"In corroboration of this point we may notice the following table of
average marks for each of nine families in each study.
"In this table, we must examine the ranks, rather than the marks, of
the different families in each subject, so as to eliminate the variation in
standards of marking. These families rank very nearly the same in the
various studies. For example, family C is first in every study except

arithmetic and there it is third. Family G is second in every subject
except arithmetic and there it is fourth. Family I is cither third or fourth
in every subject but one, and family B is last in every subject except one.
TABLE 19
Averages for each family in each subject.
"There is no evidence, at least from these figures, for the notion that
special abilities in certain studies run in families. Mental traits running
in families are very likely more specialized than abilities in school studies
which involve large groups of mental functions. The children of any
given family are on the average equally good or equally poor in all studies.
Ability in school work is apparently inherited to the same extent as
physical features since the coefficients of correlation for children of the
same parents are approximately the same for both physical and mental
traits." (Starch, '15, pp. 609-610.)
Schuster and Elderton ('07) calculated the resemblance in

scholarship between brother and brother and between father and
son among the Oxford honor men and found a coefficient of correla-
tion of .40 for the former and .31 for the latter. Miss Elderton
further determined the correlations between cousins from records of
about 300 families and found a coefficient of .27.
Miss Emily S. Dexter 1 made a study of the scholarship records of
185 pairs of brothers, sisters, brothers and sisters, graduates of the
University of Wisconsin, and of 69 similar pairs who were gradu-
ates of the high school at Ashland, Wisconsin. She reports the
following coefficients:
1
The study was carried out under the direction of Professor ITenmon and reported
iu a thesis in the library of the University of Wisconsin, 1915.
THK INHERITANCE OF MENTAL TRAITS
TABLE 20
Miss Dexter concludes "that inheritance, to a much greater extent

than training is responsible for the degree of resemblance found."
"If it were largely training, we would expect to find the resemblance
greater between brother and brother, and sister and sister, than between
brother and sister, but such is not the case. In the high school the cor-
relations for the three groups are much the same, but, as has been pointed
out, that may be due to a great extent to chance, for the groups are small.
However, in the case of the university, where the groups average nearly
three times as large as in the other school, we find the resemblance be-
tween brother and sister to be greater than between brother and brother,
or sister and sister.
"Again, there is the question as to specialized abilities, and also that of
general mental ability rather than specialized abilities as a basis of ex-
planation for the close resemblance found. Thorndike, as has been said,
finds that heredity is highly specialized. This study, however, seems to
show a stronger tendency toward general mental ability, if by that we
mean approximately equal ability in all subjects. It seems, also, to give
almost no evidence of alternate inheritance; that is, of one individual's
inheriting ability in one line while his brother inherits ability in another.
In other words, a student who is above the average, either of his family
or of the school, in one subject, is usually also above in most, and in many
cases all, other subjects."
Similarities of Twins in Special Mental Traits. The two prin-

cipal investigationson this phase of mental heredity were made by
Galton and Thorndike. Galton made a general comparison of two
groups of twins, one group of 35 pairs, which were reported as being
very similar, in fact so similar that they were frequently reported
as indistinguishable, and another group of twenty pairs of dis-

tinctly dissimilar twins. The conclusion formulated was to the
effect that the former twins remained very similar all through
in spite of different environments, while the latter twins re-
life
mained different all through life in spite of similar environments.

Concerning certain of the twins Gal ton reports:
"i. One parent says: 'They have had exactly the same nurture from
their birth up to the present time; they arc both perfectly healthy and
strong, yet they arc otherwise as dissimilar as two boys could be, phys-
ically,mentally, and in their emotional nature.'
"2. 'I can answer most decidedly that the twins have been perfectly
dissimilar in character, habits, and likeness from the moment of their
birth to the present time, though they were nursed by the same woman,
went to school together, and were never separated till the age of fifteen.'
"3. 'They have never been separated, never the least differently
treated in food, clothing, or education; both teethed at the same time,
both had measles, whooping-cough, and scarlatina at the same time, and
neither had any other serious illness. Both are and have been exceed-
ingly healthy and have good abilities, yet they differ as much from each
other in mental cast as any of my family differ from another.'
"5. 'They were never alike cither in body or mind and their dissim-
ilarity increases daily. The external influences have been identical; they
have never been separated.'
"9. 'The home-training and influence were precisely the same, and
therefore I consider the dissimilarity to be accounted for almost entirely
by innate disposition and by causes over which we have no control.'"
('83, p. 170, Everyman's Library Edition.)
Gal ton's general impression of his results is as follows:
"We may, therefore, broadly conclude that the only circumstance,

within the range of those by which persons of similar conditions of life
are affected, that is capable of producing a marked effect on the char-
acter of adults, is illness or some accident that causes physical infirm-
ity. . The impression that all this leaves on the mind is one of some
. .
wonder whether nurture can do anything at all, beyond giving instruc-

tion and professional training. There is no escape from the conclusion
that nature prevails enormously over nurture when the differences of
nurture do not exceed what is commonly to be found among persons of
the same rank of society and in the same country." ('83, pp. 168 and 172.)
Thorndike's investigation ('05) was made by more accurate

methods. He applied the tests, mentioned in the following table,
to 50 pairs of twins and found the following correlations:
THE INHERITANCE OK MENTAL TRAITS
In the A-tcbt R .
69
In the a-t and r-e test R .
71
In the misspelled word test R 80
In addition R .
75
In multiplication R 84
In the opposites test R 90
now these resemblances are due to the fact that the two members
"If
of any twin pair are treated alike at home, have the same parental
models, attend the same school and are subject in general to closely
similar environments, then (i) twins should, to the age of leaving home,
grow more and more alike, and in our measurements the twins 13 and
14 years old should be much more alike than those 9 and 10 years old.
Again (2), if similarity in training is the cause of similarity in mental
traits, ordinary fraternal pairs not over four or five years in age should
show a resemblance somewhat nearly as great as twin pairs, for the
home and school conditions of the former will not be much less similar
than those of a pair of the latter. Again (3) if training is the cause,
twins should show a greater resemblance in the case of traits much sub-
ject to training, such as ability in addition or in multiplication, than in
traits less subject to training, such as quickness in marking off the A's on a
sheet of printed capitals, or in writing the opposites of words.
"On the other hand (i) the nearer the resemblance of young twins
comes to equalling that of old, (2) the greater the superiority of twin
resemblance to ordinary fraternal resemblance is, and (3) the nearer twin
resemblance in relatively untrained capacities comes to equalling that in
capacities at which the home and school direct their attention, the more
must the resemblances found be attributed to inborn traits.
"The older twins show no closer resemblance than the younger twins,
and the chances are surely four to one that with an infinite number of
twins tested, the 12-14 year-olds would not show a resemblance .15
greater than the 9-11 year-olds. The facts are: (Thorndike '14, III,
pp. 248-249).
TABLE 21
The resemblances of young and old twins compared.

The Influence
of Uniform Environment Upon Different Original
Abilities.All studies cited thus far have attempted to measure
the amount of similarity in related persons as compared with un-
related individuals on the assumption that the environment was
roughly constant for all, that whatever resemblances existed be-
tween pairs of brothers and sisters or between other types of rel-

atives greater than that between any pairs of persons selected by
chance from the population at large, is considered to represent
the actual amount of similarity in the original inherited natures of
the individuals springing from the same ancestry. The problem
may, however, be pursued further from a different angle, namely, by
specific control of the environment or some portion of it. Thus it is
possible to keep some particular part of the environment uniform
for a group of persons of widely different abilities and to measure
to what extent the original differences remain constant, increase,
or decrease. If the differences remain constant, or increase, the in-
ference would be that the ultimate differences of achievement would
be primarily due to the inherited differences of capacities. If the
differences decrease materially and finally disappear, the original
differences would be mainly due to the effect of environment and
opportunity.
A number of such experiments have been carried out. An in-
vestigation made by the writer (*n) in which 8 persons multiplied

mentally 50 3-place numbers by a i-place number each day for
14 successive days showed the following amounts of improvement:
TABLE 22
Hence, both the greatest absolute and the greatest relative gain
was made by the group with the highest initial records.
Similar results have been found in the practice experiments of
substituting numbers for letters as described in the author's Ex-
periments in Educational Psychology, Chapter X. The follow-
ing table gives the highest five and the lowest five records from
among twenty persons. Each person practiced 120 minutes.
TABLE 23
Average number of letters transcribed.
Again the largest gain was made by the group having the greatest
initial ability.
Results pointing in the same direction have been obtained by
Thorndike, Whitley, and others. For example, Thorndike ('10)
found in the case of practice of nineteen persons in adding, the fol-
lowing results:
TABLE 24
The effect of equal amounts of practice upon individual differences in column
addition of one-place numbers. After Thorndike ('10).
The statistical studies of scholastic histories of pupils through

various periods of school life, which were discussed in a preceding
chapter under the heading of correlations of abilities at different
times of in the same individual, all tend to corroborate the ex-
life
perimental facts here presented. The scholastic records show to a

remarkable extent the uniformity with which each individual main-
tains his position throughout his educational career. A very in-
teresting tabulation was made by L. J. Coubal and E. VanLande-
gend to show the progress made by pupils in grades 4, 5, and 6
1
in one school in the four fundamental operations in arithmetic.
Progress was measured by the Courtis tests, Series B, month by

1
Under the direction of Professor Henmon, and reported in two theses in the library
of the University of Wisconsin, 1917.
month through an entire year. The records in the four opera-

tions for each pupil were combined into a single score. Figures 32,
60 1
40
80
20
10
Sept.
N
Oct. Nov. Dec. Jan. Feb. March April May]
FIG. 32. Progress in the four fundamental operations in arithmetic as
measured by the Courtis tests, Series B, given at monthly intervals. The
heavy continuous line represents all the pupils of the 4th grade. The four broken
lines represent these pupils divided into quartiles.
Sept. Oct. Nov. Dec. Jan. Feb. March April May

FIG. 33. Same as Fig. 32, for the 5th grade. Pupils divided into three groups
instead of four.
33, and 34 give the curves

of progress for the respective grades.
The pupils in each grade were divided into groups according to
their final performance. Thus the pupils in grade 4 were divided
THE INHERITANCE OF MENTAL TRAITS
into four groups while those in the other grades were divided into
three groups. The results reveal the significant fact that the best
groups in each grade made the greatest progress, the poorest groups
made the least progress and the intermediate groups made average
progress. The graphs for the various groups in any grade gradu-
ally spread apart during the course of the year, indicating that the
differences increase rather than decrease or remain constant. The
more gifted pupils profit more by their school work than the less
gifted.
All experimental results point in the direction that practice does
not equalize abilities; in fact, equal practice tends to increase differ-
To i

FIG. 34. Same as Figs. 32 and 33, for 6th grade.
ences in achievement and skill rather than to decrease them. The

more gifted individuals profit more, both relatively and absolutely,
than the less gifted. This experimental fact is one of the most pro-
found whole problem of heredity and
bits of evidence regarding the
environment. The men
not only start with greater initial
talented
capacities but seem also to be capable of more intense application
and more zealous desire to improve. "To him that hath shall be
given" is psychologically true in the sphere of intellectual training
as well as in the sphere of morality and religion. The man with

ten intellectual talents will acquire far more than the man with one
talent. If we may generalize for life as a whole, equal opportunities
for all abilities in all. Men may be born free

do not produce equal
politically,but they are not born equal mentally; they may be
born equal in opportunities in a democratic society, but they cer-
tainly are not equal in their ultimate achievements in life.
Influence of Different Environments upon Various Original
Abilities. Extensive inquiries into the effects of various environ-
mental conditions upon the native ability of human beings have
been made in other fields besides the expeiimental one which has
been surveyed. Such investigation as the study made by Dr.
Rice ('97 and '02) upon the effects of various factors in school life
upon the attainments in spelling and in arithmetic, the studies of
places of birth of American men of science made by Cattell ('06),
or the study of the places of birth of eminent men of letters made
by Odin, and similar investigations by De Candolle ('73), Jacoby
('81), and Ellis, have been extensively referred to as bearing upon
the problem of environmental forces in their interplay with heredi-
tary capacities.The real significance and argumentative weight of
such data seem to the writer to be uncertain and duplex in their
meaning. Cattell, for example, has pointed out that the number of
eminent scientific men born about 1860 in Massachusetts per one
million population was very much greater than the number of
eminent scientific men born in proportion to population in other
states. To cite a few instances, he has computed that per one
million population there were born eminent scientists as follows in
various states:
Mass 108 8
Conn 86 9
.
R.I 25.6
N.Y 470
Wis 45-0
111 24.0
Ala 2.0
Miss i.o
Similar figures are given for other states, and the inference made
by Catteli is that the environment of Massachusetts and similar
states has been much more conducive to the development of scien-
tific men and that the number of such men could be determined
practically by the control of the proper educational stimuli.
Odin, in his study of 5,233 noted French men of letters living
during the period 1400 to 1830, found the following distribution
according to places of birth:
1,229 born in Paris

" "
2.264 other large cities
" "
1.265 small cities
" "
93 country districts
From this it has been inferred that if France as a whole had

been as fertile as Paris' in the production of eminent men of letters
there would have been approximately 54,000 great men of letters
instead of less than 6,000. The difficulty, however, with both
studies is that Paris and Massachusetts have been more productive
of eminent men not necessarily on account of better educational
and environment, but possibly also because of the fact that
social
eminent men of letters and science have by virtue of the location
in them of educational institutions, scientific and other intellectual
centers, necessarily been attracted to these places, and conse-
quently their children were born in these localities. The facts as
such may actually be used in the support of heredity as against
environment as much as they have been used in support of en-
vironment as against heredity.
Likewise, the study of Rice with regard to the factors affecting
efficiency in school subjects is uncertain. Rice, on the basis of
extensive tests in spelling and arithmetic in various parts of the
country, arrived at the general opinion that practically all external
conditions of home and school such as foreign or American parent-
age, home of teaching, size of class, and time
study, methods
devoted to study, made practically no difference whatever in the
ultimate achievement of the pupils, and the implication is made
that the final efficiency depends primarily upon heredity. The
obvious uncertainty of such data as these is that while the facts
in toto may imply such a situation, it is also quite certain that such
a massing of data in this manner obliterates the effect of individual
factors. Favorable conditions may be offset by unfavorable condi-
tions and thus obscure the entire situation. To infer that good
teaching and poor teaching make no difference in the ultimate
results obtained, or that the amount of time given to study makes
no difference in results, are conclusions that are quite likely to be
unsound. The reason for the inference drawn by Rice is probably
the fact that good teaching in some schools may be accompanied
by other factors which tend to counteract its effect, whereas poor
methods of teaching in other schools may be accompanied by
favorable or unfavorable circumstances in other respects. The
massing together of returns from many schools is bound to ob-
literate the effects of the individual conditions. The only certain

way to ascertain the effectiveness of one factor or another would
be to control all conditions, or to be able to allow for them definitely,
with the exception of the one factor whose efficacy is to be deter-
mined. Thus in order to determine whether or not the different
methods of teaching a given subject make a difference, it would be
necessary to take a class of pupils in a given school and divide it up
into two or more groups on the basis of equal initial capacities and
to have each class taught, preferably by the same teacher and un-
der the same general circumstances, by a different method. Then at
stated intervals the two groups should be compared by reliable
measures. In this manner definite results could be obtained as
to the effectiveness of the various methods, conditions, or amounts
of time devoted to a subject. Consequently, Rice's figures as they
stand are of little worth so far as proving the forcefulness of dif-
ferent environmental factors in the production of results is con-
cerned.
A similar criticism applies to such studies as that made by Spill-
man with regard to the place of birth of various men prominent
in public and business life, such as presidents of the United States,
governors, and railroad presidents. Spillman ('09) has pointed
out, for example, that 23 of the first 25 presidents of the United
States were born in the country, that 41 out of 45 governors, and
47 out of 62 cabinet officers were born in the country. It is unsafe
to argue that because a large percentage of the presidents of the
United States or other prominent persons were born and reared
in rural districts, that country life is more productive of ability
and ambition. To argue with any certainty on the basis of such

facts, it is necessary to know the ancestral antecedents of the
persons springing from various localities and sections of the popu-
lation.
General Interpretation. The general impression from all ex-
perimental, statistical, and historical material thus far accumulated
on the problems of mental heredity would seem to be somewhat
as follows: Barring paupers, invalids, and those suffering from
want of food and shelter due to conditions beyond their personal
control, and referring to all others living in the same community
at the same time, the ultimate achievement of any given individual
is due to his original ability, probably to the extent of 60 to 90%,
and to actual differences in opportunity or external circumstances

only to the extent of 10 to 40%.
The facts of heredity bear down so heavily that the impression

gained of the large part played by it leads one almost to a fatalistic
philosophy. One is almost inclined to believe that persons become
what they do largely on account of their hereditary capacities, and
that they are not in the least responsible for their own outcome;
that if a person is born with great capacities he will achieve high
distinction, and if he is born with mediocre or slender capacities,
he will not achieve anything beyond his limits no matter what he
may do. While it is certainly true that no one may achieve a
position higher than his original capacities will permit, it does not
follow that a mechanical, fatalistic view needs to be taken. Nature
predominates enormously over nurture only in the relative and
not in the absolute sense. This distinction must always be borne
in mind in studies of heredity. In fact, in the absolute sense,
nurture predominates enormously over nature. A Newton born
among Australian bushmen would no doubt have become a re-
markable bushman, but never a world-renowned scientist. The
necessary stimuli of environment must be at hand to train and
develop original capacities. The difference between relative and
absolute achievement may be illustrated in any of the experimental
results concerning the effects of equal practice cited in a preceding
section. The fact that all individuals improve by practice shows
absolute gain in performance or skill. The fact that the gifted ones
maintain their lead or even gain in their lead is relative achieve-
ment. Before practice, no child can write after practice, ail normal
;
children can write with more or less excellence. This is absolute

gain. Before practice, some children have greater original capaci-
ties for learning to write; after practice, these same children main-
tain the same superiority. This is relative gain. A Newton and
an ordinary bushman born and reared among bushmen would
probably be superior and ordinary bushman respectively. A
Newton and an ordinary bushman born and reared in .New York
City at the beginning of the 2oth century would probably become,
respectively, the one a great scientific, professional or business
man, and the other an ordinary person, able to get on, earn a
living, and enjoy life within the ordinary limits. The original
abilities of ancient civilized peoples were probably very little
different from the original abilities of modern civilized peoples.
The differences are probably due to the transformation of the
environment which is constantly being brought about through
the efforts of man. A Newton born in a modern civilized com-
munity would have greater and different stimuli than one born
in an ancient or uncivilized community. His ultimate eminence
would be determined by his environment.
The pessimistic air may further be dispelled by noting the fact
that hardly one person in a thousand makes allthe absolute gain
possible forhim even in a single capacity. It has been proved over
and over again in numerous abilities which have been used daily
in one's occupation that by a little special practice each day their
efficiency may be enormously improved. Consequently, while the
possibilities of each individual are limited by his original inherited
equipment, each one may develop his capacities far beyond the
usual degree of attainment. While experimental evidence indicates
emphatically that under equal opportunities the more gifted surpass
the less gifted, yet rarely does anyone do his best or attain his limit
even in a single capacity. Life is a matter of competition; let every-
one compete to the fullest extent of his inherited ability.
CHAPTER VH
THE MEASUREMENT OF MENTAL CAPACITIES
Problem. Strictly speaking, it is impossible to measure directly

the original equipment of a human being unmodified by environ-
mental causes. The nearest approach would be the preparation of
a complete inventory, and an exact measurement, of all the capac-
itiesthat an individual possesses at birth. Even then, pre-natal

conditions have entered into the growth of the organism. The next
nearest approach would be a measurement of all capacities which
are not directly or specifically trained by school, occupation, or
special circumstances. In fact, no one can live and possess capac-
ities without any modification of them from outside causes; hence
the best that we can do is to measure as many capacities and abil-

itiesas possible which have been modified least by special exercise
or training, and then to consider them as approximately represent-
ing a person's original abilities, or, to make such allowances as we
can for the influence of external causes. No human being up to
the present time has been measured in all respects at any given
point in his growth by thoroughly accurate methods. A great
many persons, however, have been partially measured in a great
many capacities by more or less accurate or inaccurate methods at
various stages of their growth.
General Value. John Stuart Mill has said the "greatest thing
in the world is man, and the greatest thing in man is mind." To
this statement we might possibly add that the greatest achieve-
ment of science would be the measurement of the mind. The im-
,
port and value of definite means for measuring the capacities of

human beings would touch all phases of human life in which in-
telligence is involved. If we had accurate means of describing a
given person's capacities in all directions in terms that could be pre-
cisely defined and understood, we would have an instrument for
evaluating human beings far beyond our present possibilities. We
would then be able to obtain a precise notion of the capacities of
an individual. Consider for a moment what the advantages would
be! In all sorts of human relations, men are called upon con-
stantly to pass judgment upon their fellows concerning their fit-
97
ness, capacity, and promise of success for this or that particular

line of work. The one thing which is probably most important of
aside from the special training in a given field, is the intelligence
all,
and native ability which a person possesses. What aptitudes does
a person have for this or that type of work? Enormous waste in
the energies of men are due to mal-adaptation of individual to
task. The business world is rapidly turning toward psychology for
help, and if psychology is to give the help it will have to be in the

form of adequate measurements of the capacities of human beings.
Sound vocational guidance, in which much interest has recently
sprung up, will have to be founded upon a sound vocational psychol-
ogy whose development lies largely in the future. Courts are
recognizing that responsibility for conduct rests upon mental
maturity and intelligence, and that these must be determined first
before proper adjudication may be made of an individual's behavior.
Psychological laboratories have therefore been established in recent
years in connection with juvenile courts. The immigration office
finds it necessary to make intelligence examinations, even if crude,
in order to exclude those distinctly unfit. In normal times a con-
siderable number, 80 to 100 per month, are returned to the coun-
tries whence they came on account of mental deficiency.
One of the
large problems of the school is the proper adjustment
of work and progress to the natural ability of the pupils and, in
particular, the discovery of the morons and borderline individuals
so that they may be taken care of in special classes or otherwise to
the best advantage to themselves and to the other pupils in the
school. Intelligence examinations would be useful not only in
connection with the relatively small percentage of backward and
defective pupils, but also in connection with the normal and supe-
rior individuals. Such tests would be valuable in conjunction
with the measurement of attainment in school subjects specifically,
so that a child's progress and rate of advancement could be deter-
mined on the basis of both types of measurements. Precise methods
of evaluating the actual capacities of pupils would be of decided
value in making possible a more accurate promotion or retardation
of pupils according to their abilities, and a more accurate prescrip-
tion of work to be done and of the progress that can most profitably
be made. The school has paid relatively more attention to the
backward pupils by putting them into special classes than to the
superior ones. And yet the latter will be the ones who will con-
tribute most to the advancement of society as a whole. Why,
THE MEASUREMENT OF MENTAL CAPACITIES 99
should there not be special classes for the gifted pupils so that
they might be led to reach their fullest intellectual growth and
thus return to society the most that they are capable of?
Methods of Measuring Original Capacities. In general two
types of methods have been developed, at least in part, and used
for determining the native ability of human beings. The term
"native" of course, must be understood to signify not pure, native
ability unmodified by experience, but native or original only in the
sense of not being directly affected by specific training. The one
method consists of a considerable variety of reactions to questions
and situations which a child would be able to make as a result of
normal growth in a normal environment. The tests developed on
this principle are the Binet-Simon tests and the various modifica-
tions of them.
The second general method has proceeded on the basis of meas-
uring,by fairly precise methods, certain special mental functions
from year to year, and of determining thereby the mental status
and growth of the individual. Thus, for example, many capacities
might be measured by a definite psychological technique from year
to year, and certain norms might be established for each year so
that we could say that a given individual's memory has been de-
veloped to the norm or average of a child of ten. Similar tests and
norms could be developed in as many different mental capacities as
would seem to be necessary in order to obtain a fairly complete
evaluation of an individual's natural abilities. This second general
method has not as yet been developed to the same degree of com-
pleteness as the Binet-Simon type ('05) with respect to either the
selection of the particular capacities that should be tested, or the
types of tests that ought to be used, or the technique by which
they should be given. Brief consideration will be given to both
plans of measurement.
The Binet-Simon Scale. This series of tests is arranged in
groups according to years. Thus there is a series of tests for every
year from age three up to twelve or fifteen, and in some of the
revisions even to adult life. These tests were first prepared by the
French psychologist, Binet, and the French physician, Simon,
who collaborated for a period of twelve or fifteen years in the
selection of tests and in assigning them to the proper years accord-
ing to the growth and development of the child. These tests were
first published in 1905 and since then were revised by the original
authors in 1908 and in 1911. A number of investigators have

attempted to revise them and to adapt them to the conditions in

their respective countries, and to improve them so that they would
be more reliable and more accurately graded according to the
growth of children from year to year. In this country, the chief
revisions and improvements have been made by Goddard, Kuhl-
mann, Yerkes, Terman, and others. Probably the most satis-
factory and careful revision of the original Binet tests is the one
recently prepared by Terman and known as the Stanford Revision.
This revision consists in the elimination of some of the original
tests, in the addition of a considerable number of new tests, in the
readjustment of other tests up or down the scale of years according
to their difficulty, and particularly in the development of a more
precise technique for giving and evaluating the tests, so that ex-
aminers may be guided specifically in the administration of them.
The following is a complete list of the tests in the Stanford re-
vision. The detailed directions for giving and scoring the tests
together with extensive results, are given in Tennan's The Meas-
urement of Intelligence.
THE STANFORD REVISION AND EXTENSION

Year III. (6 tests, 2 months each.)
1. Points to parts of body. (3 of 4.)
Nose; eyes; mouth; hair.
2. Names familiar objects. (3 of 5.)
Key, penny, closed knife, watch, pencil.
3. Pictures, enumeration or better. (At least 3 objects enumerated
in one picture.)
(a) Dutch Home; (b) River Scene; (c) Post-Office.
4. Gives sex.
5. Gives last name.
6. Repeats 6 to 7 syllables, (i of 3.)
Ai. Repeats 3 digits, (i success in 3 trials. Order correct.)
Year IV. (6 tests, 2 months each.)

1. Compares lines. (3 trials, no error.)
2. Discrimination of forms. (Kuhlmann.) (Not over 3 errors.)
3. Counts 4 pennies. (No errors.)
4. Copies square. (Pencil, i of 3.)
5. Comprehension, ist degree. (2 of 3.) (Stanford addition.)
"What must you do?" "When you are sleepy?" "Cold?"
"Hungry?"
6.
Repeats 4 digits, (i of 3. Order correct.) (Stanford addition.)
Ai. Repeats 12 to 13 syllables, (i of 3 absolutely correct', or 2 with x
error each.)
THE MEASUREMENT OF MENTAL CAPACITIES IOI
Year V. (6 tests, 2 months each.)

1. Comparison of weights. (2 of 3.)
3-i5; iS-3; 3-iS.
2. Colors. (No error.)
Red; yellow; blue; green.
3. Esthetic comparison. (No error.)
4. Definitions, use or better.
(4 of 6.)
Chair; horse; fork; doll; pencil; table.
5. Patience, or divided rectangle. (2 of 3 trials, i minute each.)
6. Three commissions. (No error. Order correct.)
Ai. Age.
Year VI. (6 tests, 2 months each.)

1. Right and left. (No error.)
Right hand; left ear; right eye.
2. Mutilated pictures. (3 of 4 correct.)
3. Counts 13 pennies, (i of 2 trials, without error.)
4. Comprehension, 2nd degree. (2 of 3.) "What's the thing for
you to do?"
(a) "If raining when you start to school?"
it is
(b) "If you find that your house is on fire?"

(c) "If you arc going some place and miss your car?"
5. Coins. (3 of 4.) Nickel; penny; quarter; dime.
6. Repeats 16 to 18 syllables, (i of 3 absolutely correct, or two with
error each.)
i
Ai. Morning or afternoon.
Year VII. (6 tests, 2 months each.)

1. Fingers. (No error.) Right; left; both.
2. Pictures, description, or better.(Over half of performance de-
scription.) Dutch Home; River Scene; Post-Office.
3. Repeats 5 digits, (i of 3. Order correct.)
4. Ties bowknot. (Model shown, i minute.) (Stanford addition.)
5. Gives differences. (2 of 3.)
(Fly and butterfly; stone and egg; wood and glass.)
6.Copies diamond. (Pen. 2 of 3.)
Ai. i. Names days of week. (Order correct. 2 of 3 checks cor-
rect.)
Ai. 2. Repeats 3 digits backwards, (i of 3.)
Year VIII. months each.)

(6 tests, 2
1. Balland field. (Inferior plan or better.) (Stanford addition.)
2. Counts 20 to i. (40 seconds, i error allowed.)
3. Comprehension, 3d degree. (2 of 3.) "What's the thing fo*
you to do?"
(a) "When you have broken something which belongs to some*

"
one else?
(b) "When you are on your way to school and notice that you
are in danger of being tardy?"
(c) "If a playmate hits you without meaning to do it?"
4. Gives similarities, two things. (2 of 4.) (Stanford addition.)
Wood and coal; apple and peach; iron and silver; ship and
automobile.
5. Definitions superior to use. (2 of 4.)
Balloon; tiger; football; soldier.

6. Vocabulary, 20 words. (Stanford addition. For list of words
used, see record booklet.)
Ai. i. First six coins. (No error.)
Ai. 2. Dictation. ("Sec the little boy." Easily legible. Pen. i
minute.)
Year IX. (6 tests, 2 months each.)

1. Date. (Allow error of 3 days in c, no error in a, b, or d.)
(a) day of week; (b) month; (c) day of month; (d) year.
2. Weights. (3, 6, g, 12, 15. Procedure not illustrated. 2 of 3.)
3. Makes change. (2 of 3. No coins, paper, or pencil.)

10-4; 25-4-
i5-p;
4. Repeats 4 digits backwards, (i of 3.) (Stanford addition.)
5. Three words. (2 of 3. Oral, i sentence or not over two co-
ordinate clauses.)
Boy, work, money, men; desert, rivers, lakes.
river, ball;
6. Rhymes. rhymes for two of three words, i minute for each
(3
part.) Day; mill; spring.
Ai. i. Months. (15 seconds and i error in naming. 2 checks of 3
correct.)
Ai. 2. Stamps, gives total value. (Second trial if individual values
are known.)
Year X. (6 tests, 2 months each.)

1. Vocabulary, 30 words. (Stanford addition.)
2. Absurdities. (4 of Warn.
Spontaneous correction allowed.)
5.
(Four of Binct's, one Stanford.)

3. Designs, (i correct, i half correct. Expose 10 seconds.)
4. Reading and report. (8 memories. 35 seconds and 2 mistakes in
reading.) (Binet's selection.)
5. Comprehension, 4th degree. (2 of 3. Question may be repeated.)
(a) "What ought you to say when some one asks your opinion
about a person you don't know very well?"
(b) "What ought you to do before undertaking (beginning)
something very important?"
(c) "Why should we judge a person more by his actions than

"
by his words?
6. Name 60 words. (Illustrate with clouds, dog, chair, happy.)
Ai. i. Repeats 6 digits, (i of 2. Order correct.) (Stanford addi-
tion.)
Ai. 2. Repeats 20 to 22 syllables, (i of 3 correct, or 2 with i error
each.)
Ai. 3. Form board. (Healy-Fernald Puzzle A. 3 times in 5 min.)
Year XII. (8 tests, 3 months each.)

2. Abstract words. (3 of 5.)
Pity; revenge; charity; envy; justice.

3. Ball and field. (Superior plan.) (Stanford addition.)
4. Dissected sentences. (2 of 3.) (i minute each.)
5. Fables. (Score 4; i. e., two correct or the equivalent in halt
credits.) (Stanford addition.)
Hercules and Wagoner; Maid and Eggs; Fox and Crow;
Farmer and Stork; Miller, Son, and Donkey.
7. Pictures, interpretation. (3 of 4. "Explain this picture/')
Dutch Home; River Scene; Post-Office ; Colonial Home.
8. Gives similarities, three things. (3 of 5.) (Stanford addition.)
Snake, cow, sparrow; book, teacher, newspaper; wool, cotton,
leather; knife-blade, penny, piece of wire; rose, potato, tree.
Year XIV. (6 tests, 4 months each.)

2. Induction test. (Gets rule by 6th folding.) (Stanford addition.)
3. President and king. (Power; accession; tenure. 2 of 3.)
4. Problems of fact. (2 of 3.) (Binet's two and one Stanford addi-
tion.)
5. Arithmetical reasoning, (i minute each. 2 of 3.) (Adapted
from Bonser.)
6. Clock. (2 of 3. Error must not exceed 3 or 4 minutes.)
6.22; 8.10; 2.46.
Ai. Repeats 7 digits, (i of 2. Order correct.)
Average Adult. (6 tests, 5 months each.)

2. Interpretation of fables. (Score 8.) (Stanford addition.)
3. Difference between abstract words. (3 real contrasts out of 4.)
Laziness and idleness; evolution and revolution; poverty and
misery; character and reputation.
4. Problem of enclosed boxes. (3 of 4.) (Stanford addition.)

6. Code, writes "Come quickly." (2 errors. Omission of dot
counts half error. Illustrate with "war" and "spy.") (From
Healy and Fernald.)
Ar. i. Repeats 28 syllables, (i of 2 absolutely correct.)
Ai. 2. Comprehension of physical relations. (2 of j.) (Stanford
addition.)
Path of cannon ball; weight of fish in water; hitting distant
mark.
"Superior Adult." (6 tests, 6 months.)

2. Binet's paper-cutting test. (Draws, folds, and locates holes.)
3. Repeats 8 digits, (i of 3. Order correct.) (Stanford addition.)
4. Repeats thought of passage heard, (i of 2.) (Binet's and Wis-
sler's selections adapted.)
6. Ingenuity test. (2 of 3. 5 minutes each.) (Stanford addition.)
The mental maturity or intelligence of a child is expressed in

terms of a quotient which represents the relation between his
mental age and his chronological age and is obtained by dividing
the former by the latter. Thus, a child 10 years old with a mental
age of 10 would have an intelligence quotient (IQ) of i.oo; a child
10 years old with a mental age of n, would have an intelligence
quotient of i.io, or a child 10 years old with a mental age of 8
would have an intelligence quotient of ,80. If the quotient is
under i.oo it means that the child is below the average and if it
is above i.oo, it means that the child is above the average. Ter-
man has suggested the following classification of intelligence
quotients with their approximate meanings:
"IQ Classification
Above .40i "Near" genius or genius.

i .20-1 .40 Very superior intelligence,
i .10-1 .20 Superior intelligence.
.90-1 10 .
Normal, or average, intelligence.
.80- .90 Dullness, rarely classifiable as feeble-mindedness.
.70- .80 Border-line deficiency, sometimes classifiable as dullness,
often as feeble-mindedness.
Below .70 Definite feeble-mindedness.
"Of the feeble-minded, those between .50 and .70 IQ include

most of the morons (high, middle, and low), those between .20 or
THE MEASUREMENT OF MENTAL CAPACITIES IO$
.25 and .50 are ordinarily to be classed as imbeciles, and those

below .20 or According to this classification the
.25 as idiots.
adult idiot would range up to about 3-year intelligence as the
limit, the adult imbecile would have a mental level between 3 and
7 years, and the adult moron would range from about 7-year to
n-year intelligence."
Terman ('17) has attempted to estimate the boyhood intelligence
quotient of Sir Francis Galton from such records as are available
of his youth, and believes it quite certainly to have been 2.00.
For example, on the day before his fifth birthday he wrote the
following letter to his sister, the statements of which have been
corroborated by other general evidence:
"MydearAdele,
"
I am 4 years old and I can read any English book. I can say all the
Latin Substantives and Adjectives and active verbs besides 52 lines of
Latin poetry. I can cast up any sum in addition and can multiply by
2,3,4, 5,6, 7, 8, (9), 10, (n).
"
I can also say the pence table. I read French a little and I know the
clock.
"
Francis Galton,
"
February 15, 1827."
At the age of 10, young Galton wrote the following letter which
represents maturity of judgment and intellectual interest worthy
of a high school or college student:
"
December 30, 1832.
"
My
"
Dearest Papa:
It is now my pleasure to disclose themost ardent wishes of my heart,
which are to extract out of my boundless wealth in compound, money
sufficient to make this addition to my unequaled library.
The Hebrew Commonwealth by John 9
A Pastor's Advice 2
Hornne's Commentaries on the Psalms 4
Paley's Evidence on Christianity 2
Jones Biblical Cyclopedia 10
"
27
To illustrate concretely the manner of determining the mental
age of a child, the following record of the examination of a boy 14

years and n months old is cited. The test numbers refer to the
preceding list.
Year III. (6 tests, 2 months each.)

i. Passed.
2.
3-
4-
5-
6.
Year IV. (6 tests, 2 months each.)

i. Passed.
2.
"
3-
"
4-
"
5.
6. Failed.
Year V. (6 tests, 2 months each.)

i. Passed.
2.
3. Failed.
"
4.
5. Passed.
"
6.
Year VI. (6 tests, 2 months each.)

1. Passed.
2. Failed.
"
3-
"
4-
5. Passed.
6. Failed.
Ai. Passed.
Year VII. (Failed.)
This boy passed alf the tests of the third year and twelve addi-
tional tests scattered through the years IV, V, and VI, for each
of which he receives two additional months of credit. Hence his
mental age is three years plus 24 months or five years, and his
intelligent quotient is .34. He falls into the class of imbeciles.
A different plan of evaluation has been prepared by Yerkes and
Bridges ('15) in the construction of their point scale from the

materials contained in the original Binet tests. This modification
has proved quite satisfactory in practice. Haines has constructed
from the same material a similar point scale which is adapted for
use with the blind. Pintner and Paterson ('17) have assembled
and standardized a series of fifteen performance tests chiefly of
the form-board type which are especially valuable for use with the
deaf and the mute.
It must not be assumed from the apparent simplicity of the
Binet tests that any novice can use them. On the contrary it
requires considerable training and psychological insight to use
them properly. Some persons by reason of lack of tact and sym-
pathetic attitude are temperamentally unfitted for mental testing.
is desirable that at least a
It year of training in experimental psy-
chology should precede practice work in giving the tests; and
before the examiner may be confident of administering them
FIG. 35. Distribution of intelligence quotients of 905 pupils. After Ter-

man ('16, p. 66).
satisfactorily he should have practice under supervision with

thirty to forty cases. The examining room should be plainly fur-
nished and free from disturbing interruptions or distractions.
The confidence of the child should be obtained before the examin-
ation is begun and his efforts should be encouraged so that he may
react to his best advantage. Above all, the exact formula for giv-
ing and scoring each individual test must be followed rigorously
if the tests are to have reliable diagnostic value.
Measurements Obtained from the Use of the Binet-Simon Tests.
Goddard tested some 1,500 pupils by means of the revision of the
Binet-Simon scale prepared by himself. He found the distribution
of abilities as shown in the following table in which are specified
io8 EDUCATIONAL PSYCHOLOGY
foe number more years ahead or behind their

of pupils one or
-Chronological age in mental maturity. The remarkable finding
is the enormous range of the mental ages of pupils of the same
chronological age, extending from those who are 4 or 5 years re-

tarded to those who are 4 or 5 years accelerated.
TABLE 25
A similar investigation was made by Terman J
i6) of 905 pupils

(
showing a distribution in terms of the intelligence quotient as

given in the accompanying diagram, Figure 35. Kuhlmann ('12)
tested some 1,300 defective individuals in the institution at Fari-
bault, Minnesota, and found mental ages for various chronological
ages as follows:
TABLE 26
CHRONOLOGICAL NUMBER OF AVERAGE CHRON- AVERAGE
AOE OLOGICAL ACES MNTAL AGE
I- S 7 4-6 2.6
6-10 85 8.7 3-8
11-15 194 12.9 5-i
16-25 353 20.0 5-5
26-60 367 36.5 5-5
Many uses of such measurements as these may be made in

school. Thus, for example, the number of retarded pupils in our
schools is usually stated to be very large and the number of ac-
celerated pupils is usually stated to be very small. The percentage
of retarded pupils is placed in many schools anywhere from 25
to 50%. An intelligence examination of 483 pupils in Kansas City
(Dougherty '13), showed that the percentage of pupils as dis-

tributed both according to intelligence and according to over- and
under-age plans was as follows:
RETARDED NORMAL ADVANCED

Chronological Age 49% 45% 6%
Mental Age 21% 49% 30%
It appears thus that the large percentage of so-called over-age
pupils is not due to lack of capacity but to various other causes.
The abilities of pupils seem to be distributed approximately in a
normal, symmetrical manner. The large majority of pupils ar*

in the center or at-age with about equal numbers retarded or
accelerated. of the over-age pupils in the numerous age-
Many
grade have recently been complied are not mentally
statistics that
arrested but entirely normal or average.
Measurements of Special Capacities in Relation to General
Intelligence. The second general method of measuring original
capacities is as yet largely in the experimental stage. However,
substantial beginnings have been made and a number of investiga-
tions have been carried out which indicate with considerable
assurance that it would be possible to select a series of tests of
specificmental functions which would be correlated closely with

general ability and could, therefore, be used as symptomatic meas-
ures of general intelligence. An extensive investigation on this
order was made by Simpson ('12) in which he attempted to solve
two problems one of which was to determine what sort of tests
would be indicative of intelligence, and the other was to apply
them to two groups of persons widely separated in general in-
telligence. He proceeded accordingly to apply some 13 or 14 tests
to two groups of persons, one group of 17 individuals composed of
professors and graduate students of Columbia University, which
represented the upper end of the intelligence scale, and the other
group of 20 persons whom he found at the Salvation Army Indus-
trial Home and at the mission on the Bowery in New York City,
which represented the lower end of the intelligence scale. These
tests were applied from two to four times each and were then com-
pared with estimated intelligence as ranked by the impressions of

several persons. The correlations of these various tests with es-
timated intelligence were as follows:
no EDUCATIONAL PSYCHOLOGY
TABLE 27. After Simpson ('i 2)
Estimated intelligence and Ebbinghaus Test 89

Hard Opposites 96
Memory for Words 93
Easy Opposites 82
"A" Test 21
Memory for Passages 35
Adding 72
Geometric Forms 07
Learning Pairs 34
Completing Words
Drawing Lines . 20
Extending Lines
From these correlations it appears that some of the tests serve

very well as indicators of general ability, such as the various mem-
ory tests, the Ebbinghaus test, and the opposites tests. Certain
other tests, however, seem not at all to be symptomatic of general
capacity since their correlations are approximately zero, such as,
for example, the drawing of lines and geometric forms.
A similar study was made somewhat earlier by Burt ('09), in
which he found the following correlations of various functions
with estimated intelligence:
TABLE 28. After Burt ?og)
General Intell gence and Attention (Dotting) 72

Apprehension (Pattern) 75
Adaptability (Mirror) 60
Memory (Words) 67
Discrimination (Alphabet) 70
(Cards) 54
Reaction (Tapping) 45
"
(Dealing) 36
Perception (Pitch) 38
(Lines) 23
Touch (Two pt.) 03
Weight Discrimination . 16
In an investigation made by the writer a series of eight tests

was applied to a group of 15 high school pupils. Each test was
given very carefully to each child individually according to a
uniform technique and repeated four times on four different occa-
sions. The difficulty with a great deal of testing work has been
that the tests have not been applied under sufficiently constant or
rigorous conditions or repeated sufficiently often to yield a fairly
THE MEASUREMENT OF MENTAL CAPACITIES III
accurate measurement of the particular capacity in question. At

the close of these tests each of the 15 pupils was asked to give a
rating of the other 14, placing the one that was estimated to have
highest intelligence as i, and so on down to the i4th. The pupils
knew one another welland so were able to give fair opinions. From
these ratings a combined rank of general, estimated intelligence
was obtained and these ranks were then correlated with the ranks
in individual tests as well as with the combined ranks in all tests
put together. The results are shown in the following table:
TABLE 29
Estimated Intelligence and Memory of Words 52
" " "
Memory of Passages 56
" " "
Opposites 08
" " "
Mental Addition 50
" " "
Arithmetical Reasoning 83
" " "
All Tests 69
" " " " "
except Opposites 89
Thus it will appear that the tests individually, with the ex-
ception of the Opposites, as well as collectively, agree very closely
with the combined estimates of intelligence given by the 15 pupils
of one another. The estimates of intelligence of pupils agreed very
closely among themselves. The pupil who stood highest in the
estimates of his fellows was estimated first by all pupils but one.
The ranking of the others, of course, did not agree as closely, but the
agreement was so close that the combined ranking yielded a rather

reliable rating. The chief discrepancies between the estimated rank
and the test rank occurred in the case of two of the fifteen pupils.
The one was a boy who was estimated considerably lower by his fel-
lows than was his rank in the tests. His estimated rank was 14
while his test rank was 5. The other pupil was a girl whose esti-
mated rank was considerably higher, 5, than her test rank, n.
Upon inquiry it was discovered that the boy was not liked well
by his associates, while the girl was unusually popular. Their true
mental ability was probably indicated more correctly by the tests.
The pupil who stood first in all tests combined, and was given
first place by all his comrades except one, completed the high school
at fourteen and maintained an excellent record in the university.

The promise of these and other investigations is sufficiently
great to make it possible to develop a series of properly selected
tests with a definite technique to measure the general ability of
human beings with considerable trustworthiness. It would be

necessary, after the selection of the tests, to develop definite norms
for each test and for each particular year from infancy up to adult
life. The advantage of a series of tests of this kind over the Binet
type, would be that they could be applied and evaluated with

greater precision; that they would measure more directly certain
fundamental mental capacities; that they would be less dependent
upon particular environmental conditions, and that they would
yield more objective and scientific results. The specific test method
of measuring intelligence gives greater scientific promise and will
in the future probably replace the Binet method to a large extent.
PART II
THE PSYCHOLOGY OF LEARNING: A. IN GENERAL

CHAPTER VHI
ANALYSIS OF PROBLEMS
Analysis of the Learning Process. What are the mental and

neural processes involved in various types of learning such as
writing, reading, spelling, adding, solving problems, operating type-
writers, grasping laws of nature, retaining facts, playing tennis,
riding a bicycle, sawing to a straight line, speaking a foreign lan-
guage, and the like? Probably all forms of learning can be reduced
to one relatively simple, schematic type: Reception of impressions
through the senses; assimilation, analysis, and combination of proc-
esses in the mind; and redirection of impulses to produce a reaction;
or in brief, stimulus, association, response. A child learns to avoid
a disagreeable stimulus by receiving the sensation, associating it
with disagreeable consequences, and reacting by avoiding the stim-
ulus. The chick learns to avoid disagreeable caterpillars by the
same process. Visual and gustatory stimuli are brought to the ap-
propriate centers of the brain and there associated so that whenever
in the future the visual stimuli of the caterpillar are brought in, the
disagreeable taste associations will also arise which will cause an in-
hibition of the muscular actions concerned in pecking at the cater-
pillar. Apupil, on the first day of school, is shown certain black

marks on a chart or in a book and is told "hat" which is to cause
"
him also to say the word hat." The psychological series of events
is as follows: First, the visual stimulus from the page transmits im-
pulses to the visual centers in the brain and simultaneously the audi-
tory stimulus from the pronunciation of the word by the teacher
transmits impulses to the auditory center in the brain; second, con-
nections between these visual and auditory stimuli in the brain and
arousal by the auditory stimuli of images and meanings of the object
"hat" which have been established through previous experiences
before coming to school; and third, a redirection of impulses to the
motor centers to attempt to speak the word "hat." A little later
the pupil is given a pencil and is asked to make these same black
marks which have the name "hat." The psychological series con-
sists of, first, visual stimuli from the form of the letters to the visual
centers in the brain, second, the establishment of connections in the

i ue
Il6 EDUCATIONAL PSYCHOLOGY
brain between the visual centers and motor centers for the hand,
and third, redirection of impulses from the motor centers to attempt
to write the word "hat." Then from the muscular movements of
the hand and arm, made more or less by trial and error, kinesthetic
sensations are derived and associated in the mind with the visual
stimuli of the word "hat." These two sets of sensations become
associated and direct the motor responses in carrying out the writing
act. Still later the pupil is given (i) the visual or auditory stimulus,
"If you buy a pencil for three cents and give the clerk five cents,
how many pennies should you receive back?" which (2) arouses a
variety of association processes between various numbers such as
multiplication, division, addition, or subtraction, and out of the
mass of associations one is selected, namely, five minus three equals
two, and this in turn (3) directs the impulse to the motor centers
to say "two." All learning, even including reasoning, is probably
of the same fundamental type. The only difference is that there
are more elements involved in each of the parts of the three-series
connections and that, owing to the larger number of elements
aroused, a selecting or picking out of certain elements rather than
of others takes place. Learning facts of history, economics, or
science may be described in the same general schematic manner.
The facts are either read in a book, heard spoken by the teacher, or
observed directly. These sensory impressions are associated, dis-
sociated, and combined in various ways, which in the course of
time usually lead to some form of reaction either of speech or of
larger muscular activity.
The preceding examples of school learning depend for the most
part upon simple associations, that is, upon the law that things ex-
1
perienced together or in close succession tend to come back together.

Thus after a certain number of repetitions the sight of certain black 1
marks will set off promptly the reaction of speaking the word "hat."
But in reality the process of learning is almost never as simple as
this. While it is true that association bonds must be set up be-
tween situations and responses, a single situation is almost certain
to present to the mind of a child of school age a multiplicity of
aspects. As a consequence we find, instead of a single bond joining
the response to the situation, a number of bonds each joining the
response to a different part of the situation. Thus the word "tri-
angle" may be associated with an equilateral triangle of red card-
z
board, /s inch thick, 8 inches on each side, showing a dull gray
edge and weighing one ounce. Innumerably different combina-
ANALYSIS OF PROBLEMS 117
tions of bonds may accomplish this. There may be, for example,
a major bond leading to a reaction to the redness, a secondary bond
connecting the reaction with the size and symmetry of the sides,
and minor bonds emanating from the thickness and color of the
edge, while the weight and texture of the paper, the shade of red
and, most important of all, the number of sides and the angles may
not emeirge from the complex at all. Clearly such a set of bonds
would be worse than useless in the presence of a right-angled triangle
indicated in a book by black lines on a white page with the angles
labeled by letters and with a base }4. inch long. By dint of numer-
ous experiences with a variety of triangles and with the help of the
teacher who points out the essential three-straight-sidedness, the
characteristics peculiar to a triangle finally emerge more or less
clearly from the complex and become associated with the various
reactions appropriate to "triangle." The false bonds are either
destroyed or greatly weakened. Association is still the basis of the
process, but there is in addition the dissociation from one another
of the various characteristics which make up the complexes called
objects. This is conveniently spoken of as learning by analysis arid
abstraction. When complete, the process of analysis and abstrac-
tion, which makes possible the reaction to parts of situations rather
than wholes, clearly is an enormous advance over simple associative
learning. One association thus properly attached to the significant
part of a situation may function without any further effort in a great
variety of similar but otherwise entirely novel situations. This
is probably the essence of reasoning. But again we must note that
the process is rarely so simple as has been outlined. Rarely are the
preliminary analysis and abstraction so complete that a reaction
is transferred without delay to a very novel situation. Besides,
the attention may be distracted from the often subtle and incon-
spicuous but significant element in the new situation by the novel
and striking but irrelevant features. Sometimes some of these ir-
relevant features touch off a reaction which is entirely inappropriate.
For example, all but the very brightest pupils in a class, which has
learned to compute with facility the area of triangles from printed
problems and diagrams in a book and which knows how to measure
accurately a straight line, would be completely at a loss to know
what to do if given a 66-foot tape measure and confronted by an
area of ground in the shape of a scalene triangle measuring 4 by 7
by 10 rods, covered with flower beds in a setting of greensward and
surrounded by an ornamental iron fence three feet high. The
Ii8 EDUCATIONAL PSYCHOLOGY
sagacious few, if well taught, will react unerringly to the significant

element in the situation. Such are the processes involved when a
child has learned to isolate the significant clement from the situa-
tion: "If you buy a pencil for three cents and give the clerk five
cents, how many pennies should you receive back?" so that it

touches off the subtraction reaction rather than that of addition,
multiplication, or division.
Common and Special Elements in the Learning Process. If we
grant that the stimulus-association-response series is the schematic
pattern of learning, it will be convenient to discuss the psychology
of learning in two the psychology of learning in general
parts: first,
in which the elements common to various types of learning^ will
be examined; and second, the psychology of the learning of school
subjects in particular in which the special elements and processes
peculiar to each type of learning will be examined.
Practically all types of learning of whatever sort, have certain
processes in common. They have in common certain elements of
sensation and perception which are involved in the reception of the
stimuli in any sort of learning. Associated bonds are formed in
certain fundamentally similar ways no matter what the mental
processes are between which the bonds are formed. Retention, as-
similation, analysis, abstraction and generalization have certain uni-
form characteristics. Likewise the redirection and reaction proc-
esses follow certain general principles. But on the other hand, each
type of learning has its own special sensory material presented and
its own particular manner; it has its own special bonds
perceived in
which must be formed between its peculiar elements; and it has its
own type of reaction occurring in its own individual way. Thus
in learning to read, the series is, first, visual-auditory stimuli of
words and groups of words; second, association of visual-auditory
impressions and the memory of the objects which they represent;
and third, the response of speaking. In learning to spell, the series
is, first, visual-auditory stimuli of letters in a certain succession;
second, association of the stimuli in their particular orders; and

third, response in speaking or writing.
The Psychology of Learning in General versus the Psychology
of Learning of School Subjects. The procedure of learning in
general can probably not furnish the process and technique of the
learning of school subjects. The process, the technique, and the
economy of the learning of school subjects must be worked out
experimentally in detail for each particular subject. All that the
ANALYSIS OF PROBLEMS 119
procedure, technique, and economy common to all types of learning

can do is to furnish a broad general background. Thus the sub-
stitution experiment in the author's Experiments ('17), Chap-
ter furnishes certain conclusions with regard to the distribution
X,
of time according to which the establishment of the bonds in-
volved in this particular type of learning can be made most
economically. This experiment shows that it is more profita-
ble to work at the task 10 minutes twice a day than 20 minutes
once a day, or than 40 minutes once every other day. It would,
however, be folly to attempt to generalize from such an experiment
merely and to say that it is better to teach writing 10 minutes at a
time twice a day than 20 minutes once a day, or than 40 minutes
'
every other day. All that the experiment indicates is the general
principle that short periods of work distributed at certain intervals

of time are productive of greater progress in learning material of this
sort than longer periods distributed at longer intervals. What the
length of the most economical periods and the intervening intervals
would be in the case of writing, reading, or any other school sub-
ject, cannot be inferred on a priori grounds from a general prin-
ciple, but must be determined in detail for each particular type

of material and for the particular conditions under which the
learning must take place. All that the general principle can do is to
point the way to a more or less probable solution, but the particular
direction and the course of the path must be determined from fur-'
ther observations. The factors and laws of the mind as set forth
in general psychology can therefore not be carried over bodily
into the psychology of a particular type of learning. General
psychology can furnish its experimental technique and its funda-
mental laws which will serve as guides in the development of the
psychology of special types of learning. From this point of view,
the psychology of school subjects and the pedagogy of these sub-
jects resulting therefrom, which is likely to be the only sound
pedagogy worthy of the name, are as amenable to experimental
attack according to rigorous, scientific procedure, as the problems
in other fields of psychology have been amenable to the technique
of experimental methods.
Program of Problems. According to our analysis, then, the
following problems result:
A. The psychology of learning in general.

i. How are the stimuli received?
a. How do sensory defects interfere?

b. What are the factors and conditions of observation
and perception?
2. How are they associated, analyzed and combined in the
mind?
a. What is the rate and progress in the formation of
associative bonds?
b. What factors and conditions promote or hinder the
most economical formation of the bonds?
c. What are the effects of one set of associations upon
other sets of connections? (Transference of
training.)
3. How are they redirected into responses?
B. The psychology of school subjects in particular.
1. What are the specific psychological processes involved
in the learning of each particular subject?
2. How may the capacities in each subject be measured?
3. What factors and conditions promote or retard the
learning in each particular subject?
CHAPTER IX
THE RECEPTION OF STIMULI: A. SENSORY DEFECTS
Importance of Normal Sense Organs. The first point of con-

tact with external stimuliis through the sense organs. It is obvious
that this point of contact should be as perfect as possible so that
the stimuli which are to furnish the material for learning, may be
received as accurately and completely as possible. The eye and
the ear are the most important avenues of information. Defective
eyes and ears necessarilymake school work difficult and disagree-
able. A
child with defective eyes or ears misses a great many
things which the normal child can see clearly or hear distinctly.
The sad aspect of the matter is that a great many children have
sensory defects of which neither they themselves nor their parents
or teachers are aware. They assume that every one else sees or
hears just as they do and consequently their attention is not called
to it. These defects, however, often become serious and remain as
permanent hindrances throughout life. Many of them, if discovered

early, may either be kept from becoming aggravating or be reduced
very considerably. Furthermore, sensory defects, and in particular
visual defects, have far-reaching consequences upon a child's life
as a whole through the production of headaches, nervousness, and
dislike for school work. It is even claimed that truancy is in some
instances indirectly traceable to visual defects on account of the
dislike for school work produced by them. Dr. J. H. Cliborne has
given a retrospect of his own school days during which he suffered

from visual defects, as follows:
"I now know I have always carried about 1.50 diopters of hyper-
metropia; in my very early days, possibly more. Books and school were
to me a nightmare, a source of unutterable disgust. I drove myself to
my tasks with the scourge of duty; I never took one moment's joy or
pleasure in the acquisition of knowledge, unless it was the satisfaction
of a task accomplished or conquest gained. I have no memory of a sense
of pleasure connected with my studies at school or college. The only
pleasant memories I have are those connected with outdoor sports, or
facts gained through observation, or in the lecture-room through my
ears; and from my boyhood I could never understand why we were

forced to read from books all that we learned.
"Early in life pondered over the easiness of the task of those who
I
never sat at the feet but who followed the tracks of the peripatetic philos-
ophers. Verily, my school and college days would have been a joy to me
had my cars and my distant vision been my means of acquiring knowl-
edge; and yet I never had a headache in my life at school nor in after
years until after the commencement of presbyopia. I was nervous to
the point of madness at times, and the more nervous I was the more
diligent I became, and the nearer I put my nose to my book. I have
frequently observed that my right eye was crossed after prolonged study,
or after a long written examination; this was also at times observed in
my case by a fellow-student. That the difficulty lay in my hypermetropia
I have no manner of doubt. I had inherited a love of learning, I felt
sure, and I had a right to the assurance, and my hatred of close applica-
tion was a mystery to me. I created a frown by my accommodative
strain, which has ever been a part of me. Prolonged application to
books would be followed often by sleeplessness or violence in the field at
play. I learned for these reasons the art of complete concentration, but
at what an expense of nervous energy." (Swift '12, pp. 94-95.)
As concrete illustration of the manner in which undiscovered

visual defect may interfere with school work, we may cite the
following case from the psycholgical laboratory of the University
of Pennsylvania:
"On a certain afternoon in March, 1896, Miss Margaret T. McGuire, a

grade-school teacher in the Philadelphia public schools, went to the
psychological laboratory of the University of Pennsylvania, accompanied
by a lad of fourteen a well mannered, intelligent lad, industrious in his
school work; one of the favorite pupils, in fact. Yet this lad was the
'
'bogy of the teachers who for seven years had had him in their classes:
he was a chronic bad speller. This does not mean that he misspelled
some words sometimes. He misspelled every word always, and did it in
the same careful and serious manner with which he recited the history
lesson he loved. His reading was as bad as his spelling; he was absolutely
incapable of getting through a single sentence correctly, a, an, and, the,
and a few three-letter words being the net result of his seven years'
schooling. He read saw for was, water for weather; wrote hlat for thai,
soas, for soap, and other picturesque combinations of the sort in endless
variety. His case seemed hopeless. Dr. Witmer made a long
. . .
examination, the result of which was the discovery that Charles Oilman
had an ocular defect never, in all these years, so much as suspected by
either his parents or his teachers: at the distance of about three feet the boy
THE RECEPTION OF STIMULI 123
saw everything double: 'he lacked the power to direct the two eyes co-
ordinately upon the same point in space, the left eye looking a Little
higher than the right/ A page of ordinary print was thus a blur; when-
ever he attempted to write, the words doubled tinder his pen. Curiously
enough, he had never mentioned this peculiarity he seemed to think it
the natural process of vision. And he had repeated three whole years of
school on this account alone. ... lie was fitted with glasses and later
operated upon; then for the first time in his life the printed page and the
words he was tracing with his pen were clear. But his reading and writing
and spelling were just as bad as ever! The oculist had removed the
icfcct he had not removed the effect of the defect: that was in the boy's
mind. And it was here that the psychologist came to the rescue by show-
ing justwhat the effect was and how to remedy it.
"Now, it is an obvious truism of daily life that in order to recognize a
thing when we see it again, we must have seen it, at least once, clearly and
distinctly: a mental image of it must have been left in the mind. Read-
ing issimply a rapid-fire recognition process by means of the stored
mental images of words. Charles Oilman had no stores of images of
words, for he had never seen any he had seen only blurs of words. He
was even worse off than the child just groping its way through the primer,
for he had to unlearn the blurs he had patiently acquired through those
seven years when nobody knew what his trouble was; then word by word,
he had to restock his mind with the images of words shown him through
his glasses. ... In spite of this handicap, the boy learned to read, write,
and spell, and was finally graduated from the grammar school only three
years later than he should have been; which was better than not being
graduated at all." (Carter '09.)
Types of Visual Defects. The most common forms of visual

defects are myopia, hypermetropia, astigmatism, strabismus, and
color-blindness.
Myopia, or near-sightedness, is usually due to the fact that the

eyeball is too long and consequently the rays of light entering the
eye are focused at a point somewhere in front of the retina. As a
result the rays of light are again spread out when they reach the
retina and therefore do not form a distinct image.
Hypermetropia, or far-sightedness, is usually due to the fact
that the eyeball is too short and so the rays of light entering the
eye are not sufficiently refracted in order to form a clear image
when they impinge upon the retina. The image would be formed
at a point back of the retina if the rays of light were extended.
In some cases myopia and hypermetropia may be due to improper
refraction of light by the crystalline lense or the cornea. These
abnormal conditions cause a considerable strain upon the muscles
of the eye which attempt to accommodate the lens in order to

form as clear an image as possible.
Astigmatism is due to the fact that the curvature of the cornea
or of the lens, usually of the former, is not the same in all meridians.
The result is that if the lens is accommodated for rays of light from
some parts of the field of vision, it will not be accommodated for
rays of light from other parts of the field of vision, and consequently
a portion of the field of vision will be distinct and another portion
will be blurred. The curvature of the cornea or of the lens in the
case of astigmatism be compared to that of an eggshell. It
may
is different in different directions whereas in the normal eye it is
similar in all directions like that of a sphere.

Strabismus refers to the lack of perfect balance in the external
muscles of the eyes, so that the two eyes do not focus upon the same
point at the same time. This difference in point of focus may, of
course, vary anywhere from perfect coincidence, as it should be in
the normal eye, to a very large deviation, commonly known as cross-
eyedness, which can be observed readily by looking at a person's
eyes. The history of Charles Gilman cited on preceding pages was
a case of strabismus.
Color-blindness consists in the confusion of certain colors, nearly
always red and green. Confusion of the other two fundamental
colors, yellow and blue, almost never occurs. The cause of it is
more or less speculative and be due to the absence, or the im-
may
proper functioning, of the color elements in the retina. Color-
blindness is obviously a drawback in any type of school work in
which colors are concerned, such as drawing, map work, domestic
science and manual arts and all scientific studies in which color
discrimination is involved.
Causes of Visual Defects. Visual defects are due in general
to two causes, (i) heredity, and (2) the strained use of the eyes for
fine distinctions at close range, particularly under poor illumination.
Color-blindness and strabismus are probably inherited in nearly
every instance. The other types of defects which relate to the
formation of the image are proably due in part to hereditary con-
ditions in the sense organs and in part to overstrained use of the
eyes. Reading, which has been the great promoter of civilization,

has also been, in a certain sense, a deteriorator of the eyes through
the strain put upon them by the fine distinctions that must be
made at close range and at a tremendously rapid rate. Durr [as
J
reported by Whipple ( io, p. 139,)] has attempted to explain the

great prevalence of myopia in Germany as compared with other

countries by the excessive demands made by the German school
system. He has estimated the number of hours devoted to study
and to exercise by the typical boy during the years 10 to 19 in
different countries as follows:
Hours study Hours exercise

Germany ............ 20,000 650
France .............. 19,000 1,300
England..' ....... 16,500 4,500
Most hygienists maintain that myopia is an acquired condition

whereas anatomists are more inc'i ,cd to regard it as an inherited
condition. Cohn (Whipple 'm, p. 139) reports that myopia oc-
curred in gymnasia in the following increasing percentages during
the six years of study: 12.5, 18.2, 23.7, 31.0. 41.3, 55.8.
Frequency of Visual Defects. Recent years have brought a
considerable number
of investigations as to the percentage of
children with defective vision. Whipple made an examination of
the vision of 1,000 white and 100 colored children in Jefferson City,
Missouri, in which he fourd the following percentages of visual
defects:
Taussig ('09) has summarized the percentages of visual defects

in various cities in different countries as follows:
1
This percentage is probably too high since it was discovered that the colored chil-
dren took peculiar pride in reporting headaches because they seemed to consider it a
sign of intellectual keenness.
126 EDUCATIONAL PSYCHOLOCiV
TABLE 3 1. After Tausbig ('og)
Heidelberg, Germany (1870) 35 .0%

Edinburgh, Scotland (1904) 43 2
Dunfermline, Scotland (1907) 170
Cleveland, well-to-do district (1907) 32 4
"
congested district (1907) 71.7
Massachusetts, except Boston (1907) 19 9
Boston and environment (1907) 30 7
Boston (1908) . . ... . .
23 o
New York City (r 906) . .
31 3
New York City, Borough of Manhattan ... ". . ro 2
Chicago (1909) .. .
19-4
Jefferson City, Mo., either eye (1908). . . .
3(1 5
" " '*
22 7
both eyes (1908) . . . ...
St. Louis County, Mo., either eye less than 20/20 30 6
" " " "
both eyes less than 20/30 (1909) 14.3
" " " " both 28
eyes less than 20/40 (1909)
Additional results from other cities as reported by Gulick and

Ayres ('08, p. 83) are as follows:
TABLE 32
Bayonnc, N. J . . .
77%
Camdcn, N. J. (1906) ... .27 7
Milwaukee (1907). . . . .... 14.7
Minneapolis (1908) .... . . . . .... 23 9
Pawtucket, R. I. (1901) . . . . n.i
Utica,N. Y. (1897) 10.9
Worcester, Mass 19.1
The large variations in the percentages quoted for different

citiesprobably do not represent actual differences in the prevalence
of visual defects. They are probably due mainly to differences in
standards adopted by various examiners. Whether an eye is re-
ported as defective or not is, in the milder forms of defect, an
arbitrary matter. From the purely mechanical standpoint, an
absolutely perfect eye is undoubtedly very rare. Therefore the
matter resolves itself largely into the question as to whether the
deviation from perfection is sufficient to interfere appreciably
with normal distinct vision. general statement we may say,
As a
according to the quoted tables, that approximately 25% to 33%
of the school children have visual defects sufficiently serious to
demand some attention. Color-blindness fortunately is relatively
rare. It is estimated that 4% or of men and less than
5% of i%
women are color-blind.
Visual defects seem to increase measurably with successive

/cars in school. Thus in Whipple's table the percentage of de-
fect among high school pupils is 1 1% higher than that among pupils
in the first three grades. An extensive comparison was made by
Gulick and Ayrcs ( o8) in New York City which showed the follow-
7
ing percentages grade by grade:
TABLE 33
Grade Per Cent of Visual Defect
2 20.2
3 21.9
4 25.8
5 24.8
6 24.5
7 26 9
8 32 3
Remedial Measures for Avoiding Visual Defects. The first

and probably most important suggestion is the examination of
the eyes of pupils at least once in two years or preferably once a
year. This would serve as a means of discovering the visual de-
fects so that measures could then be adopted for the proper care
of those pupils suffering from them. Excellent results have been
shown in various cities in which general sensory examinations
have been introduced. Taussig reports that in Boston the per-
centage of visual defects dropped from 30.7% to 23% and in
New York City from 31.3% to 10.2% as a result of the introduc-
tion of visual examinations. The schools require compulsory
attendance, but they have not taken sufficient steps to make it
possible for the children to remain in school to their greatest profit.
In the next place, the proper lighting of schoolrooms is highly
important. This matter is being taken care of, however, at the
present time by school architects in a much more thoroughgoing
manner than was formerly the case. Many of the older buildings
are poorly arranged and wretchedly lighted. The amount of win-
dow area to floor area usually recommended as satisfactory, is
approximately one to five or six. As typical of the inadequacy
of the lighting in older school buildings we may note that in a
test of the amount of light at different desks in one of the old school
buildings in Madison, made by Cohn's light tester, showed that in a

room of five rows of scats the two rows next to the inside
showed a decided insufficiency in illumination.
Numerous other precautions may readily be exercised by the

school to avoid the aggravation of existing visual defects. Such
measures are the breaking up of the school program to spread the
severe use of the eyes, reduction of closework in the early years of
a child's prohibition
life, of work in poor artificial light, adjust-
ment of the size of the desk to the size of the pupil, the instilling
of the habit of resting the eyes even for the short interval of a few
minutes in the midst of cyc-straining work. The proper printing
of books and the use of appropriate paper is being looked after
by publishers much more carefully to-day than formerly.

Frequency of Auditory Defects. The presence of a greater
or less amount of deafness in one or both ears often interferes very
considerably with normal school work. Whipple has reported for
the 1,000 white and 100 colored children in Jefferson City, Missouri,
auditory defects as follows:
White Colored
Defective hearing (whisper test) 7-7% 7-o%
" "
" "
(one ear) 64 4.0
(both ears) 11.3 17.
Taussig ('09) has reported the following percentages of de-

fective hearing in various cities:
TABLE 34
Edinburgh, Scotland (1904) 12.2%
Dunfermline, Scotland (i97) 4-o
Cleveland, well-to-do district (1907) 5.2
"
congested district (1907) r 8
Massachusetts, except Boston (1907) 58
Boston and environment (1907) 77
Boston (1908) 7 '
New York City (1906) 20

New York City, Borough of Manhattan ... i o
Chicago (1909) . . . ... . -'
7
St. Louis County, Mo., either ear defective (1909) . . . .
7 .>
" " " " both ears 2.2
seriously defective (1909)
Additional results reported by Gulick and Ayres ('08) are as

follows:
TABLE 35
Bayonne, N. J 2 .
5%
Camden, N. J. (1906) 4.1
Minneapolis (1908, 7.7
Pawtucket (1901) 4.3
Utica, N. Y. (1897) 6.6
Worcester, Mass 6.6
As a general statement we may say that approximately 5% to

10% of pupils from defective hearing of a sufficiently serious
suffer
character to interfere with their school work and to require medical
attention.
Effects of Sensory Defects upon School Work. number of A
inquiries have been made to determine the amount of hindrance
which sensory defects have upon the proper performance of school
work. Dr. Cornell has reported the following results for three
schools in Philadelphia showing the average school marks of normal
and defective pupils:
TABLE 36. After Cornell
Allison School 219 children, both sexes, 6 to 12 years old

Average
Normal child 75
Average child 74
General defectives 72.6
Adenoids and enlarged tonsils 72
Deaf 67.2
Ninth Street Primary School 84 children, both sexes, 6-10 yrs. old.
Language Arithmetic Spelling Average

64 cases normal children 72.9 75 5 75.4 74.6
84 cases average children 70.5 74 72.8 72.4
21 cases general defectives 63.3 70 64.8 66
8 cases adenoids 60. o 66.7 65 63.9
No cases deaf.
Claghorn School 252 children, both sexes, 12 to 15 years old.
Geography
and
Language Arithmetic History Average
179 cases normal children 74.4 72 76.6 74.3
252 cases average children 72.7 70 76 5 .
73 . i
73 cases general defectives 71.4 65 . i 76. 2 70.8
Whipple in his study in Jefferson City found that among pupils

of good vision, 26% did unsatisfactory work and among pupils
with defective vision 38% did unsatisfactory work. Smedley
reported that in Chicago there were 18% with auditory defects
among pupils above grade, and 25% among pupils below grade.
It is difficult to say how much of the scholastic inferiority of those
having sensory defects is due to these defects and how much is

due to inferior native ability. It has been shown by Pintner and

Paterson that deaf children are fully three years behind
('16)
normal children in learning the digit-symbol test which, as given
by them, does not depend upon the use of language. Goddard ('14)
in an intensive investigation found blindness perceptibly asso-
ciated with feeble-mindedness, and deafness with neuropathic
taint.
Musical Discrimination. Besides the various degrees of deaf-
ness, the sense of hearing is of direct interest to the school from
the standpoint of musical instruction. The ability to sing, and
to some extent, to appreciate music, depends in part upon the
accuracy of the discrimination of pitch. This ability varies widely
among people, and considerable inaccuracy in musical discrimina-
tion is a distinct difficulty in learning music. The results of recent
experiments seem to indicate rather definitely that accuracy in
pitch discrimination can probably not be improved by practice
or by The voice in singing and the hand in playing the
teaching.
violin, areguided by the accuracy of the ear. If the ear is not
accurate the individual is unable to guide his voice or his fingers
with a precision necessary for the production of music.
Seashore ('01), who has studied this problem carefully, has
suggested that pupils whose discrimination is two vibrations or
less have a sufficiently fine ear to become musicians; pupils whose
discrimination lies between 3 and 8 vibrations, which includes the
large majority of people, have a sufficiently accurate discrimination
for ordinary musical instruction and enjoyment; pupils whose
discrimination lies between 9 and 17 vibrations should have musi-
cal instruction only if they have special inclination or desire for it
and singing in school should be optional for them; and finally,
pupils whose discrimination lies at 18 or above, should prob-
ably not be required to study music or to attempt to produce
music.
Physical Defects. Beside the sensory defects there are several
types of physical defects whose frequency is high and whose inter-
ference with school work is serious. The prevalence of these de-
fects is indicated in the following table, adapted from Gulick and
Ayres('i3,p. 38):
Average 41.5 20.1 10.8 n.8 88

CHAPTER X
THE RECEPTION OF STIMULI: B. PERCEPTION AND OBSERVA-
TION OF THE SENSORY MATERIAL
Problems. Next to the normal operation of the sense organs

come the actual perception and observation of stimuli as pre-
sented to the sense organs. What material you learn and how
you learn it depends upon what and how you perceive or observe.
Obviously what you perceive or observe depends upon the normal
operation of the sense organs. That has long been recognized.
But it has not been so fully recognized that it depends also upon
the mental apprehensionjiLlifiêjagpry stimuli. Perception and
observation do not depend alone on what is presented to the sense
organ, but also upon how the stimuli are taken into the mind.
The specific problems involved in it are:
(1) How accurate is the observation of stimuli?

(2) How large is the range of stimuli observed at a given time?
(3) How may the accuracy and range of observation be im-
proved?
(4) How are the stimuli interpreted?
Accuracy of Observation. Much of the difficulty in learning

a given material is due to inaccuracy and error in the observation
of the material. The word to be spelled, the letter to be written,
the plant to be described, the experiment to be reported, the map
to be drawn, may all be done and learned incorrectly in part be-
cause they are perceived and observed inaccurately and in-
completely.
Recent experiments on the "fidelity of report" have called at-
tention to the prevalence and nature of inaccuracies in observation.
The observations are not made with sufficient care and attention
to impress a faithful image of the object upon the mind. Inac-
curacy in perception and infidelity in report have hardly been

known to exist and therefore have not been sufficiently guarded
against because, under ordinary conditions of learning in school
as well as under ordinary conditions in life, there is seldom an op-
13*
portunity for comparing directly the observations as received in

the mind with the original stimuli as actually presented to the
sense organs.
The recent investigations on the fidelity of report have been
carried out by presenting to a group of observers a series of events
either in the form of more commonly by means of a
real actions or
picture. The picture

exposed to the observers for a short period,
is
say thirty seconds or a minute, after which they are asked to write
a report of their observations. This is usually supplemented by
an interrogatory report consisting of answers to questions regarding
the picture. Experimental inquiry into these matters has been
stimulated primarily from the practical importance of determining
the reliability of witnesses in court. Incidentally, the results have
an exceedingly significant bearing upon the accuracy of observation
involved in learning. The main results of these experiments have
been summarized by Whipple ('10, pp. 304 ff) as follows:
"The Aussage psychology is that an errorless
chief single result of the
report not the rule, but the exception, even when the report is made by
is
a competent S (subject) under favorable conditions. Thus, in 240 reports,

Miss Borst found only 2% errorless narratives and 0.5% errorless deposi-
tions. The average subject, when no suggestive questions are employed,
exhibits a coefficient of accuracy of approximately 75%.
"Generally speaking, attestation does not guarantee accuracy: on the
contrary, though the numbers of errors is nearly twice as great in un-
sworn as in sworn testimony (according to Stern, 1.82 times, according to
Borst, 1.89 times as great) there still remains as high as 10% error in
sworn testimony.
"Reports of children are in every way inferior to those of adults: the
range is small, the inaccuracy large, and, since the assurance is high, the
warranted assurance and reliability of assurance are both very low.
During the ages 7 to 18 years, the range, especially the range of knowl-
edge, increases as much as 50%, but the accuracy, save in the deposition,
does not increase as rapidly (20%). This development of capacity to
report is not continuous, but is characterized by rapid modification at the
age of puberty.
"The one factor that more than others is responsible for the poor
reports of children is their excessive suggestibility, especially in the

years before puberty.
"Not all the features of the original experience are reported with the
same frequency or with the same accuracy: there is, rather, a process of
selection, both in the process of observation, and also, probably, in
memory and in the formulation of the report. In general, we may say
that persons and their acts, objects, things, and spatial relations are re-
ported with considerable accuracy (85-90%), whereas secondary features,

especially quantities and colors, arc reported with considerable inaccu-
racy (reports on color have an error of from 40-50%).
"All authorities agree that the use of interrogatory, whether of the
complete or incomplete form, increases the range and decreases the
accuracy of the report.
"The introduction of leading or suggestive questions very noticeably
decreases the accuracy of report for children, and, unless the conditions
of report are quite favorable, even for adults. The greater suggestibility
of children is shown by Stern's results in which the inaccuracy of boys
and girls, ages 7 to 14 years, was from 32 to 39%, as against 10% in-
accuracy for young men aged 16 to 19 years."
The reasons for inaccuracies in reports of observation must be

sought in various directions. The chief ones are (i) insufficient
attention to the material observed so that only a vague impression
is produced which may easily be disturbed and modified by other
stimuli, (2) meagejrness of ideas and experiences with which the

observed material may be connected and interpreted, (3) low
retentiveness of the impressions so that other impressions can
readily distort them, (4) faint imagery in terms of which to picture
the objects, (5) lack of conscientiousness in keeping apart the
observed and the inferred items as a result of which missing or
doubtful elements are supplied by unconscious inference, (6) the
effect of suggestion through which the mind is prone to seize upon
any slight hint and to fit it into the story.

The Range of Observation. The amount of material observed
by different persons within a given limit of time varies over an
astonishingly wide range. Individual differences in the capacity
for apprehending stimuli from the outside world are probably as
large as those in other mental abilities. One person may report

several times as much of a scene or series of events as another.
The pupil with a wide scope of apprehension and observation has
a tremendous advantage over one with a narrow scope of observa-
tion. This range probably depends fundamentally upon the span
of attention, quickness of assimilation of items, retentiveness, and
previous knowledge about the facts to be observed. The span of
attention as measured by rapid exposure methods, ranges in normal
persons from three or four objects to eight or nine. Such experi-
ments are carried out by exposing momentarily before the observer
cards with varying numbers of objects and by recording the num-
ber of objects noticed.
Some years ago a number of interesting investigations were

made by G. Stanley Hall ('83) and others to determine the range of
observation and knowledge of pupils entering school. The returns
showed a surprisingly narrow range of information and the great
extent to which it was determined by the environment in which
the child lived. The educational value of such inquiries has been
to emphasize the importance of a wide variety of immediate con-
tact and experience with real objects by observing, manipulating
and using them.
As a concrete illustration, the author made the following simple
experiment to indicate the range and fidelity of such observations
as would be made in a class in biology. Some plants in a jardiniere
were exhibited before a class of thirty-nine students for thirty sec-
onds with instructions that they were to observe them as carefully as
possible and that they would be asked immediately afterwards to
record their observations. The shortest report (A) was only fifty-
seven words in length, and one of the longest and best (B) was 131
words in length. One of the most erroneous ones is given under (C).
B is more than twice as complete as A, and fully as free from error.
C is quite typical with regard to kind and frequency of misstate-
mcnts. The erroneous portions are italicized and the corrections
are given in parentheses. All three persons had had an elementary
course in botany.
Report A:
"There was a high upright geranium plant having no buds but broad
leaves. This was surrounded by. low plants with drooping stems and
bearing many pointed, small leaves. The leaves had a pinkish center,
surrounded by a pale green band, the contour very irregular and the
general effect bushy. There were no buds on these plants either."
Report B:
"From a brown jardiniere arose one stalk of a geranium bearing nine
big green leaves. The leaves spread out in all directions and are round
in shape with large scallops. Lower was another plant with much smaller
leaves and more bushy. It had three large branches, one leaning over
the pot on each side and one across the front. The leaves of this plant
were more oblong in shape, rounding at the base and reaching a point.
The color was a pale pink in the center and back to the base, shading to
a deep red or wine (pink) color towards the tip end and the whole edge of
the leaf was green. The stem was much more delicate than the stem
of the geranium and the leaves were much more numerous."
Report C.
"
The plant was in a brown bowl. There were varieties of plants. The
one had one large stalk with five (nine) branches growing out of it. The
leaves were large and heart-shaped (rounding). The other plant was lower
and drooped, and had more leaves. The leaves were oblong (heart-
shaped) and smaller than the leaves of the first plant. They had a center
of very light green (pink) and were outlined by darker green. The leaves
were smooth and glossy (velvety). The leaves of the tall plant were
1
notched.'
It is obvious that the reporter of B has a great advantage over

either A or C both in the quantity and in the quality of his observa-
tions. In a given period of time B will learn much more material
and assimilate it in more correct form than either A or C. The
objection might be raised that a report formulated after the obser-
vations have been made is bound to be erroneous and that it would
be fairer to have all three persons stand before the plant and re-
cord their observations at the time. To this, however, we may
reply that the range of facts observed and recorded in a given
period of time would be just as wide and that observed items are
not really assimilated mentally until they can be adequately
thought out and expressed. The simple experiment here reported
probably represents quite fairly the sort of things that occur in
ordinary observation of material in learning.
Improvement in Observation. Granting the importance of
a wide range and a high degree of accuracy in the perception of the
material to be learned, what may we do to increase the accuracy,
scope, and fidelity of observation? Probably the only advice to
give at the present time is the rather obvious suggestion: Insist
on greater accuracy and attention in observation. This may be
accomplished by definite discovery of errors and inaccuracies in
the observations themselves. Experiments have shown that many
people, especially children, do not realize their inaccuracies and
that calling attention to the discrepancies between objects and
mental impressions of them results in a material reduction in
unreliability. Whipple ('10, pp. 309 ff.) has summarized the experi-
mental results on the possibility of improvement in observation
by repeated tests with the same groups of persons as follows:
"
Simple practice in reporting even without special training or conscious
effort to improve, facilitates and betters the report, as is shown in Ta-
ble 47, from Miss Borst. It will be noted that the tendency to oath and
warranted tendency to oath are both particularly improved by practice,

and that there is also an appreciable improvement in range, accuracy,
warranted assurance, and reliability of assurance, whereas assurance and

accuracy of assurance are scarcely affected. Similar practice-effects
may be discerned in their deposition. From these results, it is clear that
the several coefficients of report may vary more or less independently."
EFFECT OF PRACTICE UPON COEFFICIENTS OF REPORT (NARRATIVE)
Range
Number
Accuracy
of Report (Test) 12345
(BORST)
39.0
86.6
390
87.7
42.3
92.9
40.3
88.2
42.0
90.0
Assurance 96.6 96.4 97.8 97.9 98.6
Warranted assurance 84.0 87.0 91.0 88.0 89.0
Reliability of assurance 87.5 89.4 92.6 89.8 90.3
Accuracy of assurance 97.0 98.0 98.4 98,6 99.2
Tendency to oath 43.0 59.8 62.8 61.9 72.1
Warranted tendency to oath 402 53.2 58.5 57.5 66.5
Unwarranted tendency to oath .. . 2.8 6.6 4.3 4.4 5.6
Reliability of oath 93.0 88.8 92.5 93.0 91.7
Norr: The effect of practice in these tests is somewhat obscured by the fact that thj
firbt and third tcbts were made after a 3-day, the others after a y-day interval.
"The capacity of children to observe and report in a detailed and

accurate manner may be improved by systematic training. This educa-
tion may be best secured by appeal to zeal, interest, enthusiasm, or desire
for improvement on the part of the child; more formal training of an
intellectual type, e. g., suggestions for systematic observation, specific
training in sense-perception, instruction designed to augment appropriate
apperceptive-masses, etc., is much less effective.
"The inadequacy of the child's report is due, not so much to poor
memory, as to the fact that he fails to perceive many features in the
original experience, that he fails to put into words even what he does
perceive, and especially to the fact that he is absurdly uncritical (his
assurance, indeed, commonly reaches 100%)."
Tests such as these, but made with the material of school studies,
would probably be very useful in bringing about more concen-
trated attention upon, and greater reliability in, observation.
Thus a plant or a flower in a course in biology, might be exposed
for a definite period of time to a class of pupils who would then be
asked to write as accurate a description of the object as possible.
This description could then be definitely compared, point by point,
with the original object a.nd in this manner the errors and inaccu-
racies would be discovered and noted. Difficulty in acquiring the
material in various school subjects is no doubt traceable in larger

part than we realize to the incompleteness and the unreliability
of the perception of the material or stimuli to be acquired. It
would be an experiment worth making to determine to what extent
the difficulty of a pupil in learning to spell is due to actual incom-
pleteness of the observation and perception of the letters in the
words.
Interpretation of Stimuli. To a large extent, observation is
interpretation. The same identical sense impressions are inter-

preted very differently by different observers. This may be dem-
onstrated perhaps in extreme form in such tests as the one with
the ink blots outlined in the author's Experiments, Chapter XIII.
The first ink blot in that series signified to eight persons the follow-
ing different things: map, bear, trees, lake, cloud, chfld, bat, man
running. The same mental processes occur in a less variable man-
ner in kinds of observation. Incoming stimuli are interpreted
all
by the association processes aroused in the mind. On this basis

the traditional doctrine of apperception has been formulated and
from it have been derived such pedagogical corollaries as, "Link
thenew to the old," or "Proceed from the known to the unknown."
The theory of apperception has been very clearly expressed by
James in the following statement:
"The matter is this: Every impression that comes in from

gist of the
without, be a sentence which we hear, an object of vision, or an efflu-
it
vium which assails our noses, no sooner enters our consciousness than it is
drafted off in some determinate direction or other, making connections
with the other materials already there, and finally producing what we
call our reaction. The particular connections it strikes into are deter-
*
mined by our past experiences and the associations' of the present sort
of impression with them. If, for instance, you hear me call out A, B, C,
it is ten to one that you will react on the impression by inwardly or out-
wardly articulating D, E, F. The impression arouses its old associates:

they go out to meet it; it is received by them, recognized by the mind as
'the beginning of the alphabet/ It is the fate of every impression thus
to fall into a mind preoccupied with memories, ideas, and interests, and
by these it is taken in. Educated as we already are, we never get an
experience that remains for us completely nondescript: it always reminds
of something in quality, or of some context that might have surrounded
it before and which it now in some ways suggests. This mental escort
which the mind supplies is drawn, of course, from the mind's ready-made
stock. We conceive the impression in some definite way. We dispose
of it according to our acquired possibilities, be they few or many, in
the way of 'ideas.' This way of taking in the objects is the process of
apperception. The conceptions which meet and assimilate it are called
by Herbart the 'apperceiving mass.' The apperceived impression is
engulfed in this, and the result is a new field of consciousness, of which
one part (and often a very small part) comes from the outer world, and
another part (sometimes by far the largest) comes from the previous con-
tents of the mind." ('99, p. 157.)
The doctrine of apperception as here stated by James is simply

a statement of the psychology of perception as nsna|)y accepted.
The importance of the applications of the doctrine to teaching has
perhaps been overemphasized in the educational writings of the
recent past and in the pedagogical methodology that has been
worked out in accordance with its corollaries. The applications
of the theory as represented by the injunction, "Link the new to
the old," is no doubt sound from the psychological side and useful
from the pedagogical side as a general guiding principle. Illustra-
tions of the principle would be the teaching of a topic in geography
by connecting it up with the known facts of geography in the
immediate environment of the child, or of teaching laws of chem-

istry by relating them to familiar facts and problems that have
arisen within the child's own experience, or of teaching the spelling
of a new word by pointing out its similarities to words already
known, or of teaching forms of a foreign language by referring
them to related forms in the language previously acquired. Such
a procedure is unquestionably valuable whenever it can be em-
ployed. However, some of the enthusiastic advocates of the
doctrine have been somewhat blinded to the limitations of it. If
we regardJtearning as a process of establishing rnnnections h^fwepn
elements of the learning-material, we can conceive of three possible
kinds of bonds to be formed: (i) between two known elements
which had previously not been connected, (2) between a known
and an unknown element, (3) between two unknown elements.
It is obvious that a great deal of learning consists in the formation
of the third type of connections. The doctrine of apperception
can apply only to the first and second type of connections and
these can very probably be formed more readily according to the
natural workings of apperception because some of the elements
had previously been acquired.
Much of the discussion in favor of the doctrine of apperception is
really based upon the greater practical value of the known and
nearer at home, and upon the urgent need of knowing something
about the immediate environment rather than about a distant

time or territory, whose history or geography may be of little value
to the child, than upon greater ease in the formation of bonds be-
tween a known and an unknown element. The main thing in educa-
tion is not to proceed from the known to the unknown, but rather
to acquire the unknown. If this can be done by linking the new
to the old, well and good, but the chief object is the linking of the
new. Much of the so-called proceeding from the known to the un-
known or of the linking of the new to the old, is more or less fruit-
less, since it neither proceeds to the unknown nor links anything
new. reawakening of the known and of the

It usually consists of a
old. Learning fundamentally the acquisition of new sets of
is
stimuli-association-response series. When a pupil first attempts to

write, he must acquire new neural connections in securing control
of his hands. When he begins to learn the meaning of printed
symbols, he is confronted with new stimuli among which new con-
nections must be established. Much of the so-called teaching ac-
cording to the theory of apperception consists in setting up prob-
lems concerning things with which the child is already familiar
and thus in arousing in him a desire to learn something new. This
is, no doubt, good teaching, but the important part in learning is the
new element to be sought and the new associations to be built up.
CHAPTER XI
THE RATE AND PROGRESS OF LEARNING
Problems. The chief problems to be considered are as fol-

lows:
(1) How
rapidly are new associative bonds formed?
(2) rate of acquiring new connections and new materials
Does the
continue uniformly per unit of time or per repetition?
(3) Do variations in rate occur in a uniform manner?
(4) What
causes will bring about variations in rate?
(5) Does the rapidity of learning occur in a similar manner in
alltypes of learning?
Such problems as these may profitably be raised with regard to
any sort of learning. If we consider the learning of a language we
may ask, How rapid is the progress in acquiring the meanings of
the words, knowledge and use of grammatical forms and ability to
translate? Is progress uniform or are there times of rapid advance
occurring in alternation with periods of little or no gain? What
conditions will promote the learning of the language most ef-
fectively? If we could answer these and similar questions con-
cerning any type of learning we would be able to control its
progress far more economically than we are able to at the pres-
ent time.
The Curve of Learning. The rate and progress of learning
may be expressed in terms of the amount done per unit of time, or
in terms of time required per unit of work. The relation between
these two variables is represented by the curve of learning in
which one function, usually time, is represented along the base
line, and the other, usually amount accomplished, is represented
along the vertical axis. Figure 36 represents a typical curve of
learning in which progress is measured by the amount achieved per
five minutes of time. It represents the rate of forming associations
between numbers and letters in transcribing letters into numbers
as specified in the author's Experiments ('17), Chapter X.
Characteristics of Learning Curves. Most of the experimental
work on the course of learning curves has been done chiefly with
various kinds of skills such as telegraphy, typewriting, tossing
141
balls, mirror tracing, substitutions, and the like. Little has been
done on the progress of analytical types of learning, on the advance
in the acquisition of facts of a science or of the history of a country,
260
240
^220
200
1,100
g 140
m
! 120
| 100
J 80
5 10 15 20 24
Five Minute Periods
FIG. 36. A curve of learning showing the progress of one person in learning
to substitute numbers for letters in the experiments outlined in Chapter X,
Experiments in Educational Psychology.
or on the rate at which a child learns to read or to write. Conse-

quently most of our generalizations up to the present time have
been based upon curves that represent the acquisition of skill.
120,
gioo
>
4 8 12 18 20 24 28 32 80
Weeks of Practice
FIG. 37. Improvement in telegraphy. Individual E. L. B. After Bryan
and Harter ('97, p. 49).
Such curves seem to have in common two general characteristics,
although it is doubtful whether they are universal in all types of

learning: (i) An initial period of rapid progress, and (2) successive
periods of no progress, or plateaus, followed by periods of rapid
progress. Theoretically, a curve of learning may have two initial

directions: (i) rapid progress followed by slower progress, or (2)
slow progress followed by more rapid progress, that is, a convex or a
concave form with all possible shapes between these extremes. The
large majority of curves of learning derived to date are of the former
sort. All the illustrations reproduced in Figures 37 to 41 have this
general shape. In the substitution test referred to in Figure 36

the author found that among twenty curves obtained from as
many individuals, thirteen were of convex form, six were practically
straight lines rising from left to right, and one was of concave
form. Hence, initial rapid gain seems to be a very common feature
Slowest Main Line Rate
10 20 90
Weeks of Practice
FIG. 38. Improvement in telegraphy analyzed. Individual J. S. After

Bryan and Harter ('99, p. 350).
in curves of skill. Other types of curves are shown in Figures 42

to 44.
The early period of rapid progress may be due (i) to the fact that
the first elements of a new set of materials or a new set of associa-
tions may be picked up rather easily and quickly because of their
in a
simplicity, (2) to the probability that the first stage of practice
new type of learning makes available various elements or activities
already in the possession of the learner, (3) to the initial zeal in
beginning a new task, (4) to the large opportunity for progress in
the beginning, (5) to the physiological limit in many types of skill
such as typewriting, mirror tracing and writing numbers for letters,
and (6) to an absolute limit of the number
bonds that the task
of
presents to the learner. Thus typewriting has a physiological

limit in the rapidity with which the fingers can be moved in striking
the keys. Progress cannot go on indefinitely at the original rate.
Typewriting also has an absolute limit in the number of strokes
to be learned.
Book explains the initial period of rapid progress thus:
"After what has been said our explanation of the general features of
our curves can be brief. The first rapid and continuous rise is due to
the fact that the learner making progress along many different lines at
is
once. Rapid strides of improvement arc possible and made simulta-
6 10 X5 20 25 80 85 40 45 60 65 60 65 70 75
Days
FIG. 39. Progress in learning Russian. After Swift ('08, p. 198).
neously in every department of the work. The learner is not only forming
and perfecting letter associations but syllable, word and phrase associa-
tions as well. He simultaneously improving his method of dealing
is
with every problem that the writing presents; locating the keys, directing
and controlling his fingers, * spelling' or initiating the movements, get-
ting his copy, learning to deal with special difficulties, learning to keep
attention more closely and economically applied to the work, etc. The
curves will rise rapidly and continuously so long as many of these possibil-
ities of improvement exist. As they grow less numerous the rate of gain
will likewise decline until, as still more skill is acquired, a state is reached
where most adaptations or short cuts in method have been made; fewer
1234 5 7 8 9 10
Successive Trials
11 12 13 H
FIG, 40. Improvement in tracing a star outline when seen in a mirror.
Continuous lines represent reduction in seconds in successive trials. Dotted
line represents reduction in errors.
10 20 30 40 CO 60 70 80 00 100 110 120 130 140 160 100 170
Amount of Practice in Hours
Fro. 41. Improvement in typewriting by the sight method. After Book

('08 plate, opposite p. *t.
special habits remain to be developed; fewer adaptations are possible.

Those possible have become harder and harder to make, because they
must be made in the realm of higher habits where the learner has had
less experience. Every man has had experience with the first stages of
learning, but little with the later stages because most people touch
lightly many things and are masters of nothing. There being now fewer
adaptations to make, and the process of finally perfecting all the special
associations being so gradual and slow, the learning curve becomes, as
1100
1000
ooo
800
700
600
500
400
800
200
100
101520258085404560
FIG. 42. Progress in ball-tossing. The horizontal axis represents days.
The vertical axis represents the number of balls caught. After Swift ('08, p. 1 74)
.
the expert stage is approached, almost horizontal. In the later stages of

learning the sole gain must come from an occasional adaptation and
from a further perfection of the present habits and methods of work."
(Book, '08, pp. 99 f.)
Swift and Batson have each published curves based on the in-
crease in skill in ball-tossing which purport to be of the concave
type. A
careful examination of the original data shows that the
apparently concave form is due in each case to peculiari-
in reality
ties in the method of plotting. Fortunately some of the data pub*-
THE RATE AND PROGRESS OF LEARNING 147
lished by Batson permit of being plotted strictly according to the
principle laid down at the beginning of this chapter, namely, that

vertical distances represent amounts of performance and horizontal
BOOr
460
400
5300
120
100
2200
60
1 40
20
100 1
11 21 81 41 51 61 81
Successive Trials
4 8 12 16 20 24
Days Practice
FIG. 43. Lower graph shows Batson's original curve. Upper graph shows
the reconstruction of his curve as stated in the text.
distances represent equal amounts of time or practice. This method

of plotting yields the perfectly typical convex learning curve shown
in Figure 43.
The types of learning so far investigated have been for the
most
part of a relatively simple sort. Other types of learning may be ex-
pected to bring to light curves of very different shape. This is par-

ticularly true of forms of learning which depend chiefly upon an-
alysisand selection or in cases where there is no physiological or
absolute limit within ordinary attainable bounds, such as, for ex-
ample, learning facts of history.
Two rather extensive studies on analytical types of learning are
now available. They reveal a very characteristic type of curve.
The first investigation, by Ruger ('10), was based upon the number
of successive solutions of a given mechanical puzzle that could be
0123456789
Time
10 11
Spent in Practice
12
FIG. 44. Curve to show the progress in solving puzzles. After Ruger (*io).
as reconstructed by Thorndike ('14), III, p. 342.
performed by an individual within a certain period of time, Fig-

ure 44. It yields a strikingly concave curve in marked contrast to
those we have previously examined. Unfortunately, practice was
not continued long enough to reveal the complete curve of this type
of learning. A second study, by Hull, was based upon the rate of
evolution of abstract ideas as shown by their increase in ability
to function, Figure 45. The material used was an elaborate
system of Chinese characters combined with nonsense syllables.
In this case the work was carried to the point of perfection. He
found as a consequence not only the initial concave section shown by
Ruger but a later period of diminishing returns. Taken alone, this
last section strikingly resembles the learning curves of the simpler
processes and its course is doubtless determined by the same causes.

The initial probably due to the
plateau or period of slow progress is
necessity of making a preliminary analysis of the material used be-

fore proceeding with the remainder of the process. Clearly the
elements common to many situations, as in Hull's investigation,
for example, must be perceived as separate elements before they
can be perceived as common elements.
It seems also quite likely that in learning facts of history 01 facts
of science, in which there is no physiological limit and in which
12 24 38 48 60 72 84 9G 108 120 132 144

Minutes of Work
Fio. 45. Progress of two individuals in generalizing abstractions or forming
concepts. After Hull ('i;).
the number of items that may be learned is practically unlimited,

the course of the curve, at least for a considerable distance, is con-
cave. This is hinted at in the results obtained with the author's
tests in geography and history. Thus in the former test, given at
the end of the school year to some 1,300 pupils, and in the latter
test given to some 2,000 pupils, the average scores for the ends of the
respective years were as follows:
Grade
Geography
5678
25 46 72
American History 7 20 38
These scores substantially mean that so many geographical or

historical items were known to the pupil. Both sets of figures show
a larger gain from the second year to the third year than from the
first to the second. These data furnish of course only fragmentary
portions of the curves of learning in these subjects. But so far as

any form may be inferred from them, it is probably of the con-
cave sort.
It seems probable that future experimentation will yield similar
forms of curves in other types of learning. Apparently there are
types of learning in which continued training brings increasing
returns. Thorndike states that:
"Negative acceleration (that is rapid rise or convex form) of any great

amount is far from being a general rule of learning. On the contrary,
it may well be that there are some functions, such as amount of knowledge
of history or geography, or of foreign languages, or of fiscal statistics,

*
where, by any justifiable score for amount of knowledge* the rate of
improvement in hour after hour of practice would rise, giving a pro-
nounced positive acceleration. Each item of information may, in such
cases, make the acquisition of other items easier; learning some one fact
may involve knowledge of a score of new facts in the shape of its relations
to the facts previously learned. So knowledge may roll up like a snow-
ball, its sum being, say, as the cube of the amount of time spent. What
'
we may call the knowledge functions' do, as a rule, show, to say the
very much less of the diminishing returns from increasing practice

least,
than do the functions of skill in some single line of work which figure so
often in the experimental studies of practice." ('14, II, p. 257.)
Whether or not plateaus occur universally in all types of learning,

and whether they are really unavoidable stages in the course of
learning, is an open question. They have not been found to occur
as generally as the initial rise even in curves of skill. Bryan and
Harter found periods of slow progress in three-fourths of their
subjects. Book found them in two of his three persons, and Swift
reports none. In the twenty curves obtained in the author's sub-
stitution test, eleven contained plateaus and nine did not. The
practice periods totaled 120 minutes. It is possible that some of the

curves might have shown plateaus if the practice had been con-
tinued longer.
Batson, who undertook an investigation for the purpose of study-
ing plateaus, found none in the ball-tossing curves, although the
training was continued for a long time, but found a pronounced
plateau in learning to throw shot into a pocket.
Plateaus may be caused by lagging in energy, by loss of atten-
tion, interest, and effort, fatigue, by periods of mechanization,
by
and the like. Rapid progress after a plateau may be due to a re-
cuperation in physical energy, in attention, interest, and effort,

to the acquisition of new methods of learning and doing the task
concerned, and to better use of the bonds which have been made au-
tomatic by the preceding practice. Bryan and Harter believe that
the plateaus in the learning of telegraphy were due to the establish-
ment of a hierarchy of habits. During the initial period of progress,
the simple elements such as the signals for letters, were acquired
firstand when these all had been learned, there came a dead level
during which the connections became automatized, and then rapid
progress was again possible by virtue of the acquisition of combi-
nations of letters into words and words into phrases. Their own
statement follows:
"A
hierarchy of habits may be described in this way: (i) There is a
certain number of habits which are elementary constituents of all other
habits within the hierarchy. (2) There are habits of a higher order which,
embracing the lower as elements, are themselves in turn elements of
higher habits, and so on. (3) A

habit of any order, when thoroughly ac-
quired, has physiological and, if conscious, psychological unity. The
habits of lower order which arc its elements tend to lose themselves in
it, and it tends to lose itself in the habits of higher order when it appears
as an element therein.
"
2. The Order of Learning the Habits of the Telegraphic Language.
"
The synchronous curves of Figure 38 and the experience of operators
agree in showing that from an early period letter, word, and higher habits
make gains (a) simultaneously, but (b) not equally.
"
(a) The simultaneity in these gains is shown in Figure 38 by the fact
that from the point where the curves diverge, each continues to rise.
This is perhaps to be explained by the fact that from an early stage the
learner practises with sentences, taking them as slowly as necessary. In

this way there is incidental practice of every language unit and of every
language unit in its proper setting.

"
(b) The curves of Figure 38 show also, however, that for many
months the chief gain is in the letter and word habits, that the rate of re-
ceiving sentences, is in this period, mainly determined by the rate of re-
ceiving letters and words, and that rapid gain in the higher language
habits does not begin until letter and word habits are well fixed. This ob-
jective result is supported by the introspective evidence of operators. In
the first clays one is forced to attend to letters. In the first months one is
forced to attend to words. If the learner essays a freedom for which he is
unfit, suddenly a letter or word which is unfamiliar explodes in his ears
and leaves him wrecked. He has no useful freedom for higher language
units which he has not earned by making the lower ones automatic. The
rank and file of operators are slaves to the machinery of the telegraphic
language. They must copy close. They cannot attend much to the sense
of the message as it comes, but must get its form, and re-read for the
sense. Only when all the necessary habits, high and low, have become
automatic, docs one rise into the freedom and speed of the expert.
"
3. The Plateaus.
"
We are now prepared to offer an explanation for the salient peculi-
arity of the receiving curve its plateaus.
"
A
plateau in the curve means that the lower-order habits are ap-
proaching their maximum development, but arc not yet sufficiently
automatic to leave the attention free to attack the higher-order habits.
The length of the plateau is a measure of the difficulty of making the
lower-order habits sufficiently automatic."
The explanation of plateaus probably depends upon the nature

of the. learning process in which they occur. The theory of the
hierarchy of habits would probably not apply to such a task as
mirror tracing.
Experimenters are divided in their opinions concerning the in-
evitableness or the usefulness of plateaus even in those types of
learning in which they frequently occur. Bryan and Harter, Swift
and others believe that they serve a beneficial purpose. Swift,
for example, says:
"The real advance in the early stages of learning is made during the
periods of seeming arrest of progress. The manifest advance, that
which is revealed by the curve or by examination marks, which is the
same thing, is discouragingly brief. By far the greater part of the learning
period is spent on plateaus when both teacher and pupil, failing to under-
stand the situation, feel that they are marking time. Yet it is during
these days of retardation that the valuable and solid acquisitions arc
being made. Americans who spend several years in Germany pass
through a long period of discouragement. Though they study the lan-
guage faithfully, and avail themselves of every opportunity to practice
conversation, they seem to make absolutely no progress. The length of
this plateau-period varies with different persons, but all experience its
oppressiveness. Now the most curious feature of this plateau, aside from
itsoverpowering monotony, is the suddenness with which it finally dis-
appears. Several have told the writer that they went to sleep one night
unable to understand anything, as it seemed to them, and utterly dis-
couraged, and awoke the following morning to find that they had mas-
tered the language, that they could understand practically everything
that was said to them. The word associations and national peculiarities
of thought sequence had been automatized during the long period when
no visible progress was being made." ('06, pp. 310 f.)
Other investigators believe that plateaus are not necessary

stages in the course of learning, but that they are due to causes
which may be avoided by introducing new stimuli or new methods
of attack in learning so that continued progress may be possible.
Plateaus are apparently not universal in all types of learning,
nor are they found in all persons in the same type of learning.
Whether they are useful stages in the learning process is a moot
question. Ifthey are not necessary, it would be highly important
for education to prevent their occurrence in the learning of school
material (i) by removing the conditions which bring them about,
and (2) by providing stimuli at the points at which they are apt
to occur so as to continue upward in the course of learning. Further
experimentation will have to be made to furnish a definite solution
of the problem. _,
Factors Affecting Progress, a. Length and Distribution of the
Periods of Work. How long at a time,"and how often, should the
learner work at his task in order to make the maximum progress
for the time devoted to it? of learning probably has
Every type
an optimum length and frequency of periods of practice. Ebbing-
haus ('85), in his pioneer study of memory, found that it was better
in learning nonsense syllables to distribute a given amount of
time over three days than to spend it all on one day. Sixty-eight
repetitions made in immediate succession were not as advantageous
for later relearning as thirty-eight repetitions distributed over
three days. Practically all investigators who have touched upon
this phase of learning have found a principle of similar nature.
Jost ('97), also working with syllables, found, for example, that
two repetitions a day for twelve days were better than four repeti-
tions a day for three days. Some of the results of both Ebbinghaus
and Jost imply that in some instances a decreasing amount of
time on successive days would be more economical than an equal
amount on all days; that instead of distributing 24 repetitions
by having four on six successive days, it would be better to have
eight on the first day, six on the second, four on the third, three
on the fourth, two on the fifth, and one on the sixth day.
Lueba and Hyde ('05), in an experiment on learning to transcribe
English words into German script, found that of four plans of
distributing time, twenty minutes twice a day yielded the slowest
gain, while twenty minutes every third day yielded better, and
twenty minutes every day or every other day yielded the best
results.
Miss Munn ('09) made an investigation of practice in a sub-

stitution test consisting of transcribing 4,000 letters into other
letters according to a key. Her distributions of time and results
are given in the following table :
TABLE 38
Practice in substituting letters for other letters according to a key. After
Munn ('09)
Average Time (Seconds)

First 200
No. of Letters Last 200
Subjects Distribution of Work Approximate Letters
23 200 letters: a day for 20 successive clays 41 5 13 4

4 800 letters a day for 5 successive days 57.5 17.1
(400 in m., 400 in p. m.)
a.
4 looo letters a day for 4 successive clays 47.00 16 5

4 2000 letters a day, seven days apart 39 5 182
4000 letters in one day (1000 at a sitting) 38 5 18 5
4 3000 letters a day (at one sitting) 44 oo
. 21.1
The highest degree of efficiency was reached by the 20-day

group who reduced their time for the last 200 letters to 13.4 seconds.
A definite comparison is a little difficult to make owing to the
large differences in initial ability among the various groups.
In the substitution test carried out by the author, Figure 46,
ten minutes twice a day was productive of the greatest progress,
twenty minutes once a day was productive of almost as rapid
progress, forty minutes once a day was productive of considerably
less progress, while 120 minutes at one time produced scarcely
half as much progress as the ten-minute or twenty-minute periods.
The was the same.
total time in all four distributions
Dearborn ('10), who reported an earlier experiment with the
same substitution test, divided the subjects into two groups work-
ing ten minutes once a day and ten minutes twice a day respectively.
He found a small advantage in favor of the former group.
Pyle ('13), working with a substitution test, reports that:
"
Generally speaking, daily practice seems to give better returns than
the same number of periods distributed on alternate days or in twicc-a-
day periods. However, there is some evidence that in the early stages of
habituation, the second practice on the same day gives good returns and
that, later on. alternate âys may be the best distribution."
Kirby ('13) carried out a practice experiment in addition and

division with 1,300 pupils in the third and fourth grades. The
pupils practiced addition in 22.5, 15, 6, and 2-minute periods, and
division in 20, 10, and 2-minute periods with the following gains:
1234 56789 1011 12 1314 15161718 19202122 2324

Successive Five Minute Periods
FIG. 46. rractice in writing numbers for letters according to a key. After
Starch ('12).
TO min. curve = group working 10 min. twice a day.
" " = " " " " tl
20 20 once .
" " =* " " "

40 40 every other day.
" " = " " "
120 120 at one time.
TABLE 39
Addition Division
Per Cent Gain over the Per Cent Gain over the
Period 22.5-Minule Period Period 2o-Minute Period
22.5 20
IS- 21 % IO 10-5%
6. I % 2 77 %
2. 46.5%
The superiority of the 2-minute period is probably exaggerated,

as Thorndike has suggested, by the greater opportunity for out-
side practice and longer continuation of regular school work, since
was extended over a larger number of days.
this period
Thorndike ('n) compared the improvement in multiplying
mentally with three-place numbers continuously, for two to twelve

hours with Miss Whitley's ( n) subjects who did three similar
J
problems a day for twenty days. The outcome was in favor of

the distributed practice, but probably only slightly so when allow-
ance is made for the effect of fatigue in the continuous work.
In general, relatively short periods of work in simple associative
learning are probably the most economical. It would be unwise
in the absence of more extensive experimental studies, to generalize
to types of learning and particularly to the learning of school
all
subjects. What the most productive periods for learning reading

or spelling or Latin or English composition are, will have to be
determined experimentally in each case. All that we can say at
present is that each type of learning probably has its optimum
length and distribution of practice periods. Lyon has stated the

matter in the following manner as a result of his studies on this
problem:
"With reference to the problem of the most favorable distribution of

single reading. ... I would say that the most general statement that
can be made, taking materials and methods of presentation into con-
all
sideration, is that the most economical method is to distribute the read-

ings over a rather lengthy period, the intervals between the readings
being in arithmetical proportion. For example, with one individual, in
memorizing a poem of twenty stanzas, the highest retentiveness was
obtained by distributing the readings as follows: two hours, eight hours,
one day, two days, four days, eight days, sixteen days, thirty-two days,
etc. The practical bearing of the results obtained on education in general
is that when associations have once been formed, they should be recalled
before an interval so long has elapsed that the original associations have
lost their color and cannot be recalled in the same shape, time, and order.
In general it was found that the most economical method for keeping
material once memorized from disappearing was to review the material
whenever it started to fade. Here also the intervals were found to be,
roughly speaking, in arithmetical proportion. For similar reasons the
student is advised to review his lecture notes shortly after taking them,
and, if possible, to review them again the evening of the same day. Then
the lapse of a week or two does not make nearly so much difference.
When once he has forgotten so much that the various associations orig-
inally made have vanished, a considerable portion of the material is
irretrievably lost." ('13, p. 161.)
b. Forgetting. Learning is, in a certain sense, a fortification

against forgetting, and from the practical side, it is important to
THE RATE AND PROGRESS OF LEARNING '57
know how frequently and in what manner the fortifications should

be strengthened in order to resist the attacks of forgetting. Only
a few experimental studies have been made on the rate and factors
of forgetting. Ebbinghaus ('85) learned nonsense syllables until
he could give them once correctly, and then measured the rate of
forgetting by the amount of time required for relearning them at
different intervals after the original memorizing. Radossawlje-
witch ('07) used nonsense syllables and poetry, Bean ('12) used
i
series of letters, and Magneff used poetry. The curves of for-
5 10 15 20 25 80 85 40
Days
FIG. 47. Curves of forgetting.
getting obtained by these investigators are presented in Figure 47.

They agree in showing a very rapid rate of loss at first, followed
by a very gradual decline afterwards. In retaining syllables,
Ebbinghaus found that he forgot as much in the first twenty min-
utes as in the following thirty days; in remembering a poem which
had been learned to the point of two perfect repetitions, Rados-
sawljewitch found that his subjects forgot as much in the first
two days as in the next twenty-five days.

Inquiry into the rate of deterioration of connections through
made by Book ('08), Rejall, 2 Swift
lack of practice have also been
1
As reported by Radossawljewitch.
*
Reported by Thorndike, II, p. 309.
and Schuyler ('07) in the case of typewriting, and by Swift in the

case of tossing balls. These seem to indicate much greater per-
manency in sensori-motor connections than in the memory of syl-
lables or poetry.
The experimental work on forgetting is too limited as yet to
permit of much definite application in the practical procedure of
learning school material. The one suggestion that may possibly
be made would be this: Since the rate of forgetting is very rapid
at first and more gradual later on, it probably would be highly
âdvantageous to have relearning of a given material come very
frequently at first and more rarely later on. Thus the facts of a
lesson in history or the newly acquired words of a spelling lesson
should be reviewed the next day or perhaps on the
preferably
same day, then again two or three days later and then a week or
ten days later, and so on.
The effect of long vacations upon the retention of school material
has been investigated only partially. Measurements of skill in
arithmetical operations in June and September show heavy losses
(see page 403) and raise the questionv as to whether long vacations
are really profitable or_ detrimental.
^cl Concentration, Effort, and Zeal. "Practice makes perfect,"
isonly a half truth. Only practice with zeal and effort is likely tc
bring improvement. A great deal of practice and repetition may
continue day in and day out without the slightest gain. While
the factor of zealous attention and interest has long been recog-
nized as a matter of common -sense observation, its real value,
however, has never been appreciated until experimental studies
pointed out its actual significance. Bryan and Harter have called
attention to this point in a very emphatic manner as follows:
"Afact which seems to be highly significant is that years of daily prac-

tice in receiving at ordinary rates will not bring a man to his own max-
imum ability to receive. The proof of this fact is that men whose re-
ceiving curve has been upon a level for years, frequently rise to a far
higher rate when forced to do so in order to secure and hold a position
requiring the higher skill. That daily practice in receiving will not
assure improvement is further seen in the fact that in many cases in-
ferior operators after being tolerated for years are finally dropped be-
cause they do not get far enough above the dead line. One conclusion
seems to stand out from all these facts more clearly than anything else,
namely, that in learning to interpret the telegraphic language, it is in-
tense effort which educates. This seems to be true throughout the whole
length of the curve. Every step in advance seems to cost as much as the
former. Indeed, each new step seems to cost more than the former. In-
quiry at the telegraph school and among operators indicates that between
sixty and seventy-five per cent of those who begin the study of telegraphy
become discouraged upon the plateau of the curve just below the main-
line rate. As a rule, ordinary operators will not make the painful effort
necessary to become experts. Facts of an analogous character will be

recalled from other fields.
'
"The physiological, psychological and pedagogical implications of this

conclusion are manifestly important. If in our educational methods in
the past, we have often made the pace that kills, there is possibly the
danger on the other hand that we shall make school work all play, and so
eliminate the intense effort which is necessary for progress." ('97, p. 50.)
A great deal of learning is done without any real zeal or effort

toward improvement. The usual way in which a great many
children learn to play the piano illustrates how much practicing
and learning consists in dawdling with more attention upon the
clock than upon the music sheet. A great deal of learning of school
material is done with the same lack of interest and effort.
^. Specific Practice in the Functions to be Improved. One of the
striking discoveries of experimental investigations is the very
rapid progress in specific functions when the practicing is done on
the particular connections to be established. A surprisingly large
percentage of pupils make little or no progress in an entire year's
work in such subjects as reading, writing, and the like, while the
remaining pupils make only a part of the progress that they could
make if their efforts were squarely directed at the material to be
learned or at the associations to be established.

The numerous practice experiments that have been conducted"
in psychological laboratories during the last two decades furnish
overwhelming evidence of the tremendous improvement ob-

tained under experimental conditions. Only a few examples will
be cited.
The writer found that eight persons, practicing mental multi-
plication of three-place numbers by one place-numbers for about
15 minutes a day for 14 successive days, made euormously large
gains as shown in the following table:
i6o EDUCATIONAL PSYCHOLOGY
TABLE 40
Improvement in mental multiplication. After Starch (*ii)
These subjects gained in approximately four hours of practice

per person over 100%, varying, of course, from one person who
made little gain up to two persons who gained nearly 200%.
Wells found the amounts of gain from 150 minutes of practice
in addition on the part of ten adults as follows:
TABLE 41
Improvement in addition: adults. After Wells ('12)
The gains show approximately a doubling in efficiency in the

course of thirty days.
Dearborn used vocabularies and poetry in learning experiments
and reported the following results:
TABLE 42
Improvement in ability to memorize. After Dearborn (*io)
TOTAL PRAC- AMOUNT NUMBER OF TOTAL TIME RE- TIME RE-
JECT TiCETiMB LEARNED I *YS OF AMOUNT QWRRD ON QUIRED ON
IN HOURS DAILY PRACTICE LEARNKD FIRST DAY MOSTFFI-
Lcarning the English meanings of French or German words:

i 6 1/3 50 21 1050 30 13
2 6 35 20 700 30 12
3 6 35 18 630 30 14
4 8 i/io 30 22 660 33 15
5 7 2 /3 3 20 6 40 15
learning poetry:
7 3i/3 32 15 480 38 7
8 32/3 18 16 288 30 8
1
10 4 17 13 221 30 12
Similar results have been reported by Bair in tossing shot, by

Swift and Batson in tossing balls, by Whitley in drawing lines in a
maze, by Wells in tapping, by Kline, Wells, and Whitley in can-
cellation tests, by Thorndike, Wells, and Kirby in adding, by
Swift, Book, Rejall in typewriting, by Ebert and Meumann,
Winch, Sleight, Dearborn, and Fracker in memory in fact in all
experimental work in which practice enters. Improvement of
mental functions through practice is well-nigh universal and the
amount of improvement through specific training under experi-
mental conditions is almost incredible, particularly when we con-
trast with it the gains made in school functions in from 50 to 150
hours devoted to a subject in the course of a year.
The average gain made by pupils in school activities in the
course of a year's practice as indicated by the standard scores
derived from measurements with tests and scales is shown in the
following percentages of gain at the end of the eighth grade as
compared with the end of the seventh grade:
1
Approximate.
l62 EDUCATIONAL PSYCHOLOGY
TABLE 43
Based upon published scores for the various tests
These gains are surprisingly trivial when compared with the

gains, often running over 100%, reported in connection with
experimental investigations of practice.
Definite experimental results arc not at hand to substantiate
the following assertion, which may seem doubtful but which is
not impossible from the present indication of other measurements,
such as those presented by Dearborn in Table 42 or by the writer
in Table 9 (Experiments), namely, that the average high school
pupil could learn in 20 minutes a day for thirty days, all the Latin
words (500 words) that he would need in an entire year of Latin.
He could learn in 30 minutes a day for one-half the school year,
all the Latin words (approximately 2, coo) that he would use in his
entire study of four years of high school Latin.
The difficulty with the material of school subjects is that we do
not, and in some instances we cannot, specify with sufficient def-
initeness just wherein the improvement is to be made. can We
point out specifically to a child whether or not he spells a word
correctly and what part of the word may be incorrect, but we have
not until recent years made any attempt at determining which
particular words a child really ought to know how to spell. The
pupil was given a list of words selected more or less on the basis
of their unusualness and difficulty rather than upon the basis of
usefulness or frequency of occurrence. The idea seemed to be that
if he learns to spell a sufficient number of difficult and useless words,
he will know how to spell all otherwords in the English language.

The school has virtually said, "Learn to spell," but has not said
what a Even in a subject in which the

child should learn to spell.
associative bonds be precisely defined so that they can be
may
directly attended to, we have not done so. The same situation
obtains in practically all other subjects with the added difficulty
that in some subjects the material is of such a nature that specific
directions and specific material or specific bonds to be formed,
cannot easily be isolated. This is particularly true of such a sub-
ject as English composition which represents the opposite extreme
from spelling and arithmetic. The child is told to improve his
style,or his language, or his expression, or his originality, or his
imagination; but he is not told very definitely how he may do this,
or just what he is to do. The school should, therefore, aim to
specify exactly what sort of learning is to be done so that a definite I
notion on the part of the learner may be formed of the precise bonds
and connections to be made.
.
e.JDejlnite Knowledge of Success and Error. Much experimental
work implies that the feeling of satisfaction resulting from success-
ful trials of a task facilitates the formation of the connections con-
cerned. It seems obvious therefore as a practical matter that pre-
cise knowledge of the success or failure on the part of the learner
is
exceedingly important. It will not only serve to arouse the
feeling of satisfaction but also help to define the particular bonds
to be established. Feelings of dislike on the part of the learner
toward the material to be learned undoubtedly interfere with the
rapid formation of the connections, and frequently the feeling
of dislike isaccompanied by an attitude of unwillingness or stub-
bornness indicated by such statements as "I know I can't learn
languages; I never could." "I never was able to get mathematics."
"I can't memorize anything." A concrete case that came under
the writer's observation was that of a man considerably older than
the average university student, who in the experiment on the trans-
ference of training (Chapter XI, Experiments in Educational
Psychology) reported that he was unable to learn vocabulary and

that the net result of half an hour's work on the first list was ten
words. The average student is able to learn the entire list of thirty
words in from twelve to fifteen minutes. He further stated that
he had always had great difficulty in learning languages. In order
to ascertain, if possible, the real status of his memory and other
abilities, he was tested by Terman's revision and some additional
tests, all of which indicated that he was of average intelligence and
that his memory was not defective, but approximately average.
He was informed of the results of the tests, that his defective

memory was largely illusory, and that probably his real trouble
lay in his contrary attitude toward certain tasks, which was also
indicated by his own statements concerning his work. The general
effect upon his later attitude in learning was a distinctly whole-
some one. This case is cited because it exemplifies many similar
instances of persons who fed incapable of learning certain things.

/. Interest in Improvement. One important element in the re-
markably large amount of gain through practice in specific func-
tions, isthe fact that the progress is directly observable and

definitely measureable which in turn produces a real zeal toward
improvement and toward outstripping the preceding records.
In an experiment such as the substitution test or the practice ex-
periments in arithmetical operations, the observation of a definite

gain is possible so that the learner can see just how much progress
he is making. The practical value of this suggestion would be the
creation of circumstances for the learning of school subjects similar
to the conditions of learning in laboratory experiments by intro-
ducing forms of measurement through which the progress may be

determined at frequent intervals, so that the pupil may see what
progress he is actually making.
In a certain elementary school a series of standard tests 1 was

applied every month throughout the entire school year. Tests in
reading, writing, spelling, and arithmetic were given at monthly
intervals to determine the progress made. Each pupil knew his
own record in each test and compared it from month to month.
This condition developed a remarkable interest and zeal in striving
to surpass the record of the preceding month. The condition
created thereby was similar to that of a learning experiment under
laboratory conditions. Each pupil kept his own score and knew
what gain he was making each month. This condition had a re-
markable effect upon the total progress made during the school
year as shown in the accompanying graphs. These results show
that the pupils made on the average a gain in some studies
twice as great as that made ordinarily in the course of a school
year. This gain cannot be attributed to familiarity with the test
material since, in the case of reading, different passages were used
in each successive test; in writing, a different sentence was used
1
The tests were made in The Alice School, Hibbtng, Minnesota, by Principal L. J.
Coubal, and reported in an unpublished thesis in the library of the University of Wis-
consin. The tests were carried out under the direction of Professor V. A. C. Henmon.
THE RATE AND PROGRESS OF LEARNING I6S
each time; in spelling, the author's six lists were used in rotation,
one at a time; and in arithmetic the three sets in the Courtis Series
A were used and rotated so that there was a recurrence of the same
40
ao
10

Fin. 48. Progress in speed and comprehension of reading combined into
single scores as measured by monthly tests (Starch reading tests) upon 4th grade
class. The straight,
pupils. The continuous curve represents the progress of the
broken line is the progress for schools generally based upon the standard scores
for June of the 3rd grade and June of the 4th grade.
test every three months, but very unlikely that this contributed
it is
any appreciable share toward the gain shown. It would seem,

therefore, highly desirable if there could be introduced into the
schoolroom a similar atmosphere of motivation such as obtains

Other
FIG. 49. Progress in speed and quality of writing (Thorndike scale).
facts same as for Fig. 48.
in learning experiments in the laboratory. The knowledge of one's

actual ability and of the actual amount of gain serves as an ex-
ceedingly powerful spur for the learner to surpass his own previous
succeeds like sue-
performances. The popular dictum "Nothing
i66 EDUCATIONAL PSYCHOLOGY
cess" may be partly justified by such experimental results as the

ones here cited. To see oneself gaining tends to stimulate greater
efforts toward gain. The educational measuring scales and tests
FIG. 50. Progress in spelling (Starch test) of 3id grade class. Other facts
the same as for Fig. 48.
ought to serve a useful purpose at this juncture. They will pro-

vide means whereby the pupil may be able to see for himself in
definite terms the gains he is making.
.10
Addition
Subtraction
a
Si
Multiplication
Division

FIG. 51. Progress in the fundamental operations in arithmetic (Courtis
testsL Other data same as for Fig. 48.
g. Mental Imagery. After the early studies on mental imagery

became known, there followed considerable theorizing as to types
of persons and types of learners, and with it came the resulting
endeavors to make applications of these theories to methods of

teaching. Thus it was said that if a pupil has difficulty in learning
to spell or in learning a foreign language he may be devoid of, or
weak in visual imagery; or if he has trouble in learning to write,
he may be short on motor imagery; or if he finds it hard to learn
the pronunciation of words, he may be defective in auditory and
motor imagery; and he fails in the academic subjects he was sus-
if
pected of being devoid in visual and auditory images and strong

in motor imagery and should therefore turn to manual training.
The proposed remedy was that the material to be learned should
be presented to a different sense organ so that the pupil might use
the imagery natural to him. All these inferences may possibly
be true; but later additions to our knowledge of mental images
make us more hesitant regarding the real part played by them in
learning and concerning the actual differences produced by present-
ing material to different sense avenues.
Before we can make changes
in practice we must be sure of the
principles upon which the practice is to be based. It is important,

therefore, to examine at least the following three considerations:
In the first place, more careful studies of the sorts of images
employed by different individuals show that the classification of
persons into visuals, audiles, motiles, and so on, is fundamentally
misleading. Studies by Betts ('09) and others have helped greatly
to clarify the matter by showing that pure types rarely exist.
During the last six years several hundred students have performed
the imagery tests outlined in Chapter VII of the author's Experi-
ments ('17). Among this entire number not more than two or three
persons were found whose images either were practically all of
one type, or who had one or another commonly prevalent class
almost entirely missing. The facts for 95% of all persons are sub-
stantially as set forth on page 45 (Experiments), namely, that nearly
all persons have all types of images which are combined in different
individuals in varying proportions. Mankind as a whole does not
fall into sharply or even vaguely divided groups of visuals, audiles,
and so on. They are not found except in rare instances.

IB the second place, recent inquiries indicate that images of the
class most natural or predominant for a given person may be
aroused by stimuli sense. For example,
coming through another
auditory stimuli arouse visual images as well as, or even more
may
readily than, auditory images if visual images are more natural to
the individual. Miss Abbott ('09) found in a detailed investiga-
tion with four subjects, on the memory consciousness in orthog-

raphy that "irrespective of the method of presentation and the
manner of learning, the typical mode of recall for all observers
was through the visual imagery of the letters" (p. 153). Conse-
quently it docs not follow that, even if a person has a strong lean-
ing toward one or another type of imagery, it is necessary or even
advantageous to present the material to be learned through the
sense avenue of the favorite type of imagery.
Colvin and Myers ('09) made an extensive series of tests with
school children to ascertain to what extent visuals retain visual
material best, audiles auditory material, and motiles motor ele-
ments. He concluded that:
"There seems to be a fairly definite relation between the effectiveness

of memory in the case of a particular ideational type and the memory
material which is most suited to that type. In particular the visual type
retains best material with a visual content and the auditory and motor
types, to a less degree, material with an auditory or motor content, as the
case may be."
While this conclusion is probably justified, the differences among

the various types of elements retained by the different groups of
pupils are in most instances small.
In the third place, the most advantageous mode of sensory
presentation for a given person is not necessarily the one that
corresponds to his dominant type of imagery. Much of the experi-
mental evidence that has been accumulated on the problem of
modes of presentation of the learning material is conflicting in
character. It is uncertain whether visual, auditory, motor, audi-
tory-motor, or visual-auditory-motor presentations are most

advantageous. Henmon ('12) has reported an experiment which
has thrown important light upon the problem. He employed four
methods of presentation, visual, auditory, visual-auditory, and
visual-auditory-motor; and used three sorts of material, concrete
nouns, two-place numbers, and nonsense syllables presented to
six subjects with one, two, or three repetitions.
"In the visual presentation the subjects read the stimuli directly from
the rotating drum and immediately wrote down as many members as
could be recalled and in the order presented. The subjects were asked
to repress movements of articulation. In the auditory presentation the
experimenter read the stimuli from the drum, the subject keeping his
eyes closed and repressing movements of articulation. In the visual-

auditory presentation the subject both saw the stimuli and heard them
read by the experimenter. In the visual-auditory-motor presentation
the subject himself read the lists aloud."
His results are summarized thus:

"
i. Auditory presentation is clearly superior to visual presentation in
immediate memory of adults, a result attributable to the greater ease and

freedom of visualization with auditory presentation and the greater
effort of attention required.
"
2. This superiority of auditory over visual presentation holds for all
materials (nouns, nonsense-syllables, numbers), for all subjects irrespec-

tive of image type, and for one, two and three presentations. Thit
result is not in accord with the opinion commonly held that visual pre-
sentation is superior, especially with meaningless material.
"3. Combined visual-auditory presentation is slightly inferior to the
auditory alone and decidedly superior to the visual alone. The advan-
tage of a combined method is very much less than that shown in earlier
investigations.
"4. Visual-auditory-motor presentation is slightly inferior to the
auditory and the visual-auditory presentations and superior to the visual
alone. Articulation or vocalization is of little value for immediate mem-
ory.
"
5. The correlations of abilities with different forms of presentation
are positive and very high, superiority with one indicating practically
the same degree of superiority with another." (After Henmon '12.)
A fair general impression of the present status of our knowledge

of imagery in relation to learning would be that distinctions among
types of pupils have been overemphasized and that much of the

endeavor to adjust methods of teaching accordingly has been mis-
directed.
h. Fatigue. As a final important factor in the progress of learn-
ing, we must consider fatigue. Educational and psychological
literature has been replete with discussions regarding the part
which fatigue plays in the reduction of mental efficiency. While
psychological research has provided information
considerable
concerning the course of continuous work and concerning the
changes in the efficiency of the worker as measured by cross-
sectional tests at various stages of work, it has not furnished as yet
much definite knowledge concerning the control of the work of the
pupil in school. Pedagogical literature has been generous in
pointing out signs of fatigue and serious consequences of overwork
and in suggesting remedies for avoiding exhaustion, yet we are

not sure whether the so-called symptoms arc indications of real
fatigue or whether any serious or even mild fatigue effects ever
result from the work as carried out in the great majority of schools.
In discussions of fatigue it is important to bear in mind two
distinctions in the meaning of the term, namely, fatigue in thei
sense of decrease in the capacity to do work, and fatigue in the
sense of decrease in interest in, or willingness to, work. The two
are plainly different and do not necessarily go together. The one
is actual loss in efficiency; the other is a feeling of ennui or weari-
ness. Much of our thinking about the problem has been confused
by a failure to distinguish between these two meanings. Fatigue

in the former sense probably has been greatly exaggerated as an
educational problem. Perhaps only in exceptional individuals is
there injurious overstrain due to mental work. The discussion of
this topic will, therefore, be abbreviated.
The experimental methods by which the phenomena of fatigue
have been investigated will first be mentioned briefly. They may
be divided into two classes: (i) Indirect methods, and (2) Direct
methods.
(i) Indirect methods. The principle, upon which the indirect
methods have proceeded, has been to measure some physiological
or psychological functions at different points during the course of
work in order to compare the efficiency of those functions on the
assumption that the difference in them would be indicative of
efficiency in general. One of the first methods was that employed
by Griesbach, who determined the two-point threshold upon various
parts of the skin at various times of the day on the belief that a
decrease in sensitiveness or a widening of the threshold indicated
a reduction of general mental efficiency. He made extensive com-
parisons among school children for the purpose of determining
the amounts of fatigue produced by various types of school work,
and formulated an elaborate series of conclusions with regard to
them. For example, specific fatigue values were assigned by him
and his followers to the different school subjects. Vannod states
that mathematics, Latin, and Greek produce most fatigue, and that
French and geography produce least. The difficulty, however,
with results of this type is that while the two-point discrimination
upon the skin varies under different mental and physical conditions,
it is a rather unsafe basis upon which to make sweeping generaliza-
tions concerning the general mental efficiency of a person. In

fact, the closeness of the agreement of the size of the two-point

threshold with the actual amount of fatigue is too uncertain to use
this function asa symptom of general mental or physical fatigue.
A number methods have been employed, such as
of other indirect
the rate of tapping with a stylus, the variation in blood pressure,
in pulse, in respiration, the range of visual accommodation, sensi-
tiveness to pain, and so on. The same criticism applies to these as
to the two-point discrimination. These functions may have con-
comitant variations within rough approximations, but they are too
distant to be precise indications of mental efficiency.
The use of the ergograph as developed by Mosso and his co-
workers has probably been the most successful and useful method
for studying problems of fatigue. As such it is, however, a direct
method for investigating muscular work and fatigue and only a very
indirect and doubtful method for investigating mental fatigue.
Other indirect methods of a more distinct psychological char-
acter have also been employed. These have consisted of the meas-
urement of certain mental functions at various intervals in order
to determine how much variation there may be in these functions
and to regard them as indications of mental efficiency in general.
Such tests have been made upon memory, various types of associa-
tion processes, perception as measured by cancellation tests, and the
like. These tests have a certain superiority over those mentioned
in the preceding paragraph since they deal at least with psychologi-
cal functions, but they likewise do notdirectly measure the course
of work as it actually occurs. They have, however, been useful
in comparing efficiency in the same mental capacities at various
^Joints during the course of a day.
A considerable number of researches by means of cross-sectional
test methods ha\4 been carried out upon school children as well
as adults. Thus, lor example, Sikorski ('19) tested pupils before
an^p after school in writing from dictation, and compared the
number of errors made. Bolton ('02) measured the memory span
for digits during the early and the later part of the school day.
Laser ('94) made a test with pupils in addition and multiplication
at hourly intervals. Friedrich ('97) tested 51 pupils in addition,
multiplication, and in dictation exercise at hourly periods. Eb-
binghaus with the aid of the teachers, gave tests at hourly
('97),
intervals inimmediate memory of numbers, in addition and multi-
plication, and in supplying words and syllables omitted from sen-
tences. Ritter (IQOO) used tests in dictations of words, numbers,
and sentences, and tests in cancelling letters and words. Thorndike

(1900, 'n and '12) used, early and late in the school day, tests
in adding, multiplying, cancelling certain words in a printed text,
and memorizing numbers, letters, and geometrical forms. Heck
('13) measured the performances of pupils in adding, subtracting,
multiplying and dividing at four points during the school day.
Miss King i used tests at five points during the school day in adding
and multiplying and in answering questions of a general informa-
tional character.
Practically all of the investigations here mentioned that were

carried out reliably, agree, when interpreted fairly, in showing that
efficiency in the various functions examined is changed very slightly
or unappreciably during the course of a school day. Not all of
the investigators, however, interpret their results in this manner.
Thorndike has pointed out a very important misconception in the
interpretation put by some of the experimenters upon their data,
namely, that of counting simply the number of errors made at
different times of the day instead of expressing efficiency in terms
of both amount and accuracy of work done. This point may be
illustrated in the case of Friedrich's results presented in Table 44.
TABLE 44. After Friedrich
The results obtained by Friedrich concerning the accuracy of school work at

different periods of the day
1
An unpublished study reported by Thorndike. ('14, III, p. 93).
If in this table we compare simply the percentage of errors the

efficiency of the pupils was over five times as great at the beginning
of the school day as at the end in the dictation test, and over two
times as great in the computation tests. If, on the other hand, we
consider the column giving percentage right we find that the ef-
ficiency changed but very little.
To point out further how inconsiderably the performance of

pupils changes in the course of a day we may note the following re-
sults from Heck:
TABLE 45. After Heck ('13)
The amount of work increased slightly while the accuracy de-

creased slightly from the first test to the last.
(2) Direct methods. The most fruitful direct methods of measur-
ing continuous mental work have been the various types of mental
calculations, addition and multiplication.
particularly These
methods have been used by Krapelin, Thorndike and Arai, Starch
and Ash, and others. As an illustration of one type of mental
addition, the writer has used a method consisting in the mental
addition of 6, 7, 8, and 9 in rotation by beginning with a given
number and adding each of these numbers in turn to the answer
last obtained, as described in Chapter XVI, Experiments ('17).
The advantage form of calculation is that it affords suf-
of this
ficient difficultyand thus fully taxes the efforts of the individual
and makes possible a minute record of the amount and accuracy of
work done during succeeding short intervals of time. Figure 52
shows a curve obtained by this method, covering a period of con-
tinuous work of two hours.
As an illustration of mental multiplication, we may cite the
experiment carried out by Miss Arai under the direction of Thorn-
dike. She used the method of multiplying mentally four place
numbers by four place numbers, as 4,962 times 7,584. She trained
herself for a considerable period of time in this type of mental mul-
tiplication in order to reach

an approximate limit of practice. Then
she did the following experiment:
"On March 3, 4, 5, and 6, that subject did the mental multiplication

from ii A. M. to ii P. M. without any pauses except the two or three
seconds between the examples for recording time. But the subject had
taken a heavier breakfast than usual at 10 A. M. and a light supper
after 1 1 P. M. Her health was in good condition and she slept soundly at
night. The contents of her consciousness during the experiments were
very simple, all desires being completely subjected to the one desire to
get true fatigue curves." (Arai, '12, p. 37.)
The remarkable result of all experiments with purely mental

functions has been that mental efficiency is reduced only very
slightly even after two or more hours of very difficult, uninterrupted
* 1 IK 2^
Hours
FIG. 52. Mental work curve. Upper curve shows number of additions
made per half minute period. Lower curve shows number of errors made.
Work was continued for two hours. After Starch and Ash ('17).
work. Thus in the curve, Figure 52, the reduction in the number of
additions made per thirty seconds, was only from 14.0 down to
13.4, or a loss of only 4.3%. Arai found even in the course of 12
hours of such difficult mental multiplication as she carried out,
that her efficiency was reduced only by about one-half. Other
investigators have shown in general the same facts.
Seashore and Kent ('05) measured continuously, for as long as
two hours, the threshold of hearing by recording the audibility and
inaudibility of a sound varied about the limen. The intensity of
the sound was changed at a uniform rate. As soon as it became too
faint tobe heard the subject gave a signal to the experimenter who
at once increased the strength of the sound. As soon as it could be
heard again the subject again responded. Then the sound was de-
creased again, and so on without break. A sample curve is shown
in Figure 53. Ten records were obtained which showed that "con-
THE RATE AND PROGRESS OF LEARNING I7S
tinuous liminal or moderately faint sounds do not seem to lower the

efficiency of the ear in a two hour test" (p. 100).
It would seem, therefore, on the basis of experimental work, that
fatigue in the sense of decrease in product achieved is practically a
negligible element in school work. The actual capacity to do work
with the same degree of accuracy is practically undiminished in the
course of a school day. Such symptoms of fatigue as have been
frequently enumerated in pedagogical writings, are apparently only

superficial signs of monotony, of lack or diminishing of interest,
or of being bored by school work, and not actual signs of loss of
capacity to do the work. Such statements as "I simply cannot
work any longer" made after a half or whole hour's work, are il-
lusory and probably signify chiefly a weariness with the work
' ' ' ' ' '

'
20 80 40 60 60 70 80 90 100 liO
Minutes
FIG. 53. Continuous record of the measurement of the threshold of hearing.
After Seashore and Kent ('05).
which, if it must be kept up by force of conditions, can usually be

continued without difficulty or harm and usually without being
seriously boresome.
The feeling of interest or satisfaction in doing work does decrease
very materially as the work goes on. Thorndike (17) for example
found that the satisfyingness of such work as grading compositions
decreased in the course of two hours to about one-half and in the
course of four hours to about one-third of the amount of satisfac-
tion at the beginning of the period.
The feeling of weariness, from the practical side of school activ-
ities as well as of mental work generally, is, however, an important
item. In a certain sense it is a real thing. Even if it is illusory

it does interfere with the smooth continuation of work. But it is
very likely a than an actual loss of capacity

less serious situation
to do work would be. Practically it resolves itself into a problem
of maintaining interest rather than relieving depreciation of ef-
ficiency.
CHAPTER XII
HOW TO STUDY
Waste in Studying. Since studying is learning under school
conditions, it would seem worth while to make such suggestions
as can be made concretely to assist pupils in this important phase
of the psychology of learning. It may seem preposterous to give
advice about something concerning which each pupil is presumably
proficient after years of practice in it, and furthermore to attempt
to give suggestions on studying may seem to many to be nothing
more than an "unprofitable delineation of the obvious." It is,
however, very certain that there is an uncalculated waste of energy
and a still more prodigal waste of time in so-called studying. If
we may judge from the possibility of improvement in reading ca-
pacity alone, and from the larger accomplishments attained under
favorable conditions of work, we may venture to guess that the
average student could accomplish his work just as efficiently or
more efficiently, in two-thirds, or less, of the amount of time
ordinarily consumed, by developing more economical methods
and habits of studying. Improvements in proper procedure in
studying have shown how much more may be accomplished in the
same length of time or even in a shorter period of time. Vicious
habits of dawdling in school work are acquired, which may have
their permanent effect throughout the individual's life.
Is Studying Worth While? This question is worth raising in
view of the belief, prevalent among students, parents, and grad-
uates, that after all it does not matter much whether a pupil does
well in his studies or not, that the boy who does poorly in the
grammar grades or the high school will outgrow his negligence and
come into his own when he gets into his college or professional
course, or that when he gets into the real business of life he will
outstrip his more studious mates. To what extent are these beliefs
true or false? To what extent is early scholastic performance
indicative of similar or different performance later on? To what
extent is scholastic performance prophetic of performance in life?
A considerable amount of statistical material has been accumu-
lated in the attempt to answer these questions. Some of this
176
HOW TO STUDY 177
material was presented in the latter part of Chapter IV under the

heading "Correlations between Early and Later Mental Abilities."
These correlations were found to be high. Dearborn, for example,
found that of 47? pupils, whose records were traced through the
high school and college, only two who were in the lowest quarter in
the high school rose to the top quarter in the university. It should
be noted further that these two were just barely poor enough in
the high school to be classed into the lowest quarter and that
they rose just barely enough to get into the top quarter in the uni-
versity. The chances that the pupil who is doing poor work in the
high school will later come into his own are exceedingly small;
apparently he has been in his own all along, or, if not, he had better
have got into his own as soon as possible.
President Lowell ('10) made a study of the records of the grad-
uates of Harvard College for a period of twelve years. He found
the following situation:
Men graduating with various Percentage graduating with distinction from

honors The Law School The Medical School
A. B.'s with highest honors 60 92
A. B.'s with great honor 40 87
A. B.'s with honor 22 76
A. B.'s without honor 6> 36
A. B.'s without honor, of men who
had entered college with condi-
tions 3
The 250 Yale men who graduated from the Harvard Law School
in 1900-1915 were divided into nine groups according to their
scholarship at Yale. These nine groups, with the exception of one,
finished the Harvard Law School in the same relative order of
scholarship that they had held at Yale.
To many persons a more important problem is the relationship
between scholastic attainment and success in business or profes-
sional work. Foster ('16) has summarized in an interesting manner
much of the evidence pertaining to this problem. He made a
study of the Harvard College class of 1894. He asked three men,
the dean of the college, the secretary of the alumni association,
and a member of the class, to name the most successful men of the
class. They were free to use their own interpretation of success
except that they were not to include men whose success appeared
to be due chiefly to family wealth or position. The three judges
agreed on twenty-three men. Foster then obtained their records
in college and compared them with the records of twenty-three

other men
chosen at random from the same class. The former had
nearly four times as many highest grades as the latter, namely,
196 A's as compared with 56 A's. By a similar plan three judges
selected the most successful men among the graduates of the
University of Oregon for the period of 1878 to 1901. Of the grad-
uates designated as successful, 53% had been good students and
17% had been weak students. Of the graduates designated as
unsuccessful, 12% had been good students and 52% had been weak
students.
A study of the alumni of Wesleyan University showed that of
the living graduates for the period of 1860 to 1889, 50% of the men
who had graduated with honors were listed in Who's Who, and
only 10% of the men who had graduated without honors were in
Who's Who. Among the living graduates for the period of 1890
to 1899, 60% of the men graduated with highest honors were listed
in Who's Who, 30% of the men elected to Phi Beta Kappa were
listed in Who's Who, while only 11% of the graduates without
superior scholarship were found in Who's Who. (Nicholson '15).

E. G. Dexter investigated the records of the living graduates of
twenty- two colleges and found that 5.9% of the honor scholars
and only 2% of all graduates were listed in Who's Who. Further-
more, 56% of the Yale valedictorians were found in Who's Who.
Their chances were, therefore, more than twenty-five times as
great as those of other graduates. The records of 13,705 living
graduates of two New Englandcolleges revealed the fact that
5.4% of those who constituted the highest tenth were listed in
Who's Who while but 1.8% of those in the fourth tenth were there
listed. Who's Who is,of course, not an absolute criterion of suc-
cess; it is, however, a rough measure of success.

A tabulation of the Oxford University men who entered the law
or the ministry showed the following percentages of men who
attained distinction in their respective professions:
Men with varying honors Percentages attaining distinction

In the law In the ministry
Men with ist class honors 46% 68%
" " "
2nd 33 37
" " " "
3rd 22 32
" " " "
4th 20 29
" "
pass degrees 16 21
" "
no degrees 15 9
HOW TO STUDY 179
As a matter of correct interpretation of these extensive statistics,

itmust not be assumed that success or failure is solely attributable
to the amount of devotion to school studies. The uniform manner
with which the early scholastic records agree with the later records
of the same persons, or the pronounced tendency with which scho-
lastic attainment correlates with business or professional attain-
ment is probably due to a common cause, namely, original ability
or make-up of the individual. At their face value, these figures
mean that the person who does well in his school work also tends
rather strongly to be a person who will do well in his business or
professional work. However, this array of facts is impressive and
ought to be brought emphatically to the attention of high school
and college students. They ought to have a tonic effect upon their
efforts. While our native make-up determines to a large extent
what we shall become, yet rarely does any one utilize or develop
to the fullest extent even the limited measure of ability that he
possesses. The laggard can find little consolation in the hope of
somehow redeeming himself later on.
Types of Studying. For the sake of convenience, we may divide
studying into three types:
1. The Reading Type of Studying. In the elementary school
probably eight-tenths and in the high school and the university

probably two- thirds of all studying consists essentially in reading.
2. The Laboratory Type of Studying. This type obviously
consists of the manipulation of apparatus, the observation of

material, the recording of observations and experimental data, and
the interpretation of these data.
3. The Analytical or Reasoning Type of Studying. This type
is involved in those subjects in which the amount of reading is
relatively little, but in which the task consists in a thorough mas-
tery of a relatively small amount of text. Such studying is ob-
viously involved in mathematics and in a few other types of
difficult reading, as for example, certain branches of philosophy
and the speculative and theoretical aspects of the sciences.
Problems. Every type of studying is different and, in a sense,
every lesson has its own special material and presents its own
problems on how to study effectively. It may seem futile to at-
tempt to give general advice on how to study. Yet upon further
analysis, it appears that there are several elements common to

ailtypes of learning. These elements are (i) the control of atten-
tion in securing the most favorable attitude of work, which would
l8o EDUCATIONAL PSYCHOLOGY
be involved in all types of mental work, (2) common principles in

the assimilation and retention of the material, and (3) proficiency
in reading. Problems involved in all of these elements would be:
First, what are the specific processes common to ail types of study-
ing here referred to, and second, how may these various processes
be facilitated?
Control of Attention. One of the chief, if not the chief, source
of waste in studying and in fact in all mental work, is the reluctance
in beginning an intellectual task. There seems to be in many in-
dividuals an almost insuperable inertia to overcome before work
is,or can actually be, begun and continued without constant self-
pushing. The common feeling is a dislike to begin work. "I
don't like to study my history," or "I just hate to write this theme,"
"I don't see why he makes us do this," represent states of mind
frequently found among the average pupils and to some extent
even among the better pupils who often have severe struggles with
such a tendency.
In papers on "Difficulties and Hindrances in Studying and
How to Overcome Them" collected by the author from about
one hundred university juniors and seniors, 56 mentioned lack of
concentration, 26, dislike for or lack of interest in the subject,
23, getting started, 9, mind-wandering, 5, failure to organize mate-
rial, and 4, day dreaming. These may all be classed as internal
psychological difficulties centering around the problem of getting
the mind to work at the task. Practically every student mentioned
one or another or several of these four difficulties.
Besides attributing this situation to indolence or to stupidity,
is there anything in the way of concrete suggestions and help that
can be given to overcome this mountain of difficulty? I believe
there are two general procedures which may be followed. One is to
grit one's teeth and to it"; that is, simply to force oneself
to "go
by sheer voluntary effort to begin the task. The other is to put
oneself into physical surroundings and into a frame of mind in
which it will require a minimum, or at least a smaller amount, of
voluntary effort. voluntary effort consumes a large
Strictly
amount of mental energy and, if it must be continued for a long
time, is very wasteful of one's strength. The second is distinctly
the more advisable plan to adopt. With the help of such a control
of external conditions as is possible, the following means of di-
recting one's energy may, therefore, be suggested:

(i) Put yourself into the proper physical or bodily attitude of
HOW TO STUDY 181
work. Sit up to your desk or table at which you customarily work.

This in itself will help to start the mental machinery agoing and
make it easier for the mental processes to operate.
(2) Work
in surroundings in which there are absolutely no dis-
tractions as far as possible. Some persons can work under very
distracting conditions, but these are exceptions, and if one has
difficulty in beginning work, he should go alone into a separate
room, shut the door, and sit facing away from the windows, and
have nothing to look at or to attract his attention. A certain life
insurance agent of one of the largest companies in America adopted
the plan of selling to no one except by special appointment in his
own office from which all possible distractions had been removed.
There was nothing on the walls and nothing in the room but a desk,
a telephone, and a couple of chairs. There was nothing on the
desk except a life insurance policy, which was placed there at a
certain time of the interview. The purpose was to secure condi-
tions under which there were absolutely no distractions what-
ever, and the only thing to think about was the purchase of a life
insurance policy. For a time there was a calendar of the com-
pany hanging above the desk. He found that many clients would
remark, upon leaving the office, about the interesting dates des-
ignated on the calendar. There was nothing else to distract their
attention and consequently these stood out in the minds of the
clients, and, therefore, appeared interesting. He then removed
the calendar to a rear wall so that even the dates might not dis-
tract. All these features were a part of his carefully prepared
sales plan. This man was one of the most successful life insurance
salesmen among all the agents of that company. In a certain
month he had the record of selling the largest number of policies
of all the salesmen of this large company a record that was
achieved after only eighteen months of experience in selling life
insurance policies, immediately after graduation from college. It
would, of course, be absurd to attribute his remarkable success
to this one element, but it was nevertheless a very important part
in a carefully prepared plan of salesmanship.
The removal of distractions, or what amounts to the same thing,
the selection of a place for study where there are no distractions,
isone of the most useful suggestions that anyone can adopt for
developing concentration in work without a constant and exhaust-
ing tax upon the worker's voluntary efforts. In the course of time,
't may be possible to work under even distracting circumstances,
but probably no one, except the rare, absent-minded genius,

can work as well among distractions of sights and sounds, and
in the presence of other people, as under the complete absence of
such stimuli. No one is in a position to appreciate the great ef-
fectiveness in intellectual work under complete absence of dis-
tractions until he has tried it. The average pupil wastes an in-
estimable amount of time by having to study in the presence of
other members of the family who may be conversing or moving
about, and every word or action or stimulus of any sort is bound
to enter the mind and to divert the association processes to some-
thing else. Even though they are very minor, they require a few
seconds, if not longer, to cause the thought process to return again
to the subject-matter to be studied. Only those persons who have
compared their own working efficiency under distracting condi-

tions with their efficiency under ideal conditions can appreciate
the enormous difference in the amount that can be accom-
plished.
Begin work. Don't continue to think, "Oh, I just hate to
(3)
do but instead go to your desk in your secluded room, sit
this,"
down, take hold of book, paper, pencil, or whatever may be needed,
and begin to write, or read, or figure. In short, if you have diffi-
culty in overcoming inertia, just begin to go through the motions
of work. This will automatically start the mental processes going
relative to the work to be done, and before you realize it, you will
be in the midst of the task, reading, thinking, and writing in an
interested manner concerning the problems at hand. The external
mechanical movements will act as stimuli for the inauguration of
associative processes, and are likely to start mental activities
without a great deal of voluntary effort.
A prominent story writer relates that he had difficulty in be-
ginning his writing and in working out his plan necessary to finish
up the details after the plot of the story had been conceived. This
aspect of story writing is work and probably not a matter of in-
spiration; it involves close application and sometimes drudgery.
He found that he was able to get into his writing by simply sitting
down, taking a pencil and paper, and beginning to write whatever
came to his mind, whether it was very pertinent to his story or
not. Going through the motions started his thought processes
agoing, and very shortly his associative and imaginative proc-
esses were almost automatically producing pertinent and excel-
lent ideas.
HOW TO STUDY 183
In likemanner, begin to study a lesson by taking the book, turn-

ing to the page, and simply looking at the print. Some voluntary
effort, of course, must be exercised, if only to take hold of the book,
but it is more economical to do so than continuously be thinking
"How I hate to do it." This thought will automatically be driven
out by the processes started by simply going through the motions
of beginning work. The more voluntary effort and force one may
be able to exercise in not thinking about dislike for the task and
in beginning the motions of the work, probably the better and
the sooner one is able to start, but this voluntary attention should
normally pass very quickly into automatic attention and interest.
In the papers previously mentioned, the students stated that
they overcame their difficulties of going at their tasks and keeping
"
at them, besides "exercising will-power which was mentioneO
most frequently, by "setting a certain hour to begin," by "doing,
work in a limited time," by "doing the work under pressure," by
"dividing number of pages so that they could tell how many would
have to be read every fifteen minutes," by "copying a sentence
which helps to get the mind on the subject," by "starting directly
for if I wait at all a million things would come up which were more
interesting," by "having a definite schedule of study," by "plan-

ning the day," by "repeating with lips what is read," by "read-
ing aloud," and by "studying in one place."
Common Elements in the Assimilation and Retention of the
Material. At least five or six specific suggestions applicable to
any kind of studying may be given.

(1) Take a problem solving attitude. Know definitely what
you want to find. Ask questions and then look for the answers.
(2) Understand what you want to assimilate and letain perma-
nently. To go through reading matter in a perfunctory manner
will not cause retention of it except after long, wasteful, and fre-
quent re-readings. A certain psychologist, in conducting experi-

ments in memory with words and syllables, had dictated over and
over a great many times certain series of materials, so that they
had been completely memorized by several subjects, but he himself
was unable to repeat the material from memory. The reason was
that he. himself had never paid strict attention to the memorizing,
and had read them over and over again purely in a passive, in-
attentive manner.
Do not try to memorize ideas blindly; memorize understandingly.
Some material in school must be memorized mechanically, but
much more of it can be learned with a thorough conception of its

meaning.
(3) Organize ideas with reference to certain larger ideas and prin-
ciples. Organize your ideas and think out their relation to general
principles. Grasp in as large units as possible and note the rela-
of details.
^tion
In th Writer's Experiments ('17), page 190, is given an ex-
periment in whictitwo series of facts of apparently equal kind

and difficulty are presented for memorizing. Each list is com-
posed of five dates of history, five Greek words with their
English meanings, and five numbers with their sums. In one
list ideas are arranged in miscellaneous order; in the
these
other they are grouped by subjects, the five historical dates
list
are in one group, the five Greek words in the second group, and
the five sums in the third group. The time required by ten sub-
jects for memorizing the first set was an average of 14 minutes
and 3 seconds; the time required for memorizing the second series,
which was arranged in order, was on the average 9 minutes and n
seconds. The comparison shows a very decided advantage in
favor of learning the material in organized form.
(4) Recall at brief intervals the essential ideas of what you have
read. Stop at the end of each paragraph or two, shut your book
or your eyes, and recall the essential ideas you have read. Say to
yourself "What did I read about?" Then try to answer the ques-
tion. Note here what was snid about forgetting in the last chapter.
The is the occasion and stimulus which
chief value of examinations
they afford for recalling and organizing the material covered. In

some respects the most valuable studying done by pupils is done
in preparation for examinations. The value of the principle of
recall in learning or memorizing has been thoroughly demonstrated
by laboratory experiments.
Then each day or two, relate the recent material in a given
subject to the earlier material in that subject. That is, review in

your mind at short intervals, the larger essentials of all the ma-
terial covered up to date. The principle of recall in this form is
used far too little in studying. These suggestions would be applica-
ble to every type of reading which has to be done rather carefully.
It would, of course, not be advisable to do so in materials such as a
novel in which the ideas in detail need not be retained.
(5) At moment and as frequently as possi-
the earliest possible
ble, use the ideas that have been acquired, either by telling them
HOW TO STUDY 185
to some one else, or by thinking them over in your mind in con-

nection with other related materials or situations. This will give
them meaning in new ways and from new angles, and will help
to fix them permanently by virtue of the principle of recall.
(6) In committing material to memory, learn by wholes rather
than by parts. Poetry or prose can, as a rule, be memorized more
quickly if the material is read through as a whole from beginning
to end than if it is memorized in small sections of two or three lines;
and what is more important, when this method is employed, the
retention is more permanent. With many persons who are ac-
customed to memorizing by the part method, there is frequently
no saving of time in the first learning partly because the whole
method is new to them and partly because the learner often doubts
the advisability of using the whole method.
There are three reasons why the whole method proves in the
long run to be more economical: (i) Learning by parts establishes
many useless and interfering connections. Thus in committing
the first two lines of a poem the association is established between
the last word of the second line and the first word of the first line.
But not the order in which the lines are to be recalled.
this is
Rather the connection should be established between the last
word of the second line and the first word of the third line as is
done in the whole method. Consequently every portion memorized
by itself forms at least one detrimental connection and in a long
selection a very considerable number of such associations are
formed. These derailing paths probably account for the fact that
pupils in reciting a poem become stalled usually between the
portions learned piecemeal. Reading the material over as a
(2)
whole gives a view of the entire selection and will serve to give
meaning and correlation of the parts in the whole. It will help
to organize the ideas as a whole. (3) Learning by parts is apt to
produce great unevenness among the various portions of the ma-
terial in the degree of perfection of the memorizing. Some parts,
especially the earlier ones, will be repeated needlessly a great many
times and result in much greater over-learning of those parts than
of other parts. One point of caution in using the whole method
should, however, be noted. When the learner reads over the entire
selection to be memorized he does not make much visible progress
until, after a sufficient number of repetitions, he is able to repro-
duce most of the material. This situation is likely to be discourag-
ing, particularly to children. Perhaps the most effective manner
of employing the whole method is to learn the material in relatively

large sections instead of as a complete whole, particularly if the
selection is very long.
Improvement in Reading Ability. The average child, as well

as the average adult, reads far too slowly, and in fact, far more
slowly than he is capable of reading. About one-fourth of univer-
sity students read less rapidly than the average 8th grade pupil
does, and about one-fourth of 8th grade pupils read less rapidly
than the average 5th grade pupil. Experiments indicate that by a
moderate amount of definite practice, with conscious effort to
improve, the speed of reading may be increased from 50% to
100% without loss in the comprehension of the ideas read. The
moral would be: Force yourself to read more rapidly, which will
be accompanied by greater concentration of attention and in the
course of time this more rapid reading will become habitual, so
that the comprehension will be just as complete as at the slower
rate of reading. Consult the latter part of the chapter on "Read-
ing" for a more detailed discussion of these points.
Concrete Rules for Studying. Whipple has presented a scries
of thirty-eight rules which ought to prove valuable for increasing
effectiveness in studying. Some of these rules involve points that
have been previously presented in this chapter. Their specific
character makes them commendable for the student's considera-
tion and observance. They are as follows:
SUMMARY OF RULES. After Whipple fi6)
Keep yourself in good physical condition.

1.
Attend to, remove or treat physical defects that often handicap

2.
mental activity, such as defective eyesight, defective hearing, defective

teeth, adenoids, obstructed nasal breathing.
3. See that external conditions of work (light, temperature, humidity,
clothing, chair, desk, etc.) are favorable to study.
4. Form a place-study habit.
5. Form a time-study habit.
6. When possible, prepare the advance assignment in a given subject
directly after the day's recitation in it.

7. Begin work promptly.
8. Take on the attitude of attention.
9. Workintensely while you work: Concentrate.
10. But don't let intense application become fluster or worry,
n. Do your work with the intent to learn and to remember.
12. Seek a motive or, better, several motives.
HOW TO STUDY 187
13. Get rid of the idea that you are working for the teacher.
14. Don't apply for help until you have to.
15. Have a clear notion of the aim.
1 6. Before beginning the advance work, review rapidly the previous
lesson.
17. Make
a rapid preliminary survey of the assigned material.
Find out by trial whether you succeed better by beginning with the
1 8.
hardest or with the easiest task when you are confronted with several
tasks of unequal difficulty.
19. In general, use in your studying the form of activity that will
later be demanded when the material is used.
20. Give most time and attention to the weak points in your knowledge
or technique.
21. Carry the learning of all important items beyond the point neces-
sary for immediate recall.
22. You must daily pass judgment as to the degree of importance of
items that are brought before you, and lay special stress on the per-
manent fixing of those items that are vital and fundamental.
23. When a given bit of information is clearly of subordinate im-
portance and useful only for the time being, you are warranted in giving
to it only sufficient attention to hold it over the time in question.
24. Make the duration of your periods of study long enough to utilize
"warming-up" but not so long as to suffer weariness or fatigue.
25. When drill or repetition is necessary, distribute over more than
one period the time given to a specified learning.
26. When you interrupt work, not only stop at a natural break, but
also leave a cue for its quick resumption.
27. After intensive application, especially to new material, pause for a
time and let your mind be fallow before taking up anything else.
28. Use various devices to compel yourself to think over your work.
29. Form the habit of working out your own concrete examples of all
general rules and principles.
30. Form the habit of mentally reviewing every paragraph as soon as
you have read it.
31. Don't hesitate to mark up your own books to make the essential
ideas stand out visibly.
32. Whenever your desire is to master material that is at all extensive
and complex, make an outline of it. If you also wish to retain this
material,commit your outline to memory.
33. In all your work apply your knowledge as much as possible and as
soon as possible.
34. Do not hesitate to commit to memory verbatim such materials as
definitions of technical terms, formulas, dates and outlines, always pro-
vided, of course, that you also understand them.
35. When the material to be learned by heart presents no obvious
l88 EDUCATIONAL PSYCHOLOGY
rational associations, it is perfectly legitimate to invent some artificial

scheme for learning and recalling it.
36. In committing to memory a poem, declamation or oration, do not

break it up into parts but learn it as a whole.
37. In committing to memory, it is better to read aloud than to read

silently and better to read rapidly than slowly.
38. If your work includes attendance at lectures, take a moderate
amount of notes during the lectures, using a system of abbreviations, and
rewrite these notes daily, amplified into a reasonably compendious out-
line, organized as suggested in Rule 32.
Supervised Study. Teachers have come to recognize in recent

years the waste of time and the blind direction of energy, or pos-
sibly lack of energy, in so much of the studying done by pupils
that a widespread movement has gotten under way for the super-
vision of studying. The plans for supervising studying are carried
out in so many different ways that hardly any one plan can be
designated as typical. The results accruing from the general ef-
forts in this direction have been in most cases beneficial. Con-
tinued experimentation during the next few years with various
plans of supervised study will lead to a more general agreement
as to the most effective manner of administering it.
In a recent inquiry of supervised study in schools on the Pacific
coast, Proctor ('17) found that forty-two high schools employed
it in one form or another. Of these forty-two schools, thirty-one
reported the use of a lengthened period distributed as follows:
(a) 60' period, divided 30-30, No. of cases 3

60' period, divided 35-25, No. of cases i
60' period, divided 40-20, No. of cases 15

21
(b) 70' period, divided 40-30, No. of cases 4
70' period, divided 35-35, No. of cases 2
6
(c) 80' period, divided 40-40, No. of cases i i
(d) 85' period, divided^ 45-40, No. of cases 2 2

(e) 90' period, divided*45~45, No. of cases i i
Total 31
Regarding the effects of supervised study, Proctor reports that:
"Twenty-six of the 31 principals employing the lengthened period

ttaidthat study habits had been improved; one could discover no apparent
HOW TO STUDY 189
effect; two said that only the slow students had been helped, the brighter
ones were not; and two had no data on which to base their opinions.
"Wherever the plan had been in use long enough to make possible the
compiling of statistics as to the effect of supervised study on scholarship,
there was practically unanimous agreement that the number of failures
had been reduced and the standards of scholarship had been raised. The
high school at Snokomish, Washington, reports that the average per-
centage of failures in elementary algebra for the two years prior to the
adoption of supervised study was 28%. But for the two-year period
following the adoption of supervised study the failures in the same
subject were reduced to 17%. Hoquiam, Washington, reports that the
average marks of the students range 10% higher and that the number of
honor pupils has been doubled since supervised study was introduced.
The principal of the Arcata high school, California, reports that the
average mark of the freshman class has been raised from 78% to 82JHz%
during the first year of supervised study. Santa Cruz, California, com-
paring the year 1914-15, the last under the old plan, with the year
1916-17, the second year under supervised study, finds that the increase
in the total number of high marks has been 157%; the decrease in low
failures, 188%. Reno, Nevada, reports a decrease of 45% in the number
of failures, and an increase of 24% in the number of students making
excellent marks."
J. Stanley Brown, principal of the high school at Joliet, Illinois,

reports, as quoted by Hall-Quest ('17), a decided reduction in the
percentage of failures after the introduction in the high school of
supervised study, as indicated in the following table:
TABLE 46. After Brown and Hall-Quest ('17, p. 386). Supervision of study
apparently was begun in 1912 although I have not been able to find a
definite statement by Hall-Quest to that effect.
Table of percentage of failures
Breslich ('12) made an experiment to determine the effect of

directed study by dividing an algebra class into two sections, one
190 DUCATIONAL PSYCHOLOGY
of which was conducted in the usual manner of recitation work

and home study, and the other was conducted by confining all of
the work to the recitation period of 45 minutes. This time was
devoted partly to study under the supervision of the teacher and
partly to recitation work. The two sections were made up of pupils
of approximately equal ability as indicated by the marks for the
preceding semester's work, which averaged 81.4 for those who

constituted the home study group, and 79.4 for those who con-
stituted the supervised study class. The home study class de-
voted approximately two hours to each lesson including the 45
minutes for the recitation period. The experiment was conducted
for a period of fourteen lessons. At the end of that time, the same
examination over the work that had been covered was given to
90
I
3-
75
1 2 8 4 56 7 8 9 10 11 12 13 14 15
Weeks
FIG. 54. The broken line represents the supervised group. The continuous
line represents the unsuperviscd group. After Minnick ('13).
both sections. The supervised study class made an average of 65.5,

and the home study group made an average of 62.8. Hence
the supervised study class obtained as good results as the home
study class, or slightly better, in spite of the fact that the former
spent only about two-fifths as much time upon the work.
Minnick ('13) of Bloomington, Indiana, divided a group of thirty-
six pupils in plane geometry into two divisions, providing super-
vision of the study to one division and none to the other division
for a period of fifteen weeks. The results show an advantage in
scholarship for the supervised group as represented in the curves

of Figure 54.
Hall-Quest quotes similar results of improvement in scholarship
as indicated either in the reduction of the percentage of failures
or in the higher scholastic marks as reported by Loveland at Potts-
town, Pennsylvania, by Rickard, at Oakland City, Indiana, and
from the high school at Pueblo, Colorado.
CHAPTER XIII
TRANSFERENCE OF TRAINING IN SPECIFIC MENTAL FUNCTIONS
Problems. To what extent does training in one mental function

or set of functions modify the operation of other mental functions?
To what extent will training in mathematical reasoning modify
reasoning ability about political events or bargains, or vice versa?
To what extent will training in remembering faces and names
modify the remembering of the prices of goods or words in a lan-
guage, or vice versa? To what extent will improvement in pro-
ficiency in such mental capacities as are involved in school studies
modify proficiency in any other specific or general activities or
interests in life? These problems open up all the ramifications of
the traditional controversy concerning mental discipline, the educa-
tional value of school subjects, and their related discussions. The
fundamental problem, however, is not, does training transfer?
The task is more complex and suggests rather the following three
fundamental problems: (i) To what extent does training transfer,
(2) To how how
distantly related functions does training
closely or
transfer, and (3) How
does the transfer take place? Thus we see
that transference of training is one of the three or four most im-
portant perennial problems in the entire field of education. So
many problems in the administration of schools, in the construction
of courses of study, in the emphasis upon various aspects of school
subjects, in fact, the ultimate values of education as a whole, de-

pend primarily upon our attitude toward the problems of mental
discipline and transference of training.

In order to think about these matters clearly, it is necessary
to distinguish between two quite different aspects of the discussion,
namely, (i) The pure disciplinary or training value in the improve-

ment of a mental function irrespective of the material through
which it is trained or which is acquired in the training; (2) the
intrinsic value of the information or material acquired in the proc-
ess of the training irrespective of the training referred to in (i).
The one is acquisition of training; the other is acquisition of con-
tent. The difference between these two aspects may be illustrated
thus: The learning of shorthand will furnish practice of certain
XQI
types of memory and associative processes; it will also supply the

individual with certain symbols for recording ideas. The former
would be the pure training value of the mental functions, the latter
would be the informational or instrumental value of shorthand.
If one should never expect to use the symbols of shorthand for
recording ideas, to what extent would the practice in memory

and associative functions modify memory and associative processes
in other reactions in life? How much value, accordingly, may we
attach to the practice of these mental processes? In thinking
about these problems, we must distinguish sharply between the con-
tent or informational aspect of a given type of learning and the
pure improvement value in mental functions to be derived from
the learning. Viewed from the standpoint of the school, the situa-
tion presents two problems: (i) To what extent does training of the
mental capacities involved in a given school subject carry over and
produce efficiency in other subjects or in other activities in life,
and (2) are certain school subjects more capable of improving the
mental functions generally and of carrying the improvement over
to other responses of behavior?
The Effect of Improvement in Specific Mental Functions upon
other Mental Functions. The influence of improvement in one
function upon others may be one of help, hindrance, or indifference.
Which and how much, can be determined only by recourse to
it is
facts. Until twenty-five years ago the problem was discussed

wholly as a matter of opinion. During the last twenty-five years,
STcbhsiderable number have been made on many
of researches
aspects of the problem so that the controversy may be dealt with

in a more definite and factual manner than was formerly the case.
The experimental technique of research in the field of transference
of training TrasTSeClT practically the same in all investigations, and
has consisted (i) of testing the strength of a variety of mental
capacities, (2) of training one or more capacities for a specified
period of time, and (3) by finally testing again the same capacities
tested before the training in order to determine what changes may
have been produced in them as a result of the intervening training.
The tests referred to under (i) and (3) are conveniently called "end
tests" or the "test series" and the work under (2) is usually called
the "training series." This plan has been followed in. the large
majority of transfer experiments. A different plan, however, is
possible and has been employed in a few studies. This consists
of giving training to a group of persons in some particular function
TRANSFERENCE OF TRAINING
and then giving them practice in another function. Their progress

in this second function is then compared with that of other individ-
uals who have not had training in the first function.
1
a. James Experiment on Memory. The first experimental in-
vestigation was made by James and published in 1890 in his Prin-
ciples of Psychology. This experiment is of interest and impor-
tance chiefly because of its historical significance in opening the
problem by an experimental approach. James attempted to de-
termine the effect of training in learning one kind of poetry upon
memorizing other kinds of poetry. He first made the experiment
upon himself by memorizing in the course of eight daysî^lines
of Victor Hugo's Satyr. This required a total of 131 5/e minutes.
He then spent some twenty minutes a day for 38 days in learning
the first book of Milton's Paradise Lost. At the end of this time he
again memorized 158 lines from Victor Hugo and found that it
took i5iX minutes. This loss in time was surprising and James
explained it by saying that he was fagged out by other work at the
time of the second test on Victor Hugo and that he was not really
in fit condition for such an experiment. He then repeated the ex-
periment with four students in a similar manner by using different

poetry. The results of these early experiments are given in the
following table:
TABLE 47. After James ('90, 1, p. 667)
This experiment in its essential details, was repeated by Peterson

('12)with two subjects, one of whom showed gain and the other loss.
b. Reaction Time. The next series of experiments was under-
taken by Gilbert and Fracker, who attempted to determine the
amount of^transference of training from one type of reaction to
other types of reaction. Three subjects were tested first in simple
reaction to sound, to electric stimuli, to touch, to visual stimuli,
and likewise in complex reaction to stimuli involving discrimination
and choice. The training series consisted of simple and complex
reaction to sound only and continued for twelve days. The results
obtained in this experiment are given in the following table which
shows the percentages of gain made in each of the end tests:
TABLE 48
The spread of improvement in reacting to various sensory stimuli. After
Gilbert and Fracker ('97)
1
Average of J. A. C. and G. C. F. only.
J. C. P. was practiced only in reaction time, while the other two were
practiced in both reaction and reaction with discrimination and choice. All
figures of the above table represent per cent of gain by practice.
Each of the forms of reaction shows on the whole a distinct gain
in the second end tests. How much of this gain is actually due to
the training series cannot be definitely determined. of the Many
earlier investigators did not make control tests, that is, they did not
repeat the end tests on another group of subjects who did not take
the practice series but who took only the end tests separated by an
interval equal to that consumed by the practice series. It is obvious
that a certain portion of the gain in the end tests is due to the fact
that when the second end tests are made, some advantage is derived
from the familiarity or practice in having done the end tests once
TRANSFERENCE OF TRAINING 195
before; consequently, the actual amount of improvement in a prac-

tice experiment can be determined only by subtracting the amount
of gain made by a control group which has not done the practice
series in order to obtain the residual amount of improvement ac-
tually transferred from the training series.

Another important item frequently omitted in the early investi-
gations is a statement of the actual amount of progress made in
the practice series itself. This element is significant because it is
possible thereby only to determine the amount of gain made in the
end tests as compared with the improvement in the training series
itself in order that some definite conception may be formed of the
amount of gain made in the practice series which is transferred to
the end tests. Thus in the reaction experiments of Gilbert and
Fracker, the gain in the practice series is shown in the first and
fifth columns. It will be noticed that the average gains in the end
tests in simple reaction to electric shocks, to touch, and to color
was about as great as in the training series itself, that is, in simple
reaction to sound. It was 17%, n%, and 20%, or on the average
1 6% in the former, as compared with 16% in the latter. In case
of the complex reactions, the average gains in the reactions to
electric shocks, to touch and to color were 40%, 17%, and 22%,
or on the average 26%, as compared with a gain of 50% in the
practice series. On the face of it, 100% of the practice effect in
simple reaction to sound was transferred to the other forms of
simple reaction, while 52% of the practice effect in complex re-
action to sound was carried over to the other types of complex
reactions. Actually the amounts of transfer effects are probably
considerably less; how much we do not know since Gilbert and
Fracker made no control tests.
c. Perception and Discrimination. Thorndike and Woodworth
('01) made aninvestigation to determine the transference of prac-
tice in estimating areas, lengths of lines, and weights to estimating
areas, lines, and weights of different sizes. They also measured

the effect of practice in perceiving words containing certain letters
upon the accuracy and quickness of perceiving other words con-
taining different letters. The results of this experiment are sum-
marized in the following manner by Thorndike:
"Individuals practiced estimating the areas of rectangles from 10 to

100 sq. cm. in size until a very marked improvement was attained. The
improvement in accuracy for areas of the same size but of different shape
due to was only 44% as great as that for areas of the same
this training
shape and For areas of the same shape, but from 140-300 sq. cm.
size.
in size, the improvement was 30% as great. For areas of different shape
and from 140-400 sq. cm. in size, the improvement was 52% as great.
"Training in estimating weights of from 40-120 grams resulted in
only 30% as much improvement in estimating weights from 120 to 1800
grams. Training in estimating lines from .5 to 1.5 inches long (resulting
in a reduction of error to 25% of the initial amount) resulted in no im-
provement in the estimation of lines 6-12 inches long.
"Training in perceiving words containing V
and 's' gave a certain
amount of improvement in speed and accuracy in that special ability.
In the ability to perceive words containing 'i and 't/ 1
and 'p/ 'c' andV
'a,' V
and 'r,' 'a' and n,' c
T
and V, mispelled words and A's, there
was an improvement in speed of only 39% as much as in the ability
specially trained, and in accuracy of only 25% as much. Training in
perceiving English verbs gave a reduction in time of nearly 21% and in
omissions of 70%. The ability to perceive other parts of speech showed a
reduction in time of 3%, but an increase in omissions of over ioo%."
The experiments in marking out words and in estimating weights

were repeated with two persons in substantially the same manner
by Coover. ('16.)
"Two n
days in marking out words con-
reagents were trained for
*
taining e and columns of the Outlook' Magazine. Each

s in selected
reagent looked over 12,000 words in each day's practice.

'Tests were taken before and after training, in marking out
'(i) Words in 'Outlook' columns containing e-s, i-t, s-p, c-a, e-r.
'
(2) Words on manuscript pages containing a-n, l-o, e-r.
'(3) Common nouns in 'Outlook' columns.

1
'(4) Words in 'Outlook columns containing e-s."
Coover's results showed a gain of 44% in the training series and

f m
33% the end tests, or 75% as much as in the training series.
This a larger transfer effect than that of Thorndike and Wood-
is
worth whose results, however, were based on five persons and

showed a gain of 37.7% in the training series and of 17% in the
end tests, or 48% as much as in the training series.
Coover's experiment in estimating weights was carried out by
training two persons with a set of seventeen blocks ranging from
40 to 120 grams. Each person made 1,700 judgments. The per-
sons were tested, before and after the training, in estimating ten
common objects averaging 67.5 grams in weight but falling within
TRANSFERENCE OF TRAINING IQ7
the limits of 40 and 120 grams, and in estimating ten common

objects averaging 552.7 grams but all exceeding 120 grams.
The experiment yielded a gain in accuracy of estimating weights
of 23% in the training series and of 29% in the end tests with the
set of ten smaller objects but a loss of 100% with the larger ob-
jects. This loss was due to the very large loss of one subject which
far outweighed the gain of the other subject. The gain in the
estimation of the smaller weights was greater than in the training
series itself. Thorndike and Woodworth's experiments showed
a gain of 45% in the training series and of 38% in the end tests
with the smaller weights and of 16% with the larger weights.
Kline had nine persons practice for fourteen days from 30 to 45
minutes daily in canceling e's and t's on pages of prose. Before
and after the practice he tested them in canceling nouns, verbs,
prepositions, pronouns, and adverbs. Eight other persons were
tested ig like manner without doing the practice series. Kline
found that the practiced group did not gain as much as the un-
practiced group. This he explains by the introspective statements
of his subjects that "there was a tendency to cross out words
containing e's and t's rather than the required part of speech."
The detailed results follow:
TABLE 49
The spread of improvement in marking letters. After Kline ('09, p. 10)
NOUNS VERBS PREPOSITIONS PRONOUNS ADVERBS
PRACTICED
GROUP
After practice 14.0 6.0 28.00.5 2 3 6.3 0.6
Before practice 28.6 17.3 4 4 4 5
Differences. . 7.4
. 4.7 -2.0' 2 1 2.5 2 1 -1.3 3.0
UNPRACTICED
GROUP
Second period 30.4 1.3 6.0 7.0 26.6.1.7 5.0 0.6 4.0 0.7 7.0
First period . . 23.5 5.0 16.62.6 10.5 4.6 0.3 13.7 2.013.0
Differences... 6.9 6.7 2.61.0 -2.0> 10 00.9 0.4-0.3 9.7 6.0
1
sign indicates loss at second period.
Bennett ('07) tested a group of sixteen pupils in discriminating

between shades of red, yellow-green, and orange, and differences
Ufi EDUCATIONAL PSYCHOLOGY
in the pitch of tones before and after training twice a week for five
months in discriminating shades of blue. The accuracy in the
four end tests showed the following gains:
Coover and Angell tested four adults in discriminating intensi-

ties of brightness before and after training in discriminating in-
tensities of sound consisting of seventeen series of forty judgments
each. The end tests without the intervening training were also
given to three other subjects. The four trained persons rose from
56.9% of right judgments before the training, to 66.0% after the
training, while the untrained persons dropped from 65.5% right
judgments to 61.7%.
d. Sensori-motor Association. Bair ('02) attempted to measure
spread of practice, not by testing certain capacities before and
after training in some other capacity, but by training the subjects
ina certain function and then determining the effect of this train-
ingupon the progress in the subsequent training of other functions.
His experiments are described thus:
"(i) Six keys of a typewriter are labeled with six symbols (letters or
figures). Fifty-five of these letters or figures, in chance order, are now
shown one by one, and the subject on seeing one taps the corresponding
key. The time taken to tap out the series is recorded. Six different
symbols are then used with a new series composed of them, and the sub-
ject's time record is taken as before. This is continued until twenty
different sets of symbols have been used. Although the symbols have
been changed each time, there is a steady improvement, ranging for the
four subjects in the following decrease in time: 62 to 52, 95 to 85, 71.5 to
58, 65 to 56. The major part of this gain could not have been due to
merely getting used to the machine or to the general features of the
experiments, for the fourth subject was already used to these and still
gained about nine-tenths as much as the other three.
"(2) The other experiment consisted in taking daily records for
twenty days, by means of a stop-watch, of the time required to repeat
the alphabet from memory. Each day's experiment was as follows:
First, the alphabet was repeated as rapidly as possible forward; sec-
ond, the letter n was interpolated between each of the letters; third,
the alphabet was repeated backward interpolating n between each two
of the letters. At the end of twenty practices in each order the subject
repeated the alphabet first forward interpolating instead of n the letter

x and repeating three times; secondly, interpolating r and repeating
three times; then lastly, repeating backward and in like manner inter-
polating x and r and repeating three times. There was improvement
in the test series, the effect of the twenty days' training with the training
series being to put the abilities in the test series as far ahead as three
days of the direct training would have done."
Scholckow and Judd investigated the effect of knowledge of the

principle of refraction upon learning to hit a target under water.
"One group of boys was given a full theoretical explanation of refrac-

tion.The other group of boys was left to work out experience without
theoretical training. These two groups began practice with the target
under twelve inches of water. It is a very striking fact that in the first
series of trials the boys who knew the theory of refraction and those who
did not, gave about the same results. That is, theory seemed to be of no
value in the first tests. All the boys had to learn how to use the dart, and
theory proved to be no substitute for practice. At this point the condi-
tions were changed. The twelve inches of water were reduced to four.
The differences between the two groups of boys now came out very
strikingly. The boys without theory were very much confused. The
practice gained with twelve inches of water did not help them with four
inches. Their errors were large and persistent. On the other hand, the
boys who had the theory, fitted themselves to four inches very rapidly."
(Judd, '08, p. 37.)
Webb ('17) used the plan of determining the effect of acquired

skillupon the acquisition of other skills. He employed 54 rats and
21 humans in learning mazes in various orders. He measured the
results in terms of the number of trials required, the number of
errors made, and the amount of time needed to learn the mazes.
The following table gives the savings in learning a second maze
as compared with the learning of the first one:
TABLE 50. After Webb

Average percentage of saving in transfer
RATS HUMANS
200 KDUCATIONAL PSYCHOLOGY
"
Webb concluded thatthe learning of one maze has a beneficial
"
effect in the mastery of a subsequent maze situation (page 50)
and that "the degree of transfer is dependent in part upon the
degree of similarity of two maze patterns" (page 53). Webb
further attempted to ascertain whether a new habit has a retro-
active effect upon habits previously formed. He had his subjects
learn one maze, then a second one, and then return to the first one.
His findings were inconclusive.
>
Coover ( i6) reports an unpublished investigation by Carrie
W. Liddle designed to measure the effect of practice in discrimi-
nating and sorting cards bearing colors or geometric signs upon
discriminatingand sorting cards with different colors or signs.
"Each set of 102 cards contained six colors, or six designs, was shuffled
so that no color or device repeated itself, and was sorted into six com-
partments. The first six cards of the pack determined the order of colors
in the compartments according to which the rest of the pack was to be
sorted. Nine reagents took part and the experiment continued two
semesters. There was transference of practice-effect from one set of
colors to the other set of colors, and to the geometric forms; and from one
set of geometric forms to the other and to the colors. Increased powers
of discrimination and attention were thought to be the causes of trans-
ference."
Bergstrom ('94) had found previously that training in sorting

cardsby one method interfered with sorting them by a different
method. The same situation is shown by the card-sorting experi-
ment in the author's Experiments, Chapter XV.
Coover and Angell ('07) attempted to ascertain the effect of
practice in card sorting upon typewriter-reactions. They trained

four persons in card sorting on 15 days scattered through a
period of 40 days. During that time the subjects sorted 4,200,
3,800, 5,200,and 4,000 cards respectively. Before and after this
training they were given practice in typewriter-reactions. Three
other persons, as a control group, were given practice in typewrit-
ing at two periods separated by an interval of 45 days. The re-
sults are interpretedby the authors as indicating transfer, but it
is doubtful whether there is any transfer and, if there is, how much.
The practiced group reduced their time for the first 100 typewriter
reactions, before the training in card sorting, from 84.4 seconds,
with an average of 2.3 errors, to 62.3 seconds, with 6.3 errors, for
the last 100 reactions after the training in card sorting. The
TRANSFERENCE OF TRAINING 2OI
unpracticed group reduced their time from 106.3 seconds, with

3.3 errors, to 80.6 seconds, with 2.3 errors. The trained group
reduced its time by 26% but increased in errors, while the untrained
group reduced its time by 25% but decreased in errors. There is
obviously no appreciable transfer.
e. Memory. More extensive researches have been made in the
field of memory than in any other single aspect of the problem of
transference of practice. One of the most elaborate investigations
was made by ELert and Meumann. They measured the amount
of transfer from memorizing a series of nonsense syllables to vari-
ous other types of memory, such as immediate memory for num-
bers, letters, words, permanent memory of prose, poetry, etc.
The end tests were made at three different times, before the be-
ginning of practice series, about the middle, and at the close of the
practice series. The results showed very considerable gains in
these other types of memorizing. The difficulty in interpreting
their results, however, is the fact that they did not make the cross
section tests with a control group according to which a deduction
could be made for the gain in the end tests themselves. Dearborn
repeated the end tests on a group of subjects to ascertain the
amount of allowance to be made. His results together with those
of Ebert and Meumann are shown in the following table. Dear-
born found a very considerable amount of gain in these end tests.
His comments are as follows:
"The results indicate that a considerable part of the improvement

found must be attributed to direct practice in the test series, and not
to any 'spread* of improvement from the practice series proper. There
is further, at times, lack of correlation between the amount of improve-
ment made in the practice and that made in the test series; occasionally
a larger percentage of gain is made in the latter than in the practice itself.
This again indicates the presence of direct practice in the test scries.
"Some at least of the remaining general improvement found is to be
explained simply in terms of orientation, attention, and changes in the
technique of learning.
"These results seem to render unnecessary the hypothesis proposed by
Ebert and Meumann to account for the large extent of the general in-
fluence of special practice, which their experiments seem to indicate."
Three subjects of Ebert and Meumann were trained in learning

64 sets of i2-syllable series; they gained 70%. Three others were
trained with 48 sets. They gained 50%.
2O? EDUCATIONAL PSYCHOLOGY
TABLE 51. After Ebert and Meumann ('05) and Dearborn
1
Quoted by permission from an unpublished table prepared by Dearborn.
Thus the high percentages of Ebert and Meumann are reduced to

an average residual transfer of 22%.
Fracker has reported a rather extensive series of investigations
on transfer in memory in which the training series consisted of
memorizing various combinations of four degrees of loudness in a
sound. These four loudnesses were presented in the various possible
combinations and the responses of the subjects consisted in indi-
cating the proper order in which the sounds had been received.
The end tests consisted in determining the memory capacities for
various combinations of four shades of gray, 9 tones, 8 shades of
gray, 4 tones, geometrical figures, 9 sets of numbers, arm move-
ments, and poetry. The results are summarized in the following
table which also indicates the amount of deduction to be made due
to the improvement in the end tests made by the control group.
TABLE 52
Transference of training in memory. After Fracker ('08)
The improvement made in Training Scries by 8 subjects was 21%.

End Tests:
Similar to Training Series:
Four Grays, 8 trained subjects,
" "
36% 4 untrained,
"
4% 32%
Nine Tones, 8 22 4 n 11%
Nine Grays, " " "
8
"
19 4
"
10 9%
Four Tones, 8 10 4 -2 12%
22% 6% 16%
End Tests:
Unlike Training Series:
Geometrical Figures, 8 trained subjects, 5%
Nine Numbers, 8
8
"
"
"
"
44"
13%
04
4 untrained,
"
8
o 4%
i%
Movement,
Poetry, 8 74" 2
i
5%
6% 3% 3%
An interesting result emphasized by these data is the fact that
the transfer to the types of memory similar to that involved in the
training series is considerably greater than the transfer to the
memory functions unlike the training series. The average residual
gain in the four similar memory processes is 16%, whereas in the
four unlike memory processes it was only 3%.
Sleight made a careful and extensive investigation on transfer-
ence of training in one sort of memory to other sorts of memory.
He believed that previous researches had not used enough subjects
to be statistically reliable. He therefore carried out his first re-
search with 84 pupils from three girls' schools, averaging 12 years
and 8 months old. Ten cross sectional tests were made before,
in the middle, and after the training series, as follows: (i) Re-
membering and reproducing the location of points in circles, (2)

two series of six dates each and their corresponding events, (3)
series of eight syllables, (4) a stanza of from eight to twelve lines
of poetry, (5) learning a passage of prose, (6) reproducing the con-
tent of a passage of prose, (7) remembering locations on a map, (8)
remembering dictated sentences, (9) memory span for letters, (10)
remembering names.
The pupils were divided into four groups of approximately equal
ability as determined by the ten tests before the training series.
One group was then trained in learning poetry; another in learning
tables of multiplication, denominations, squares, fractions, etc.;
X>4 EDUCATIONAL PSYCHOLOGY
a third in reproducing the thought content of prose selections of

scientific, geographical and historical material; and the fourth
group had no special practice. The training period lasted four
days a week for six weeks, practicing 30 minutes each day.
The chief results arc presented in Table 53. I have computed
the percentage of gain made by each group in Section III, that is,
the end tests made at the close of the training series, over Section
I, the tests made before the training series. These percentages are
given in the last column. Sleight has not made such a percentage
comparison, but has used a different, and possibly fairer, plan of
computing the data. I have, however, made this computation in
terms of percentages as these will be more intelligible to the reader
unfamiliar with statistical methods. The average percentages at
the bottom of the table show only slight gains on the part of the
trained groups, 2, 3, and 4, over the untrained group. The average
gain of group 2, trained in poetry, over group i, untrained, was
3.3%; of group 3, trained in arithmetical tables, over group i was
2.6%; and of group 4, trained in prose, over group i was 4.0%.
The amounts of transfer are very small. Sleight failed to indicate
the improvement in the training series themselves so that it is im-
possible to compare the transferred amount with it.
TABLE 53
The numbers in the following table are the average scores made by each
group in each test. Group i had no special practice, Group 2 was practiced in
learning poetry, Group 3 in learning tables, and Group 4 in learning prose
substance.
The column under Section I gives the scores before the training, under
Section II about the middle of the training, under Section III after the training.
After Sleight. ('11,^413.)
TRANSFERENCE OF TRAINING 20$
TABLE S3 Continued
Average % of gain ofa Group

it
i in all tests
" tl
.......................... 13-0
^ .......................... 16.3
u tt u
17.0
Sleight, by his method of computation, found only

a few in-
stances of significant amounts of transfer. His conclusion is that
"There appears to be no general memory improvement as a result
of practice, nor any evidence for the hypothesis of a general memory
"
function (p. 455).
After the conclusion of these experiments, Sleight repeated the

same investigation with some modifications, on a group of young
women, 18 to 19 years old. The results were substantially the
same.
Coover made a study of the effect of training in reproducing im-
agery of a simple kind upon ability to reproduce imagery aroused by
materials of various sorts. The tests made before and after the
training were as follows: (i) recognition or choice of one of two
letters previously shown, (2) reproduction and recognition of letters
presented in groups of 12, (3) discrimination of intensities of sounds,

(4) memory of visual symbols. The training consisted in practice
in discriminating intensities of sound, and extended through a
period of 48 days. These intensities of sound were produced with a

sound pendulum (wood) instead of with a fall phonometer (steel)
as inend test number (3). The results of the investigation show
small or doubtful effects of transfer.
"The training on discrimination of sound did not result in improve-

ment in efficiency with the training material. But, according to intro-
spective evidence, it effected changes in the processes employed. Quan-
titative analysis showed that the practice-effect of the evident exercise of
retention and reproduction of auditory and other imagery 'spread* to
the tachistoscopic test of Recognition or Choice of One of Two Letters,
and to the test on the Complete Learning of series of visual symbols,
both of which involved retention and reproduction of imagery."
Dearborn made some experiments to measure the effect of prac-

vocabulary and poetry upon ability to memorize
tice in learning
various sorts of material as specified in the following table. He
did not make the end tests on a control group and hence it is im-
possible to determine how much of the gain in the end tests was due
to the practice series. Judging from other experiments these gains
would have to be reduced by one-half or one-third. An interesting
comparison may be made between the gain in the end tests and the
training series. The average gain in learning French and German
vocabulary was 57%, whereas the average gain in the end tests in
learning French, German, or English verse was only 19% or one-
third as much. Practice in learning Paradise Lost made no im-
provement in learning chemical formulae.

TABLE 54
Transference of practice in memorizing German and French vocabularies and

English poetry and prose. After Dearborn (1910), p. 385
Bennett ('07) had one person memorize 16 lines of In Mcmo-

riam a day for 28 consecutive days. This person was tested before
and after this period of training by learning a list of 15 names of
places each day for five days in which he showed a gain of 58 %.
Another person memorized two stanzas of Faerie Quccne a day for
35 consecutive days. Before and after this period he was tested in
learning a list of 30 digits each day for five days and in which he
showed a gain of 22%. No control tests were made on other
persons without training in learning the poetry.
Winch tested a group of 34 girls, averaging 13 years of age, by
having them learn a passage of historical prose. On the basis of
this test he divided them into two groups of equal ability. Group
A. memorized 18 to 20 lines of poetry each day on four days scat-
tered through a period of two weeks. Group B meanwhile worked

sums. At the end of that time both groups were tested with his-
torical prose. Group A rose from a total score of 1,497 to 2 J55> or
37%, while group B rose from 1,497 to 1,890, or 27%.

Winch ('08 and *io) next tested another class of 34 girls in the
same general manner, except that the before and after test was
made with geographical passages and that the poetry for the train-
ing series was somewhat simpler. He also carried out a similar
experiment with a third class of girls, using a historical passage
for
the end tests. The results in each case showed a greater gain in the
practiced group.
The results from the author's class experiments (Chapter XI V
Experiments) indicate that an improvement of 27% in learning

Italian vocabulary is accompanied by an improvement of only
8% in learning French vocabulary or less than one-third as much.
Improvement in transcribing letters into numbers was accompanied
by only 12% as much gain in transcribing numbers into symbols.
Attention. Coover made a scries of tests on transference which
he under the head of attention, but it is doubtful, as he himself
lists
states, whether they are measures of attention any more than

many other tests that have been reported under other headings.
At any rate, attention probably played an important part in most
of the tests that Coover employed. The following were the nineteen
end tests made before and after training.
I. Reaction
1. Simple sensory to sound (50) i
2. Compound
a. With discrimination
(1)Marking out small a's (100) 2
(2)Marking out o's (100) 3

b. With discrimination and choice
(1) Card-sorting (200) 4
Typewriter-reaction
(2) (200) 5
(3) Controlled reaction (50) 6
II. Sensible discrimination of sounds (90) 7
HI. Reproduction
1. Unequivocal (Rote memory)
a. Successive presentation
(1) Memory of sound intensities (50) 8
(2) Memory of consonants (50) 9
(3) Memory of Arabic numerals (50) 10
(4) Memory of visual signs (10) n
(5) Memory of associated pairs (50) 12
b. Simultaneous presentation
(i) Learning i2-letter-rectangles
(a) Free (10) 13
(b) With distraction (10) 14
Equivocal Word-completion
2. (10) 15
Free 2-minute trains of ideas
3. (3) 16
IV. Extensive threshold of visual attention
1. Free (15) *7
2. With distraction (10) 18
V. Maximum voluntary activity tapping (5 30") 19
(The figures in parenthesis indicate the number of reactions, memory units
or experiments, in the test.)
"These tests were taken by 10 reagents, 8 of whom took training be-

tween the first and final series which were separated by an interval of
55 days. The first scries of tests occupied 12 days during a period of

36 days; the second, or final, 10 days during a period of 21 days. Each
pair of tests was separated by an interval of about 66 days.
"The pairs of tests were also taken by two reagents of a group of 21
control reagents. There were thus two sets of control reagents: The two
who took all the tests, and the 21 each of whom took only one pair or a
few pairs of tests." (Coover, '16.)
Different ones of the subjects took training in different material:
"During the 55-day interval between the tests, two reagents (MN.,
Le.) took training 18 days on Test 17; 25 1 2-letter-rectangles were pre-
sented daily, making in all 450 experiments each. Two reagents (Rt. and
SI.) took training 18 days on Test 13; 20 1 2-letter-rectangles were pre-
sented daily, aggregating 360 experiments each. One reagent (Ly.) took
training in simple reaction to sound for ndays, 1,100 reactions in all.
(Le., who took training on test 17, also took training in this simple re-
action to the extent of about 500 reactions.) Two reagents (He., Cr.)
took training on memory schemes for about 14 days. And one reagent
(al.) took training on Test 17 for 8 days, almost consecutive, to the ex-
tent of 200 experiments."
The results obtained from the various end tests are rather intri-
cate and difficult to present in tabular form and somewhat doubtful
as to their meaning so far as improvement in attention is concerned.
Coover attempted to interpret their meaning from the standpoint
of control of attentionby comparing the variability in the perform-
ance of the persons before and after training on the assumption that
reduction in variability indicated better attention. For a detailed
consideration, Coover's original report must be consulted. His
general conclusion was that "as a measure of attention our tests
are inadequate, and the question of transference of improved con-
"
ditions of attention remains open (page 183).
g. Analysis and Ingenuity. Ruger ('10), in connection with his
study of learning to solve puzzles, made observations on transfer

of practice. His results are difficult to summarize in brief form.
It will have to suffice, therefore, to say that he enumerates general
factors of transfer in solving puzzles as follows: (a) The ideal of
efficiency, that is, "the active search for methods of control;"

(b) a high level of attention was a precondition of success; (c)
attitudes "The change from the self-conscious to the problem-
attitude occurred sometimes automatically, and sometimes de-
liberately by means of an ideal. The most powerful stimulus to
change of attitude and so of its transfer was personal success; it

did not matter much whether it was accidental or planned;" (d)
methods of attack. As to special factors he mentions: (a) Related
"
ideas Geometrical concepts played an almost negligible part in
the work of solution; " "The greatest transfer in the way of related
ideas was that from similar puzzles; " (b) motor habits "The mere
presence, in the case of change of conditions, of motor habits ap-
propriate to the new conditions did not necessitate positive trans-
fer," "The degree of positive transfer varied directly with the pre-
cision of analysis of the similarity of the new case to the old,"
"
In some cases a generalized formula developed in connection with
the first case was essential to effective transfer of motor habits to
later modifications of the first case," "Transfer was more effective
in those cases where the formula or general rule was developed in the
few trials, and where the formation of perceptual-motor habits
first
had been controlled and inter-penetrated by it from the start,

than when the generalization had been arrived at after those habits
had been set up."
h. Cross
+f.
Education. Cross education refers to the transfer of
practice from one organ of the body to^ bjj&terally symmetrical
organs, as for example the spread oTtraining from the right hand
to the left hand. A number of investigations have been made on
this problem which show that such transfer takes place to a very
great extent. Scripture, Smith, and Brown ('94) state that im-
provement in the strength of grip with one hand produced 80%
as much gain in the other. They also report that Volkmann found
that improvement in discrimination with the left arm was accom-
panied by approximately 80% as much gain in the other arm and
that other instances showed similar gains.
Davis ('98-'oo) measured the effect of practice in tapping Vvlih
the right great toe upon the rate of tapping with the right hand,
the left hand and the left great toe. He found that the left toe
improved 151% as much as the right toe, with which the prac-
ticing had been done, the right hand 100% as much, and the left
hand 83% as much. He also found that practice in gripping a
dynamometer with the one hand improved the other about 70%
as much. Practice in hitting a target 100 times with the right hand
improved the left hand about 75% as much.
Woodworth ('99) reports that practice in hitting dots with the
left hand improved the right hand about 50% as much and Swift
found that practice in tossing balls with the right hand caused the
left hand afterwards to improve in the same exercise more rapidly

than it would otherwise have done.
The writer ( io) measured the amount of transfer of improve-

J
ment in tracing with the right hand a star outline as seen in a

mirror to tracing the same outline with the left hand. The
test was made by having a tracing made first with the left hand,
then a series of 25 to 100 tracings with the right hand, and, at
the close, a tracing again with the left hand. The amount of
transfer to the left hand is approximately 90% of the total amount
of gain made by the right hand. The left hand improves nearly
as much as the right hand although all the practicing had
been done by the right hand. The results of experiments in cross
education are somewhat uncertain in their meaning so far as
transfer of training is concerned. Improvement in one organ,
which uniformly accompanied by a very large improvement
is
in bilaterally symmetrical organs, is probably due to the fact that

many common processes are involved in doing a task with two
bilateral organs. For example, in practicing with the hands, many
of the same sensory and neural processes are involved. Thus, the
same visual processes and the same visual brain centers would be con-
cerned. Likewise, it is also probable that neural innervations going
to the right hand in practice also go to the left hand. These would
tend to improve the control of the left hand without actual practice
with the left hand. The data on cross education probably have
only a distant and doubtful bearing upon the problem of transfer.
Criticism of the Technique of Experiments on Transfer of Train-
ing. There are three important elements in the technique of
experimentation in this field which have not been recognized by
investigators from the beginning and are not recognized by all
investigators even to-day, (i) The first is the length of the end
tests. These have been too long in some investigations to give
as full opportunity as possible to transference. This was one
of the difficulties in James' original experiment and was recog-
nized by James himself. (2) In the second place, the end tests
have not always been repeated on a control group of subjects.
This istrue of nearly all of the early studies. (3) In the third place,
many investigations do not mention the amount of improvement
made in the training series itself with which the gain in the end
tests may be compared. Failure to observe these precautions
makes impossible an accurate, quantitative interpretation of
many of the early researches and even of some of the recent ones.
Summary. It may perhaps be unwise in view of the intricacy

of the researches andtheir partial incongruity to attempt to sum-
marize general conclusions. However, a brief resume will help
to clarify the reader's thinking about these problems, (i) Practi-
cally every investigation shows that improvement in one mental
or neural function is accompanied by a greater or less amount of
modification in other functions. (2) This modification is in most
instances a positive transfer, that is, an improvement. Negative

transfer, that is, loss of efficiency in other functions, or interference,
has been reported principally among sensori-motor habits. (3)
The amount improvement in the capacity trained is probably
of
never accompanied by an equal amount of improvement in other
capacities, with the possible exception of a few isolated instances
whose actuality may be questioned. Thus, for example, Thorndike
and Woodworth found that the gain in various types of perception
or discrimination closely related to the type in which the training
took place was from o%
to about 40% as great as that made in
the particular kind of perception trained. In memory, Fracker's
results showed that the improvement in different sorts of memoriz-
ing, so similar to the training series that they were all but identical
with it,was about 75% as much as that made in the training
series;while the improvement made in the forms of memory rather
different from the training series was only about 15% as much
as that in the training series. Up to about 1890 when James re-
ported the first investigation on the problem of transference, it
was tacitly assumed by many writers that a very large share, if
not all, of the training derived from one sort of exercise was carried
over to other sorts of exercise. After the first investigations be-
came generally known, many writers went to the other extreme
and assumed that all training is entirely specialized and that
nothing carries over from one kind of practice to any other kind
of practice. As a general estimate, on the basis of experimental
work done thus far, the amount of transference between the ex-
tremes of 100% and o% of transfer lies nearer to the zero end and
is probably in the neighborhood of 20% to 30% of transfer to
closely allied functions and from that point on down to of o%

transference to more unlike functions. (4) In the fourth place, the
improvement spread to other functions diminishes very rapidly
in amount as these other functions become more and more un-
like the function specifically trained. This diminution occurs at a
surprisingly rapid rate.
How Does Take Place? If improvement in one

the Transfer
mental function accompanied by, or produces improvement in,
is
other functions, how may the change in these other mental func-
tions be explained? How does change in one function carry over
to others? Two general theories have been proposed: (i) The
theory of identical elements or special connections, and (2) the
theory of generalization or common capacities.
The theory of identical elements has been advocated by Thorn-
dike and may best be stated in his own words:
"The answer which I shall try to defend is that a change in one func-
tion alters any other only in so far as the two functions have as factors
identical elements. The change in the second function is in amount that
due to the change in the elements common to it and the first. The
change is simply the necessary result upon the second function of the
alteration of those of its factors which were elements of the first function,
and so were altered by its training. To take a concrete example, im-
provement in addition will alter one's ability in multiplication because
addition absolutely identical with a part of multiplication and because
is
certain other processes, e. g., eye movements and the inhibition of all
save arithmetical impulses, are in part common to the two functions.

"Chief amongst such identical elements of practical importance in
education arc associations including ideas about aims and ideas of method
and general principles, and associations involving elementary facts of
experience such as length, color, number, which are repeated again and
again in differing combinations.
"By identical elements are meant mental processes which have the
same cell action in the brain as their physical correlate. It is of course
often not possible to tell just what features of two mental abilities are
thus identical.But, as we shall see, there is rarely much trouble in

reaching an approximate decision in those cases where training is of
practical importance." (Thorndike/i4, II, pp. 358-359.)
The theory of generalization has been advocated by Judd in

the following manner:
"The important psychological fact ... is that the extent to which

a student generalizes his training is itself a measure of the degree
to which he has secured from any course the highest form of training.
One of the major characteristics of human intelligence is to be de-
nned by calling attention, as was pointed out in the chapter on science,
to the fact that a human being is able to generalize his experience.
James has discussed this matter by using the example of the animal
trained to open a particular latch. The animal becomes acquainted with
the necessary movements to open one door, but he never has the ability
to generalize this experience. He cannot sec that the same method of
opening doors is applicable to many other latches. The result is that the
animal goes through life with one particular narrow mode of behavior,
and exhibits his lack of intelligence by his inability to carry this single
type of skill over to the other cases which arc very familiar to the trained
human intelligence.
"James goes on to say that the same distinction appears when we con-
trast a trained scientific mind with the ordinary mind. The ordinary
thinker docs not see how to deal with a situation in terms of scientific
principles. James cites the his own experience with a smoking
example of
student-lamp. He discovered accident that the lamp would not
by
smoke if he put something under the chimney so as to increase the air
current, but he did not realize that what he had done was only one par-
ticular example of the general principle that combusion is favored by a
large supply of oxygen. The general principle and its useful application
belong to a sphere of thinking and experience which the untrained lay-
man has not yet mastered." (Judd, '15, pp. 413-414.)
A
The theory of identical elements is based on the doctrine that

learning or changes in mental capacities consist of the establish-
ment of specific connections or associations between various
specific elements. One form of exercise has influence upon another
capacity whenever connections established in the former may also
be used in the latter. In a certain sense the theory of identical
elements describes or explains transfer of training in a tangible,
concrete manner. In a certain other sense it does not explain
transfer of training at all or else it implies that there is no general
training in the sense in which formal disciplinarians use the term.
If special training is general or helps in performing various mental
activities only to the extent to which the special training has
elements in it which occur also in these other activities then there
is no spread of training to such activities in which no elements
are found which also appear in the capacity specifically trained.

The formal disciplinarian assumes that training of one sort affects
capacities of other sorts irrespective of identical elements or simi-
larity to the activities developed. In the last analysis the contro-
versy comes down to a question of fact, namely, to how dissimilar
activities does any given form of training spread? The theory of
identical elements, when the term identical elements is used in a
liberal manner, has the advantage of describing the situation in con-
crete, definite concepts and lends the interpreta-
itself fairly well to
tion of experimental results'. The discussion of the formal discipli-

narian is usually not in as tangible terms but is likely not to be very

differentfrom the statements of the experimentalist when the former
reduces his argument to specific terms.
The theory of generalization attempts to explain spread of
improvement in terms of the recognition of application of an ex-
perience obtained in one connection to other connections and is
probably more satisfactory to the formal disciplinarian. In the
author's opinion there is no necessary opposition between the
theory of identical elements and the theory of generalization.
The essential difference is in the emphasis upon the conscious rec-
ognition of identical elements in as many situations as possible.

Judd has emphasized this in connection with teaching:
"The first and most striking fact which is to be drawn from school
experience that one and the same subject-matter may be employed
is
with one and the same student with wholly different effects, according to
the mode of presentation. If the lesson is presented in one fashion it
will produce a very large transfer; whereas if it is presented in an entirely
different fashion it will be utterly barren of results for other phases of
mental life. It is quite possible to take one of the objects of nature study,
for example, and to teach it in such a way that it becomes an isolated and
utterly formal possession of the student. This has been illustrated time
and time again by the instruction which has been given in birds and
plants. A teacher can teach birds and plants in such a way as to arouse a
minimum of ideas in the student's mind. The training may be as formal
in these content subjects as it ever was in language instruction. On the
other hand, the same subject-matter may be taken by a different teacher,
and under other methods can be made vital for the student's whole
thinking. Thus the teacher who is dealing with birds as a subject of
nature study and secures an interest on the part of his students for the
world in which these birds live, through an examination of the structures
and habits of the birds, will have in this subject-matter one of the most
broadly interesting topics that can be taught. In exactly the same way a
teacher who knows how to make use of the materials given in a Latin
course may render this subject very broadly productive, as contrasted
with the teacher who merely gives the formal aspects of the subject.
Formalism and lack of transfer turn out to be not characteristics of sub-
jects of instruction, but rather products of the mode of instruction in
these subjects." (Judd, '15, pp. 412-413.)
two theories are not necessarily antagonis-

It seems then that the
tic but when sanely interpreted are useful supplements to each
other. The theory of identical elements has helped to make the
2l6 EDUCATIONAL PSYCHOLOGY
discussion of formal discipline or transfer of training concrete, and

the theory of generalization will help to emphasize the conscious
recognition of the identical elements in as many situations as
possible. Some writers have assumed that transfer is limited
to a conscious recognition of elements. This, however, is dis-
proved by some experiments with human beings and particularly

by the experiments with animals such as those reported by Webb.
CHAPTER XIV
TRANSFERENCE OF TRAINING IN ABILITIES IN SCHOOL
SUBJECTS
To what extent does the training of the capacities exercised by

school subjects carry over to capacities concerned in other school
subjects, and especially to the capacities involved in the usual
activities of life? This question brings the problem of transfer-
ence of training directly face to face with the issues of education
and is the form in which it is usually concerned in discussions of
mental discipline. It is of more special concern to the liberal

phases of education in the high school and the college for the reason
that the subjects taught in the elementary school, in the vocational
courses in the high school, and in the professional courses in the
university are directly pertinent to the common needs of life or
to the various occupations and professions. Most of the discus-
sion has, therefore, centered about the training value to be de-
rived from the traditional academic work of the high school and
the college.
General Opinions. The beliefs concerning transfer of training of
the capacities employed in school subjects have been largely matters
of opinion and not matters of fact. These opinions, held by persons
prominent in educational affairs, have been for the most part

rather uniformly confident in the faith that the exercise of the
mind upon the materials of the school subjects produces a very
profound improvement in mental powers as a whole. Typical of
such opinions are the following:
"But my opinions of the supreme educational value of the great dis-

ciplinary studies have not changed, and will never change.
"As a result of mylong experience in watching their effects on our
students I am absolutelyand irrevocably sure that certain subjects train
in thinking straight and reasoning clearly.
"I am absolutely sure that Latin and Greek, higher mathematics,
philosophy, the critical study of the literatures of different nations (and
the better the literature, the better the training it gives, Greek, Latin,
English, and French literatures leading all others in this respect, and in
the order named), economics and politics, especially on their theoretical
217
side, and English composition are thinking subjects of very high educa-
"
tional value." (Thomas, M. C., Old Fashioned Disciplines," Journal
of the Association of Collegiate Alumna, May, 1917, p. 588.)
In connection with the Classical Conference at Ann Arbor,

Michigan, 1909, Dr. Harvey Wiley sent out questions to 100
scientific men asking what they considered to be the value of Latin
and Greek as preparation for scientific pursuits. He received 35
replies distributed as follows:
Favorable to the study of Latin and Greek 14

Unfavorable to the study of Latin and Greek 17
Favorable to Latin, but not to Greek 4
Among the opinions expressed in this connection, Professor

R. P. Biglow made this statement:
"To summarize my opinions in the matter of a scientific education, it

seems to me that the essentials are of two classes: First, a thorough
by a scientific man, namely, the
training in the use of the tools required
modern languages and mathematics; second, a training in the scientific
method, especially as applied to the branch of science in which he desires
to specialize. If to the curriculum, the studies of classics can be added
without interfering with these essentials, then it seems to me that in some
cases it would be desirable as a means of culture."
Professor Neff of the University of Chicago regretted the time

he spent on Latin and Greek:
"I think everyone realizes as he grows older that he has his limitations.
I,for one, regret very keenly that I took a great deal of Latin and Greek
and did not spend far more time on advanced mathematics and physics.
I am, however, not now wasting any time in vain or useless regrets on
this account,but simply doing the best I can with the knowledge that I
have acquired."
The opinions of prominent business men were reported at the

same conference:
Mr. William Sloane, a New York business man:
"I believe that the slow processes of translation of the classics make
good training for the boy who has chosen a business career."
The Hon. J. W. Foster, of Washington:
"The mere routine labors of the translation of Greek and Latin authors
into one's vernacular, the effort to ascertain their exact meaning and the
choice of the words which correctly express that meaning, constitute a

mental training which will be invaluable to the future lawyer or public
Probably the most notable assemblage of opinions ever brought

together was that presented at the conference on classical studies
held at Princeton University in June, 191 7. 1 These opinions were
quoted from some 300 persons prominent in public life, business,
universities and colleges, schools, the ministry, law, medicine,
engineering, science, journalism, modern literature, history and
related branches, fine arts, and oriental studies. These statements
were substantially unanimous in bearing testimony to the value
of the classical languages.
But perhaps as many opinions on the other side from men of
equal intellect could be gathered. The point is that mere opinions
cannot yield a final appraisal of the training value of school studies.
Both favorable and unfavorable opinions are bound to be very
nearly worthless because at best they are apt to be prejudiced by
personal likes or dislikes and by exceptional instances of benefit
or lack of benefit from the pursuit of this or that particular subject,
and most no general observer has at hand sufficient,
of all because
complete evidence concerning the problem. Experi-

reliable or
mental and statistical data are hard enough to interpret because
of the complication of factors in the production of any type of
training to say nothing of the settlement of the controversy by
general impressions.
We cannot determine by ballot the shape of the earth, or the
value of a patent medicine, no matter how many testimonials may
be presented on the one side or the other. Men prominent in
life have testified to the benefits of patented remedies which
science has shown to be not only valueless but harmful.

Specific Estimates of the Value of School Studies. The writer
attempted to obtain specific estimates of the value of school sub-
jects according to the best judgment that could be exercised by
persons who are concerned with the work of public education.
These were obtained not with a view to contributing anything
toward the solution of the problem but for the purpose of examining
more precisely the drift of the consensus of opinions held by persons
immediately in charge of school work.
Any branch of learning may have thret possible values a
1
Reported in Value of the Classics, Princeton University Press.
disciplinary value, a utility value, and a cultural value. Thus

the study of English has a certain amount of disciplinary value in
training the mental capacities involved in the learning and under-
standing of the material studied; it has an obvious utility value in
acquiring the ability to use English correctly and effectively; and
it has a certain cultural value in acquainting the student with the
thought and life of mankind!

Estimates of the three values of each of the subjects listed in
the table were made as carefully as possible by fifty-eight super-
intendents, principals and teachers. In making these estimates
it was assumed that the pure, disciplinary value of the first year
of high school English as taught in the average way be equal to 10

and that all other values be estimated in terms of this assumption.
If the disciplinary value of algebra was considered to be twice as
great as that of English, hour for hour devoted to each, then it

should be estimated as 20. Or if the utility or cultural value of
English was considered greater or less than its disciplinary value,
the rating should be indicated accordingly. It was further assumed
that these were to be the values for the average boy or girl in the
high school.
At first glance it would seem that such judgments would be
rather uncertain and variable, and, as a matter of fact, they were
quite variable. Nevertheless, viewed from the statistical stand-
point, the judgments present a normal distribution spreading over
a wide range but clinging in large numbers about a central point.
For example, the judgments of the disciplinary value of American
history ranged as extremes from 3 to 30, with the largest number
of estimates on 10 and a gradual decrease in the number of estimates
on values farther and farther removed from 10. The median
estimate was 10 and the probable error was 3.5, that is, one-half
of the estimates were between 8 and 15. So that the judgments
about the various values, even though quite variable and usually
accompanied by a feeling of uncertainty so far as the individual
judge was concerned, were as reliable and as normal as judgments
about most matters are.
The following table gives the median judgment for each of the
values listed:
TABLE 55
Estimated values of school subjects. After Starch ('17)
An examination of the table reveals some interesting compari-

sons. The highest disciplinary value is assigned to geometry with
a rating of 20, or twice as high as that of English. Algebra is next
with a value of 19. It is rather surprising to find gymnastics and
football rated as high as they are. It is also interesting to note
that the disciplinary value of a pupil's earning his way through
school is rated higher than that of any of his studies. The lowest
disciplinary value is assigned to sewing and typewriting.
In the case of utility values, the highest rating is given to English,
cooking and sewing (30), a value approximately three times as
great as their disciplinary value. The lowest utility value is as-
signed to football (6) and the next Again the
to algebra (9). utility
value of a pupil's earning his way through school (36) is placed

above that of any of his studies.
In the case of cultural values, the highest rating is given to
music (25) and the next to English (22). The lowest value is as-
signed to shorthand, typewriting, and football (5). The cultural
value of a pupil's earning his way through school (18) is placed
below only music and English.
A similar study with reference only to the disciplinary aspect of
college studies was previously reported by Thorndike ('15). Esti-

mates obtained from 100 teachers are summarized thus:
"Philosophy (for freshmen) 8; English composition, 10; German,

Chemistry and Logic, n; Physics, 13; Latin, Greek, Mathematics, 16.
Waiting on table is rated at 3; athletics is rated at about 7; work for the
college paper at 8 or 8>; tutoring at 13 or i3>; and regular productive
work in the world as teacher, business man or skilled laborer at 14^-"
(Page 281.)
Such tabulations of opinions are valuable only in showing in

more accurate terms what teachers and educators think about
the question and not in really answering it. However, so long
as schools are operated by opinions, a combination of opinions
may be better than individual ones as guides of educational policies.
To what extent individual opinions are consciously or unconsciously
prejudiced by personal interests is shown by the fact that the
teachers overestimated by nearly one-half the value of their own
specialties as compared with the average values assigned by
teachers as a whole.
Experimental and Statistical Inquiries, a. Arithmetic. The
writer made an
investigation to measure the effect of improvement
in mental multiplication of three-place numbers by a one-place
number, doing 50 problems a day for 14 days, upon other types of

arithmetical processes. This experiment was carried out with
eight subjects who constituted the training group, and seven,
subjects who constituted the control group. The results are given in
the following table, which shows the percentages of gain of the
second end tests over the first:
TABLE 56. After Starch ('11)
The residual gain on the average was 29%. The average gain
made by the trained group in the practice series, comparing the
first day with the i4th day, was 112%. Hence the gain trans-
ferred to the allied arithmetical operations was only 26% of the
gain in the practice series itself. From one point of view, this
seems to be a very considerable amount of transfer, but when we
note that some of the end tests were as similar to the training
series as they could be without being identical with it the transfer
is small. We might expect almost a complete carrying over to
the closely similar operations but the largest amount of residual
gain took place in the multiplying of two-place numbers by a
one-place number and in the adding of three-place numbers; but
even there it was only slightly larger than the transfer to the other
operations.
Winch conducted a series of experiments to determine the amount
of transfer from improvement
in numerical computation to arith-
metical reasoning. In each experiment the class was divided into
two groups of approximately equal ability as shown by a previous
test in arithmetical reasoning. Then one-half of the class was
trained in "rule" sums after which a final test in arithmetical
reasoning was given alike to both groups.
The first class, composed of 1 3-year-old girls from a poor neigh-
borhood, showed improvement in numerical accuracy but no
transfer to arithmetical reasoning. The second class, composed of
lo-year-old girls from a poor neighborhood, showed considerable

improvement in accuracy and a doubtful transfer to reasoning.
The third class, composed of lo-year-old girls from a good neigh-
borhood but poor in arithmetic, showed transfer in three sections

but the opposite in a fourth section. In the fourth class, composed
of lo-year-old boys, both practiced and unpracticed groups showed
about equal gains in arithmetical reasoning. Winch concludes
from these experiments that improvement may take place in
numerical computation without any certainty of improvement in
arithmetical reasoning.
In a later experiment conducted on the same general plan, he
used 72 ten-year-old boys. Here again he found no evidence of
transfer from improvement in numerical computation to arith-
metical reasoning.
Carrie R. Squire made an experiment regarding the transfer of
neatness: "At the Montana State Normal College careful experi-
ments were undertaken to determine whether the habit of produc-
ing neat papers in arithmetic will function in reference to neat

written work in other studies; the tests were confined to the
intermediate grades. The results are almost startling in their
failure to show the slightest improvement in language and
spelling papers although the improvement in arithmetic papers
was noticeable from the first." (Bagley, The Educative Process,
p. 208.)
This experiment was repeated under the direction of Ruediger
with the difference that along with the specific training in neatness
in one particular study a general practice of neatness in daily life
was held up before the class as an ideal to be striven for. Care was
taken not to discuss neatness in the other classes. Sample papers
were taken in the one subject concerned and in two other subjects
before and after the training. The seventh grades of three schools,
located at widely different places and comprising 39 pupils in all,
were used in the experiment. The two schools which showed
an appreciable improvement in neatness in the study where special
training was given also showed considerable, though less, improve-
ment in the other studies. Thus the two schools showed an average
improvement of 4.75 points in the study where training was given
and 3.1 points, or 65% as much, in studies where nothing was said
about neatness.
b. Grammar. Grammar has been regarded as a highly effica-
cious instrument for training the functions of the mind. Thus
Commenius stated: "I presume that no one can raise any objection
to my placing (Latin) grammar first, since it is the key of all knowl-
edge." Locke said on the other hand: "I would fain have anyone
TRANSFERENCE OF TRAINING 22$
name to me that Tongue, that anyone can learn, or speak as he

should do, by the rules of Grammar. Languages were made not
by Rules, or Art, but by Accident, and the Common Use of the
People."
Claims for Grammar have been that it
1. Disciplines the mind.

2. Prepares for the study of other languages.
3. Gives command of an indispensable terminology.
4. Enables one to use better English.
5. Aids in the interpretation of literatures.
The Committee of Ten (1893) said: "The study of formal gram-

mar is valuable as training in thought, but has only an indirect
bearing on the art of writing and speaking."
What are the actual facts so far as any are available at the
present time? Briggs ('13) attempted to determine the extent to
which the various claims made for grammar are substantiated.
He outlined the following claims and devised an elaborate set of
tests to measure the effects of training in grammar.
"It is held that grammar trains children:
A. With rules and definitions:

1. To see likenesses and differences.
2. To critically test a definition.
3. To thoroughly apply a definition.
4. To make a rule or definition.
B. With reasoning:
5. To test reasons.
6. a. To take from a mass of data all that are necessary

and to use them in reaching a judgment.
b. To demand all necessary data before drawing a
conclusion.
7. To reason in other fields, e. g., arithmetic.
8. To reason syllogistically.
9. To detect "catches."
As illustrations of the nature of the tests we may cite the follow-
ing instances. For measuring the observation of likenesses and

differences, Briggs used such a test as this:
"One-half of the following z6 words are alike in one respect and in that
respect unlike all the others in the list. Find these eight words and mark
them with a check (V)."
biscuit pirate mountains men
oxen geese fathers-in-law factory
scholars knives vessel table
pole frame children mice
(8 are plurals)
To determine ability to judge definitions and to amend them,

he used as tests such statements as the following for shoe:
1. A portion of clothing.
2. Something black made of leather.
3. Something to wear on the feet.
4. A necessary article costing from $1.00 to $5.00 or $6.00.
These were given to 25 or 30 pupils in each grade from two
tests
to seven in the Horace Mann school. Each class was divided into
two divisions. Then for three months, three times a week, the
children of Division I were taught formal grammar. During the
same three months, the children of Division II had work in com-
position and language. They were then given the second set of
tests similar to the preliminary tests, after which the conditions
were reversed. Division II then had formal grammar and Divi-
sion I had language and composition work. At the conclusion of
this period, the first set of tests was again given to all of the chil-
dren. The upshot of the whole investigation is summarized by
Briggs in the following manner:
"As a result of this experiment it may safely be asserted that these

particular children after the amount of formal grammar that they had,
do not, as measured by the means employed, show in any of the abilities
improvement that
tested, may be attributed to their training in formal
grammar."
Hoyt ('06) made a study to determine the relation between the

knowledge of grammar, ability to interpret English, and ability
to write English. He employed three tests: One for grammar
consisting of ten questions on four stanzas of Gray's Elegy; the
second for testing ability to interpret English, consisting of a
statement of thought in four other stanzas of Gray's Elegy; and
the third for ascertaining ability in composition, consisting of
writing a composition in forty minutes. These tests were made
with 200 pupils in a high school in Indianapolis. All papers were
marked by two examiners according to the percentage method.

Correlations were then computed among the different tests, which
were as follows:
Grammar and composition 23

Grammar and interpretation 28
Interpretation and composition 32
These very low and indicate that a greater or

coefficients are
less amount knowledge of grammar is accompanied to only a
of
slight extent by greater or less ability to write a composition. The

same statement holds for the relations between the other compari-
sons of The fact that the pupil who knows more or less
abilities.
grammar writes respectively a slightly better or worse composition

is quite likely due to the fact that he is a better or poorer
pupil
rather than to any aid which knowledge of grammar may render
him in writing a composition.
Hoyt concludes that "... the teaching of grammar is of
avail in strengthening one's ability to use language."
little
The writer ('15) made a series of tests in formal grammar and

in correctness of English usage. The test in formal grammar
consisted of three parts: First, a passage in which the parts of
speech of as many successive words as possible were to be indicated
in three minutes; second, a passage in which the cases of nouns
and pronouns were to be indicated in three minutes; and third, a
passage in which the tenses and modes were to be indicated in
three minutes. The test for usage consisted of 100 sentences each
of which was stated in two ways. Both might be correct, both
might be incorrect, or one might be correct and the other incorrect.
Pupils were allowed fifteen minutes in which to indicate the cor-
rect expressions. The results are summarized in the following
tables.
The tests were made upon 54 university Juniors and Seniors and
146 high-school pupils. They gave the results shown in Table 57,
in which the scores for knowledge of grammar are the numbers of
the parts of speech, tenses, cases, and modes indicated correctly
in the specified period of time, and the scores for correctness of
usage are the numbers of sentences designated correctly in the

specified period of time.
VEARS OF FOREIG: NUMBKR OF AVERAGE SCORES FOR AVERAGE SCORES FOR

LANGUAGES bruoENfs KNOWLEDGE OF GRAMMAR COKRECTNbS Of USAGE
UNIVERSITY JUNIORS AND SENIORS
o 2 48.0
25 12 47-8 71. 1
6-9 25 58.6 75-5
10 15 IS 63.4 75- 7
HIGH SCHOOL PUPILS

o 12 14.7 32.2
8 weeks 50 20.8 43-0
1 year 18 25-5 43-4
2 years 39 24.8 4S.9
3 years 27 28.6 47-7
UNIVERSITY JUNIORS AND SENIORS

YEARS OF LATIN
O IS 45-8 70.9
i3 ii 56.1 75-7
4 14 57-5 74-3
5 or more 9 51.8 76.1
Another test for correctness of usage, consisting of sentences

like the set of one hundred but arranged in the order of increas-
was made on another group of 146 university
ingly difficult steps,
students and 92 high-school pupils. This test yielded the results
given in the following table. The scores are the numbers of the
highest steps passed. The higher the score is, the greater is the
ability of using English correctly.
TABLE 58
YXA&S or LATIN NUMBER OF PUPILS AVERAGE SCORES
UNIVERSITY STUDENTS
o 47 IO.I
1-6 99 10.2
HIGH SCHOOL PUPILS

78 9.0
14 9-3
These tables agree in showing one very significant result, namely,

that the study of foreign languages materially increases a pupil's
knowledge of English grammar but only slightly increases his
ability in the correct usage of the English language. Notice, for
example, the upper part of Table 57. The students who had
10 to 15 years of foreign languages made a score in grammatical
knowledge of 63, as compared with a score of 47.8 made by the
students who had years of foreign languages, a difference of
2 to 5
32.6% former group. For correctness of usage, the

in favor of the
corresponding difference is only 6.4%. The two students with no

foreign languages made high scores because they were exception-
ally good students, but they are too few in number to be con-
sidered. The high-school pupils show a gain in grammatical
knowledge of 37.5% from the 8-week group to the 3-year group
and a gain in usage of only 10.9%. The twelve pupils with no for-
eign language made low scores because they were exceptionally poor
pupils. This is indicated by their low scholarship records, by the
fact that many were over-age, by the fact that they avoided the
foreign languages, and also by the large difference between their
scores and those of the 50 pupils who were just beginning foreign
languages. Eight weeks of foreign languages could hardly have
produced such a big gain. Their higher scores must be due largely
to a difference in original nature.
c. Foreign Languages. Extensive and confident claims have been
made for the value of general mental training to be derived from the
study of languages. Thus Lodge states the value of the study of
Latin as follows:
"
Far above every other subjectit trains (i) the process of observation,
(2) the function of correct record, (3) the reasoning power and general
intelligence in correct inference from recorded observation. To this
should be added its great value in developing the power of voluntary
attention.
"The value of Latin as a practical subject has to do particularly with
the effect of the language in the cultivation of English style. In the
English vocabulary a very large proportion of words in everyday use are
of Latin origin, and it has been estimated that two-thirds of the Latin
vocabulary of the classical period has in some form or other come over
into English speech. For the correct use of synonyms in English and
the habit of expressing one's thoughts clearly, concisely, and cogently, a
discriminating knowledge of Latin is indispensable, and while not every
pupil in the school may be expected to develop a good style, nevertheless
he should be given the necessary foundation for it.
"When we turn to literature, we find that Latin is influential every-
where particularly in our classical authors by allusions, by quota-
tions, by actual domestication. Many of our great English writers are
permeated with Latin. We cannot expect that all will desire to feed
theirminds on the works of our greatest authors, however much we

might prefer it; but certainly we should not deprive them of one of the
most important elements in their enjoyment should they be so minded."
(Lodge, p. 388, in Principles of Secondary Education, Edited by Paul
Monroe.)
Swift measured the progress in learning a new language made by

pupils with different amounts of previous language study to de-
termine, if possible, the advantage in beginning a new language to
100%
oo
80
bo 70
M 60
60
A -One Year of Latin and German.
B-One Year of Latin.
C- Spanish Only.
2 3 4 5 6 7 8 9 10 11 12 13 14 16
Weeks
FIG. 55. Progress of pupils in learning Spanish. After Swift ('06).
be derived from the previous study of another language. The ex-

periment was carried out with two classes, composed of 24 boys
and 24 girls in a St. Louis high school, who were beginning the study
of Spanish. Weekly tests were made to measure their progress and
individual abilities in learning Spanish. The classes were taught in
the usual manner and the pupils knew nothing of the purpose of the
tests. A record of progress was kept for the first 15 weeks. The
results are shown in the following graphs (Figure 55), which in-
dicate the relative progress of the three groups, namely, those who
had previously studied one year of Latin and one year of German,
those who had only one year of Latin, and those who had never
studied a foreign language before. The "Latin" and the "Latin
and German " groups stood considerably higher at the beginning
than the "Spanish only" group, but the "Spanish only" group
gained gradually so that at the end of the 15 weeks it had made up
about two- thirds of the difference. Swift concludes:
"The number included in these tests was too small to serve as a basis
for anything more than tentative conclusions, but the results certainly
open the question whether the advantage to beginners of a new language,
so generally thought to accrue from the study of Latin, may not be due,
chiefly if not solely, to grammatical information that would be carried
over from one language to another, and which would naturally help
enormously at the start. In acquiring facility in the use of the Spanish
gender, to cite one example, Latin would aid materially, since the ma-
jority of Latin feminincs are feminine in Spanish, and a large part of
Latin masculines and neuters become masculine in Spanish. The de-
clension of Spanish adjectives for gender and number, and their agree-
ment, in these respects, with their nouns, would give Latin students a
further advantage. The teacher of the Spanish classes noted that more
frequent and detailed explanations of case were needed by those who
had not studied Latin. The order of words, also, was more readily
mastered by those familiar with the Latin arrangement. Finally, in
learning the conjugations and in understanding the significance of tenses,
the assistance of the information acquired under these topics in Latin
was found to be especially great. The indications, however, are that the
higher records made by the Latin and German pupils were the result of
the substance of language information obtained from these studies rather
1 "
than of so-called
any 'mental
'language* or discipline. (Swift, '06,
pp. 250 ff.)
The writer ('15) made a comparison of the scholastic records

of university students who had entered the university with two
to four years of Latin with the records of those who had entered
with two to four years of German. The average grade for the
four years of college work of each of the graduates of the Col-
lege of Letters and Science of the year 1910 was computed. The
median mark of the 104 students who had entered the university
with Latin was 85.7 and the median mark of the 45 students who
had entered with German was 84.0. Hence the difference between
the two groups was only 1.7 points.
The explanation for this small advantage of Latin over German
may be sought in three directions: First, the disciplinary difference

between Latin and German is either zero or very small. Second,

whatever difference they may have produced originally may have
tended to disappear in the four years of college work, owing to the
freedom of electives, pursuit of different courses, disciplinary effect
of other studies, etc. Third, the small difference in scholastic
records may be due to an original difference in the students them-
selves, owing to the possibility that one language may attract a
better class of pupils than another.
To determine what part, if any, the first two factors played, the
average grade of each of the 738 Freshmen of the year 1909-1910
was computed. The median grade of the 416 Freshmen who had
entered with Latin was 82.4 and that of the 322 Freshmen who had
entered with German was 81.0. Hence the difference between the
two groups was only 1.4 points, or approximately the same as that
for the graduates.
The next problem was to compare the grades of these two groups
in specific subjects as follows:
Median grade in modern languages of 362 Freshmen who had entered

with Latin 84 5
.
Median grade in modern languages of 293 Freshmen who had entered

with German 82.3
Difference in favor of the Latin group 2.2

Median grade in Freshman English of 54 students who had entered with
Latin only 83 .9
Median grade in Freshman English of 97 students who had entered with
German only 82.7
Difference in favor of the Latin group 1.2

Median grade in first-year French of 27 Freshmen who had entered with
Latin only 81.5
Median grade in first-year French of 34 Freshmen who had entered with
German only 82 .o
Difference in favor of the German group 5
The very small. The claim of language

differences again are
teachers, socommonly made, that beginners in French who have
had Latin are much superior to those who have not had Latin, or
that students in English with previous training in Latin are superior
to those without such training is ill founded.
Another tabulation was made to show the scholarship records of
Freshmen in relation to the amount of foreign language studied,

irrespective of what the languages were.

YEARS OF FOREIGN NUMBER OF MEDIAN GRADE IN ALL
LANGUAGES SIUUKNTS FILL.SUMAN bruoiEs
o 25 81.8
1-2 224 81.9
3-4 195 83-05
5 6 155 84.0
The next problem was to measure the extent to which a pupil's

English vocabulary increased through the study of Latin. The
is
method employed measured the percentage of words of the entire

English vocabulary, as well as the approximate absolute number of
words, whose meaning a person knows. The test was made with
189 university students and with 46 Juniors in the Madison High
School.

%
Size of English vocabulary of 139 university students who had studied
Latin 60.9
Size of English vocabulary of 50 university students who had not studied
Latin 58.2
Size of English vocabulary of 14 high-school Juniors who had studied
Latin 54.7
Size of English vocabulary of 32 high-school Juniors who had not studied
Latin 50.2
The between the Latin and the no-Latin groups are

differences
surprisingly small. Nevertheless, the study of Latin does produce

an appreciably larger English vocabulary. This advantage becomes
less in university students with whom it is partly counterbalanced
by the increase in vocabulary due to wider experience.
Partridge compared the standing in the regents' third year Eng-
lishexamination of 783 pupils by dividing them into groups ac-
cording to the number of years of Latin they had studied. His
tabulation is as follows:
TABLE 62. After Partridge ('15)
The entire 783 papers divided on a basis of the number of years Latin was
studied
Number years studied o i 2 3

Number papers written 181 123 220 259
Average standing (percentage) 65 65 69 76
The pupils with no Latin may have had one or more years of
other languages and consequently Partridge presents the following
table for students who had studied Latin only:

Includes papers of pupils having Latin only and no other language to their
credit. Total number of papers, 167
No. of years studied o i 2 3

No. of papers written 28 25 42 72
Average standing (per cent) . .
63 61 69 78

Includes papers of pupils having German or French only and no other lan-
guage to their credit. Total number of papers, 176
1
No. of years studied o i 2 3
No. of papers written 28 41 57 50
Average standing (percentage) 63 61 65 68
Partridge believes that the "superiority of the classical over the

non-classical pupils is due not solely to initial natural ability, but
to the training received in Latin." He has, however, failed to show
the differences in initial ability and consequently any inference of
this sort is doubtful.
Harris made a study of the effect of knowledge of Latin upon
ability to spell English words by submitting a list of 50 words of
Latin origin to 324 freshmen in the University of Illinois. He gives
the following table:
TABLE 65. After Harris ('15)
He further submitted to the same group of students 10 words

of Latin origin which were to be defined. This test gave the follow-
ing result:
1
It is obvious that the o columns in these two tables will contain record of exactly
the same pupils.
He also compared the grades in Rhetoric of students who had hau

various amounts of Latin as follows:
Harris concludes: "From these various results the conclusions

in so far as these students are concerned, are obvious. In all fields
the four-year Latin students showed a marked lead, and in all but
the spelling which I have considered above there is a steady re-
trogression although for the practical purposes the one-and-two-
year Latin students might be classed together."
The interpretation of these figures is by no means so obvious.
Harris has made no allowance for the native superiority of the stu-
dents with more years of Latin study. In fact, the probability is
that, if we may infer from other studies in which such a deduction
has been made, a large part of the superiority is due to original
nature. Harris's results as they stand prove little or nothing con-
cerning the effect of training in Latin.
F. M. Foster performed a similar experiment at the University of
Iowa; 503 freshmen, about equally divided between the sexes, were
given a spelling test of forty words of Latin derivation. The results
are given in the following table:
TABLE 68. After Foster ('17)
Number of years of Latin o i 2 3 4

Average % of errors (girls) 23 28 25 24 17
Average % of errors (boys) 39 37 29 28 27
As in the study made by Harris there appears to be a decided re-

lationship between ability to spell words of Latin derivation and the
number of years devoted to Latin. In this case, however, it hap-
pened that Professor Irving King had previously given intelligence
tests to these same students by means of which it is possible to se-
cure a more accurate notion of the forces really producing the
better scores of the Latin students. The following table shows the
relation between the index of intelligence (that is, the percentage
above or below the average adult intelligence), the number of years
spent in the study of Latin and the ability to spell words of Latin
derivation for the two extreme Latin groups.
TABLE 69. After Foster ('17)
Best 14 of 4 year Latin girls, mental ability av. 25.8, spelling av. 5% error
This table shows clearly that students who chose to study Latin
had on the average a distinctly better native intelligence than the
non-Latin students and that the ability to spell words of Latin der-
ivation was to a considerable extent due to this superior intelligence
rather than to the study of Latin.
The secretary of the College Entrance Examination Board made
an extensive tabulation of the records of the classical and the non-
classical students who took the examinations in 1914, 1915, and
1916. The classical students are theones who offered Latin or
Greek, or both; non-classical students are those who offered neither
Latin nor Greek. A total of 21,103 candidates are concerned in the
following table which is based on the marks in all subjects except
Latin and Greek (reported in Value of Classics, 1917, p. 366):
COMBINED RATINGS IN ALL THE NON-CLASSICAL SUBJECTS

Candidates who obtained a rating of 90 to 100:
2.95% of all the classical candidates.

2.05% of all the non-classical candidates.
The classical students show a superiority of 44%.

17.31% of all the classical candidates.

The classical students show a superiority of about 40%.

51.96% of all the classic-til candidates

The classical students show a superiority of about 27%.

A committee in connection with the Princeton Conference made a

comparison of the honors received by classical and non-classical
students upon graduation from high schools, academies and colleges.
The table is based upon 2,799 classical and 5,606 non-classical
students from 19 high schools and academies, and upon 4,092
classical and 2,003 non-classical students from 17 colleges and
universities:
"The combined data from the nineteen high schools and academies
reporting yield the following results:
"Students receiving High Honors at Graduation were 18% of all the
classical students, but only 7.2% of all the non-classical students.
"That is: the classical students show a superiority ol 150%.
"Students receiving Honors at Graduation were 32.1% of all tha
"That is: the classical students show a superiority of 36.7%.
"Students receiving Honors or Prizes for Debating, Speaking or
Essay-writing were 8.8% of all the classical students, but only 3.5% oj
all the non-classical students.
"That is: the classical students show a superiority of 150%."
"The combined data from the seventeen colleges and universities
reporting yield the following results:
"Students receiving High Honors at Graduation were 17.3% of all
the classical students, but only 6.6% of all the non-classical students.
"That is: the classical students show a superiority of 162%.
"Students receiving Honors at Graduation were 46.5% of all the
"Students elected to Phi Beta Kappa were 16.8% of all the classical
students, but only 8.9% of all the non-classical students.
"Students winning Prizes or Honors for Scholarship in Other than
Classical Subjects were 13.5% of all the classical students, but only
9.3% of all the non-classical students.
"Students serving on the Editorial Boards of Student Newspapers and

Magazines were 15.1% of all the classical students, but only 9.2% of all
the non-classical students.
"Students acting as Members of Intercollegiate Debating Teams were
5.1% of all the classical students, but only 3.2% of all the non-classical
students.
"
That is: the classical students show a superiority of 59.4%." (Reported
in Value of Classics, pp. .381-383.)
These statistics are interesting enough, but they represent a

mingling of training and native ability in the superiority shown and
as such do not afford conclusive proof for the efficacy of classical
training.
Wilcox ('17) made an inquiry with the endeavor to ascertain the
amount of difference in original capacities. If the superiority of the
Latin students is due to their study of Latin, we ought to find that
they were not superior, or at least not as much superior, to the
other students before they undertook Latin. If, however, we should
find that the Latin students were as superior before they studied
Latin as afterwards, we may infer that Latin had nothing to do
with their superiority.
Wilcox tabulated the records of pupils in the Iowa City High
School graduating during a period of ten years. He tabulated
separately the grades made by all the students with Latin or Ger-
man, but not with both. These results are shown in the following
table in which the numbers were obtained by transposing the
symbols E, G, M, P, and F into numerical values of 4, 3, 2, i, and
o respectively.
TABLE 70. After Wilcox ('17)
Median grades in English of Iowa City High School students who studied
Latin or German
The comparison is graphically shown in Figure 56. It will be

noticed that the superiority of the classical group is found in the
freshman year and continues throughout the course.
A comparison was also made of the English grades of students
having four, three, and two years of Latin. This is shown in
Median grades in English ofIowa City High School students who had 4, 3,
and 2 years of Latin
A graphical comparison is shown in Figure 57. It is evident

that those who were destined to take four years of Latin were
already in their freshman year clearly superior to those taking
less Latin.
Wilcox made a similar investigation of the graduates of the
high school of Cedar Rapids, Iowa.
1234
Years of High School
FIG. 56. Median grades in Eng-

1284
Years of High School,
Fie. 57. Median grade in Eng-

lish in the four successive high-school lish in the four successive high-school
years of students taking Latin or Ger- years of students who had 4, 3, or 2

man. After Wilcox ('17). years of Latin. After Wilcox ('17).
Comparisons were made of the grades of 150 graduates having

Latin,German or no foreign language. This is shown in Table
72.
TABLE 72. AfterWilcox'('i7)
Median grades in English ofCedar Rapids High School students who studied
Latin, German or no foreign language
It will be observed that in the freshman year the classical group

was superior to the other two, but that by the senior year there was
very little difference in the three groups.
In Table 73, comparison is made of the English grades of stu-
dents having four, three or two years of Latin.
Median grades in English of Cedar Rapids High School students with 4, 3,

or 2 years of Latin
Here again the people with four years of Latin maintain their
superiority throughout the course.

"
Conclusions: It seems evident, so far as the Iowa City and Cedar
Rapids high schools are concerned, that the frequently demon-
strated superiority of students who have had Latin is not due to the
special discipline or training secured in the study of Latin. It is
probably due to the fact that, as a whole, the students who elect
Latin are somewhat superior to those who refuse to take it."
Perkins ('14) made an investigation to determine the effect of
emphasizing the derivation of English words from Latin words in
the instruction of Latin in the commercial course in the Dorchester,
Massachusetts, high school. This investigation was designed to
eliminate as far as possible the differences in original abilities
between pupils with Latin and without Latin. His report follows:
"Obviously, the first step was to select two sets of pupils of equal
ability,one set in the second year of Latin, and the other in the second
year of a modern language. Accordingly, we chose pupils such that each
group had virtually the same average mark in Latin, on the one hand,
and modern language, on the other, and also in English, with the result,
in actual figures, that the non-Latin group in the two studies averaged
0.5 of i% the higher. To make doubly sure that the Latin pupils were
not favored, the non-Latin group were taken from the section of Mr. Mur-
dock, a classical scholar, who in his English teaching emphasizes the
Latin element in the language. There were twenty-one pupils in each
set, all in the second year class of the school.
"Five measurements were made, one in spelling, one of the use of
words in sentences, the third in definitions and parts of speech, the
fourth in the meaning of words and spelling, and the fifth in excellence
in vocabulary.
"Miss Humphrey selected the words in Nos. 1-4, and the subject in
No. 5. In Nos. i and 2 the words were taken from the 600 or 800 deriva-
tives in the notebooks of a fourth-year pupil of the class, who was ex-
cluded from the measurements. Moreover, to be fair to the non-Latin
group, care was taken not to select words too difficult. In No. 3 the
words were taken from the 'Tale of Two Cities' which the pupils of both
groups were reading at the time in connection with their work in Eng-
lish II. Of the twenty words in No. 4, ten were taken ficm the 'Tale of
Two and ten from other sources. The subject in No. 5 was,
Cities'
1
'What do best. The papers were marked by teachers in the
I like to
English department and the results given to me. Altogether, six teachers
of English assisted in the measurements.
"To these five measurements is added a sixth in my opinion most
impressive of all. This test was made last June, shortly after I had
received Professor Holmes's letter, by Miss Gormley, with her pupils in
English II. As it happened, Miss Gormley, who was also the 'home-
room* teacher of all the pupils and consequently had access to their
marks, in making up the two groups to be composed of pupils of equal
ability,took into account not only foreign language and English II, as
was the case in measurements 1-5, but also the studies the pupils had
taken during the year. Hence we have even more reason in this case
than in the others to assume that the pupils were of equal ability. In
each set there were seventeen second-year students. The words were
taken entirely from Franklin's Autobiography and Silas Marner which
all were reading at the time. The Latin pupils were selected from the
first class Ihad had in the subject, just as they were completing the
course at the end of the second year.
"The result of the six measurements were as follows:
AVERAGES
LATIN NON-LATIN
PER CENT PER CKNT
January and February, 1914:
1. Spelling 82.5 72.6
2. Use of words in sentences 57.5 40 6
.
3. Definitions and parts of speech 69 5

.
33 3
.
4. Meaning of words and spelling 57.0 27 5

.
5. Excellence in vocabulary 36.0 6.8

June, 1913:
6. Meaning of words and spelling 63 .3 12.3
367.8 193-1
Averages 61 .3 32 18
.
32.18
Difference 29. 12%
"In No. i, the spelling measurement, the words were not difficult,
but such as ordinary pupils of sixteen should know something about,
' *
whether they had studied Latin or not âs valedictory/ competition,'
'occurrence,' 'benevolence,' 'legible.'
"In No. 2, the pupils composed
sentences containing the derivatives,
some of which, in this measurement also, ought not to be unfamiliar to
1
non-Latin pupils in their second year of English, as 'impediment,
'advocate/ 'reference/ 'anticipate/ 'subside.'
"In the third measurement, the difference in the averages of the two
groups 69.5% and 33.3% was so great that Miss Humphrey thought
that perhaps too difficult words had been selected, or at least words which
placed the non-Latin students at an unreasonable disadvantage. Cu-
riously enough, in this measurement the words were taken, not from the
notebooks of a Latin pupil, as in the first two tests, in which the difference
between the two groups was much less, but, as stated above, from The
Tale of Two Cities. Furthermore, in No. 3, the non-Latin pupils were
so far afield in giving accurate definitions, and so confused in classifying
the words as to parts of speech, that it was decided to give another test
in which they should be asked, not to define words, but to give their
meanings, with the parts of speech omitted entirely. The results in this
measurement 57% and 27.5% were virtually the same as in No. 3.
"
Since practically every second-year pupil could write at least passably
on such a subject as 'What do I like to do best* it was decided to make
the basis of comparison in No. 5, not the average of the two groups, but
the percentage of rating above the passing mark. Moreover, in this
vocabulary test, emphasis was laid, not merely upon words of Latin
origin, but upon any words out of the ordinary, from whatever source.
The wide difference in the results from the view-point of excellence in
vocabulary 36.0% and 6.8% shows clearly what I have always be-
lieved and maintained, namely, that the work in commercial Latin
necessarily gives the pupils the dictionary habit, the results of which
extend far beyond the Latin derivatives actually studied.
"Of all the measurements, No. 6, was perhaps the most convincing.
In this test, the Latin pupils, unlike those in Nos. 1-5, had had during
the last six months of the two years' course the benefit of drill in a vocab-
ulary not in the commonest use and yet valuable and even necessary to
educated people. The list of words was taken entirely from Franklin's
Autobiography and Silas Marner which the pupils had just read, and
was not of unusual difficulty, consisting of such words, for example, as
asperity, promiscuous, mortuary. Yet by referring to the results it will
be seen that to the non-Latin group of pupils such words were practically
meaningless.
"An examination of the marks on these tests may prove of interest.
Among the seventeen non-Latin students the highest grade was 30%, and
five zeroswere recorded. In the Latin group, on the other hand, the
lowest mark was 30%, while one pupil received 100%, two 90%, two 80%,
five 70%, and only three had below 50%. The difference in averages of
the two groups was 53%." (Perkins, '14, pp. 11-14.)
This investigation is interesting and one of the few whose results

were carefully worked out to make a precise comparison after elimi-
nating differences in original capacity. It is a question, however, to

what extent the Latin students may have been favored by the
manner in which the words for the various tests were selected. For
tests i and 2 they were taken from the lists compiled and studied
by the Latin pupils. Even when the words are selected from Eng-
lish sources such as Silas Marner and Franklin's Autobiography
there is still the question as to the particular words chosen for the
test. It is obviously unfair to select words which are relatively rare
and whose meaning may readily be inferred from their origin. To
what extent the words were selected fairly cannot be judged since
Perkins docs not give the lists of words used.
The writer ('17) undertook a study to determine as precisely
as possible, the relative shares contributed by language training
and by original ability toward proficiency in English composition.
A series of tests was carried out with a group of 177 university
students. These tests together with their findings are given in
Table 74.
No. 3 consisted in writing an extemporaneous composition within
a limited time. These compositions were rated by three judges by
the Hillegas Scale.
No. 4 gives the average number of words written by each group
of students.
No. 5 gives the average number of different words used in each
composition.
No. 6 was a test in speed of reading. The numbers refer to the
words read per second.
No. 7 gives the number of words written in reproducing the
thought of the passage read in No. 6.
No. 8 gives the number of A's canceled in one minute in the well-
known A-test.
No. 9 gives the scores made in canceling in one minute a certain
geometrical form on a page of similar forms.
No. 10 consisted in reading to the persons a series of ten words
to see how many they could recall immediately afterwards.
No. ii was carried out by giving a stimulus word and having the
persons write as many associated words as they could in thirty
seconds.
No. 12 consisted in giving a series of ten words and allowing
fifteen seconds to each word for writing as many synonyms as
possible.
. No. 13 was a set of tests in imaging geometrical forms.
No. 14 consisted in presenting to the subjects seven words, one

at a time, spelled orally backwards by the experimenter. The
persons wrote down the words which they could thus recognize.
A general inspection of the table shows that there is a steady in-

crease in the scores of practically every test from left to right with
increasing years of language study. Thus the 9-1 5 year group wrote
considerably better and longer compositions than the 1-2 year
group. The purpose of tests Nos. 6-14 was to ascertain to what
extent this superiority was due to original superiority of ability
or to the effect of language training. Tests 8-14 were selected par-
ticularly because the capacities to do them would probably be
affected very slightly if at all by language training. These show
on the whole a distinct superiority of inherent ability on the part
of the groups who studied languages for longer periods of years.
In order to make a crucial comparison as to how much of the
greater composition ability was due to the greater original capacity
of the pupils and how much was due to their greater training in
language, the grades received by these students in all the subjects

carried during the first year of the high school were obtained from
the entrance records of the University. The amounts of difference
in original ability of the groups who later pursued varying amounts
of language work would be definitely indicated by this method,
since at that time none had had more than one year of foreign
language. The difference in the scholastic grades at the end of the
first year of the high school between those who later pursued lan-
guages for a total of 9-15 years and those who pursued languages
for a total of 1-2 years could certainly not be due to language train-
ing.
Row 1 6 gives for the different groups the average scholarship
grades during the first year of the high school. It will be noted that
there is a steady increase from group to group. The 9-15 year
group had an average grade of 88.0, or five points higher than the
1-2 year group.
The next problem was to compare in common terms the five
points of difference in scholarship on the percentage scale with the
difference of 10.6 in quality of composition as measured by the Hille-
gas scale. To reduce these two types of measurements to commen-
surate units, fifty-eight compositions were rated by four persons
both by the percentage method and by the Hillegas scale. By a
process of equating values it was found that i.o point on the per-
centage scale equals 2.1 on the Hillegas scale. The difference of

five points, percentage scale, in original scholarship between the
1-2 year group and the 9-15 year group would be 10.5 in terms of
the Hillegas scale. The surprising result seems to be that the differ-
ence of 10.6, Hillegas scale, in quality of composition between
these two groups is approximately equaled by 10.5, the difference
in original scholarship when expressed in terms of the Hillegas
scale. The conclusion seems, therefore, unavoidable that the
difference in ability in English composition is due practically en-
tirely to a difference in original ability and only to a slight or no

extent to the training in foreign languages. [For the method of this
computation the reader is referred to the original report of this in-
vestigation. ('17.)]
The increase in length of composition and in speed of reading is
large and very probably in excess of the difference in original ability.
Training in foreign language seems to have produced a distinct
effect in greater fluency of words in writing and in more rapid per-
ception of words in reading.

Miss M. Theresa Dallam ('17), a teacher of English in the

Western High School of Baltimore, made a series of experiments
"
on her pupils to test the truth of her conviction that Latin classes
do much better work in English than the classes that have not
studied Latin." Out of 114 students she selected 34 fourth-year
students, 17 Latin and 17 no-Latin or modern language pupils,
by pairing them on the basis of their general scholarship records,
so that the two groups would be equal in mental ability. The Latin
group had an average of 78 and the no-Latin group 79. Kelley's
Silent Reading test was also made with them. The Latin group
made an average score of 31.1 and the no-Latin group 33.1. The
two groups were, therefore, very nearly equal in general ability.
The Latin group had studied the language for four years.
Miss Dallam then made seven tests: spelling, reproduction, dic-
tation, Latin derivations, definitions, compositions, and English
grammar. The results were as follows:
TABLE 75. After Dallam ('17)
Average percentages attained
Thus the Latin group made a distinctly higher average in deriva-

tions and grammar, an appreciably higher grade in compositions;
a slightly higher average in spelling, reproduction, and definitions;
and a slightly lower average in dictation.
Miss Dallam then computed coefficients of correlation for the
Latin group between their grades in Latin and each of the seven
tests, and for the modern language group between their grades in
modern languages and each of the seven tests. These correlations
were as follows:
TABLE 76. After Dallam ('17)
Coefficients of correlation
These correlations are so low that, with the exception of the ones
for grammar, no significance can be attached to them except to say
that there is practically no correlation between the various com-
parisons made and that the Latin group shows no superiority over
the modern language group in spelling, reproduction, definitions,
and a doubtful superiority in dictation, derivations, compositions,
and grammar. The differences that are shown are non-committal
and so small that they would have to be substantiated with other
groups to be conclusive.
d. Science. Claims of general training to be derived from the
pursuit of the sciences are practically as extensive and confident'

as those made for the languages. The only difference is that the
claims for the sciences have not been questioned or investigated
as much as those made for the languages. Bagley has summarized
them in the following manner:
"i. The formation of some useful specific habits, through training,
routine, rationalized practice.
"2. The acquisition of useful information, through methodical
study, instruction, and drill.
"3. The adoption of valuable ideals, or 'emotionalized standards,'

inculcated through the inspiration to be gained from the teacher, from
the lives of great scientists, and from experiences of intimate contact
with nature.
"4. The acquisition of facility in the use of facts, ideas, and methodical
thought processes, for the solution of problems, the overcoming of
difficulties, and the accomplishment of worthy purposes, through the
mental discipline afforded by properly graded practice in the solving of
scientific problems.
"5. The development of taste, and power of appreciation, to be
gained through a clear apprehension of unity, adaptation, economy,
order, and system in nature as interpreted by science.
"6. The development of scientific or philosophic insights, perspectives,
and attitudes of mind that serve as safeguards to the intelligent inter-
pretation of contemporary life, through acquaintance with systems of
'
1 1 .)
organized knowledge." (Bagley,
One of the important values claimed for the sciences is the general
training of accuracy and fidelity in observation and the transfer-
ence of this particular type of training to other types of observation.
Miss Hewins ('16) made an attempt to measure the extent to which
this improvement is general or carries over to other types of ob-
servation. She divided each of three classes in botany, composed
of 34 boys and 50 girls, in the first year of a New York high school,
into two groups, and gave them a series of tests in various kinds of
observation of a biological and non-biological nature as follows:
Series i
Practice series
MATERIAL RECORDING
1 May 15 Description of
" "
lilac flower
2 16 box-elder leaf 10 minutes
317 tt tt
jj lc s t em>
of gill-run-over-the-ground
\ eil
^ ancj flower for each
test
4 20 Description of flower stalk and flowers

of the lily-of-the-vallcy
S 21 Description of the horse chestnut flower
" " "
6 22 buttercup flower
" " "
7 23 mustard flower
8
" " "
24 dogwood flower
" " " dcutzia flower
9 27
" " "
TO 28 columbine flower
Scries 2
The practice series was continued for a period of ten days with
one section of each of the three classes. The other section of each
class answered questions from books on the material of the lessons.
At the end of the period, she gave the tests in Series 2 to both
sections of each class and obtained the following gains in scores
over Series i:
TABLE 77. After Hcwins ('16)
Gross gains in scores
(Median score in biological tests in Series i before practice 18.7)

" "
(
"
"non- " " " " " "
39.0)
Miss Hewins infers a considerable gain in both the biological and

non-biological types of observation.
"Table 77 shows that in the biological tests, the average gain of each
practisedboy was 8.06 per test for the 5 tests while the unpractised
showed a gain of 3.03. The practised girls averaged 6.41 gain per test,
while the unpractised lost 1.24 per test. In the non-biological, the prac-
tised girls gained 6.2 per test for the 6 tests while the unpractised gained
5.6; the practised boys gained 8.97 per test
and the unpractised boys
gained 5.27 per test."
"Feeling that the balance of arguments and scientific proofs were
against formal discipline when this investigation was begun, I am forced
by the results obtained to admit that in this experiment the proof seems
to be on the affirmative side.
"A valuable lesson, I think, can be drawn from one phase of this
investigation. By consulting the tables and summaries, it will be i
250 EDUCATIONAL PSYCHOLOGY"
that sometimes one division does not fall in line with the general trend,
but that a larger number outweighs the negative and shows positive re-
sults." (Hewins, pp. in and 113.)
In order to make a relative comparison of these gains on the

basis of the original scores in the first series of end tests, I have
computed the percentages of gain in these end tests as shown at the
bottom of the above table. When such a relative comparison is
made, the net gain in the non-biological observations is very small,
being only 5.4% as compared with 33.9% in the biological obser-
vations, or about one-sixth as much in the non-biological as in
the biological types of observation. The improvement in the
biological end tests can hardly be counted as evidence of spread
of training since these tests were so similar to the training series
that they were all but identical with it. The spread of training is
apparently not as great as Miss Hewins believes it to be.
e. Geometry. Geometry, particularly in the high school, is urged
to possess a considerable amount of disciplinary value:

Plato (Republic, Book 7) emphasized this opinion thus:
"
Moreover, the science (Geometry) has indirect effects which are not
small.
" ' '
Of what kind? he said.
"'There arc the military advantages of which you spoke,' I said;
and in all departments of knowledge, as experience proves, any one who
has studied geometry is infinitely quicker of apprehension than one who
has not.
" ' '
Yes, indeed/ he said, there is an infinite difference between them."
i6) made a study of the spread of training in the case of

J
Rugg (
learning descriptive geometry. He performed three groups of

tests at the beginning and at the end of the course in descriptive
geometry, in February and in June respectively, with 326 students

in the College of Engineering of the University of Illinois. The
same tests were made with 78 students in other colleges as a control
group who did not pursue the course in descriptive geometry.
The three groups of tests were of a non-geometrical, Tests i and 2,
quasi-geometrical, Test 3, and strictly geometrical nature, Tests
4 and 5. Partial illustrations of the various tests follow:
Test x.
i. Divide eighty-one by seven. .
a. Divide seventy-eight by four, etc.

TRANSFERENCE OF TRAINING 25*
Test 2.
1. Divide eight sixty-two by three.

2. Divide seven ninety-five by four, etc.
Test 3.
A test in imaging letters.
Test 4.
The Painted Cube Test.
A three-inch cube, painted on all sides, is cut into one-inch cubes.
1. How many one-inch cubes have paint on three sides?
2.
" " " " " " " " " tWO ?
" " " " " " " "
3. one side?
" " " " " " " "
4. no side?
Test 5.
Geometrical Objects' Test.
The problem: Form a mental picture of each object and count the
number of straight lines which it would take to construct each
one in space.
1. A wedge.
2. Four triangles attached to a square, bases coinciding with the
sides of the square, etc.
The main results are represented in the following table:
"
TABLE 78. Representing the results for Rights." Adapted from Rugg's
table, p. 123 ('16)
TEST 1 TEST 2 TEST 3 TEST 4 TEST 5
TRAIN- CON- TRAIN- CON- TRAIN- CON- TRAIN- CON- TRAIN- CON-
ING TROL ING TROL ING TROL ING TROL ING TROL
GROUP GROUP GROUP GROUP GROUP GROUP GROUP GROUP GROUP GROUP
Feb.scores 20.0 1738 20.00 19.3 4.67 450 059 0.28 2.32 1.92
Junescores 22.81 2006 23.00 19.15 5.66 4.58 1.50 0.68 333 1.81
Grossgain 2.81 2.68 3.00 -.15 .99 0.083 .91 0.40 1.01-0.10
% ga in 14.5 15.6 15.0 -.78 21.2 1.8 157.0 143.0 43.5 -5.0
Residual
difference.... -1.10 15.78 19.4 14.0 48.5
Averages 7.34 19.4 31.7
Rugg concludes that the residual gain indicates a considerable

transfer of improvement to all three types of abilities tested.
"The study of descriptive geometry (under ordinary class room condi-

tions throughout a semester of 1 5 weeks) in which such natural and not
undue consideration is given to practice in geometrical visualization as is
necessary for the solution of descriptive geometry problems operates:
"
(i) Substantially to increase the students' ability in solving problems
requiring the mental manipulation of a geometrical nature, the content
of which is distinctly different from the visual content of descriptive
geometry itself.
" the students' ability in solving problems
(2) Substantially to increase
?52 EDUCATIONAL PSYCHOLOGY
requiring the mental manipulation of spatial elements of a slightly

geometrical character, i. e., problems utilizing the fundamental elements
of geometry (the point, line, and plane), but apart from a geometrical
setting, and in such form as to offer no geometrical aids in solution.
"
(3) Substantially to increase the students' ability in solving problems
requiring the mental manipulation of spatial elements of a completely
non-geometrical naturc i. e., problems in which the straight line and
(
plane do not appear in any way whatever.

"
(4) The training effect of such study in descriptive geometry operates
more efficiently in those problems whose visual content more closely
resembles that of the training course itself, i. e., in those problems whose
imagery content is composed of combination of points, lines, and planes,
and in which the continuity of the manipulating movements approaches
the continuity of those in the training course. (Rugg, pp. 114-115.)
These results are not very different from those surveyed in the
preceding chapter. Rugg states the results perhaps as favorably as
the data permit, perhaps too favorably. At any rate we ought to
note that the gain in the non-geometrical tests, Nos. i and 2, is
only about one-fourth or one-fifth as much as in the strictly geomet-
rical tests,Nos. 4 and 5. We also should note, as Rugg himself
points out, that only about two-thirds of the persons in the train-
ing group gained; the remaining one- third did not gain or show
transfer.
General Interpretation. The transference of training of the capa-
citiesinvolved in the learning of school material is very small so far
as present partial data indicate. This seems to be equally true
of all school subjects, the sciences as well as the languages and
mathematics. If we represent the possible transfer effect as ranging
from o%, or none, to 100%, or an improvement in other capacities
equal to that in the capacity trained, then the amount of transfer
is much nearer to the o%
end than to the 100% end. It probably is
o or very nearly o, for all capacities which are not distinctly similar
or related to the capacities specifically trained. Thus in the author's
experiment, practice in mental multiplication improved other forms
of mental calculations about one-fourth as much, but had no effect
upon immediate memory of numbers or words. In Winch's ex-
periment, practice in arithmetical computations had either no effect
or a doubtful effect upon arithmetical reasoning. In Rugg's in-
vestigation, practice in visualization ordinarily done in a course
in descriptive geometry had only a moderate effect upon visuali-
zation of other sorts. In Miss Hewin's study, improvement in
biological observation improved non-biological observation only

one-sixth as much. In the author's investigation, the study of
foreign languages had no effect upon the capacity to write an Eng-
lish composition, the study of Latin seemed to produce a small
increase in English vocabulary and a decided increase in the knowl-
edge of English grammar, but only a very small increase in dis-
crimination in correct English. Wilcox likewise found that the
study of Latin had no appreciable effect upon work in English
classes as measured by marks. The superiority of the Latin pupils
was due to their superior native ability rather than to the study of
the language. Perkins found that Latin as taught by him with
special emphasis upon word derivations and meanings produced
a noticeable increase in the ability to spell, define, and use English
words.
f-
The transfer effects of the training of the abilities in school sub-
jects isvery much less than is commonly assumed. This is prob-
ably due, in the first place, to the fact that the improvement in the
capacities exercised specifically by the school subjects is usually
not as great as is commonly believed by teachers. The modifica-
tions produced in theminds of the pupils are considerably less than
teachers usually assume as judged by the modifications produced
in their own minds after much longer and harder study. To il-
lustrate, teachers are inclined to believe that a course in mathe-
matics has produced a much greater improvement in mathematical
reasoning, or that a course in history has brought about much
greater facility in handling and interpreting historical material,
or that a course in psychology has brought about much keener
insight into the operations of mental process, than these respective
subjects actually have produced. This is probably the result of the
teacher's naturally egotistical belief in the effectiveness of his own
work.
The small effects of transfer are probably due, in the second
place, to the fact that the conditions for securing transfer are not
as favorable on the whole in the case of school subjects as in the
case of the special laboratory experiments on transference.
The evidence on spread of training in school material tends to
support for the most part the theory of identical elements. The
effects are the largest where there is similarity or identity of material
as, for example, in the case of the effect of the study of Latin upon
the study of Spanish, or upon the knowledge of English grammar.
The fact of identity of material or similarity of procedure makes
possible a greater control of the spread of improvement through

methods of teaching whereby the identity or the use of identical
material may be emphasized in as many desirable relations as
possible. This is illustrated by the spread of the effects of Latin as
taught by Mr. Perkins.

In formulating an opinion concerning general training effects
resulting from training of special capacities, we must bear in mind
that even where the transfer effect is considerable, as much as one-
fourth to one-third as much as in the capacity specially trained,
it is obviously more economical to give practice directly to the
capacities which we want to train rather than to do it indirectly

with the hope that the improvement may be transferred to them.
Concretely, even if the study of Latin under favorable methods of
teaching does improve the spelling of English words, would it not
be more economical to study directly the spelling of the words
which are to be acquired? Knowledge of the most common Latin
words from which the largest number of English words are derived
could be obtained in a relatively short period of time, probably a
year or even less. Learning to play the piano might help in learning
to play the violin, but no sane person would devote very much time
to the piano if his sole purpose is to learn to play the violin.
Even if mathematics may cause some improvement in reasoning
about bargains, even if the study of Latin may increase English
vocabulary, or even if a study of animal psychology did make a
man a better teamster, these effects are relatively very small and
can be produced much more economically by a direct study of
bargains, or of the origin and meaning of English words, or of driv-
ing horses. Acourse in mathematics or in Latin or in psychology
will have to stand primarily on its own feet for the content that it
offers or the skill that it develops. These by-products may be useful
but they cannot be the sole purpose of the efforts put into a course.
The value meal depep 's upon the meal itself and not upon the
of a
crumbs that fall from the table. Whenever a subject loses its con-
tent value through changed social conditions it seems mysteriously
to acquire a great deal of disciplinary value.
An immense amount of confusion in the thinking about the prob-
lem of mental discipline and the value of school subjects, even on the
part of distinguished thinkers, has resulted from a failure to dis-
criminate between the effect of a certain kind of education and the
native capacities of the individuals subjected to the education.
Whenever allowance or deductions for differences in original ability
have been made, the general disciplinary effect has been found to be
much less, or, in many instances, even non-existent. To argue that
because certain great leaders of men had a certain type of education,
it must have produced their greatness does not
prove the point.
They probably would have achieved distinction if they had had any
other sort of education. If the chief argument for pursuing a given
subject is that it selects the more able pupils, it would be much more
economical to do so by a shorter and more certain method. Almost
any fifteen or twenty mental tests that can be applied in a psycho-
logical laboratory in two hours would separate much more ac-
curately the gifted from the stupid.
Finally, the upshot of the experimental and statistical inquiries
into the transference of training is that effects of training are trans-
ferred in smaller amounts and within much narrower limits than has
commonly been assumed. This does not mean that there is no gen-
eral mental discipline in any form of training, nor that the doctrine
of formal discipline has been "exploded" but rather that the actual
limits of general discipline have been more accurately defined.
These limits, to be sure, seem to be much narrower than many are
inclined to believe. So far as the value of school subjects is con-
cerned, it means that the content value of a subject must be the
prime reason and the general disciplinary value the secondary
reason for pursuing it.
Before leaving this discussion, two further points should be borne
in mind: The first is that any effect of transfer, even if very slight,
would probably be worth while if it extended to all or to a large
number of capacities. If training in botanical observation improved
all forms of observation in life ever so little, it might still be the
best form of training in observation. But the implication of the
evidence thus far at hand is that the spread seems to extend only to
rather narrow limits. The second point is, that while the trend of
the arguments here presented would be to reduce the time devoted
to some subjects, particularly in high school and college, we must
be sure that we put something better in their places. The advan-
tage of some of the subjects that would suffer reduction is that they
are well organized for teaching purposes. Some of the new sub-
stitutes are not well organized and offer neither form nor content.
Transitions should be made gradually so that the new branches may
become organized and extended, and the teachers properly trained.
PART III
THE PSYCHOLOGY OF LEARNING: B. OF SCHOOL

SUBJECTS
CHAPTER XV
THE PSYCHOLOGY OF LEARNING SCHOOL SUBJECTS
Psychology and Teaching. If education consists in making

changes in human beings, if psychology is the scientific study of
the mental processes of human beings, and if teaching consists in
the facilitation of the changes to be made by the school, then it is
obvious that knowledge of how to bring about these changes in
the most economical manner must be based upon an exact knowl-
edge of the processes involved in these changes. A reliable peda-
gogy can be based only upon a reliable psychology of the processes
concerned. Engineering did not become a science until the physi-
ical and chemical knowledge of the processes involved in a given
project were thoroughly understood. To build a Brooklyn bridge

involves a precise knowledge of the laws of gravitation, the strength
of materials, the means of supporting weights against the force of
gravity, and the like. To compound anelectric cell involves a
precise knowledge of the chemical action of certain elements. To

know how to destroy bacteria harmful to plant and animal life,
it is necessary to understand the biological processes of the
par-
ticular plant and animal life concerned.
The foundation science for sane and dependable methods in
education is psychology in its broad sense. The great difficulty
in establishing a reliable pedagogy is the fact that sure and de-
tailed knowledge of the psychological processes and laws in learn-
ing the material of the school subjects is largely unknown. In the

past the schools have proceeded largely by guess. The future will
have to map out in detail the psychological steps involved in each
school subject, and to submit these processes to direct experimental
investigation. A science of engineering was impossible until it
was discovered how the physical and chemical laws operated in
the particular conditions under which the bridge had to be built
or the cell had to be compounded. Knowledge of the law of gravi-
tation was practically useless until it was discovered how it operated
under given concrete conditions of materials, distances, and
circumstances. The psychological laws of learning will be practi-
cally useless until we shall know how they operate under the
concrete conditions of the school and with the materials to be

learned in the school. The reaction times of numerous psychologi-
cal processes have been studied in great detail for many years, but
this knowledge is practically useless in giving to the educator
scientific information by which he may proceed to facilitate
the reactions of a child in learning to write or to read. This
knowledge is useful in furnishing a general background, in
pointing the way, and in supplying a general experimental tech-
nique. But even these must often be materially modified and
adapted to the solution of a particular problem in a given field.
The old-time pedagogy has fallen into disrepute because it has
been almost wholly a matter of personal guesswork. The slate
must be wiped clean and only those principles and laws whose
truth has been fully proved can be recorded thereon.
Problems. If we grant that the method and procedure of the
school should be based upon a sound psychology of the processes
involved in learning the special materials of the school subjects,
itfollows that the fundamental tasks to be done are these:
(1) A
thorough and complete analysis of all psychological
processes involved in the learning of a given subject, or in the
acquisition of skill in it, and of the order and manner in which
these processes intermesh.
(2) The devising of means by which these processes may be
measured and tested so that the facility in their operation may be
determined quantitively.
(3) The discovery of the most economical procedures by which
each particular step in the entire process may be developed.
If we wish to determine how to memorize a poem in the most
economical and most permanent manner, it is important to know
the perception, association, and reaction processes involved, the
means for definitely measuring facility in these operations and
the means of controlling these processes most efficiently. Teach-
ing is mental engineering; it consists in managing the mental
processes concerned in learning the materials and in acquiring the

skill of the school in the most effective and most profitable
manner.
In considering the psychology and pedagogy of school subjects
in the succeeding chapters, this three-fold division of the problems
will be made for each subject and the available information of
each one surveyed so far as our present knowledge warrants.

CHAPTER XVI
READING
PROCESSES OR STEPS INVOLVED IN READING
If we trace, for analytical purposes, the successive steps from

the external presentation of the visual stimuli of the printed
words on through the complicated elaborations within the mind
and back to the external expression of reactions in the pronunci-
ation of the words, we can discern the following order or combin-
ation of elements:
(1) Reception upon the retina of the stimuli from the printed
page.
(2) The range of the field of distinct vision on the retina.
(3) The range of attention in apprehending visual stimuli.
(4) The movements of the eyes.
(5) The transmission of the visual impressions from the retina
to the visual centers of the brain.
(6) The establishment or arousal of association processes where-
by the incoming impulses are interpreted.
(7) The transmission of the impulses from the visual centers
to the motor speech centers.
(8) The transmission of motor impulses from the speech centers

to the muscles of the vocal chords, tongue, lips, and re-
lated parts.
(9) Execution of the movements of the speech organs in speak-
ing the words.
These are the steps as they occur in the developed reading
process. It is obvious that they do not follow each other in a
temporal order but that some, as for example (i) to (4), occur
simultaneously. During the early stage of learning to read there
occurs, simultaneously with steps (i) to (5), a parallel series of
steps derived through the car by which the child learns the pro-
nunciation of the word, thus: (i) reception in the ear of auditory
stimuli from the pronunciation of the word by the teacher, (2)
transmission of the auditory impulses from the ear to the auditory
center in the brain, (3) transmission of impulses between the
26X
auditory and the visual centers whereby the sound and the sight
of the word become associated. Silent reading involves only
the first six steps except in so far as incipient speech move-
ments accompany it, in which case the remaining steps enter
in part.
Such an analysis as this may seem detailed to an unprofitable ex-
tent; but, in fact, it might be made even more detailed, depending
upon the extent to which we are able to discern and describe the
minuteness of the neural and mental functions involved. The
more complete and accurate our analysis and description of the
steps is, the more sure our knowledge for managing these
processes will be; and ultimately, that is what teaching amounts
to: The efficient management of the psychophysical processes
concerned.
The next problem is, How does each of the elements in the
reading process operate? The truth is that concerning many of
them we know at the present time very little or nothing with cer-
tainty or completeness. Concerning some of them, however,
considerable definite knowledge has been accumulated in recent
years. What the differences between an efficient and an inefficient
reader are, or what the difficulties in learning to read are at each
of the steps can be inferred partly, but only partly, from our
present knowledge about these factors. It is, however, certain
that the differences and difficulties are to be found in these and
possibly additional processes. We
shall examine each of these
steps in turn and survey what definite knowledge is available.

(i) The reception upon the retina of the visual stimuli from
the printed page depends obviously on the one hand, upon the
sizeand kind of type, length of lines, indentation of lines, paper,
and illumination; and, on the other hand, upon the inertia of the
retina in receiving stimuli. We
know, for example, that for adults
and for children above 10 or 12 years of age, type smaller than
8 or 10 points is probably too small to be perceived easily. Like-
wise, type larger than approximately 10 points, spreads out upon
too large an area of the retina to be perceived quickly in as large
groups as possible. Experiments by Dearborn ('06) and others
have shown that probably the most advantageous length of line
is in the neighborhood of 2^ r 3/^ inches, and that it is better to
have the lines on a page uniform in length instead of varying in
length as is often the case in reading-texts in which illustrations
are set into the margins and the lines made to vary according to
READING 263
the space around them. We do not know what the most ad-
vantageous size of younger children who are beginning
type is for
to learn to read. We feel that it ought to be larger than for older
children or for adults, but we do not know definitely how much
larger.
So far as the inertia of the retina to the reception of the stimuli
goes, we may
infer that it varies in different individuals probably
according to the normal distribution and that it may be less in

rapid than in slow readers, but no definite measurements have
been made to ascertain the truth about it.
In an investigation, as yet unpublished, by C. L. Hull and W. R.
Ames, an effort was made to determine the relative effect upon the
eye of reading from papers of various colors, of various amounts of
gloss and of various degrees of perfection of the inked impressions.
Four kinds of paper were compared: a matte white paper, a pink
newsprint paper, a blue newsprint paper, and a very glossy white
paper. As measured by the Ingersoll glarimeter, these papers had
the following glare or gloss values respectively, 18.5%, 41.5%, 42%,
73-5%- A
fair sized book was printed, uniformly upon each kind of
paper in such a way that the successive pages followed one another
on a single band of paper. These were placed upon special reels
in such a way that the pages would be at a uniform distance
from the eyes, at a uniform angle and would have uniform
illumination.
Three measures of the changes in the eyes produced by reading
were taken:
1. The number of lines read during a i5-minute period.
2. The number of spontaneous winks per minute while reading.
These were recorded by a special device unsuspected by the sub-

jects.
3. Extent of failure to recover during a i5-minute reading period
from artificially produced diminution in the distance at which the
subjects were able to see faint parallel lines as distinct lines. This
distance was determined before and after reading by an elaborate
recording device of considerable precision. Four subjects were
used.
The final averages for each kind of paper by each of the tests are
shown in the following table. In addition, the relative rank of the
four papers is given for each of the three measures on a scale of 10.
Lastly these ranks are averaged for a final score of all three meas-
ures.
TABLE 70,. After Hull anil Ames. From a Thesis in the Library of the Uni-
versity of Wisconsin, 1917
Despite rather striking differences between the results obtained

by the third measure and by the other two, the results as a whole,
as indicated by the final average rank, show that glare is the de-
cisive factor in diminishing ocular efficiency. The two white papers
are respectively the best and the worst of the set. The two colored
papers which are intermediate in glare are also found intermediate
in ocular efficiency. These results suggest that color as such has
little or no influence one way or the other.
A microscopic examination of the texture and perfection of the
inked impressions of the various papers revealed the glossy white
papers to be the most perfect, with the matte white, the blue and
the pink in decreasing order. This order indicates that within
ordinary limits the perfection of the printed impression is of little
consequence.
(2) Concerning the size of the distinct field of vision, Ruediger
('07) made some experiments with the tachistoscope. Speed of
reading may depend upon the horizontal area of distinct vision in
the retina. Ruediger, however, concluded that little correlation
exists between the areas of color zones on the retina and visual
acuity and other qualities of sight, that the field of distinct vision
is reduced as much as one-half when the eyes are tired from reading
and that there is only a slight correlation between visual acuity

and the size of the field of acute vision.
Judd, McAllister, and Steele ('05) marked the eye near the
pupil with a flake of Chinese white and then directly photographed
the movements of the eye. They concluded that there is no cor-
relation between the size of the horizontal field of acute vision
and the rate of reading or the number of pauses in reading. The
correlations found were .06 and .10.
READING 265
(3) The range of attention in apprehending words and letters.

Dearborn, in his photographic records of the movements and
pauses of the eyes, found that the eye takes in, on an average ob-
tained from five subjects, 1.64 words at one time, or at each fixa-
tion. He believes that the word is the unit of reading. This is
corroborated by the fact that slight misspellings in words are often

not noticed and that such words as "psychology" and "physiology"
may be confused with each other because they are per-
easily
ceived as wholes. In many instances he found that long words
take no longer time to be perceived than short ones do. The
habit of grouping is apparently very important. "It is not the
short words as such but the words which cannot be easily grouped
with others, which necessitate separate fixations." He also found
that the range of attention with slow readers is often only a syllable
while with the fast reader it is words and phrases. There is thus
a large and important difference which is probably highly signifi-
cant in the development of rapid reading as will be pointed out
later.
By presenting ordinary printed material for very brief intervals,

Gray showed that the number of words which could be compre-
hended at a single exposure was much larger than the number
actually read at a single fixation often twice as large. This
means that the areas of visual apprehension overlap considerably
at successive fixations. This suggests that the number of fixations
might be reduced considerably by practice. Gray tried this out
experimentally with two children, one a good and the other a poor
reader, and found it to be true. The poor reader decreased the
number of pauses per line from 15.5 to 6.1 by a 2o-minute practice
period each day for 20 days.
In this connection the possibility of improving the range of
visual apprehension becomes important. Some years ago Miss
Aiken claimed to have increased enormously the range of visual ap-
prehension of the girls in her school by special training. G. Stanley
Hall wrote of it, "I would not have thought such rapidity and
accuracy possible if I had not seen it." Whipple, in an attempt
to verify these claims under laboratory conditions with adult
subjects, failed almost completely. Gray, attempting by the same
methods to enlarge the range of visual impressions of two sixth-
grade children for printed words, also failed completely. He re-
peated the experiment, however, with two fourth graders and
succeeded in approximately doubling it. He concludes that such
training, to be effective, should take place not later than the

fourth grade.
(4) The movements of the eyes in reading. Closely related to
the scope of apprehension is the rapidity and the precision of the
movement of the eyes. More experimental work has been done
on these two elements in reading than on any others. It has been
known for a long time that the eyes in reading or in examining
any object do not move along smoothly and continuously but

that they move by jerks and pauses apparently in a very irregular
manner. The eyes take rapid glimpses or snap shots of successive
portions of a line of print and then piece them together in obtaining
the meaning. The reason for it is that the eyes cannot see objects
distinctly while they are in motion and consequently they per-
ceive little or nothing during that time. A
person cannot see the
movement of his own eyes in a mirror. As soon as one can see them
distinctly, they are at rest again. Dearborn found that letters
fuse when passed in front of the eyes at the rate at which they
themselves move across a page. Dodge ('oo) also found by experi-
ments that, while the eyes were in motion, no sensation resulted
even though the stimulus strikes the eyes.
A fairly good conception of the nature of eye movements, pauses,
and fixations in reading may be obtained by taking an open book,
placing a mirror on one page while some one reads the opposite
page, and then observing the action of the eyes as reflected in the
mirror. Javal ('79) counted the pauses of the eyes by means of a
sound attachment to the eyelids. Laudolt ('91) counted them
by direct observation. Erdmann and Dodge ('08) counted them
by observing the eyes of their subjects in a mirror. They found
more pauses with difficult than with easy reading material, and
also more pauses in reading a foreign language than in reading
one's native language.
The first successful attempt to record eye movements was made
by Huey ('98-^00). He
attached a plaster of Paris cup to the
cornea which was connected with a light aluminum pointer. This,
in turn, rested on the smoked drum of a kymograph on which the
movements were registered. In this manner he attempted to
study the nature and rapidity of eye movements and the nature
and length of successive pauses. Dodge then developed a falling
plate camera which photographed a beam of light reflected from
the cornea of the eye. This method was also used by Dearborn.
It has the important advantage over Huey's method in that it
READING 267
eliminates the attachment of anything to the eyes themselves,

which quite likely interferes with the natural movements of the
eyes.
Readings by five subjects of the same newspaper passage

i
SUB- NUMBER OF TOTAL
jfccrs FIXATIONS TIME AVERAGE
312 152 2 361
T ST. PE(TE]RSBUR|G, No[v.) 2. Th|e Admiralty. . .

4 1037 25!
480 110 140 260
II ST. PETERSBURG, [N)ov. 2. -|

The Ad[m)iralty. .
4 1020 255
416 IW 306 138 SU
S [ST. ]PKTKR|S(B]URG, ISTovl.
<>
(Th]e Admiralty.. 5 1334 266
670 1 1 18 016
V ST. P|TERSBUR[G, Nov). 2. The A[dm)iralty. ... 3 2432 810

811 143 171 140 851 163 271
M (S[T. |PE]TER|SBUR[G, )No[v.) 2. [Th)e Adlmiralty 7 2057 203
T has telegraphed to the offi|cers of the |Baltic 3 493 164

3JO 220 160 240
II ha[s telegra)phed to th|e o(fficers] of the [Bal)tic 4 950 237

314 144 * 23S 238 263 136
S [) |has telegrap[he) (d] to th[e of)ficers| of th|e Baltic. .7 1639 234

J9l 179 384 687 629
F [has tclegr|aphe|d to the [o)fficers of the B|altic 5 1968 393

218 196 234 226 171 179 78
M |ha|s te egraphed [to) the oflficejrs of the |Ba[lt)ic 7 1301 185
Figure 58. The vertical lines and brackets indicate the locations of the eye
fixations. The numbers give the length of the fixations in loooth of a sec-
ond. After Dearborn ('06).
Figure 58 shows the locations of successive pauses, their lengths

in loooth of a second, and the distances between them as deter-
mined by Dearborn. The brackets and vertical lines indicate the
locations of these fixations. Sometimes there is a slight movement
after the fixation. The direction and distance of this movement
is indicated by the parentheses. These regressive or corrective
movements occur when the eyes have moved too far to the right.
They also occur when the eyes make the sweep back to the begin-
ning of the next line but have not gone far enough. These correc-
tive movements come more often in long lines than in short ones.
They are probably caused by the fact that the peripheral percep-
tion of the beginning of the line is not accurate. The exact location
from the beginning and from the end of the line is gotten as a
habit after several lines have been read if the lines are of uni-
form length. For this reason the lines should be of the same
20fc EDUCATIONAL PSYCHOLOGY
length, especially for beginners. The opposite condition, however,

is often found in reading texts in which each sentence is printed
as a paragraph and in which pictures are inserted in the margins.

The more objectionable when the pictures are colored,
latter is still
since the reflex action of colors influences the position of the fixa-
tions. The numbers above the vertical lines, Figure 58, indicate
each pause. It will be noticed
in loooth of a second, the length of
that these vary considerably from time to time and that on the
average they arc approximately one-fifth of a second in length.
Schmidt found on the average nearly twice as many refixation
movements in oral as in silent reading. These movements grow
progressively fewer and shorter from year to year as children

progress through the grades. Gray found frequent refixation
movements very characteristic of slow readers. They were also
much more numerous with all readers in extremely careful read-
ing.
Dearborn found that, with newspaper lines, five persons made
an average of 4.76 fixations per line, that the number of fixations
tends to decrease and with it the range of apprehension tends to
increase as a passage is read over and over again, and that the eyes
finally develop considerable precision and accuracy in the number
and the position of the fixations. He also found evidence for the
belief that the eyes form a short-lived motor habit which deter-
mines the number and position of the fixations for each line in
reading. The formation of this habit develops during the reading
of the first three or four lines of print and seems to come much
sooner for some persons than for others. The fast readers seem
to form the motor habit more quickly than the slow ones. The
decrease in the time is due to a shortening of the pause and a
widening of the scope of attention, especially in the latter parts

of the line.
The fixations may come at any point in the line, between words,
at the beginning, at the end or in the middle of a word. "The
first and last fixations generally fall within the edges of the line,
that is, a little distance from the beginning or the end words."
The first fixation in a line is usually longer than the other fixations.
There are fewer pauses in the re-reading of a given material,

even after one month, and the number decreases somewhat as the
selection is read repeatedly. Prepositional phrases, connectives,
and substantives make the greatest demand upon perception due
to the fact that a slight change in them affects the whole meaning
READING 269
of a sentence. Dashes, punctuation marks, and capitals in the

middle of a linechange the points of fixation, and disturb some
readers more than others. Reading a foreign language or reading
aloud requires longer fixation than reading one's native tongue or
reading silently. Dearborn obtained records of three boys aged
9, 10, and n years, in the third, fourth, and fifth grades respec-
tively, and four.d that the oldest one approached very closely to
the conditions of fixation of adults, \vhile the younger ones made
many more fixations or pauses. Fatigue resulting from the use of
the eyes during an entire day decreases the rate of reading by
about one- tenth.
Schmidt ('17) made an investigation of the eye movements, in
oral and silent reading, of eighty individuals including elementary,
high school, and university students. He found that they made

1.6 more pauses per line in oral than in silent reading, that the
average duration per pause was from 20 to 27% longer, and that
the perception time was from 44 to 64% longer.
By an ingenious combination of camera and phonograph, Gray
determined accurately the relation of the eye to the voice in oral
reading. He found that the eye always precedes the voice, with
some subjects as much as four words and with others much less.
As a general rule a wide eye-voice span was associated with fluent
reading.
(5) Transmission of visual nerve impulses to the visual area of
the brain. Little can be said about this process. All that we know
about it is that the velocity of nerve impulses varies in different
persons, and mny be slow or rapid in slow or rapid readers, but we
do not know definitely. Experimental work is necessary to de-
termine to what extent it may be true.
(6) The arousal of association processes. A word gradually
acquires meaning in the life of a child by its recurrence in numerous
situations and by the connection of the particular significance or
experience with the word. Thus, the child hears the word "chair"
in connection with a certain object on which he sits. He hears it
also after a while in connection with other objects which look
different from the one he is accustomed to using. But new associ-
ative bonds are formed and he knows in general what is meant
when he hears the word "chair." Later on he is shown a certain
combination of visual characters and is told the word "chair."
A new bond is then formed between the sound and the visual
stimulus of the word "chair." An important part of the act of
reading consists, therefore, in the arousal of associative' bonds as

soon as the visual nerve impulses are brought in from the eye to
the visual brain areas. The quickness of reading depends, no
doubt, upon the rapidity with which these incoming stimuli arouse
associated bonds and thereby give meaning to the word.
That the rate of reading depends upon the rapidity with which
the visual stimuli are interpreted is shown by such experiments as
Huey made in which he showed that it takes about twice as long
to read nonsense as sense material. He also found that words in
context give fewer associations than words out of context. When
the words were shown in context the associations would sometimes
go ahead of the amount seen and would anticipate what was com-
ing.
Hamilton attempted to determine the part played by
('07)
context in readingby ascertaining the time required to perceive
isolated words and words in various relations, such as in para-
graphs, in miscellaneous sentences, and in miscellaneous phrases.
His results, expressed in terms of seconds per word required for
the reading, are as follows:
TABLK So. After Hamilton

PARAGRAPHS StNTFNCES PHRASES WORDS
Mean 420 .456 .466 .660
M. V .
.047 .047 .041 .078
Per cent using paragraph
as base. .
100% 9 4% 92% 65%
Isolated words apparently require a distinctly longer time for

perception than words in context.
Hamilton also found that the first part of a word is more im-
portant in giving the meaning of a word than the latter part is,
that the marginal impression at the right aids in the perception of
the words seen in the next eye fixation, and that the upper
and lower marginal impressions probably do not aid in per-
ception. He also thinks

probableit that the interpretation
takes place during the rest period, that is, during the movement
of the eye.
The central neural and mental
activity of putting meaning into
the incoming impressions probably the most important step in the
is
whole reading process. The important part of reading really is the

reading of meaning into words. Judd, McAllister, and Steele be-
lieve that the essential factors controlling the rate of reading are
READING 271
central rather than peripheral and that it is a matter of assimilation

rather than a matter of getting material into the brain. Ruediger
has a similar opinion:
"In reading, a similar reinstatement of experience takes place as in

thought or in oral communication. The printed symbol arouses the
meaning that has through education and experience become connected
with it. It is to the rapidity with which this meaning is aroused that we
have to look for the cause of the differences in reading rate.
"Reading rate may then be taken to depend chiefly upon the rapidity
with which meaning is aroused in the mind after the symbol is seen.
This, in turn, is in the main dependent upon the person's native brain
inertia."
(?)> (8)> and (9) Transmission of nerve impulses from the visual
center to the motor speech centers and thence to the speech organs.
The motor speech centers concerned in the control of the speech
organs are highly specialized. In right-handed persons, they are
located in the left hemisphere of the brain in the region of the fis-
sure of Rolando. These processes obviously will be involved in
oral reading and in speech, but they are also active in silent reading
in the form of incipient speech movements, particularly in the
tongue and, to a less extent, in other speech parts. They are
active in the in speaking, only on a much smaller
same manner as
scale. To whatextent this inner speech is an important part in
the reading process is somewhat uncertain. It is believed by
some investigators to be an important agency for maintaining
the continuity of the thought aroused by the successive visual
glimpses of the printed line.
Disturbances in the motor speech centers are known under the
names of various types of aphasia which will not be considered here.
But even in normal persons the rapidity and facility with which
the neural centers act may have an effect upon the efficiency of
reading.
Much valuable information could undoubtedly be obtained
from a careful laboratory study of individuals with language de-
fects, such as persons who have difficulty in learning to read or
difficulty in the speech functions proper. The following

case may
be of interest in this connection. A
boy, 17 years of age, in the
second year of high school, had always had considerable difficulty
in reading. His reading was extremely slow. He was finally sent
to a private teacher of public speaking with the hope that his
reading ability might be improved. After several months of work

with this teacher, apparently no progress was noticeable. He was
then brought to the psychological laboratory for examination to
see, if possible, wherein his difficulty lay. As careful tests as possi-
ble were made to discover in which of the various steps in the read-
ing process the difficulty might be. These successive processes
were tested with such means as were available or could be devised,
beginning with the first, the reception of the stimuli upon the
retina. Inquiry showed that his eyes had been carefully examined
and found to be only very slightly defective. Glasses had been
supplied as his father was a physician. Tests were next made to
determine the range of distinct vision by observing how large an
area he could see distinctly at one time.Next, his span of atten-
tion apprehending visual stimuli was measured. This was
in
done by rapid exposures with a tachistoscope to determine the
number of words or pictures of objects he could apprehend at one
time. In the next place, tests were made to determine the con-
trol of his speech organs. It was thought that possibly his diffi-
culty in reading might be in an inability to control rapidly his

speech organs. Tests were made by having him repeat statements
from memory as rapidly as he could and also by having him repeat
short sentences from dictation. All these tests indicated that
these various functions were normal. His eyes had only slight
visual defects, his field of distinct vision was apparently of normal
size, his span of attention likewise was as large as that of a normal
individual, and the control of his speech organs was rapid and
accurate. The inference by a process of elimination was that the
chief difficulty in his reading ability lay in the central assimilation
or association processes. It seemed that visual impressions were
brought into the brain centers with normal speed and facility,
but that there was, for some reason, extreme slowness in the
mental interpretation of these stimuli. A test in silent reading
and in oral reading showed him to be equally slow in both cases.
His rate of reading was approximately that of a child at the end
of the first grade, namely, 1.5 words per second. His compre-
hension of what he read was good. His general intelligence, as
shown by other tests, was normal for his age. His work in other
school subjects was satisfactory. It seemed, therefore, probable,
although not absolutely certain in the absence of further tests, that
his difficulty lay in the central interpretation processes. This case,
similar to others, shows in an interesting manner the great intricacy
READING 273
of the reading functions, and the difficulty in determining pre-

cisely what may be the trouble in a child who is backward in a
given school subject but normal or even superior in all others.
In order to ascertain more fully the part played by the various
factors enumerated at the beginning of this chapter, a series of
tests was carried out by A. D. Mueller under the direction of the
1
author, with 36 high school pupils. Speed and comprehension of
reading were carefully measured by three different selections
according to the writer's method. Then each of the capacities
mentioned in Table 81 was measured, and the coefficients of cor-
relation computed. These show that the two elements upon
which rate and comprehension of reading probably depend most
are the visual apprehension span for related and unrelated words
and quickness of association, correlation Nos. 2, 3, 6 and 12. Ap-
parently the amount apprehended and the quickness of giving
meaning to the visual symbols are the most important factors in
reading. Correlation No. 6 corroborates Nos. 2 and 3 since the
number of eye fixations per line of print is probably inversely
proportional to the amount apprehended at each fixation.
TABLE 81
Correlations between reading ability and various elements entering into reading
ability
1
Reported in an unpublished thesis in the library of the University of Wisconsin, 1918.
These results are supported completely by a very similar ex-

periment of Gray's, although he computed no correlations. As
in the author's experiment there was a close relation between
reading ability and the range of visual apprehension for meaningful

material but, "The difference between the span of attention of
the good and poor readers disappears in a very large measure when
nonsense syllables, digits, groups of the same digit or the aussage
l
test are given."
Tin: MEASUREMENT OF EFFICIENCY IN READING
(1) Essential Elements to be Measured. In order to be able

to determine the effectiveness of different methods and procedures
in learning and teaching a given school subject, it is necessary
to be able to measure with some degree of precision and objectivity
the general ability in that function. In order to measure such
abilities it is necessary,furthermore, to determine what the essen-
tial elements in the process are which ought to be measured. These
would seem to be in the case of reading as follows:
(a) In silent reading, speed and comprehension should be
measured. The prime purpose of reading is the comprehension
of the thoughts presented upon the printed page. The second
element is necessarily the speed at which the thought processes
may be obtained.
(b) In oral reading the chief additional element to be measured,
besides speed and comprehension, is the correctness of the pro-
nunciation.
(2) Method of Measurement. Several types of tests have
been developed for measuring efficiency in the various aspects of
reading. These methods will simply be mentioned here without a
detailed, critical discussion of their relative values or of their
various techniques of administration and evaluation. Tests for
speed and comprehension of silent reading, have been devised by
Gray, Kelly, Starch ('15), Courtis, Brown, Fordyce, and others.
The writer's test consists of a series of eight passages, one for each
grade. It is administered by having the pupils read for 30 seconds
and by having them write out as full an account as possible of the
1
Gray draws the following conclusion which is at least worth consideration :-^This
seems to prove that the inherent differences in mental capacity which exist between
members of this group are not the causes of differences in the span of attention. It
appears rather that differences in training that is, acquired abilities to deal with mean*
ings are the source of differences in perceptual span."
READING 275
portion read. Speed is expressed in terms of the number of words

read per second; comprehension is expressed in terms of the number
of words written after those words have been discarded which
represent incorrect statements of thought, additions of ideas not

found in the original passage, or repetitions of ideas previously
stated. The tests by Gray, Courtis, Brown, and Fordyce measure
rateand comprehension of reading substantially in the same manner
except that in some of them comprehension is determined either
partly or entirely by answers to questions, and in the fact that
most of them employ fewer than eight test passages. For ex-
ample, Gray uses three, Fordyce two, and Courtis one passage for
all grades.
Thorndike's ('14) tests are designed to measure comprehension
in reading, either of isolated words or of paragraphs. They are
constructed on the scale principle. Similar tests have been pre-
pared by Haggerty ('17). The Kansas silent reading test (Kelly)
is designed to measure comprehension primarily and speed second-
arily.
Gray has devised a test for oral reading consisting of a
('17)
series oftwelve graded passages which are used to determine the
rate of oral reading and the number of errors in the pronuncia-
1
tion.
(3) Uses and Results of Measurements, (a) The first obvious

use of exact measurements of reading ability is a determination
of the actual abilities of the individual pupils, classes, or schools.
The first logical step in the management of any activity is a diagno-
sis of the conditions as they exist. Then, upon the basis of this
diagnosis, it is possible to prescribe more intelligently what should

be done. In order to be able to make definite comparisons of the
necessary to know the standard norms,

abilities of the pupils, it is
or averages, of achievement in the various grades. For the writer's
test these are as follows:
GRADES
Speed of reading
12345678
TABLE 82
(words per second) 1.5 1.8 2.1 2.4 2.8 3.2 3.6 4.0
Comprehension
(words written) 15 20 24 28 33 38 45 So
1
Further discussion of these tests may be found in the original sources in which these
tests were reported, or in the writer's Educational Measurements, or in Monroe's Edu-
cational Tests and Measurements,
These measurements are shown graphically in Figure 59 and

indicate the grade-to-grade progress in speed and comprehension,
showing that there is a continuous improvement in both aspects
from year to year. By reference to these norms it is possible to
express definitely a given pupil's capacities in reading. It enables
one to say, for example, that pupil A in the 4th grade has a reading
ability equal to the average ability of pupils in the 6th grade,
pupil B in the 4th grade has a reading ability equal to the average
Speed
Fro. 59. The continuous lines represent the standard attainments in read-
ing. The broken lines represent the attainments in a certain school.
ability of pupils in the 4th grade, and pupil C

has a reading ability
equal to the average of pupils in the second grade, and so on. It
is possible to describe precisely the ability of a pupil in relation to
others because the terms in which his abilities are expressed are
accurately defined. Measurements of this sort have disclosed
enormous ranges of ability in the various school subjects as
pointed out in a preceding chapter on individual differences.
They have shown that at the present time pupils are not
promoted according to ability, but rather according to the num-
ber of years they have attended school. Thus, for example, the
READING 277
pupils in a fifth grade range all the way in reading ability, from
the second or third grade on up to the 8th grade. (See Figure 16
in Chapter III.)
Another interesting comparison at this juncture is the reading
ability of the pupils of various ages in each grade. Such a study
\\ as made of the
pupils in one school by the writer, and is shown in
TABLE 83
Showing the relation between age and attainment in reading
From the fourth grade on there is a fairly regular decrease in

reading ability from the youngest to the oldest pupils in the same
grade. The explanation for it is probably the fact that the stupid
pupils are on the whole promoted too rapidly and the bright pupils
too slowly.
(b)The establishment of accurate norms by means of def-
inite measurements has meant, furthermore, a more precise
estimate of definite aims of attainment. To say that a pupil
at the end of the eighth grade should be able to read at the
rate of four words per second and report a correct thought
content expressed in at least 50 words, means something def-
inite. The pupil, as well as the teacher, will know what each
one means.
(c) The third and probably most important use of measurements
of reading ability is their employment in the investigation of the
factors and conditions affecting the learning and teaching of read-
ing. In the long run the greatest good from tests of reading ability,
or from tests of any school capacities, will come from their service
in analyzing and measuring the potency of the numerous elements
that enter into the acquisition of knowledge and skill in a school
subject. Results on this phase of the subject will be summarized

in the following section.
Speed
Grades
FIG. 60. The continuous lines represent the standard attainments in reading.
The broken lines represent the attainments in a certain school.
ECONOMIC PROCEDURE IN LEARNING TO READ

(i) Differences among Schools and Classes. The differences
in achievement in reading among classes as a whole are enormous.
Such differences are shown in Figures 59 and 60. Thus we see
that the best classes in Figure 60 have scores approximately twice
as high as the corresponding classes in Figure 59. These differ-
ences are too large to be explained by hereditary factors, but must
be due chiefly to differences in learning and in method and spirit
of teaching in these schools. What these differences in procedure
of learning and teaching are must be discovered more fully in the
future. But it is certain that some methods produce almost
READING 279
double the achievement produced by others. Thus the school

represented in Figure 60 is, grade for grade, as much as from i to
3 grades ahead of the average of schools generally, while the school
presented in Figure 59 is from i to 2 grades below the average of
schools generally. The former school has attained such proficiency
in reading, probably chiefly because of the plan inaugurated by
the principal which did away with nearly all word drills and phonics
and placed all emphasis upon trying to read and upon reading as
much as possible. He believes that a child learns to read by reading
a great deal. He has described the method of teaching as follows:
" in the initial stage materially different

Since first grade reading is
from that of the succeeding grades, it is treated separately.

"The reading is begun in the first method reader by means of the word
and sight method, and the phonics is carried on parallel with the reading,
but in a separate period. No attempt is made in the early stages to
correlate the phonics with the reading. During the entire work on the
first book, the pupils are given words and sentence drills. Each lesson is
read orally by every child after he has read it to himself. The time re-
quired to cover this book varies from two to four months, depending
upon the ability of the child. When this book is finished, the child has a
working vocabulary, and the method is changed to more extensive
reading. The word drill is now omitted, and the lesson covering from
five to fifteen pages is read over by the pupils, they or the teacher pro-
nouncing the difficult words. Next the pupils read the lesson over
silently, and the following day they are called upon to read a page each
without assistance.
"When the seventh unit reader, at about the close of the sixth month,
is in the hands of the pupils, the group reading is begun. The pupils are
seated in groups of two, and each group is provided with an interesting
reader. One pupil in each group reads aloud; when the first pupil has
read three or four pages, the second pupil reads an equal number, then
the first continues. In this way they alternate until the time is up. A
fluent reader is often placed with a poor one for the purpose of assistance.
At the close of a ten or fifteen minute period the place where the reading
is discontinued is marked by a paper, and the books are laid aside until
the next group reading period. Group reading is not, as a rule, practised
oftener than twice a week.
"Sectional Reading: This is a phase of group reading. Each group
of two or more stands before the class. The first pupil reads three or four
pages orally, the second continues the reading for three or four pages, and
so on. In this type of reading the teacher may question the child as to
what he has read, the child may reproduce it, or some listening member
may tell what he has heard.
"Silent Reading: During the study hour or seat work period, silent
reading isconducted by means of single copies of books containing in-
teresting material. Every child is given a book, and he reads as many
as he can during the period. When the silent reading period
pages silently
is finished, a mark is placed where the reading ended. At the next silent
reading period the pupil continues his reading, and so the work progresses
until he has finished the book. The silent period may be continued as long
as the teacher wishes. A record is kept of each child's reading by checking
off a book as soon as it is finished. All through the year, a unit book is
used during the regular recitation period for the drill that is necessary.
"Books are placed in the hands of the children on the first day of
school, and they are allowed to keep books at their desks to read in
school or at home as they desire.
"With this system, each child can go his own gait, reading as many
books in a year as he can. The best readers will read from thirty-five to
forty books. The average is about twenty books each." [Report of Read-
ing in Dodgeville (Wisconsin) Public Schools, by Supt. H. W. Kircher].
The efficiency obtained by this process of teaching reading is

shown definitely by means of various tests that have been given
in this school as shown in Table 84. The results obtained in this
manner from encouraging pupils to read so extensively are further
indicated by the large number of books read as shown at the bottom
of the table:
TABLE 84
Attainments in reading in a certain school
.................. . . .. . ..
Average age of pupils ............. 6. .7.1. .8.3. .9.2 10.3. 11 4. 12. 2.. 13.1
(Dodgeville Report, page 11.)
READING 261
(2) The Possibility of Improvement in Reading Ability. Huey

and other investigators report that they have been able, by special
effort and practice, to double their speed of reading. Miss Harriet
O'Shca (under the direction of Professor Henmon) conducted
an experiment with a group of high school pupils in which she
attempted to determine to what extent the rate of silent reading
could be increased by specific practice. Her plan was carried
out by giving each pupil a book, usually of high literary quality,
and asking him to spend 15 minutes each day in reading the book
until it was finished and to keep account of the number of lines
read. Her results showed a rather remarkable increase in the rate
of reading from the beginning to the end of the book. Some pupils
gained very little and a few others gained very rapidly, as shown
in the following table:
TABLE 85
Such experiments ought to be repeated again with additional

attention to the question whether or not comprehension improves
in a proportionate manner, and also whether this gain in the speed
of reading would carry over to reading in general. At any rate,
the results are significant in showing such large gains after a rela-
tively short period of practice and effort to improve the speed of
reading.
Peters ('17) undertook an experiment to determine the effect
of speed drills conducted regularly in connection with the reading
work during a school year. The pupils in grades three to six in the
public schools of Royersford, Pennsylvania, were divided into
"drill" and "no-drill" sections. The manner of conducting the
different sections is described as follows:
"The groups which were not to have the speed drills, and which were
to be used as a basis for comparison with those which did have, were
dealt with after the usual fashion in teaching reading. The other groups,
in addition to their oral reading, were given daily speed drills, without,
however, giving a total of any more time to their reading than the other
group received. So far as feasible both groups were taught reading at
about the same time of day, or else at equally desirable periods. They
used the same books and the same degree of enthusiasm was expected
to be put into both. The drills were, of course, conducted by the teacher
in charge of the class, and ran from November 7th to June 2nd. They
were on relatively easy reading matter, and mostly interesting narrative.
They occupied ordinarily from five to ten minutes of the reading period.
The group as a whole was told explicitly where to begin and how far to
read, and were then all set to silent reading at the same time with the
exhortation to see who could get it read first. After all, or nearly all, had
finished someone was asked to tell the substance of what he had read. If,
in this reproduction, he omitted anything he was questioned on it as a
guarantee against skimming."
Tests were given to both groups at four points during the school
year, comparing the drill groups with the no-drill groups as a
base. From the first test to the last, the results showed a gain in
speed of 18.7% and the trifling loss of 1.1% in quality of com-
prehension.
Freeman ('16) reports a series of tests made by K. D. Waldo
on the possibility of increasing speed. The lower grades particu-
larly made a very large gain in speed which was accompanied by a
parallel gain in amount reproduced, as indicated in Table 86.
READING 283
TABLE 86
l
Improvement in reading from September to March
(3) The Relation of Speed and Comprehension. This question

has been one of perennial interest, and a common misconception
has been held by a great many people regarding the mutual relation-
ship of these two aspects of reading ability. Many people believe
that a rapid reader comprehends relatively little of what he reads
and that a slow reader makes up for his slowness by a more thor-
ough comprehension of content. In order to obtain some specific
facts on this question, the writer obtained the results from a careful
test of reading in an elementary school in Port Townsend, Wash-
ington, and divided the pupils of each grade into six groups ac-
cording to their speed of reading, putting the slowest sixth together
and the next sixth together, and so on to the last sixth, consisting
1
From an unpublished master's thesis by K. D. Waldo, on eie in the library of the
University of Chicago.
of the most rapid readers. The results are exhibited in the follow-
ing table. The first column gives the average number of words read
by each group, the second column the average number of words
written representing a correct report of the thought, the third
column gives the speed of reading in terms of the number of
words read per second, the fourth column gives the percentage
of words read in relation to the number of words written.
TABLE 87
Relation between speed and comprehension
These results indicate ina striking manner that the rapid reader
comprehends relatively almost as much out of what he reads as the
slow reader, and, absolutely, he grasps nearly as many more ideas
in a given period of time as is proportional to the extra ground
covered. Specifically, the table shows that the average speed of
reading of the slowest group was 1.2 words per second, whereas
the speed of reading of the fastest group was 4.9 words per second.
The percentage of comprehension in relation to the amount read
was 46% for the slowest group and 37% for the fastest group. In
other words, the percentage of comprehension is almost as large for
the fast group as for the slow group; or, comparing the first and
second columns, we note that the fast group read almost exactly four
times as fast as the slowest group and wrote three and one-third
times as much as the slowest group. In other words, the ratio of
the speed of the reading between the fast and slow group is one to
comprehension is one to three and one-third.
four, while the ratio of
The inference is then that the rapid reader derives relatively almost
as much out of what he reads as the slow reader. Absolutely he
obtains several times asmany ideas. Concretely, the comparison
may be made in another way: Of two persons belonging re-
still
spectively to groups one and six, each reading for one hour, the
READING 285
fast reader would cover four times as much ground and derive three
and one-third times as many ideas as the slow reader. The fast
reader, therefore, has an astounding advantage over the slow reader.
These results consequently give no corroboration for the common
belief that an inverse relation exists between speed and compre-
hension in the fast and the slow reader.
Similar results have been presented by Judd. These results
are shown in the following diagram, Figure 61:
"For the purpose of this study of the relation between rate and quality,
all ofthe individual records of Cleveland pupils were divided into classes.
First the speed records were arranged in order from the most rapid to the
slowest. The most rapid of these records \vere designated by the simple
* 1
term 'rapid.' In this class of rapid records were included the most
rapid 25% of all the records. In like fashion the slowest 25% of all the
records were set aside and designated as 'slow.' This left half the rec-
ords, or the middle 50%, which were designated as of 'medium speed.'
In like manner the 25% of all records which were qualitatively the best
were designated 'good'; the 25% which were qualitatively the worst were
designated 'poor,' and the term 'medium' was applied to the middle 50%.
"It becomes a very simple matter to assign all records in each grade
to the appropriate class and determine the percentage of the grade which
falls into this class. Diagram 59 gives the results, the percentages being
in each case the nearest whole number to the calculated figure, and the
size of the circle being proportionate to the size of the class indicated.
"These figures serve to emphasize the fact that good readers are
usually not slow and poor readers are usually not fast. It is evidently
not safe to attempt to lay down any absolute rule. There are good read-
ers who are slow. In some cases such readers may be temperamentally
slow. But even making allowance for such individual peculiarities, the
figures show that good reading and slow reading are not incompatible.
In like manner there are a certain number of children who read rapidly
but retain little of what they read. With the figures in hand a teacher
can profitably study her class and determine somewhat more completely
than it is possible to do for the whole school system what are the special
explanations of each individual type of ability.
"For the purpose of this survey the general fact that high rate and
good quality are commonly related, and that low rate and poor quality
are commonly related, is of great importance."
King ('17) tested the reading ability of 94 college students,

half of whomwere asked to read slowly and carefully and half to
read rapidly and carefully. They read for ten minutes and, by
observing a clock, the fast group were to read twice as rapidly as
the slow group. Comprehension of the material was tested by

answers to questions. The results showed that the accuracy of
comprehension of the fast group was 44.5% and that of the slow
group was 53.3%. In another test the subjects were divided into
groups of naturally fast and naturally slow readers as determined
by a preliminary test. The returns showed that the fastest 25% of
the group had a comprehension of 50.2%, the slowest 25% had a
comprehension of 48%, and the middle 50% had a comprehension
FIG. 61. Relation between speed and quality of comprehension. After

Judd' (16, p. 155).
of 46.5%. King interprets his results in favor of the slow readers.

As a matter of fact they show, however, the same relation between
speed and comprehension as found by other experiments. The dif-
ferences in comprehension between the fast and the slow readers
are very small, being slightly in favor of the latter group; but, when
one remembers that the fast readers in the first experiment read
twice as much advantages are distinctly on the side of the

text, the
fast readers. Whipple and Curtis ('17) in their study of skimming
in reading also found that the slowest reader was the poorest re-
producer and the best reproducer was one of the fastest readers.
READING 287
(4) Relation between Oral and Silent Reading. Considerable

attention has recently been given to the importance of relative
stress upon oral as against silent reading or vice versa. The belief
held by most of the investigators of this problem is that the schools
have placed too much emphasis upon oral and not enough upon
silent reading.
Superintendent Oberholtzer ('14) of Tulsa, Oklahoma, made a

from grade to grade
series of tests to ascertain the relative increase
in the speed of silent and oral reading. Tests were given to 1,800
pupils. The following figures give the average speed of oral and
silent reading in terms of words read per second:
TABLE 88
Speed in oral and silent reading. After Oberholtzer ('14)

WORDS READ PER SKCOND
GRADE ORAL SILENT
3 2.1 2.3
4 2.3 2.6
5 2.4 3.1
6 2.8 3.9
7 3-i 4-7
8 3-9 4-3
These results indicate that in the third grade the speed of oral and
silent reading is very nearly identical, but that silent reading in-
creases thereafter considerably faster from grade to grade, so that
in the eighth grade the rate of silent reading is approximately one
word per second, or about 25% faster than oral reading.
Mead ('15) tested 112 pupils in five classes in the sixth grade in
both oral and silent reading. He made six tests, each two minutes
long, and determined the speed by the number of lines read and the
comprehension by the number of "points" reproduced in writing.
His results are as follows:
TABLE 89
Relative ability in silent and oral reading. After Mead ('15)
Av. No. Av. No POINTS PFR CENT REPRODUCED
LINES READ REPRODUCED OF AMOUNT READ
Silent reading 39-4 16.4 38. 7
Oral reading 33.6 12.1 32.9
Each of the five classes did better in silent reading than in oral
reading. Mead concludes: "From the results of these five classes
O -f O i-< O OS
0<S
fc 00
3> U
c
Sti>
OJ f fS O t^- 00 O O 00 >
>S*
READING 289
we are more convinced than ever that our schools devote altogether
too much time to oral reading and too little to silent."
Within the following two years, Mead repeated the same tests
with 340 pupilsin grades three to ten, excepting the ninth, and ob-
tained corroborative results. "Fifteen out of seventeen classes
did better by the silent method of reading. Seventy per cent of the
children taken separately did better by this method." The de-
tailed facts are given in Table 90.
Pintner made eight tests, two minutes in length, with 23 pupils
in the fourth grade and found the following results:
TABLE 91
Relative ability in silent and oral reading. After Pintner ('13)
Av. No. LINES Av. No POINTS

READ REPRODUCED
Silent read ing 28. 18.
Oral reading 20. 15.
Thus it appears in all of the tests that silent reading has the lead
over oral reading in both speed and comprehension. H. A. Brown
('16) and others believe that there should be no oral reading as
such after the third grade, that silent reading should be empha-
sized instead, that above the third grade teaching to read should be
teaching to study, and that there should be a great deal of sponta-

neous silent reading.
Superintendent Llewelyn ('16) of Mt. Vernon, Indiana, at-

tempted to increase among his pupils the rate of reading and to
stimulate interest in reading. He adopted the plan of giving oral
reading three times per week instead of five times, of supplying a
motive for silent reading by asking questions to test the knowledge
of silent reading, of using the two extra recitation periods for live
questions and discussions of what had been read, of stimulating the
reading of library books and of having frequent book reviews. Each
book was assigned to two pupils so as to make the discussion more
interesting and lively.
No quantitative tests or comparisons were made, but the results
reported were to the effect that the plan produced a love of reading,
that teachers became more effective because they had to prepare
for the giving of suggestions, that oral reading did not deteriorate
and that reading was much more extensive as indicated by the fact
that the class read about ten times as many books as before.
(5) Phonics. The tendency in recent years has been in the

direction of less emphasis upon phonics and upon formal drills in
general. Various investigators believe that extensive emphasis

upon phonics and articulation in oral reading tends to establish
slow habits of pronunciation and interferes with the proper develop-
ment of speed in silent reading. Final experimental evidence on
this question as on many other questions is lacking.
Currier and Duquid attempted to decide by experiment whether
it was advantageous to teach phonics or not. No definite compara-
tive tests, however, were made, but their impressions were that the
phonic classes concentrated on the word and the sound at the ex-
pense of the sense, that their reading was less smooth and slower
and that their ideas were confused. On the other hand, they re-
ported that the no-phonics classes enjoyed their reading, that they
read more swiftly, more expressively, and more for the sense of the
material but that they did not read quite so accurately. The
ability to attack new words was about the same. Experiments
of this sort ought to be carried out more extensively and compari-
sons of the results should be made by means of more precise, quan-
titative measures.
(6) Comparisons of General Methods or Teaching Reading.

Numerous methods of teaching reading have been advocated by
publishers and educators, but no one knows with certainty the com-
parative merits of these methods nor which ones are most economic-
ally productive of the best development in reading ability.
Superintendent Harris ('16) of Dubuque, Iowa, in conjunction
with H. W. Anderson, undertook an experiment to determine the
relative merits of three systems of teaching reading. The teachers
had felt for some time that they were not securing the results in
reading that they might reasonably expect. The experiment is
described thus:
"As a first move to remedy this unsatisfactory condition, the superin-

tendent instituted a trial of the Beacon system at the school hereafter
designated as School A, and of the Horace Mann system at school B.
The present year was the second in which these systems have been thus
used. The Aldine system was continued in use in the four other schools,
C, D, E, F, mentioned in the report following. The teachers who worked
with the Beacon system were enthusiastic in its favor; but their opinions,
no matter how enthusiastically declared, were not sufficient to secure
general agreement. With this state of affairs existing, the problem re-
solved itself into how to raise the question of the efficiency or worth of
READING 2QI
the various systems of primary reading out of the realm of mere opinion
and place upon bed rock by scientific evaluation of results actually
achieved.
"In order to accomplish this it was decided to test:
"i. The mechanics of oral reading in the second half of Grade I (lA)
and in the two divisions of Grade II (2B and 2A).
"
2. The silent reading in Grade II: for (a) rate; (b) comprehension.
"These tests were given in the following schools:
"A Where the Beacon system is used;

" "
"B Horace Mann system is used;
" "
"C D E F Aldine system is used.
"It was believed that these tests would show the results obtained by
the Beacon system and the Horace Mann system during the first two
years of their use, and offer an opportunity for comparison with each
other and with the Aldine system."
Oral reading was tested with Gray's oral reading scale, and silent
reading was tested by Starch's tests measuring the speed and com-
prehension of reading. The results obtained are given in the fol-
lowing tables.
TABLE 92
Results of the oral reading test
"These scores seem to indicate that in each grade, School A, where

the Beacon system is used, excels all other schools in the simple mechanics
of oral reading. In fact, the scores of Grade IA at School A
are better
than the Grade IIB scores of both the C and D, and not particularly far
behind the Grade HA
scores of the Aldine group of schools. The figures
also indicate that the results at School B, where the Horace Mann system
is used, are slightly better than at the schools where the Aldine system is
employed.
"The results of the Oral Reading Tests seem to show conclusively that
the pupils trained by the Beacon System are very greatly superior in the
mechanics of oral reading to those trained under the Aldine or the Horace
Mann systems of reading."
2Q2 EDUCATIONAL PSYCHOLOGY
Results of the silent reading test
TABLE 93. Rate of reading
C D E F
.8 .9 1.7 2.
.8 .9 1.7 1.9
2. 1.7 1.8 2.3

1.9 1.6 2. 2.2
"Table 93 shows the average and median rate of silent reading. It

shows that in Grade IIB the Beacon pupils at School A read at the
average rate of 2.2 words per second, while the non-Beacon group in the
same class read at the rate of 1.3 words per second. The Aldine pupils
read at the following rates: C, .8; D, .9; E, 1.7; F, 2. The Horace Mann
pupils at School B read at the rate of 1.7 words per second. This shows
that in this grade the Beacon pupils read .2 of a word faster than the
nearest competitor (School F) and that they read more than twice as
rapidly as two of the Aldine schools. The difference between the Beacon
group of pupils and the best Aldine pupils is not significant, however.
"In Grade IIA the Beacon pupils at School A read at the rate of 3
words per second, while the non-Beacon pupils in the same class read at
the rate of 2.3 words per second. The table shows that the rate of reading
in Grade IIA is clearly faster at School A (Beacon pupils) than at any
other school; the nearest competitors being School F and the non-Beacon
fijroup at School A, where the pupils read at the rate of 2.3 words per
second."
TABLE 94. Comprehension of reading

NON-
BEACON BEACON H. M. ALDINE
A A B C D E F
Average 26.2 20.5 18.4 10.1 10. 20.6 17.
TT_3 |
\Median 30. 18.1 20. 9.4 8.5 19. 17.5
f Average 31.5 19.9 15.1 20.1 n.8 24.3 22.3

IIA
[Median 31.9 17.5 19. 19.9 8. 23.3 18.3
"Thus, the Beacon pupils in School A, Grade IIB, on the average

reproduced 26.2 words, while the non-Beacon pupils in the same class
reproduced 20.5 words. The nearest competitor to the Beacon group of
pupiJ-. is the group at School E, which reproduced on the average 20.6
word's 5.6 words behind the Beacon group. Schools C and D made the
remarkably low grades of 10.1 and 10.
"In Grade IIA, the Beacon group of pupils reproduced 31.5 words,
while the non-Beacon group in the same class reproduced only 19.9 words.
The nearest competitor in the Aldine group of schools was School E,
READING 293
where the average number of words was 24.3. The results obtained
through the Horace Mann system seem to be below the average, this
class reproducing only 15.1 words.
"The results of the Silent Reading Tests seem to show conclusively
that the pupils trained by the Beacon System are far superior to those
trained under the Altline or the Horace Mann systems of reading.
"It was realized that objections might be raised to the results herein
shown, on the ground that the teacher rather than the system was the
strong factor in the results. While it is highly improbable that out of six
different corps of teachers, the group teaching the Beacon System would
be uniformly better, in each section tested ; yet, as an absolute check upon
this phase of the matter, the Silent Reading Tests were given in Grade
IIIB, which entered school before cither the Beacon System or the Horace
Mann System was placed on trial in any school and which therefore could
not have had, in any one of the six schools, its initial training in either of
the two systems named." (Starch's Reading Test, Series A, No. 3, was
used.)
TABLE 95
Showing median rate and comprehension of silent reading in Grade HIB

A B C D E F
Rate 2.9 1.8 19 1.9 2 . 1.8
Comprehension 24. 27.5 12.3 29. 27.5 19.5
"This table sho\vs that the pupils at School A read more rapidly than
those at any other school, their median rate being 2.9 words per second,
while the nearest approach to this rate was 2 words per second in School
E. However, comprehension three schools excel Grade IIIB at
in
School A. A reproduced 24 words correctly, while those
Pupils at School
at Schools B, D, and E reproduced 27.5, 29, and 27.5 words respectively.
"Furthermore, in the tests made in Grade II, while the Beacon group
at School A excelled all other groups in rate and comprehension of silent
reading, several of the other schools excelled the non-Beacon group at
School A in both these points. Thus, in the rate of silent reading, in
Grade IIB, the non-Beacon group in School A was excelled by Schools B,
E, and F, and in Grade IIA, the non-Beacon group at School A was
equalled by School F; and in comprehension, the non-Beacon group at
School A was excelled by School E in Grade IIB, and by Schools C, E, and
F Guide IIA.
in
"These facts show rather conclusively that it was not the superiority
of the teaching which determined the results of the tests, since teachers
in other schools than School A, both in Grade II and in Grade III, with
pupils trained under the old systems, secured results as good as or better
than those secured by the teachers at School A with pupils whose first
training also had been under the old system."
2Q4 EDUCATIONAL PSYCHOLOGY
Whether or not the results would be generally superior in other

schools and under other teachers cannot be inferred perhaps from
these results with complete finality. The experiment, however,
isinteresting and is here cited chiefly for the purpose of showing
what should be done in the way of scientific comparisons and tests
to determine the most proficient methods of learning and teaching
reading.
Gray made a comparison of the attainments in reading in 44
schools, 26 of which had used the Aldine method, 17 the Ward
method, and one a method of its own. The results showed no con-
sistent or uniform superiority of one method over another. The
average test scores were approximately the same.

Waldo ('15) compared the Howe system with the Ward system.
The latter had been used up to the sixth grade in all the schools
except one in which the former had been used. The tests were
not carried out in a sufficiently careful manner to warrant reliable
conclusions.
Hendricks made a comparison of schools in which the Rational
method had been used with schools in which no special method
had been used. He found the former schools superior. This does
not necessarily prove the superiority of the Rational method but
probably the advantage of well-organized systems over less well
organized systems. The factors making for efficiency are so
numerous and intricate that much more extensive and far more
careful experiments will have to be made to demonstrate compar-
ative values of the different systems or methods of teaching reading.
(7) Suggestions for Improvement in Reading Ability. Ex-
perimental results have brought about a radical shift in emphasis
upon the aims to be accomplished in reading. In the first place,
there has been a shift from emphasis upon oral reading to emphasis
upon silent reading because facility in reading, in the sense of
thought-getting, can be developed to a much higher degree of
proficiency in silent reading and because nearly all the reading done
by the average adult is silent reading. In the second place, there
has been a distinct shift from emphasis upon slow reading to em-
phasis upon rapid reading because tests have shown that rapid
reading does not mean a corresponding loss of thought, as assumed
by many teachers, but, instead, rapid reading is accompanied on
the whole by an almost equal ability in comprehension. The rapid
reader will derive almost as many more ideas in a given period of
time as is proportionate to the greater amount of text covered.
READING 295
Pupils were formerly told that they must not read fast but that
they should read slowly because they would then get the thought so
much better. In the third place, there has come along with these
two changes a shift from the mechanics of reading to the content
of reading. Former aims of reading are fairly represented by the
following answers, given by pupils who had finished the grammar
school, in response to the question, What is your idea as to what
the reading lessons were for? Some of the answers were: "To learn
to pronounce," "To help us in reading before people," "Just to
pass away the time," and "I thought it was to learn us to use better
language." (Briggs '13).
Accordingly then the aim in reading to-day is the development
of speed in reading and a parallel gain in thought-getting. Recog-
nizing this change in aim, what definite suggestions can be made
to facilitate improvement in these aspects of reading ability?
Experimental results are as yet too few to make many specific
recommendations with complete confidence. However, several
important suggestions may be offered.
(a) As to the speed of reading: Force yourself to read more
rapidly. Continuous effort and practice in this direction will very

materially increase the rate of reading as shown by experiments.
Probably most adults, as well as most children, read far more
slowly than they are capable of reading. So far as we may judge
on the basis of experimental investigations of the reading process,
speed of reading depends chiefly upon the rapidity of the assimila-
tion and upon the span of attention and less upon the other steps
in the reading process enumerated at the beginning of this chapter.
The rapid reader assimilates more swiftly and grasps more words
at each fixation.
Forcing oneself to read more rapidly than one's customary rate
with proper comprehension, but in the course
will at first interfere
of persistent practice the more rapid visual and mental activities
will become habitual and the comprehension will probably then
come up to its normal amount.
(b) As to comprehension of reading: Grasp the thought with
concentrated attention, (i) Speeding up the rate of reading tends
also to stimulate greater concentration of attention upon the
whole thought content. That both speed and comprehension of
reading may be very greatly improved by practice, by reading a
great deal with the definite aim of improvement, is shown by such
results as have been obtained in the Dodgeville schools and else-
where. The remarkable rapidity of reading was accompanied by

an equally remarkable ability in thought-getting.
(2) Stop frequently to recall the essential ideas read. Compre-
hension will be greatly assisted by stopping at short intervals and
asking oneself the question, What have I really read? What are
the essential ideas? This will not only stimulate reading for
thought-getting but will also help to fix ideas in mind and to relate
them to larger units.
(3) Acquire the habit of looking for the essential ideas. This is
very important in efficient reading. Skill in so doing will greatly
facilitate speed of reading by covering more ground and by know-
ing what may be read very hurriedly or even omitted.

(4) Tests at frequent intervals. Measurements by means of the
standard reading tests or by means of improvised tests patterned
after one or another of these testing plans should be given at fre-
quent intervals for two reasons. In the first place, they will afford
the pupil himself a definite basis for discovering his reading abil-
ity and, by keeping his own record from test to test, they will
furnish a powerful stimulus to the pupil to surpass his own preced-
ing attainments. This point was elaborated more fully in Chapter
XI. In the second place, intensive tests with emphasis on both
rate and comprehension will give the pupil practice in the phases
of reading in which the school has in the past not furnished ade-
quate training. The great emphasis upon oral reading has tended
to slow habits of reading and placed the primary emphasis
instill
upon the mechanics of reading rather than upon thought-getting.

Tests of reading ability, for this purpose, may be improvised and
given as often as desirable by having the pupils in a class turn to
a specified page, read with their maximum capacity for a limited
interval, say half a minute, a minute, or several minutes, note the
point of stopping, and then write a full account of the thought
content, or answer questions. Such a procedure might profitably
become a regular part of the instruction in reading.
CHAPTER XVII
HANDWRITING
PROCESSES OR STEPS INVOLVED IN THE ACT OF WRITING
An analysis of the various steps involved in writing (or copying)

similar to that made of the reading process reveals the following
elements:
(1) Reception upon the retina of the form of the letters to be
written.
(2) Transmission of the visual' impressions from the retina to
the visual centers of the brain.
(3) Recognition or perception of the letters through the visual
and other association processes.
(4) Transmission of nerve impulses from the visual centers to
the motor centers of the fingers, hand, and arm.
(5) Transmission of nerve impulses from the motor writing
centers to the muscles of the fingers, hand and arm.
(6) Execution of the muscular movements involved in the
writing act.
(7) Return kincesthetic nerve impulses from these muscular
movements back to the kinaesthetic centers in the brain and
thence through steps (5) and (6) to help in correcting and con-
trolling the writing movements.
(8) Return visual impressions of the letters or marks as actually
executed back through steps (i), (2), (3), (4), (5), and (6), in
helping to correct and control the writing movements.
What do we know concerning the manner of operation and the
importance of each of these steps in the complete writing process?
Steps (i), (2) and (3), while important when the child first learns
to write, practically drop out in the skilled writer in whom the
ideational and visual processes in the brain and the kinaesthetic
sensations from the writing act itself serve directly to control steps
(5) and (6). In the practiced writer they simply serve as a general
control in securing the proper alinement, size, and spacing of the
letters. The practiced writer can write about as well with his
eyes dosed as with them open. The chief difference is in such
297
features as alinement, spacing and heaviness of strokes. The

letters themselves can be formed practically as gracefully and as
quickly with the eyes closed as with them open. Miss Downey
('08) found, in her experiment to determine the effect of different
distractions upon the writing habit, that the visual factor has an
obvious function in acquiring new coordinations but has little
effect upon the fully formed habit. The visual perception of the
form of letters, while important, is probably not as important in
learning to write, even in the beginning, as it is in learning to read,
because the writing act depends largely upon the development of
muscular control. The visual perception of the form of the char-
acters to be written and of those actually written must serve as
a guide in the attempts at writing, but quickness in visual per-
ception is not as important in the writing act as in the reading act
because the writing act is much slower even in the skilled writer
than the visual perception of the forms is, whereas reading depends
directly upon the rapidity of visual perception. Then, also, the
child has usually learned to recognize the forms of the letters
when he begins to learn to write. At any rate its importance is
rather secondary in the writing process.

The chief elements in the writing process are those connected
with the steps (5) to (8), particularly those connected with the
steps (5) and (6). Development of skill in any muscular move-
ment which is not instinctive or at least not as mechanically pre-
cise as an instinct, proceeds by trial and error. Attempts are made
at carrying out the desired movements. At first some of them
sucecd, most of them do not; and through continued trials the
erroneous attempts decrease and the successful ones increase until
perfect control is established. The adult scarcely realizes the
utter lack of control in the early attempts on the part of the child
in making the writing movements. The nearest approach that
the adult can make toward realizing the actual difficulties of the
child consists in such an experiment as the tracing of an outline
as seen in a mirror. The spatial relations are so completely new
and different thata person has little or no conception of the direc-
tion in which to move. Figure 62 shows a reproduction of the
first tracing of a star outline as seen in a mirror. It does very
little good to reason about it. One may think he is going to move
in a certain direction but finds upon making the movement that
he is going in an entirely different direction. The child learns to
write very much after the same manner. He proceeds by trial and
HANDWRITING 299
error. The movements somehow become more deeply

successful
fixed in the nervous connections and consequently more and more
numerous. The correct movements finally become associated
with the visual perception of form and direction so that the move-
ment can be carried out at will with precision and grace.
It is a matter of common observation that a child beginning to
write not only makes the movements very slowly but also with
much excess pressure. The latter point was investigated by Meu-
mann. He had children and adults write on a little platform which
was so supported that the amounts and changes of pressure during
the writing were transmitted to a light lever which traced them ac-
FIG. 62. Record of tracing of a star-outline as seen in a mirror.
curately on a moving drum. It is natural to find that men tend to

exert a greater pressure than women. Gross, working in Kraepe-
lin's laboratory, found that men average nearly twice as great a
maximum pressure as women. These records also serve to illustrate

the fact that each letter has a fairly characteristic pressure rhythm.
Freeman and others have shown by exact methods that the rate
at which the writing point moves differs greatly in various parts of
a given letter. In general the up and down strokes are the most
rapid while the sharp turns and angles are the slowest. The speed
curves of different persons writing the same letter consequently
show striking similarities.
Sex Differences. The striking differences between the hand-
writing of different individuals has led many people to believe that.
there was a definite relation between the handwriting and the per-
sonality of the writer. The greater pressure exerted by men as a
class suggests that the sex of the writer determines at least certain
characteristics of the writing. This has been investigated inde-
pendently by Binet, Downey, and Starch who find that untrained
subjects can determine the sex of the writer in from 65 to 75% of
the specimens (50% being pure chance). The author made a com-
parison of the writing ability of 2,113 boys and girls in the Madison
schools and found the differences exhibited in Figures 63 and 64.
It appears from these graphs that the median of the girls is above
that of the boys in both speed and quality, but particularly in
quality. The difference in speed is very slight.
Miss Downey ('10) attempted to determine sex differences in
handwriting by selecting from envelopes by a chance method 100
samples of men's writing and 100 samples of women's writing and
by asking thirteen persons to judge whether a given sample had been
written by a man or by a woman. The thirteen persons made re-
spectively the following percentages of correct judgments: 60, 60,
61, 64, 66, 66, 63, 68.5, 70, 70.5, 71.4, 71.5, 77.5. These show
that on the average a judgment of sex as revealed in handwriting
is correct 67% of the times, or two out cf three times. Miss Downey
women showed less variability
also reports that the writing of the
and more conventionality than that of the men. Those samples of
writing by women which were called masculine were generally from
persons accustomed to doing a great deal of writing.
Correlation of handwriting with other traits. A
good deal of mis-
leading character interpretation is based upon various features of
handwriting. Experimental work needs to be done in this field, but
it is probable that there is nothing to the claim of graphologists
that handwriting reveals such traits as energy, clearness and sim-

plicity, vanity, or self-consciousness.
Gesell ('06) reports that there is a close correlation between
quality of writing and intellectual ability, but as a matter of fact
his results show, as pointed out by Thorndike, a correlation of only
about .30.
The author ('15) found for children a correlation of .31 be-
tween writing ability and general scholarship. Thorndike ('10)
reports that for adults the correlation between writing and schol-
arship is zero. The probability is that the small correlation exist-
ing in the case of children is due to the fact of receiving instruc-
tion in writing and to the attention given to it. The better pupils
HANDWRITING 301
do somewhat better in writing because they probably pay more

attention ormake more careful efforts. So far as adults are con-
cerned, poor handwriting is no indication either of high or low
intelligence, since the correlation is approximately zero.
Professional graphologists have claimed that a great number of
specific traits of writing are determined by corresponding traits
of character on the part of the writer. Thus the lines of writing
of ambitious persons are supposed to slope upward from left to
right. Hull ('19) investigated some of the more persistent claims by
computing correlations between the traits of character of 17 univer-
sity fraternity men, as judged by their fellows, and exact measures
of samples of their writing. The correlations in each case were
approximately zero, showing these claims to be entirely unfounded.
TIIE MEASUREMENT OF EFFICIENCY IN WRITING
(a) Essential dements to be measured. The two important aspects

of writing that must be measured are speed and quality, including
under the latter legibility and form or beauty.
(b) Methods of measurement. Speed of writing is now generally
measured in terms of the number of letters written per minute.
Quality may be measured by cither one of several scales, the Thorn-
dike Scale ('10), the Ayrcs Scale ('12), the Starch Scale ('19), and
others. The Thorndike Scale consists of a series of 18 steps or
qualities of handwriting, each step consisting of one or more spec-
imens of writing of the appropriate quality. Step zero represents
an attempt at writing but as such is entirely illegible and devoid
of beauty. Step 18 is a perfect copper plate specimen. The steps
from o to 1 8 represent equal units of increase in quality. The
Ayrcs Scale consists of 8 steps designated as 20, 30, up to 90.
Each step contains three specimens of equal quality, a vertical,
a medium, and a slant sample. The recent revision of the Ayres
Scale, the Gettysburg edition, contains only a medium slant
specimen for each step. The successive steps represent uniform
increments of legibility in writing. The Starch Scale is composed
of a scries of 20 steps arranged in the order of merit or excellence
of writing as judged by 400 persons. (See the original monographs
for detailed description of the preparation of these scales.)
A sample of handwriting is measured by any one of the scales by
putting it alongside the scaleand determining which step it is most
like in general quality. Speed and quality should ordinarily be
measured simultaneously in the same sample because these two

aspects of writing have a functional relationship. A test of writing
should, therefore, be made by having the pupils write a short, simple
sentence repeatedly as many times as they can in, say, two minutes,
doing it as well as they can. Speed is then measured by the number
of letters written per minute and quality is rated by one of the scales.
Freeman ('14) has prepared a set of five analytical scales for the
purpose of rating handwriting from the standpoint of uniformity of
slant, uniformity of alinement, quality of line, letter formation, and
spacing. Each scale contains samples of three successive degrees
-
84
u
FIG. 63.
12345678 -L. -L.
Grades
-L.
Sex difference in speed of writing.

-L. JL
After Starch ('13).
of merit. These methods of rating ought to be useful in calling

attention to defects in particular features of writing.
The A. N. Palmer Company has published a scries of five or six
samples of successive degrees of value for each grade. A percent-
age value for posture, movement, speed, and formation is given for
each sample. The value of the sample as a whole is expressed by the
average of these four estimates.
The Zaner and Bloser Company has also issued a set of specimens
for evaluating handwriting consisting of a series for grades one and
two, another series for grades three and four, and a third series for
grammar and high school classes. Each series has a number of
samples whose rating is expressed in terms of percentage values.
Comments concerning the defects or excellencies are appended to
the various specimens.
HANDWRITING 303
The manner in which the values of the samples were determined

isnot indicated for either the Palmer or the Zaner scales. This im
pairs their scientific value. From the practical standpoint they are
commendable in that they suggest specific attention to, and evalua-
tion of, important elements in handwriting.
(c) Results and uses of measurements. In general the results and
uses of measurements in handwriting are the same as those pointed
out for reading. It is possible by means of these measurements
to determine more precisely the actual writing ability of a pupil,
class, or school and to compare it with standard averages
for cor-
FIG. 64. Sex differences in quality of writing. After Starch ('13, p. 461).
responding grades in schools generally. These standards of at-

tainment for the ends of the respective school years are as follows:
TABLE 96. After Starch ('16, p. 83)
Standards of attainment in writing

GRADES 1 3 4 S 6 7
Speed per minute) ...

(letters
20 38 47 57 65 75 83
Quality (Thorndike scale) 6.5 8.2 8.7 9-3 9.8 10.4 10.9
Quality (Ayres scale) 33 37 43 47 53 57
8 9 9.7 10.3 10.9 11.4 12. o 12.5
Quality (Starch scale)
By reference to these standards of attainment it is possible to

define quite accurately the speed and quality of writing of a pupil
or class by saying, for example, that a given pupil in the fifth grade
isable to write 65 letters per minute at quality 9, Thorndike scale.
The value of exact measurements of handwriting, as of any educa-
tional products, consists in the diagnosis of ability as it actually
exists in different pupils and schools, in the measurement of the in-
fluence of different factors and conditions upon learning to write,

and in the determination of the mutual relationships of various
aspects of writing. Asurvey of our present knowledge concerning
these matters will be given in a later section.
ECONOMIC PROCEDURE IN LEARNING TO WRITE

What influence do the various factors, conditions, and methods
in learning to write have in promoting or hindering the develop-
ment of skill in handwriting? This question could be answered
finallyand fully only by the careful isolation of each factor under
experimental conditions and by determining its effect upon the
progress of learning to write. Substantial beginnings have been
made in the direction of answering some of these questions, but little
or nothing is known about most of them.
(i) Perception of the Forms Written or to Be Written. This

topic requires consideration of two general questions: (a) What
are the most advantageous conditions for the visual perception of
the forms to be written? (b) What sort of form or model should be
presented? The former may be answered by the observa-
question
tion of certain obvious rules, namely, that the writing surface should
be placed before the eyes at the proper distance, especially not too
near, so as to avoid eye strain, and in a position directly in front of
the eyes so that the points on the paper to be successively fixated
may be at equal distances from both eyes, thus avoiding unequal
accommodation in the two eyes. The paper should not be glazed,
so that it will not produce a glare, and for young children the sur-
face should be rather rough so that it will easily take pencil marks.
The second question is more complicated. The sort of models
to be presented is obviously highly important since imitation,
both voluntary and involuntary, probably plays a large part
in the acquisition of writing skill. The author (*u) made an
experiment in which he attempted to measure the unconscious
effect of different models of writing upon the normal writing of
adults. Four samples of writing were obtained from each of 106
university students. In order to avoid any suggestion of imita-

tion, written rather than oral directions were given stating that
they were to produce samples of their writing and that they should
proceed at once to write the passages presented without further
thought or questions. The four passages put before each person
consisted of (x) a typewritten selection, (2) an extreme vertical
HANDWRITING 305
model, (3) an extreme slanting model, and (4) a large model with
many flourishes. The purpose of the typewritten passage was to
obtain at the outset a sample of the normal writing of each person.
The other three models were taken from school copy books.
After the experiment was finished, each person was asked whether
he had tried purposely to imitate the various models. Three
persons stated that they had intentionally modified their styles
of writing. Their records were thrown out. The samples pro-
duced by the remaining 103 persons were carefully measured to
ascertain their slant and size. Slant was measured by means of
a specially prepared, transparent device with ruled lines for de-
termining the angle of inclination of certain tall letters, such as 1,
f,and p, with the base line on which the words were written. Size
was measured by determining the horizontal width of letters by
measuring the length of words and dividing by the number of
letters in the word.
These measurements showed that the average tendency for this
group of persons was to make the letters distinctly more vertical
when the vertical model was before them and more slanting when
the slanting model was before them as compared with their nor-
mal styles of writing. They also tended to write slightly larger
when the large model was before them. The amounts of these
changes were as follows:
TABLE 97
Average inclination of 1 in thenormal writing 65 i degrees

.
"
Average inclination of 1 written from vertical copy 68.8
written from slanting copy
"
Average inclination of 1 61.5
"
Change from normal to vertical 3.7
"
Change from normal to slant 3.6
"
Total range of change 7.3
Average width of letters in normal writing 4.33 mm.
Average width of letters written from large model 4.85 mm.
When we realize that the handwriting of adults is a pretty

firmly fixed habit, the amount of unconscious imitation is consid-
erable, being a total of 7.3 degrees in slant and of .52 millemeters
in width. We may infer that with children whose writing habit
is in process of formation, the element of unconscious imitation
plays a much larger part. Furthermore, it seems quite probable,

although no experimental proof is at hand, that the style and
1
quality of writing of the teacher distinctly influences the writing

of the pupils, especially so because the writing done by the teacher
in the presence of the pupils for the purpose of showing them how
to write, is likely to be more efficacious in securing imitation than
a static model in a copy-book would be. It would seem, therefore,
highly imperative that every elementary school teacher should be
a reasonably good writer.
In connection with the survey of penmanship in the Grand
Rapids, Michigan, schools, Freeman (Judd, '16) reports that
"Grand Rapids adopted about five years ago a new system of pen-
manship. Up to that time the writingwas not regarded as satisfactory.
A part of the difficulty was thought to be due to the inability of the
teachers themselves to write well enough to furnish a good example to
the pupils. Accordingly, by action of the Board of Education, all teachers
in the elementary schools were required, as a condition of promotion, to se-
cure a Palmer certificate. This rule has been recently enforced with strict-
ness and the writing in the schools is reported to be greatly improved."
Should the model presented to the pupils be vertical or slanting?

Should it be plain or contain flourishes, decorative curves and
shading? Should it be angular or rounding? The answers to
these questions are at present largely matters of opinion and con-
venience rather than matters of scientific determination. Some
years ago vertical writing came into general use because it was
thought to be more legible and less productive of spinal curvature.
But it has largely disappeared for the obvious reason that almost
everyone naturally falls into the habit of writing a medium slant,
no matter what style of writing was taught to him previously.
The average slant for adults, as shown in Table 97 is about 65
degrees with the base line or about 25 degrees with the vertical
line. Whether slant writing actually tends somewhat more to
produce spinal curvature is doubtful. The difference in legibility
between vertical writing and a medium slant writing is also prob-
ably very small. The letters should probably be of a medium
slant and should be relatively plain and free from flourishes since
these take time and add nothing to the general value of the writ-
ing, and finally, they should probably be moderately rounding
because extreme roundedness is likely to reduce speed and ex-
treme angularity is likely to reduce legibility.
Graves ('17) classified 604 samples of handwriting according to
slant and then studied the speed and quality tendencies of the
vertical, medium slant, and extreme slant group. The final aver-
ages of speed and quality are shown in the following table:
HANDWRITING 307
TABLE 98. After Graves

WORDS WRITTEN QUALITY
IN 5 MINUTES (AYRES SCALE)
Vertical 91 .6 57-98
Median slant 96 . i 48 22
.
Extreme slant 101 .7 43-58
There is revealed clearly a positive connection between slant and
speed on the one hand and poor quality on the other. That is the
"extreme slant" writers write more rapidly and more poorly than
the "vertical" writers.
(2) Length of Period of Practice. What is the most productive
practice period in learning to write? Even such a question as
this, which is capable of definite experimental solution, has been
answered only in part. The answer given by school programs in
the time allotted therein for writing, is based largely on opinions
instead of facts.
TABLE 99
Quality of handwriting at roughly the same rate in seven school systems. After
Thorndike ('10)
Median results for eighth-grade pupils
At 70-79 words 9.8... 9.8... 9.9..
Systems A and B devote no time to writing as such in grades 7 and 8.
System C devotes 50-60 minutes weekly

" "
D 73-100
" "
Systems E and G devote 60-90
'
System F devotes 75 grade

"
F 30 8.
Thorndike ('10) compared the writing in seven school systems

as given in Table 99, and concluded that time was practically
negligible. He says:
"What these facts do prove is: First, that at least three systems (C,
D, and E) get little or no better results at a time cost of about 75 minutes
a week than two systems (A and B) do at zero time-cost; second, that
one system (F) at no greater time-cost than C, D, and E gets results
about 25% better than they do; and third, that practice for quality may
secure only at the cost of speed. The teachers in A and B are better
it
paid than those in the other cities, so that the success of these schools at
no time-cost might not be generally attainable.
"Leaving F out of account, the differences of these school systems in
the method of teaching handwriting, in the time devoted to it, and in
the ideals of the system in respect to it arc of inconsiderable influence
upon efficiency. One makes its pupils write very well at very slow rates,
the others vary a little in quality with small inverse variations in speed.
On the whole, in spite of the achievement of system F, efficiency in hand-
writing seems, like spelling, and unlike arithmetic to be under present
conditions not very much influenced by the management of the schools."
(Thorndike, "Handwriting," p. 33.)
Freeman ('15) had writing tests made in 47 cities and then

compared the attainment in these schools with the amount of time
devoted to the writing-period in each school. His results arc set
forth in Figure 65. Each school is represented in the chart by a
short vertical line. This line is placed at a position above the base
line so that it represents the relative rank of that school in attain-
ment penmanship among the 47 schools.
in
These results are interesting and valuable, but it is questionable
whether they prove that time makes no difference. The difficulty
with a wholesale set of figures such as these is the impossibility
of separating the various elements and determining their ef-
fects individually upon the ultimate attainment in writing. The
schools which devoted 90 to 100 minutes per week to writing and
obtained no better results than the schools which devoted 40 to 50
minutes per week, may contain other factors which kept their
proficiency down, such as, poorer teaching, different classes of
pupils, the quality of writing done in other subjects outside of the
writing period, which probably has as much if not more influence
upon proficiency in writing than the writing period itself, and so
on. In fact, we might even imagine that if these same schools had
devoted only 40 to 50 minutes per week, they might have been
HANDWRITING 309
much worse in writing than they actually were. The real

expla-
nation may perhaps lie in the possibility that the schools
having
longer writing periods may not use the time to as good advantage
as those having shorter periods. The latter, by virtue of
having
only a short time to devote to the subject, may work more in-
tensely and profitably.

The surest way in wl Jch to measure the results obtained in dif-
ferent periods of practice in writing would be to split up a given
group of pupils into several sections and to have each section devote
a different amount of time to the writing period, say 10, 15, 20, and
25 minutes respectively. All should preferably be taught by the
10 15 20 25 30 35 40 45
Rank
FIG. 65. Relation between attainment in writing and time devoted to writ-
ing. After Freeman ('15).
same teacher. At any rate, all other elements should be kept as

constant as possible. Then a comparison by special tests in speed
and quality made at stated intervals would reveal the effect of
time upon improvement and would show what period of time
brought the optimum results.
The investigations by Thorndike and Freeman have been highly
valuable in calling attention to this problem and in showing that
some schools obtain as good results by devoting only half as much
time as other schools obtain in double the amount of time. The
general impression is that 15 minutes per day as a maximum is suf-
ficiently long for the writing period and, under proper methods of
instruction, can produce as high attainment in writing as the schools
need to produce for all practical purposes. That there is an opti-
mum length of the writing period beyond which the principle of

diminishing returns operates is quite certain, as indicated by general
learning experiments such as those cited in Chapter XL What this
optimum period for practice in writing is cannot at present be
specified with certainty.
(3) How Great Proficiency Should Be Attained. Tests made
in a large number of schools show that the average attainment at
the end of the 8th grade is writing as good as quality n, Thorndike
Scale, or quality 60, Ayres Scale, at a speed of about 83 letters per
minute. The same tests also show that many schools reach much
higher proficiency than this and that in every school a considerable
share of pupils far exceed the limits of n
or 60 in quality and 83
in speed. Are these averages of attainment in quality and speed
sufficiently high for practical purposes? And is it worth while to

develop higher proficiency in writing than these averages rep-
resent?
In answer to the first question, Freeman ('15) made inquiries
among business firms and found that the majority considered writ-
ing equal to quality 60, Ayres Scale, as sufficiently good for or-
dinary business purposes. It would seem then that the frequent
criticism from business men who say that pupils coming to them
from the public schools cannot write, is ill-founded, and based
probably on the exceptions rather than on the majority of pupils,
since about three-fourths of the pupils finishing the elementary
schools can write better than quality 40 or 50, Ayres Scale, and
one-half can write better than quality 60. The attainment of n
or 60 in quality and of 83 letters per minute in speed reached by the
average pupil upon completion of the elementary school is fully up
to the average requirement of business. The criticism coming from
business men is probably based upon the 20 or 25% of pupils
finishing the 8th grade who fall below quality 40, and many times
upon those who leave school before completing the 8th grade to
seek business employ *nent.
The accond question is practically answered by the discussion of
the first. It probably is not worth while to attempt to reach a
proficiency in writing much higher than quality n
to 12, Thorndike
Scale, or 60 to 70, Ayres Scale. Such higher skill would be gotten
by too great an expense of time and by too great a sacrifice of
speed. Furthermore, the legibility of writing of qualities above
these limits increases very little. The gain is chiefly in beauty.
The time that would be required to reach these higher degrees of
HANDWRITING y*i
skill could be devoted to better advantage to other subjects, or to

the learning of typewriting. Thorndike says:
"
Considering the fact that above quality 1 1 there is very Kttfe differ-
ence in legibility, one is tempted to advocate the heresy that children
are taught to write too well. I personally do advocate it. If school
boards would furnish, for the use of children electing 'writing' as a study
in the last two grammar grades, typewriting machines, I should certainly
advise the transfer to typewriting of a child in these grades whose writing
at 60 letters a minute consistently reaches quality 13. For, the amount
of practice required to advance such a pupil to quality 16 at a rate of
75 letters a minute would much more than suffice to advance him to
substantially errorless machine writing at that rate. The value now at-
tached to the high qualities of handwriting is of course largely fictitious.
Employers who can afford such high qualities of writing, buy machines
to produce them. For writing cash checks, simple book entries, labels,
and the like, a good plain hand or our quality 12 is entirely adequate.
For attaining the higher qualities (15-18) the machine is a more eco-
nomical tool than the pen, and in my opinion should be provided by
those schools which require such qualities. Further, such qualities should,
in my opinion, be required of children in the elementary schools, only
when they have elected writing as a vocational subject. For the data
from the adult women-teachers make it practically certain that ability
to write above quality 14 will not be exercised in life except as a part
of a clerical trade. If very, very few teachers find it worth while to
maintain qualities above 14, it can hardly be supposed that it will be
worth while for mechanics, house-keepers, farmers and dressmakers to
do so." ('10, p. 37.)
(4) Relation between Speed and Quality. To what extent

isspeed of writing accompanied by good quality? Is there possibly
an inverse relationship between the two? From general impressions
we know that if we try to write unusually well we sacrifice speed
and if we try to write unusually rapidly, we sacrifice quality. Is

there any balance between these two elements?
Quite frequently teachers overemphasize either quality or speed,
usually the former, at the expense of the latter. In Figure 66
the teachers in the 6th and 7th grades greatly over-emphasized
quality so that the speed of writing was equal only to that of the
average 3rd grade pupil. Definite tests and comparisons with
standards will reveal to the teachers many such aberrations in
emphasis.
Sackett had 36 university students write in their normal manner
and immediately afterwards he had them write the same material
3 I2 EDUCATIONAL PSYCHOLOGY
with the knowledge that it was to be used as a writing test. He

found that on the average the writing was about .5 letters per-
second slower (original rate about 1.8 letters per second) while
the quality gained 4 points on the Ayres scale.
4 5
Grades
FIG. 66. Average speed and quality of handwriting of the various grades
in a given school. The broken lines represent the school. The continuous lines
are the standard attainments.
Freeman ('14) made tests to determine what the efficiency would

be when children were told to write (i) both rapidly and well,
(2) as well as possible, andas rapidly as possible. The results
(3)
showed that trying improved quality at the expense
to write well
of speed. Quality improved 6.2% while speed dropped 3.7%.
Trying to write rapidly increased speed by 27.2% but decreased
quality 9.1%. Improvement in both speed and quality, however,
HANDWRITING 313
can be obtained when instructions are given to stress both aspects.

Apparently this is the preferable thing to do.
The author ('15) found in the case of 144 pupils the following
correlations between various characteristics of handwriting:
Speed and quality 10

Speed and legibility 12
Quality and legibility 34
Freeman also computed the correlation between speed and

quality on the basis of the writing samples of pupils in Grades 4 to
8. These he found to be as follows:
GRADE IV v vi vn vm
Correlation. . . .08 (02) -.10(04) -.14(04) -.37 -.is(s)
These correlations are either zero or slightly positive or negative

and mean that only to a very slight extent is the good writer ex-
tremely slow or the fast writer extremely poor.
Judd ('16), in the Cleveland Survey, has presented extensive
data pertaining to this question:
"After determining the speed and quality of each specimen, it becomes

possible to work out with great exactness the relation between these two
characteristics. It is evident from ordinary experience that quality
commonly deteriorates when speed is emphasized, and that speed is slow
8th Grade
7th Grade
6th Grade
5th Grade
80 40 50 60 70 80 90
Quality-Ayers Scale
FIG. 67. Average speed of handwriting at each quality of writing from 20
to 90, 10,528 cases from 5th, 6th, 7th and 8th grades. After Judd ('16, p. 72).
when one tries to write especially well. The school is constantly in the
position of seeking some reasonable balance between speed and quality.
"Diagram 67 gives the facts for the 10,528 specimens carefully studied.
In the vertical axis of this diagram are represented the different speeds;
in the horizontal axis are the various grades of quality. The results from
each grade are represented separately. Thus, beginning at the extreme
right end of the bottom line, we see from the diagram that for those
writers in the fifth grade who show the highest quality (90) the rate is
on the average 51 letters per minute. Advancing along the line toward
the left, we find that those in the fifth grade who show a quality 80 have
an average speed of 54 letters.
"The diagram shows that there is a general area between qualities
60 and So, and between speeds 60 and 80, where all the grades above the
fifth may be said to reach a level. Greater speed seems to be purchased
at an undue sacrifice of quality, and higher quality seems to result in
much slower speeds. We thus have in our results some indications as
to the probable area within which teachers will find a desirable balance
between speed and quality." (Pp. 70-71.)
(5) Methods of Teaching Penmanship. Experimental efforts

have thus far not been directed very vigorously toward ascertain-
ing the specific effect of different methods of learning to write
upon rate of improvement in it. It seems, however, very certain
that different procedures do produce enormously different results.
This is amply shown by the wide differences in attainment of the
various grades and schools, even in the same school system. Judd
found, for example, in the Cleveland Survey, that the average of
the best class was twice as proficient, either in speed or in quality,
as the poorest class. The facts are shown in Figures 68 and 69:
"Diagram 68 shows the average results for the four upper grades in
36 schools. The figure is to be interpreted as follows: In the upper
diagram, which gives the results for the fifth grades, there are numerous
small squares, each representing a single fifth grade. In each square is a
number showing the average number of letters written per minute in a
grade. Thus in the square at the extreme left of the diagram is the num-
ber '39.' This means that the average number of letters written per
minute by that fifth grade was 39. In the next vertical column of squares
are numbers ranging from 42 to 49. These indicate that there were fifth
grades showing each of the averages given.
"One of the most impressive facts which is brought out by this com-
parison is that the slowest fifth grade
is only half as fast as its fastest
fifth grade. Like statements can be made regarding the other grades.
These wide differences cannot be attributed to any native characteristics
which the children bring to the school. Such disparities might appear in
individuals, but the figures report whole grades. All the fifth grades are
going through the schools parallel with one another and arc officially
ranked as alike. The same statement can be made regarding the other
HANDWRITING 3*5
Speed Records of 56
36 Fifth Grades 75
64
64
52
139142 61
66
Speed Records of
36 Sixth Grades
60 71
81
r-t^%
47 50 60
1 1 1
88
75
Speed Records of 78
78 85
36 Seventh Grades
73 84
64 72 06
57
145150 3SL
79
79 88
76 87
Speed Records of
36 Eighth Grades 74
72
_7J. 81
3 <W_ _TO 80
64 MtOOl
FIG. 68. Distribution of grade averages in speed of writing. After Judd

('16, p. 64).
grades also. Perhaps the most obvious case is that of the eighth grade.
Children will go out of the various eighth grades into high schools with
the official assumption that they arc equally well fitted for advanced
Quality Records of
36 Fifth Grades
Quality Records of
36 Sixth Grades
Quality Records of
36 Seventh Grades
Quality Records of
36 Eighth Grades
FIG. 69. Distribution of grade averages in quality of writing as measured

by the Ayres Scale. After Judd ('16, p. 64).
HANDWRITING 317
work, and yet one eighth grade averages only 46 letters a minute, and
another averages 101. Is it not evident that there must be a difference
in emphasis on speed in writing in different schools?" (Pp. 63-64.)
"Diagram 69 shows in a manner similar to that explained in the
earlier paragraph on speed the results obtained from 36 schools. From
the figure it will be seen that in quality, as in speed, the most striking
variation exists between grades which are officially recognized as parallel.
Furthermore, there is the same overlapping of grades, several of the
fifth grades ranking higher than the average eighth grade." (P. 68.)
Here, as in so many other problems, specific experiments with

conditions rigorously controlled should be carried out in order
to determine the actual effect of each given element or method in
teaching handwriting. Such experiments could be carried out
by teaching parallel sections of a class according to different
methods, after the general plan outlined under heading (2) of
this section.
(6) Factors Affecting the Execution of the Writing Movements.

The numerous conditions affecting favorably or unfavorably the
execution of the many complex writing movements such as the
position of the body, the position of the desk, the position of the
arm, the position of the paper on the desk, the manner of holding
the pen or pencil, and the like, are important problems concern-
ing which likewise we have little scientific information. The pro-
cedures followed by teachers are based chiefly upon general ob-
servation and personal judgment. With regard to position of
body, arm, paper, and desk, Freeman has suggested the relations
shown in Figure 70. This relationship makes possible a natural
straight front position before the desk with both arms on the desk,
and with the paper tilted at an angle of about 30 degrees to the
left. This position of the paper makes it possible for the hand to
follow easily along the horizontal line of writing by simply turning
the forearm on the point on which it rests on the edge of the desk
as a pivot. Furthermore, the most natural direction of the up and
down movement of the pen point is directly toward or away from
the body, and with the paper in the position suggested, the writ-
ing will have a medium slant of about 25 to 30 degrees from the
vertical.
(?) Types of Writing Movements. What sort of writing move-

ments may be executed most economically in learning to write
and ultimately in the perfected writing process? Considerable
controversy has occurred over this question. Theoretically there
are at least three types of movements possible in the production of

letters. One would be to do the writing entirely with finger and
hand movements and to hold the arm absolutely quiet except
for the turning of the arm from left to right to follow along the
line of writing. The second would be to do the writing entirely
with the arm movement and to hold the hand and fingers absolutely
still. The third would be a combination of these two sets of move-
ments in varying proportions. Advocates of various methods of
teaching writing favor one or another type of movement. Probably
FIG. 70. Position of pupil in relation to desk and paper. After Freeman
Ci4>.
the best method is an appropriate combination as suggested in
the third type of movement. Freeman states:
"The arm movement with rest the so-called muscular movement

is an American discovery and has been vigorously exploited in commer-
cial schools since the last quarter of the last century and more recently
in certain systems of teaching in the public schools. It seems likely
that within twenty-five years this form of writing will be practically
universal in American schools. The chief advantages of the movement
are two. In the first place, it is made with the fingers relatively relaxed,
thus avoiding cramping. In the second place, the rolling movement of
the arm upon the muscle pad of the forearm produces a firmness and
evenness of line, and the fact that the movement is produced from a center
at a considerable distance from the pen point results in regularity of slant.
"The contention that every detail of the letters shall be made by the
HANDWRITING 319
movement of the arm while the fingers remain immobile is calculated

to antagonize reasonable critics. The oscillation of the arm may well
form the main basis for the upward and downward strokes of letters, but
to require that every loop and turn and joining be produced by the
movement of the arm as a whole, instead of the much more flexible hand
and fingers, is to set up an artificial requirement and one which is not
made in regard to other types of skilled movement.
"The form of movement, then, which best meets the requirements
which may be laid down as the result of experiment and of practical
experience is somewhat as follows: The hand and arm must be so ad-
justed that the hand progresses freely along the line during the formation
of the letters and in the spaces between the words. The hand must rest
upon some freely sliding point or points of contact such as the finger nails
or the side of the little finger. When, on the contrary, the pen point is
carried along from one letter to another by means of adjustments of the
parts of the fingers and the hand, the hand continually gets into a cramped
position.
"The movements of the arm and fingers should form a smooth and
easy co-ordination in which there is a condition of flexibility in the whole
member. The rotation of the arm upon the muscle pad of the forearm
as a center carries the hand along, the upward and downward oscillatory
movement forms the groundwork of the letter formation, and slight ad-
justments of the fingers complete the details of the letters. In addition
to these chief elements of the movement, the wrist may rotate to the side
to supplement the sideward movement of the arm, and the forearm may
revolve upon its axis in the movement of pronation as a corrective to the
increase in slant at the end of the line. There is no good reason for seek-
ing to eliminate any of these component movements. Each has some
part to play. Moreover, room must be left for individual differences in
their relative prominence and manner of combination." ('14, pp. 93-96.)
(8) Movement
Drills. Special drills in movements such as
ovals, vertical movements progressing to the right, horizontal
movements from left to right and from right to left, have been
advocated by various systems of penmanship with the belief that a
substantial amount of time given to such drills will establish good
form and speed in writing. To what extent such formal drill or
how much of it may actually be profitable, is open to question.
It would be an experiment worth undertaking to teach three
sections of a class of pupils for a year or more by giving to one
section a considerable amount of such drill, to the second section
none, but to devote instead the entire time to drill and practice in
writing the letters themselves, and to the third section a com-
bination of the two types of drill.
(9) Correct Form in All Writing Done by the Pupils. It is an
elementary principle of habit formation that an act to be developed

into a skillful habit, whenever carried out, should be performed
correctly or at least as correctly as possible at that stage of learning.
Otherwise the inaccurate and careless performance of the act
tends to counteract the skill already achieved. It would seem,
therefore, to be a highly desirable plan as an incentive to pupils
to write at their best at all times, to base their marks in penman-
ship to the extent of one-half upon their work in the writing-period
proper and to the extent of the other half upon the quality of
writing in all other work submitted. One important reason why
Instruction in penmanship, spelling, oral and written composition
does not carry over into the penmanship, spelling and composition
generally is that pupils are not as careful in their ordinary writing,
spelling and speaking by observing correct form as they are in
the respective class periods devoted to these subjects. Telling
the pupils that their final grades will be made up, half and half,
as here suggested, will act as a remarkable incentive toward general
improvement as shown by specific tests in the case of spelling,
which will be discussed in the next chapter.
(10) Analysis of Imperfections. One of the important by-
products of the experimental investigation of conditions and fac-
tors in the learning process is the fact that definite practice in a
specific function consciously known to the learner greatly improves

the function. Improvement in any type of skill takes place in
many instances only when practice is squarely directed towards
certain specific elements in the process. This is one reason why
persons in laboratory experiments on learning make such enormous

progress and why pupils in school make so little progress. The
function to be trained in the one case is definitely and specifically
known to the learner, whereas in the latter case, it is indefinite and

largely unknown to the learner. It is not enough to say to a pupil,
"You must write better," "write more like the copy," or "watch
me; write as I do." The
specific defects must be pointed out,
recognized by the and then overcome by definite practice.
learner
Freeman has pointed out five main types of defects or character-
istics ofhandwriting: uniformity of slant, uniformity of alinement,
quality of line, letter formation, and spacing. The scales that he
has devised for rating handwriting from these five points of view
may be used with advantage in discovering the specific defects in a
given individual's writing and in centering definite attention and
HANDWRITING 321
practiceupon them. The score card for evaluating handwriting

prepared by C. T. Gray calls attention to a similar set of elements.
The methods of judging penmanship suggested by Palmer and by
Zaner aim likewise to center attention upon defects and excellencies
in various essential aspects of writing.
CHAPTER XVIII
SPELLING
PROCESSES OR STEPS INVOLVED IN SPELLING
Ths child learns to spell by seeing or hearing the letters of a given

word, and by thinking, speaking, or writing them in the order in
which they are seen or heard. Stated in more minute detail, the
successive steps are substantially as follows:
(1) The reading of the word, that is, the sight, sound and pro-
nunciation of the word as a whole which involves all the elements
of the reading process and need not be enumerated here. (See
Chapter XVI.) These are presupposed as the child usually has
learned to read the word before he learns to spell it. At this point
the successive steps in the spelling process as such begin.
(2) Reception upon the retina (or the ear) of the visual (or
auditory) stimuli of the first letter of the word.
(3) Transmission of the visual (or auditory) impressions from the
retina (or ear) to the visual (or auditory) centers of the brain.
(4) Arousal thereby of mental images and other associations
of meaning.
(5) Transmission of the impulses from the visual (or auditory)
centers to the motor-speech centers or to the motor-writing centers.
(6) Transmission of motor impulses from the latter to the
speech-organs or to the writing-muscles. This occurs very prob-

ably even in the silent learning of spelling since silent reading is
accompanied by the so-called inner speech.
(7) Execution of the speaking or writing movements in pro-
nouncing or writing the letters.
(8) Return kinaesthetic impulses from the speech or writing mus-
cles to the sensory centers and then to the motor speech or writing
centers. This series of steps from (2) to (8) is then repeated for
the second letter, for the third letter, and so on to the end of the
word.
The steps here outlined are the ones involved in learning the
spelling of a word. In the perfected process, however, steps (2)
and (3) and possibly (4) drop out and step (5) is inaugurated
322
SPELLING 323
directly either through step (i) or through the idea or image of

the word to be spelled or written, and from then on the whole
process of writing or spelling the word consists of a circular series
of automatic connections between steps (6), (7), and (8) in which
(8) for the first letter of a word acts as stimulus to step (6) for the
second letter and so on for the succeeding letters of a given word.
Step (8) of each letter always acts in turn as the stimulus for the
series (6), (7), and (8) of the succeeding letter so that in the fin-
ished habit the mere pronunciation, sight, image or idea of the
word automatically brings about the succeeding links involved
in naming or writing the letters in correct order.
Economyin learning to spell consists largely in providing con-

ditions under which the half dozen links here outlined may be es-
tablished most easily, most quickly, and most permanently for
the words whose spelling a child should know.
Little is known directly concerning the manner of operation of
each of these factors. The most important step, if any one is more
important than any other, possibly is number (8). This link
determines what the next letter shall be in the automatic writing
of a word. In the original learning of the spelling of a word, steps
( 2 )> and (4), which together constitute the perception of the
(3)>
letters,are highly important since the establishment of the other
links depends upon the accuracy with which the letters themselves
are perceived or observed. It seems probable, although not certain
in the absence of pertinent experimental data, that a considerable
part of the difficulty of learning to spell, lies in the inaccurate

observation of successive letters of a word. The awakening of
mental images is probably very important, although our informa-
tion as to the types of imagery concerned, the extent to which
they are essential, and the methods of arousing them, is relatively
unreliable.
THE MEASUREMENT OF EFFICIENCY IN SPELLING
(i) Methods
of Measurement. On the face of it, it would
seem be an easy matter to devise a definite and objective method
to
of testing attainment in spelling. All that would seem to be neces-
sary would be the selection of a series of words and the determina-
tion of the number or percentage of these words that a pupil or
class can spell correctly. But a closer study of the possibility of
measuring spelling ability reveals a number of complicated prob-

lems. What sort of words should be used as a spelling test? How

should they be presented to the pupils? How should they be
scored? Should any word be considered equal to any other word,
or should different values be assigned to different words?
We shall not enter here into any critical discussion of the prin-
ciples involved in the construction of spelling tests, nor into a
consideration of the technique of administering and scoring them.
Some of the methods of measuring spelling ability will be men-
tioned briefly. 1
Up to 1913, tests of spelling ability were made either by series of
arbitrarily selected words which were presented either as isolated
words or as parts of dictated sentences, or by determining the
percentage of misspelled words in spontaneously written composi-
tions. Since 1913, several more or less scientific methods of
measuring spelling ability have been devised.
The writer, ('15) prepared one test consisting of 6 lists of
ico words each by making a selection of words at certain inter-
vals from the dictionary and then discarding all technical and
obsolete words. The words in each list were then arranged in the
order of length. Each list as a whole was found to be practically
identical in difficulty with every other list. Average standards of
attainment were then prepared for the various grades as shown
in the following table, which gives the percentage of words of any
one of the 6 lists spelled correctly at the ends of the respective
years:
TABLE 99
Attainment in After Starch ('15)

GRADES 12 34
spelling.
S 6 7 8
Percentage of words correct 10 30 40 51 61 71 78 85
The author has more recently prepared a different method of

testing spelling ability on the basis of the 2,626 most common words
in the English language. This list is useful as a study list as well
as a test list and will be distinctly more valuable because of this
double purpose. The plan by which these words were selected
and the method by which they are to be used will be described
later in this chapter.
Ayres prepared a list of words consisting of 10 words for each
1
For a detailed discussion, see the original monographs or the writer's Educational
Measurements, or Monroe's Educational Test* and Measurements.
SPELLING 325
grade so selected on the basis of experiments that on the average

70% of the words for any given grade would be spelled cor-
rectly by the pupils of that grade. Later, Ayres ('15) prepared
a very useful test consisting of the 1,000 most common words.
These words are split up into 26 lists of varying length and so
arranged that the words in any one list are of approximately equal
difficulty and that the successive lists from i to 26 become harder
and harder. The scale gives the average percentage of the words in
the various lists that pupils can spell correctly in the different
grades.
Buckingham ('13) prepared, on the basis of experiments, a list
of 50 words carefully scaled in difficulty according to the percentage
of pupils who could spell the words correctly.
Uses and Results. All these scales have been found useful
(2)
in measuring efficiency in spelling in different schools more ac-
curately than was formerly possible, in determining individual
differences in abilitiesamong pupils, in ascertaining progress, and
in comparing the effects of various factors in the learning and
teaching of spelling. It is hoped that they will be still more useful

in the future in discovering the most effective methods for ac-
quiring proficiency in spelling.

The facts with regard to the enormous range of individual abili-
ties in spelling and the consequent overlapping of the abilities of
the pupils in the various grades are shown in Figure 18, Chapter
III. The facts, as in other subjects, are astounding. The best
pupil in the grade spells as well as the poorest pupil in the
first
eighth grade. Certainly some adjustment of the pupils should be

made according to their capacities.
Spelling presents one of the more striking examples of mental
sex differences found in educational psychology. Investigators
uniformly report girls doing better than boys. Wallin found the
girls averaging nearly 2% better than the boys in terms of his
spelling lists. In a study previously quoted in Chapter XIV, Foster
found that 238 girls and 256 boys, all university freshmen who
were given a spelling test of 40 difficult words of Latin derivation,
made respectively 76.6% and 68% successes. Sackett gave 24
words of Buckingham's spelling list to over 7,000 school children
and found the girls about a half year of school progress ahead of
the boys. Sears reports that a test composed. of 70 words from
the Ayres list given to nearly 13,000 children in Oakland, Gate*
fornia, showed the girls superior to the boys from 2 to 6%. espe-
3*6 EDUCATIONAL PSYCHOLOGY
daily in the upper grades. He suggests that the average girl

might be graduated from a half to a whole year earlier than the
average boy.
The writer found the differences in one school as shown in Fig-
ure 71. The difference is in favor of the girls, particularly in the
upper grades.
80
70
^60
FIG. 71.
2345678 Grades
Sex differences in spelling as measured by the author's test.
ECONOMIC METHODS IN THE LEARNING AND TEACHING OF SPELLING
(i) The Words to be Learned. For a number of years the most

important problem in the economy of learning to spell has been
the question, What words should a child really learn to spell? The
words in the spelling-books have for decades been selected mainly
by the arm-chair method and have consisted to a large extent of
rare and useless words. Swin ton's Speller published in 1872, states
',
in the preface that "It omits the alphabet and the 'ababV on the
one hand, and on the other, quite a number of sesquipedalian
words common to all old-time spelling books." It further urges
as a vantage point, "The practical character of the work which
aims to set forth, not the tens of thousands of long-tailed words in
osity and ation, but the actual vocabulary used in speaking and
writing." And yet the book contains in one lesson (page 143)
such words as, lethean, pharisee, pentagon, pneumatics, theoc-
racy, anathema, dysentery, etc. In another lesson (page 144) it
contains oleaginous, farinaceous, argillaceous, lachrymose, sacer-
dotal, animadversion.
SPELLING 327
In view of this situation, recent years have brought forth a

number of very extensive studies and tabulations to find the words
which are most commonly used in writing by various classes of
persons and to discover the frequency with which these words
occur. The following are the chief tabulations thus made:
The Eldridge List. Mr. Eldridge ('n), a business man in Buf-
falo, New York, made a tabulation of 43,989 running words from
four different newspapers in which he found 6,002 different words.
2,927 words occurred each, once

1,079 twice
5i6 three times
294 four times
212 five times
151 six times
105 seven times
84 eight times
86 nine times
261 ten to nineteen times
233 twenty or more times
The Ayres List. Ayres ('13) of the Russell Sage Foundation

tabulated 23,629 words from 2,000 letters, chiefly business letters,
and found 2,001 different words.
The Jones List. Professor Jones ('13) of the University of South
Dakota tabulated 15 million running words from 75,000 themes
written by 1,050 pupils in grades two to eight and found 4,532
different words.
The Cook and O'Shea List. Cook ('14) tabulated 200,000
running words from the family correspondence of thirteen persons
and found 5,200 different words.
These four tabulations represent four distinct fields of writing,
each being the most extensive in its field; namely, journalistic,
business, juvenile and private domestic vocabulary. One im-
portant type of vocabulary has never been tabulated, namely, the
vocabulary of our best current literary writers. Children ought
not to be confined to the words which they naturally use (Jones
List), nor to adult business vocabulary (Ayres List), nor to news-
paper vocabulary (Eldridge List), nor to the vocabulary of ordi-
nary family correspondence (Cook List). An important point in
learning to spell is to learn the meaning of words, especially of
words whose use will enhance a person's vocabulary. Hence, the
writer made a tabulation (Starch List) of the vocabulary of the

best current literary authors. This tabulation is unpublished and
on file in the Library of the University of Wisconsin.
The Starch List. The writer 1 tabulated some 40,000 running
words, about 1,000 from each of forty authors in eleven current high
grade magazines. This yielded 5,903 different words as follows:
3,in words occurred each, once

1,009 twice
512 three times
280 four times
189 five times
121 six times
97 seven times
82 eight times
53 nine times
225 ten to nineteen times
224 twenty or more times
From lists, words for spelling and testing purposes

these five
were selected according to the following plan: Every word occur-
ring three or more times in the Starch List, every word occurring
three or more times in the Eldridge List, every word occurring
seven or more times in the Cook List, and every word in the Ayres
1,000 word list was selected if it also occurred in one other list
including the Jones List. A
word occurring three or more times in
the Starch List or in the Eldridge List or seven or more times in
the Cook List or any word occurring in the Ayres List was not
included if it To be included it had
occurred only in the one list.
to occur at least once in one other This safeguarded against

list.
the inclusion of words confined to one type of vocabulary only.

For example, the word "cupfuls" occurred twenty-one times in
the Starch List but in no other list. Hence, it was excluded.
The reason for selecting words that occurred three or more
times in the Eldridge List or in the Starch List was that the words
found less frequently are so rare that they constitute a very small
part of the running words of ordinary writing. This point may be
shown most emphatically by the accompanying graph, Figure 72,
on which the relative number of words of different frequencies is
indicated. A remarkably close parallel exists between the Eldridge
and the Starch Lists. The particular point to note in the graphs
1
In cooperation with L. C. De Bruin. Reported in a thesis in the library of the Uni-
versity of Wisconsin, 19x6.
SPELLING 329
is the fact that the sharp bend in both curves occurs between
words whose frequency is between two and three. After three,
the curve shoots up very rapidly. This same breaking point occurs
in the Cook List between seven and eight. It is higher in this list
because Cook tabulated a larger amount of writing. Words
occurring three or more times in the Starch and Eldridge Lists
constitute over nine-tenths of all running words.
This process of selection yielded 2,626 words. This number may
seem small compared with the number of words in former spelling
8000
2700
2400
^2100
* 1800
1500
1200
I
000
600
300
20 10 9 8 7 6 5 4 a 2 1
Number of Times Words Occurred
IMC,. 72. Number of words occurring with various frequencies in the El-
clridgc and Starch lists.
books and even in some contemporary spellers. But it is obvious

that it is not only useless but wasteful of a pupil's time to learn
words which he will never use in writing in all his years after school
and at the same time neglect to master thoroughly the words
which he will actually use. Spelling texts commonly contain from
ten to fifteen thousand words. In fact, the number of words here
that found in the special
presented, however, is even larger than
spelling lists prepared by many cities which often do not have
more than from 1,500 to 2,000 words.

It seems, however, that a spelling list ought to include all words
of reasonably common occurrence but not words of extremely
uncommon occurrence. This number seems to be approximately

2,500 or 2,600. Another reason for limiting the spelling list to this
number is the fact that the writing vocabulary of the average
adult is not over 2,500 words and probably less for a great many
people. Cook found that three out of his thirteen persons who had
each furnished 40,000 running words (the largest number obtained
from any one individual) of family correspondence had used a
vocabulary of 2,575, 1,546, and 2,330 different words respectively.
This would mean that a person could write 100 letters each 400
words in length, making a total of 40,000 running words, and not
use more than 2,500 different words. If we may assume that the
average man or woman, exclusive of persons whose occupation
involves considerable writing, such as novelists, teachers, and
journalists, writes one such letter a month, his entire correspondence
for ten years would not involve more than 2,500 different words.
Incidentally we may also point out that this number furnishes

enough words to supply two new words for each school day in
grades two to eight, or 360 a year, counting 180 days to the school
year, and 106 words for grade one. This is in accord with the
practice of many school systems of teaching not more than two
new words per day.
It is possible that such a list as that described above may need
to be supplemented to meet the needs of certain sections of the
population. Houser tabulated the words used by farmers in
California in corresponding with the state department of agricul-
ture. He found certain radical differences between this list thus
obtained and that published by Ayres.
The Placement of the Words into the Various Grades. Can this
be done on any scientific basis? Which of these 2,626 words
should a pupil learn in each grade? There are three possible ways
in which the words might be distributed into various grades:
(1) We might distribute the words according to their frequencies,
putting the most frequent words in the lower grades and the less
frequent words in the upper grades.
(2) We might put each word into the grade in which children
first begin to use it rather frequently in their writing.
(3) We might put each word into the grade in which, according
to the consensus of competent judges, such as teachers, it ought
to be taught.
The placement of the words in the present list was made partly
according to all three principles, but where discrepancies existed
SPELLING 331
with reference to any given word final decision was made according
to the third plan.
The 2,626 words were first arranged according to their frequencies
of occurrence and all those words in the list which were also in the
Jones List were then placed into the grades in which the Jones
List shows them to be first used by children. A considerable num-
ber of the 2,626 words were not found in the Jones List. These
words were placed into higher and higher grades according as they
occurred less and less frequently.
After this task was completed the entire list of words was re-
checked according to the third plan. Fifteen different lists of
words which had been prepared by various cities or school systems
for their own uses such as the Boston Minimum List, the Stockton
List, the Santa Cruz List, the Chicago Speller, etc., were used.
After each of the 2,626 words occurring in one or more of the lists,
was written the number of the grade into which the word was
placed by each list. An average of these placements was then
obtained and accordingly the word was finally placed into its
grade. For example, the word "flower" was placed by seven
lists into different grades as follows: 4, 4, 2, 3, 2, 2, 4. This gives
an average grade placement of 3.00. " Cough was assigned by six
"
lists to grades 5, 5, 3, 3, 2, and 4, with an average grade placement
of 3.66. All of the 2,626 words were thus assigned with the excep-
tion of 126 words which did not occur in any of the lists employed
and 178 words which were found in only one list. In order to make
the grade placement of these 304 words with equal confidence, a
group of seven experienced elementary teachers or supervisors
were asked to assign each of these words to some one grade accord-
ing to their best judgment. An average of these judgments was
then obtained and the words were placed accordingly.
All of the 2,626 words were then assigned to the various grades
according to the average grade placement as follows:
ALL WORDS WHOSE

AVERAGE GRADE PLACEMLNTS WERX
In Grade One From i.oo to 2.00
This process gave 360 words in each grade or two for each school
day with the exception of grade one into which the remaining 106
easiest words fell.
Measuring Attainment of the Pupils by Means of These Words.

the
The problem was next attacked was this: How many of the
that
words of a given grade may we reasonably expect the pupils at
the end of the year to be able to spell? In pursuit of an answer
to this question, six lists of 60 words were selected from each
grade list by taking, for a given list, every sixth word through
the entire 360 words of a given grade. The words of the first grade
were split up into two lists of 53 words each. These lists were then
given as a special test at the end of the school year to approximately
7,000 pupils in 28 schools in 15 cities ranging in size from very small
towns to a city as large as Seattle. The percentage of words of each
grade spelled correctly by the pupils of that grade was as follows:
GRADE
Av. percentage of words
1234 56 78
spelled correctly 56.2 63.6 77.0 80.4 83.9 85.0 82.9 80.7
Thus a very important advantage of these spelling lists 1 is that a

school or teacher can at any time test the pupils and determine
their efficiency by comparing them with the above standard aver-
ages. This can be done by selecting a list of 60 words from the

360 words for a given grade and giving them as a test. For ex-
ample, if at any time the teacher of the fourth grade desires to
compare the achievement of her pupils with the standard averages
of other fourth grades in schools generally, all she needs to do is to
turn to the words for the fourth grade and begin with any one of
the words and then pick out every sixth word through
first six
the list. will give a total of 60 words.
This At another time a
different list of 60 words may be chosen in like manner. The
same procedure may be followed in any other grade by using
the words for that respective grade.
The important advantage of this plan is the fact that the same
words which have been used as a study list may be used at any
moment as a test list and comparisons may thus be made with
the standard averages.
The averages here presented indicate that the words for all
grades above the second are of approximately equal difficulty for

the respective grades since they all are within a few points of 80%.
Published in the author's Spelling Book, 1919, by Silver, Burdctt & Co.
SPELLING 333
That the words in the second grade are not too difficult is shown
by the fact that in several schools the percentages in this grade
averaged as high as 90. The chief reason for the lower percentage
in the first and second grades is the variation in amount of actual
instruction in spelling given in different schools in these grades.
The best school among the 22 averaged around 90% in every grade
and the poorest school averaged around 60% in its various grades.
Aside from this list, numerous special lists have been prepared
partly on the basis of vocabulary studies or partly on the basis of
words commonly misspelled by pupils as reported by teachers.
Illustrations of such lists are the Boston Minimum Spelling List
(1915) consisting of 762 words, the Nicholson List, consisting of
3,070 words, prepared for the State of California, and the Chico
(California) List, consisting of 3,470 words, prepared by Studley
and Ware.
(2) The Influence of Rules in Spelling. Cook ('12) made a
test with 50 words on 70 university freshmen and on 39 high school
seniors and 30 high school freshmen. These 50 words were ex-
amples of seven rules with their exceptions. The university fresh-

men had had drill on spelling rules about seven months before
the test and the high school classes had finished the study of rules
six weeks before the test, The results are shown in the following
table:
TABLE 100
Observance of the rules. After (Cook
"In summary, it may be said was quoted by as many

that no one rule
as 50% though more than half of them had
of the university students,
memorized all these rules, and others besides, only the winter before; and
many of the students had been over all the rules in the public school. A
little less than half the high-school students had the courage to try to
give the rules they had learned only six weeks previously. In the univer-
sity group, those who gave some sort of rule to cover any part of the list
of words, averaged 4% higher in general spelling efficiency than those who
could not give any rule. So it is fair to assume that their better observ-
ance of the rules as shown by Table 100 is the result of their better spelling
ability in general, and not to any conscious application of the rules as
such. Not a single rule tested proved to be of real value, except the one
for the last two words of the list that relating to final ie." (After
Cook.)
These conclusions arc interesting and probably correct in their

chief implications. One further point, however, ought to be con-
sidered. Inability to cite a rule or unawarcness of its application
does not necessarily prove the impotency of the rule. It might be
possible that a rule played a part in the learning of words at the
time the learning took place and then had been forgotten. A
further investigation is necessary in which a comparison "would be
made between two groups, one of which had learned and applied
rules while the other one had never had any contact with rules.
(3) Length of Class Period. Dr. J. M. Rice ('97) tested the
spelling ability of about 33,000 pupils to ascertain the effects of
different factors upon efficiency in the subject, such as methods of
teaching, foreign parentage, home environment, amount of time
devoted to spelling in the school program, and the like. His re-
sults with reference to the factor of time are presented in the
table.
The results as they standwould seem to indicate that length of
class period makes no achievement in
difference in the ultimate
spelling since schools devoting 10 or 15 minutes daily do as well

as those devoting 50 minutes daily. Thus, City 15, School B,
grade IVA, devoting only 15 minutes daily to spelling, made a
record of 70.8 in the sentence test, whereas City i, School B, devot-
ing 50 minutes daily to spelling, made a record of only 61.8. Many
other similar comparisons may be cited.
SPELLING 33S
^2 CA3
Si
SPELLING 337
Cornman made a similar study and reports substantially the

same situation. A sample of his results follows:
TABLE 102. After Cornman ('02)
Records of 1898
The investigations of both Rice and Cornman are highly im-

portant but they do not afford conclusive evidence that time
makes no difference. The fact that some schools devote two or
three times as much time to spelling as other schools do and obtain
thereby no better results, does not prove that time plays no part,
for the reason that many other complicated factors enter, such as
differences in teachers, method, spirit, and the like, to make the
effect of any one factor unanalyzable. Indeed, we might infer
that the lower schools, assuming all conditions the same, might
have obtained even poorer results than they did if they had de-
voted to spelling only as much time as the better schools did.
In order to ascertain definitely the effect of different lengths of
class periods upon spelling efficiency, it would be necessary to
proceed in a more rigorously scientific manner rather than to
make on the basis of complicated, wholesale statistics.
inferences
It would be worth while to undertake an experiment by
distinctly
teaching several equally able sections of a given class under as
nearly identical conditions as possible, such as having the same
teacher, text, method, and environment, and by varying only
the time element so as to have, for example, a period of 15 minutes
for one class, of 30 minutes for another, and of 45 minutes for still
another. Comparisons by adequate tests at different times would
yield conclusive evidence concerning the effect of time upon ulti-
mate achievement. It would seem reasonable to anticipate that
longer periods of equally good class instruction would produce

greater efficiency. The fundamental problem, of course, would
be the determination of the optimum length of the class period in
relation to the desired efficiency in spelling.
(4) Methods of Teaching Spelling. Along with the question
of time, Rice and Cornman were interested also in determining the
influence of different methods of teaching. Rice does not present
definite figures, but, on the basis of extensive inquiries among the
teachers in the schools which he tested, he tried to ascertain the
facts conerning methods of instruction and environmental con-
ditions. His conclusion was as follows:
"
In brief there
,
is no direct relation between method and results. . . .
The results varied as much under the same as they did under different
methods of instruction. The facts here presented, in my opinion, will
admit of only one conclusion, viz., that the results are not determined
by the methods employed, but by the ability of those who use them. In
other words, the first place must be given to the personal equation of the
teacher, while methods and devices play a subordinate part." (After
Rice.)
Practically the same criticism, made in the preceding section

concerning the factor of time, is pertinent here. The situation is
too complex and the cooperating factors too numerous to make
such inferences without a careful isolation of the individual ele-
ments and their separate effects. Each element should be subjected
to an experimental procedure similar to the one outlined in the
preceding sections.
Cornman went further than Rice by undertaking an experiment
in which the spelling period was entirely eliminated from two
schools in Philadelphia for a period of three years. He states:
"It was decided to abandon the use of the spelling book and home
lessons in the subject, to omit also the period from the school programme
which had been devoted to its study and recitation and to investigate the
effect that the abstraction of these influences might produce upon the
spelling of the pupils of the several school grades. Several methods of
measuring results were devised which will be herein described and statis-
tically reported upon." (After Cornman.)
On the basis of these tests, made at different intervals, Cornman

found that the spelling ability of the two schools was almost as
SPELLING 339
good as that of the other schools and that the pupils improved
steadily in spelling, even though special instruction in them had
been omitted. Some of his figures follow:
TABLE 103. After Cornman
Spelling ability measured by uniform examinations for all schools,

given by the city superintendent.
50 SCHOOLS GIVING SPECIFIC WHICH FOR 3 YKARS NO SPFCIFIC

2 SCHOOI s IN
iNSFRi'cnON IN bPLLLiNO INSTRUCTION IN SPELLING I1AO IJLLN GlVEN
7th grade 73 o 69 .9
6th grade 70.4 65 i
.
5th grade 71.6 72.7

Average 71.7 69 2
.
The spelling ability of classes in the Northwest School who had for
three years been without specific instruction in spelling compared with
that of corresponding grades of previous years, who had had the full
amount of drill in spelling.
TKSTS OF JUNE, 1897 TESTS ot JUNE, 1900

8th grade 99 4
.
99 8
.
7th grade 99 i
.
98 6
.
6th grade 96 75 99 .o
5th grade 96 95 97 6
.
The spelling ability in a test in writing words in sentences of classes

which had been without specific instruction in spelling for a year and of
classes which had regular drill.
CLASSES WITH RKC.UL\R DRILL CLASSES WITHOUT RFC.IIL\R DRILL DlFFFRtNCE

TESTS or JUNE, 1897 Tbbibor JUNK, 1898
That is, half of the classes without specific instruction did better and
half of them did worse than the corresponding classes with specific in-
struction.
Cornman concludes:
"
(6) The amount of time devoted to the specific spelling drill bears no
discoverable relation to the result, the latter remaining practically con-
stant after the elimination of the spelling drill from the school programme.
" view of the economy of time, to rely
(9) It is therefore advisable, in
upon the incidental teaching of spelling to produce a sufficiently high
average result.
"
(10) The average result is what can be and is attained, as shown by
statistical evidence, by average pupils under teachers of average pro-
fessional efficiency in classes of average size, i. e., in the elementary
schools of this country as now organized. To remain strictly within the
evidence gathered by this investigation, it must be admitted that there
may be teachers of surpassing ability, who can obtain more than average
results by the method of the specific spelling drill, and other teachers of
meaner ability who need the drill to bring their pupils up to the level of
this average result. It is claimed, however, that there is no evidence
(whatever may be the wealth of opinion) to prove that such teachers
exist or to show where they may be found. Moreover, the evidence
which has been presented in this paper makes their existence at least
improbable."
The extensive work of both Rice and Cornman has been very
valuable in attacking a large educational problem by more exact
methods and in showing that there is undoubtedly an enormous
waste in the teaching of spelling, as there is probably in all school
subjects; but the results are not final proof that length of time or
methods of learning are negligible elements. Indeed both labora-
tory experiments in the learning process and the more recent and
more carefully controlled tests in spelling itself indicate quite
certainly that length of time and manner of learning do make
important differences.
The author found that there are large differences in the average
spelling attainment of the various classes of any given grade in
the same school system. The results for the 10 schools of a Wis-
consin city, Table 104, City J, were as follows:
SPELLING 341
TABLE 104
Differences in attainment in spelling in 18 Wisconsin cities as measured by the

author s test
Thus the lowest sixth grade made an average of 60.2 which is

barely up to the standard of the fifth grade and the highest made
an average of 79.0 which is above the standard of the eighth grade.
These differences are actually quite large when we remember that
they are grade averages and not scores of individual pupils. These
differences must be due in a large measure to differences in methods
of teaching.
Wallin attempted to make a comparison of the careful drill
method, devised for the Cleveland schools by Assistant Superin-
tendent W. E. Hicks, with the incidental method employed by
Cornman in the two schools in Philadelphia from which the spell*
ing period had been eliminated. Wallin reported results from the
Cleveland schools distinctly superior to those in the Philadelphia
schools.
TABLE After Wallin ('n)
45678
105.
Combined averages for the composition and column tests, all schools
GRADES AU.GEADIS
Percent 98.40 96.31 96.95 97.03 96.28 97.00
He states:
"First, the general spelling efficiency for all schools shown (97%) is
striking. It is 12.6% higher than Kratz's results (84.4%, for the fourth
to eighth grades, inclusive), 25% higher than Chancellor's (72%, for
10,000 pupils from the fourth to eighth grade), 25.48% higher than the
results in Rice's column test (71.52%), which consisted of a list of dic-
tated words, and 22.42% higher than the results from his sentence test,
which consisted of dictated sentences containing 50 test words for the
fourth and fifth grades, and 75 words for the sixth, seventh and eighth
grades. It eclipses by 25.7% Cornman's average in three term examina-
tions during three years for eighty Philadelphia schools (71.3%), and
is 27% higher than the results of these examinations in his two experi-
mental schools (70%), in which the spelling instruction was incidental.
In four column tests given to these two schools from September to June
and consisting of lists of fifty words differing from grade to grade, the
averages were 33%, 49%, 50% and 50% respectively for one school, and
49%> 57% 60% and 68% for the other; while the repetition of Rice's
column and sentence tests gave an average efficiency in 1897 of 78.9% in
one school, and 67.1% in the other, and in 1898, 73.1% and 65.4% re-
spectively, for the column test. The corresponding averages for the
sentence test were: 82.3% and 74.6% in 1897, and 76.5% and 77.9% in
1898. It will be observed that there was a loss of efficiency in 1898 except
in the case of the sentence test for one school." (After Wallin.)
Miss Fulton reports an experiment in which an attempt was

made to ascertain the effect of a specific drill in learning to spell.
One hundred words, ten each day for two weeks, were taught by
the following plan: i. "Write word upon the board/' 2. "Explain
meaning." "Children use the word in a sentence." 4. "Write
3.
word ten times while saying letters aloud at same time." 5. "Em-
phasize by intonation of voice or by colored chalk on blackboard

the difficult part of words." At the end of the two weeks a test
with the loo words was given.
During the next two weeks another list of 100 words of similar
difficulty was used with no directions except to "study the lesson."
The children were "heard" but no special effort was made to
teach the words. A with these words also was then given.
test
The results of the two plans as indicated by the tests given im-
mediately and three weeks later were as follows:
TABLE 106. After Fulton C*4?
WITH THREE WEEKS WITHOUT TFTOEE WEEKS
BRILL LATER DRILL JLATS>
Average % 98 96 73 68
Daily results 98 .. 78
SPELLING 343
Laws of Association. Skill in spelling is primarily a matter

(5)
of forming associative connections between certain arbitrary sym-
bols arranged for the most part in arbitrary order. Economy in
the learning of spelling reduces itself* to this question: Under what
conditions can these associations be made most quickly, most
effectively, and most permanently? Of the four laws of associ-
ation, frequency, vividness, primacy, and recency, the first two
are most directly applicable. Obviously, frequent repetition is
necessary to establish the connections. Frequent reviews, monthly,
weekly, and possibly daily, are indispensable.
The law of vividness operates in numerous ways. This law
states that other things being equal, the most vivid or intense
association is most apt to be recalled. It may be made of service
in two general ways: (i) By any device that will stimulate the
clearest possible attention and interest on the part of the learner
upon the spelling of words, or (2) by any device that will make
conspicuous the particular part of any word that is most likely to
be misspelled. These points will be considered more fully in the
next three paragraphs.
(6) Careful Attention upon the Successive Letters of a Word.
Pryor ('15) reports "an experiment to determine the value of
'spelling the word through' as an aid to learning. Two divisions
of the fifth grade studied the same list.
Conditions as to time,
length of period, and the like were the same for both divisions.
For one, emphasis was placed on observing carefully the order of
letters while studying. Preliminary and final tests given to both
divisions showed an advance from 50.55% to 83.39%, or an average
gain of 32.84% for the division working under the usual condi-
tions. The other division advanced from 48.58% to 89.14%, an
average gain of 40.56%."
(7) Personal Incentive to Interest and Effort on the Part of the
Learner. Aside from competitive contests in their various forms,
there are at least two ways in which personal effort may be stimu-
lated: (a) By
having at regular intervals definite tests, preferably
by means of some one of the standard spelling scales or tests so
that each pupil may know his attainment from time to time and
measure his progress. This plan will usually arouse very keen
interest in the individual to surpass his own previous record. See
Chapter XL This is a form of incentive whose actual effective-

ness has never been fully appreciated, (b) A second form of in-
centive is to arrange with the pupils that the grade in spelling will
be determined half by the work in the spelling-class and half by
their spelling efficiency in all written work. This plan was em-
ployed in a school in Potsdam^ New

York, with the result that the
pupils developed a remarkable efficiency in spelling as shown in

the accompanying graph, Figure 73. Each grade is approximately
one entire grade ahead of the general average. This is one of the
school measured by the writer's
highest records found in any
spelling test.
(8) Calling Special Attention to Difficult Parts of Words. One
of the much needed investigations of spelling is a careful collection
and classification of errors in the words used as spelling material.
loo
80
60
40
20
8 4 5 6 8
Grades
p IG yjj Averages in spelling in grades 5 to 8 in a certain school as measured
by the author's test. The continuous line represents standard attainments.
Errors in spelling usually consist in certain letters or parts of

words only. For example, the error in receive is usually in the two
letters e-i, or in separate, in the a after the p. (Such a study is
under way most common words previously referred
for the 2,626
to.) the teacher would be able to antici-
By means of such a study
pate the probable misspellings and to call especial attention to the
letters likely to be missed. Such particular emphasis may be
secured by asking the pupils to focalize the correct order of the
letters which are usually confused, by writing those letters extra
large, or by drawing a line around them, or by having them printed
in larger or heavier type, and so on. It is quite imperative to
anticipate the troublesome parts of words so as to forestall the
formation of wrong connections. Instruction in spelling should
SPELLING 345
consist in the teaching of correct forms rather than in the un-

teaching of incorrect forms.
(9) Writing the Words. We need to know how to spell words
only in writing. It is a general law. of association that connected
bonds should be established in the manner and order in which
they are to be used. It would seem, therefore, on a priori grounds
that the associative bonds between the successive letters should
be formed by exercise in the writing of the words so that the spell-
ing may become automatic during the act of writing. Current
emphasis upon writing the words in sentences is in the right di-
rection. It would be an interesting experiment to teach two
sections of a spelling-class by having in one section a great deal of
writing of the words of the spelling lesson and by having little or

no writing in the other section. The spelling ability of the two
sections would have to be compared at various times by appro-
priate tests.
(10) Context versus Column Spelling. On the basis of the
discussion in the preceding paragraph, it would seem obviously
advantageous to have a great deal of writing of the words to be
learned, especially in sentences. The belief is held by many that
the ability to spell words in isolation does not insure ability to
spell them in context. As a matter of fact, however, there is very
difference in spelling efficiency between these two situations.
little
Wallin ('n) used the same test words both in a column test and in
a dictated composition and found an average spelling effi-
test
ciency of 97.72%, in the column test, and 96.28% in the composi-

tion test, giving an advantage of only 1.44% to the former. Cook
('14) used 60 words in a column and in a composition test and
likewise found only a very slight advantage in favor of the former
test. Spelling efficiency in composition is only very slightly lower
than it is in columns. This slight loss is probably due to the dis-
traction of attention by the other factors in writing, such as punc-
tuation, grammatical form, and thought content. The argument
that spelling should be taught by using words only in sentences
is not very weighty.
(u) Teaching Homonyms Together or Separately. Suzzalloand
Pearson ('13) report an experiment in which they attempted to
determine the relative effectiveness of teaching homonyms to-
gether or apart. They used five pairs of homonyms in each of
grades 3 to 7 in the Horace Mann school. Each grade was taught
in two sections of about equal ability. One section was taught by
the together-method and the other by the separate-method. All

other conditions were kept as nearly alike as possible. All
classes were tested at the beginning and at the end of the experi-
ment. The outcome showed a decrease of 2.29 errors by the
together-method, and of 1.61 errors by he separate-method.
f
There is thus a slight advantage in favor of the former method.

Suzzallo's experiment was repeated with the same material by
Knight at Montclair, New Jersey,
in grades 3 to 7, inclusive. The
together-method reduced the errors on the average 2.63% while

the separate-method reduced them 2.24%, thus supporting Suz-
zallo's findings. On the other hand, W. F. Jones ('15) reports
that
"Experiments in teaching homonyms have been made by the depart-

mentof education at the University of South Dakota, which show that
homonyms should not be brought together until the second one of the
pair appears in the child's vocabulary. This often gives time to fix the
meaning and the spelling of the first member of the pair before the second
one appears. In such cases the homonyms give relatively little trouble."
(12) Class versus Independent Study. Suzzallo and Pearson

further undertook a comparison of progress in spelling when the
pupils studied independently, with progress when the pupils
studied under supervision. For nearly a year, one class in each of
grades 4, 5, and 6 was taught by the supervised-study plan and the
other by the independent-study plan. In the latter case, the recita-
tion period consisted chiefly in testing or lesson hearing. Their
conclusion is stated thus:
''The evidence of this experiment, therefore, from whatever angle we

study it, shows that teaching of the class-study type is far more effective,
than the independent-study typ."
SPELLING 347
TABLE 107. After Suzzallo and Pearson ('13)
Total decrease in errors
(13) Grouping Words of Similar Spelling. Is it an advantage

1
to learn words of similar spelling in groups? Wagner made an
experiment to answer this question. He divided a 6th grade class
into two sections both of which studied the words in the usual
manner with the exception that the words were presented to one
section in groups, according to their similarity, such as lineal,
lineament, linear and lineage while to the other they were presented
in miscellaneous combinations. The former section, which studied
the words in groups, raised its average percentage from 68.36 in
the preliminary test to 97.14 in the final test, or 28.78% while
the latter section raised its average percentage from 73.25 to 93.6,
or 20.35%. The former group, therefore, gained 8.42% more
than the latter, showing a decided advantage in favor of the
grouping plan.
(14) Imagery. Individuals differ in mental imagery, but recent
inquiries point out the probability that pure types of visuals,
audiles, motiles, etc., are exceedingly rare and that imagery of any
sort may be aroused irrespective of the sense organ through which
the stimuli come. Since practically every child possesses images
of all classes the safest procedure is no doubt to appeal to a variety
1
Reported by Pryor. ('15.)
of images. Note here again the discussion of imagery in Chapter

XI, p. 166.
(15) Spelling To-day and Formerly. Finally it will be of interest
to note how the spelling ability of pupils of to-day compares with
that of our forefathers, particularly in view of the claim often made
that the schools to-day do not train the pupils as thoroughly in the
fundamental subjects. One of the comparisons made in the Spring-
field test (Riley, '08)was that of spelling ability. The same 20
words that had been given as a spelling test to gth grade pupils
in 1846 in Springfield, Massachusetts, were given again in 1906
to 246 pupils of corresponding age in the same school. The pupils
in 1846 had made an average grade of 40.6%, while the pupils
in 1906 made an average of 51.2%. A similar test, made in
Cleveland in the years 1858 and 1909 showed one error less per
child in the 1909 test. Apparently the "superior" spelling ability
5n the good old days is largely an illusion.
CHAPTER XIX
LANGUAGE
PSYCHOLOGICAL PROCESSES INVOLVED IN LANGUAGE
The subjects thus far considered, namely, reading, writing, and

spelling, together with the one to be considered in this chapter, con-
stitute psychologically the complete set of language functions since
they all play a part in the communication of ideas. Reading deals

with the reception of ideas; while writing, spelling and language in
the restricted sense, have to do with the expression of ideas. The
term language as used here, and as used in the school program,
refers only to that portion of the complete language process which
deals with the organization and expression of thought in speaking
and writing.
A complete analysis of all the psychological processes involved
in the language functions would require again an enumeration of
all the elements in reading, writing and spelling. This is unnec-
essary; and nence our present analysis will be limited to the steps
immediately concerned in the expression of ideas in oral or written
form. These elements are:
(1) The arising of ideas in the mind.
(2) The simultaneous or successive arousal of symbols or word
forms corresponding to the ideas.
(3) The transmission of the nerve impulses, connected with the
ideas that arise, to the motor speech centers in oral expression, or
to the motor writing centers in written expression.
(4) The transmission of nerve impulses from the latter to either
the muscles of the speech organs or to the muscles of the hand and
arm.
(5) The execution of the speaking or writing movements.
Steps (3), (4) and (5) are identical with the corresponding ones
previously enumerated in the analysis of the writing and spell-
ing processes. The important steps for our present purposes are
numbers (i) and (2).
From the psychological standpoint, the arising of ideas in the
mind is practically identical in the normal person with the arising
349
of words in the mind, since in the normal child words and meanings
are built up together through oft repeated associations of words
" "
and meanings. Hence we shall use the term word-idea to convey
this union of symbol and meaning. Furthermore, from the psychol-
ogical standpoint, the occurrence in the mind of ideas or words
to be expressed, is fundamentally a matter of the psychology of
association. Where do the ideas come from? What causes them
to arise in themind? Why do certain ones arise rather than others?
To what extent can the occurrence of ideas be controlled? Com-
position, either oral or written, is simply the outward expression of
the ideas that do arise. The occurrence of the ideas and their
precise verbalform takes place in the mind as a part of step (i).
Obviously then, a study of steps (i) and (2) and an attempt to
answer the questions raised thereby, constitute almost entirely
a study in the psychology of association or thinking processes.
Composing fundamentally is thinking.
Let us take a typical example of oral or written composition,
such as a bit of speaking or writing. How do the ideas in this
simple composition arise in the mind? The blunt answer is, they
arise almost entirely in a mechanical manner according to the
established neural or mental connections. The first idea in a chain
arises as the result of a stimulus, either through the senses or
through previous ideas or images in the mind. No

idea probably
ever arises independently in the mind as though out of the blue
sky, but always in succession to a preceding link, stimulus or
occasion. Thinking is simply the flow of ideas, which occurs
almost wholly automatically, according to the laws of association
frequency, vividness, primacy and recency. Conversation also
is largely the automatic flow of association processes in the minds
of the participants. Each statement comes in response to the re-

mark of one of the conversers or in succession to the preceding re-
marks of the speaker himself. Even more formal composition,
such as the writing of a theme, a story or an essay, is principally
the result of association processes aroused in the mind by the sub-
ject of the theme. This occurs, even in vigorous and original
thinking about a subject, largely according to the mechanical laws
of association. Voluntary thinking or composing is controlled
probably only in two ways: (i) by effort and concentration more
ideas are likely to arise than by a purely passive attitude, and (2)
by making a selection among the ideas that do arise preference to
further chains of ideas are determined. But the arising of ideas
LANGUAGE 351
themselves occurs fundamentally according, to the mechanical,

neural and mental connections previously established.
The objection is likely to be made by the reader at this point,
that if this is true, how can any new ideas ever arise? or how can
any original thinking occur? The blunt answer is: Thinking is
original only in the sense of making certain selections, rather than
others, among the ideas that do arise and of -allowing further
associations to arise in connection with them, rather than in connec-
tion with others. In this sense, the possibility of original thinking
is
indefinitely great.
Correct English is fundamentally a matter of associating certain
words in certain orders, rather than in others. The reason why
people say, "Do it good," is because they have been told since
infancy to "Do it good." Language is an immensely intricate
net-work of associative links. Wide vocabulary in speaking or
writing due to the arising of numerous words in connection with
is
given meanings. Elegant diction is due to the more appropriate

selection of words among the larger varieties that do arise. In the
skilled speaker or writer the more appropriate words arise auto-
matically in the course of time. Greater varieties of words come

up with given word-ideas because more of them have been previ-
ously experienced and retained. Two persons may read the same
literary masterpiece, the one may retain a great deal of the actual
diction and phraseology in connection with the ideas read, while
the other may retain very little. In the course of time the former
will acquire a far wider vocabulary and a much nicer diction than
the latter because he retains much more of the actual words and
forms of expression than the other, and consequently when any
topic is brought to his mind, it arouses a far richer wealth of ideas,
words and phrases, and by virtue of this wealth he is able to make
a far superior selection. Ingersoll appropriately said of Shake-
speare:
"The moment his attention was called to any subject comparisons,

definitions, metaphors, and generalizations filled his mind and begged
for utterance. His thoughts like bees robbed every blossom in the world,
and then with 'merry march* brought the rich booty home 'to the tent
royal of their emperor.'" (P. 66 1, Modern Eloquence, Volume V.)
"Some have insisted that Shakespeare must have been a physician,
for the reason that he shows such knowledge of medicine, of the symptoms
of disease and death; because he was so familiar with the brain, and with
insanity in all its forms.
"
I do not think he was a physician. He knew too much ; his generaliza-
tions were too splendid. He had none of the prejudices of that profession
in his time. Wemight as well say that he was a musician, a composer,
becausewe find in 'The Two Gentlemen of Verona* nearly every musical
term known in Shakespeare's'time.
"Others maintain that he was a lawyer, perfectly acquainted with
the forms, with the expressions familiar to that profession. Yet there is
nothing to show that he was a lawyer, or that he knew more about law
than any intelligent man should know. He was not a lawyer. His sense
of justice was never dulled by reading English law.
"Some think he was a botanist, because he named nearly all known
plants. Others, that he was an astronomer, a naturalist, because he gave
hints and suggestions of nearly all discoveries.
"Some have thought that he must have been a sailor, for the reason
'
that the orders given in the opening of 'The Tempest were the best
that could, under the circumstances, have been given to save the ship.
"For my part, I think there is nothing in the plays to show that he
was a lawyer, doctor, botanist, or scientist. He had the observant eyes
that really see, the ears that really hear, the brain that retains all pictures,
all thoughts, logic as unerring as light, imagination that supplies defects
and builds theperfect from a fragment. And these faculties, these apti-
tudes, working together, account for what he did." (P. 665.)
Thinking and language are the two sides of the same shield.
The language used to express ideas depends upon the thinking
that goes on in the mind; and the thinking depends upon the
verbal-ideational connections established in the neural and mental
network by reading and hearing successions of words, phrases and
sentences. Language is not words; it is thinking, thinking by
means of symbols.
THE MEASUREMENT OF EFFICIENCY IN ENGLISH
(a) Methods of Measurement. Four types of measuring devices

have been prepared, each for a different phase of language, (i) For
measuring the grammatical correctness of language, the writer ('15)
has prepared a series of scales consisting of sets of sentences, each
stated in two different ways. These sets of sentences are arranged
in an order of increasingly difficult steps. The pupil in doing the
test is requested to indicate which the correct or preferred form is.
A similar scale has been prepared for testing ability in punctua-
tion. (2) For measuring technical knowledge of the grammatical
structure of English the writer has devised several tests for ascer-
taining quickness and accuracy in indicating parts of speech,
LANGUAGE 353
cases, tenses and modes. (3) For measuring general merit in

written composition, two scales have been constructed. The one,
known as the Hillegas-Thorndike ('12) scale, is composed of a
series of compositions, arranged according to a large number of
judgments in the order of steps of increasing merit from zero to

nearly one hundred. The second, known as the Harvard-Newton
scale (Ballou, '14), is composed of a series of four scales for nar-
ration, description, argumentation and exposition respectively.

Each scale contains six compositions. These were graded by 24
teachers and approximate in percentage marks the values of 45,
55, 65, 75, 85 and 95. A composition is rated by any one of the
scales by comparing its merit with those in the scale, and giving
to it the value of the step on the scale to which it is judged equal.
J
(4) Trabue ( i6) has prepared a series of language scales, each

consisting of some 10 mutilated sentences. These sentences are
arranged in steps of increasing difficulty as determined by experi-
mentation with a large number of pupils. In doing the test, the
pupil is required to supply the most appropriate words in the
blank spaces. It is difficult to say just what this scale measures,
but it probably tests the ability to think of the proper word for
a given situation.
(b) Uses and Results. Without repeating here, we need to say
merely that these tests serve the same general purposes for the
language functions as the measurements devised for the subjects
previously discussed serve for their respective subjects. Measure-
ments of ability in language have shown incredibly large individual
differences among the pupils in the same class and the resulting
overlapping of the abilities of pupils in succeeding grades. Note

the facts in Figure 25 in regard to ability in composition in a
certain Illinois high school. The overlapping is enormous and the
median gain from year to year is surprisingly small. In fact there
is almost no improvement above the first year. The little im-
that is noticeable is probably due chiefly to the drop-
provement
ping out of the poorer pupils in the lower years. Figure 26 shows
similar facts with regard to ability in recognizing correct gram-
matical forms.
Brown and Haggerty measured the progress in composition
('17)
a period of 12
ability in the case of 78 pupils in three classes during
weeks by having them write an extemporaneous composition each
week. These compositions were rated by the Harvard-Newton
scale. The median progress of the three classes is indicated in
Figure 74, being 4.2 points for Classes I and III, and 5.2 points
for Class II. These graphs indicate a noticeable improvement
100
1 2 45678
'
Weeks of Practice
10 11 ia
80
80
60
I
I
50
H 40
o
5
"2
& 90
20
10
FIG. 74. Median scores in composition for three high-school classes. Con-
tinuous line represents a first-semester freshman class; the dotted line repre-
sents a second-semester freshman class; and the dash-line represents a first-
semester sophomore class. After Brown and Haggerty ( '17, p. 524).
from the first week to the twelfth, but show no difference in ability
between the sophomore class and the freshman classes.
ECONOMIC METHODS FOR ACQUIRING SKILL IN THE ENGLISH

LANGUAGE
A great variety of experiences has been obtained and an equally
great variety of personal opinions is held by teachers with regard
LANGUAGE 355
to the most effective manner of learning and teaching English.

But there a corresponding paucity of scientific facts regarding
is
the matter so that we cannot speak with certainty concerning very

many of the factors and conditions that 'promote or retard develop-
ment of skill in the use of language.
(1) The Acquisition of Ideas. According to the psychological
viewpoint, from which we are here examining the language process,
the first and most fundamental element is the acquisition of ideas.
How this be done our present psychological knowledge gives
may
no specific than the common-sense advice of
directions other
gaining as great a variety of ideas as possible through direct per-
"onal experiences and contact with environment and through
extensive reading. The extent to which such varied experiences
and wide reading will be profitable depends mainly upon the extent
to which the experiences and facts are assimilated and retained.
Next to the experiences and ideas themselves, excellent retentive
capacity, or a very essential factor. Re-thinking the
memory is
experiences and facts is likely to be a most helpful exercise in

making new ideas a part of one's mental machinery.
(2) The Acquisition of Words and Forms of Expression. The
next important element is the acquisition of extensive vocabulary
and phraseology. On this point also little more can be said at
present than what, on the basis of general experience rather than
on the basis of demonstrated facts, common sense and good judg-
ment dictate. Wide vocabulary and precise phraseology can prob-
ably be acquired mainly (a) by reading and hearing language
which employs extensive vocabulary and proper phraseology, and
(b) through definite attempts to express one's thought by means
of these words and phrases. Probably one of the best means of
extending one's vocabulary is to take special account of new words
as one reads or hears them, and to make a special point of using
them on the first, and on every other, appropriate occasion. A
useful plan is to record these new and unfamiliar words in a special
notebook. They must, however, not be kept in the notebook only,
but they must be referred to frequently so as to make them a part
of the mental fabric of connections to be used in thinking, speaking
and writing. Pupils often are required to keep such notebooks,
but usually the words never get farther than the notebook. Special
effort must be made to use the words so that they may gradually
be woven into the automatic association processes of the mind.
Another plan is to make a list of the less familiar or unknown
35 6 EDUCATIONAL PSYCHOLOGY
words that are observed in the language of good writers, speakers

and conversationalists, and to think of as many synonyms and
equivalent phrases as possible so that when occasion for using one
of the words arises, synonymous terms will come up which might
preferably be used instead. Frequent drills and tests might very
profitably be given in accordance with this suggestion. The teacher
could give such a list of words in mimeographed form to the pupils
and then see how many corresponding synonyms in a given time
limit, say 5 or 10 minutes, each pupil can write out. The habitual
use of words is fundamentally a matter of association processes
and these can be built up only by a repeated use of the words to be
acquired.
For such instruction it would be exceedingly useful to have a
listof the 10,000 most common words employed by the best current
writers. Special drill exercises in the use and meaning of these
words could be devised by the teacher. 1
(c) The influence of knowledge of other languages. It has
been generally urged that a knowledge of other languages, par-
ticularly of those from which many English words have been de-
rived, is of muchvalue in extending the vocabulary, in developing
a finer discrimination in themeanings and uses of words, and in
facilitating the writing of English. The objective data thus far
available on this problem have been summarized in the chapter on
transference of training, to which the reader should turn for a recon-
sideration of that evidence. In general the evidence indicates
(i) that the study of Latin seems to increase the size of a pupil's
English vocabulary only slightly when Latin is taught is it ordina-
rily has been taught, or quite considerably when taught with specific
reference to the derivation of English words; and (2) that, when the
differences in the original abilities of pupils are allowed for, the
study of other languages aids either very little or nof at all in the
writing of English.
(3) Acquisition of Grammatically Correct English, (a) The
influence of knowledge of grammar. Grammar was introduced into
1
Such a list has been prepared by the writer on the basis of the vo*abulary studies
described in Chapter XVIII. The Starch List of 5,003 different words used by 40
authors was checked up with the Eldricige, Ay res, Jones, and Cook Joists to find all
additional words in them which were not in the Starch List. This made a total of 0079
differentwords found in 40,000 running words (Starch List) of 40 different high-grade
current magazine writers, in some 43,000 running words (Eldridge List) of newspaper
writing, in some 23,000 running words (Ayres List) of correspondence, in 15,000,000
running words (Jones List) of children's compositions, and in 200,000 numinjL words
(Cook List) of family correspondence.
LANGUAGE 357
the schools as a scientific study of the structure of the language

with the belief that this knowledge would aid in the acquisition of
correct usage of the language. To what extent is this belief jus-
tified? All definite investigations of 'this problem tend to show
that knowledge of technical grammar is of much less service in
developing use of correct English than has always been supposed.
This experimental evidence was reviewed in Chapter XIV on the
transference of training, to which the reader should turn, and
hence it will not be restated here. Furthermore, the studies by
Charters ('15) on grammatical errors in oral and written language
indicate that a relatively small number of grammatical rules is con-
cerned in ordinary language. In accord with these results, the
amount of time devoted to the study of grammar has been mate-
rially reduced in recent years and a great many of the topics have
been entirely eliminated. The amount of grammatical knowledge
which functions directly in establishing correct usage is relatively
small. For example, the Iowa Committee on Minimum Essentials
recommended in 1915 the elimination of the following topics:
"The exclamatory sentence; the interjection; the appositive; the

nominative of address; the nominative of exclamation; the objective
complement; the adverbial objective; the indefinite pronouns; the objec-
tive used as a substantive; the classification of adverbs; the noun clause;
conjunctive adverbs; the retained objective; the modes (except possibly
the subjunctive of 'to be'); the infinitive; the objective subject; the
participle, except the definition and the present and past forms; the
nominative absolute; the gerund nominative absolute; sentences for
analysis and parsing that involve subtle points of grammar; formal
parsing; conjugation; diagramming; person of nouns." (After Betts, '17.)
(b) The function of imitation. These investigations imply that

linguistic forms and expressions are acquired very largely through
imitation, both conscious and unconscious, perhaps especially the
latter, of the forms and expressions read or heard, particularly
those heard in one's customary environment. Language forms are
psychological habits which become deeply ingrained in the human
psycho-physical system through constant repetition. To hear
and say,from birth on, "good" for "well" in such expressions as,
"Do it good," or "I don't feel good," establishes such strong chains
of associated bonds that, in spite of better knowledge, such expres-
sions will continue to be used and not be overcome by pages of
grammatical knowledge.
The factor of conscious, intentional imitation is probably em-

ployed far too little in the development of ability in composition.
Students of art learn to paint by copying the great masterpieces.
Students of composition arc shown great masterpieces and asked
to read them but are told not to imitate them as that would pre-
vent the development of originality. Is this not incorrect proce-
dure, psychologically? Why would it not be excellent practice
in acquiring vocabulary, diction, style, and even ideas to have
pupils read and study, for example, a description of a landscape
jid then to ask them to write one of a real landscape in their
environment by making it just as similar to the masterpiece, in
plan, diction and phraseology, as the actual landscape permits.
Or why should not such a plan as the following be really effective
in acquiring the use of words. Have the teacher take such a descrip-
tion as the following one of George Washington:
"When Washington was elected general of the army he was forty-

three years of age. In stature he a little exceeded six feet his limbs were
;
sinewy and well proportioned; his chest broad, his figure stately, blending
dignity of presence with ease of manner. His robust constitution had
been tried and invigorated by his early life in the wilderness, his habit of
occupation out of doors, and his rigid temperance, so that few equalled
him in strength of arm or power of endurance. His complexion was
florid, his hair dark brown, his head in shape perfectly round. His broad
nostrils seemed formed to give expression and escape to scornful anger.
His dark blue eyes, which were deeply set, had an expression of resigna-
tion and an earnestness that was almost sad."
Let the teacher select all the important descriptive words and
phrases from it and put them in a list, such as this:
stature rigid temperance

sinewy endurance
well-proportioned complexion
stately florid
blending dignity nostrils

ease of manner scornful
robust resignation
invigorated earnestness
occupation
The pupils should first study these words and phrases to make
sure that they thoroughly understand them; then they should
write a description of some person of their acquaintance who may
LANGUAGE 359
also be known to the teacher, using as many of these phrases as

may be applicable to this person. The pupils in this case should
not first read or hear the description from which these words were
taken. After they have written their own composition, they should
then compare it with the masterpiece so that they could check up
their own use of words with the possibly better use in the model.
Such a plan would seem to be worth while experimenting with
to ascertain its proficiency in developing language ability. How
can a pupil acquire proper words to be used in describing a human
being if he does not know, and has no means of finding out, what
such words are? Such a plan would seem to have on a priori
grounds two distinct advantages: First, it would make conscious
use of imitation which is unquestionably a potent factor in the
acquisition of language; and second, it would tend to make the
learning of language definite and concrete. The great difficulty
in learning composition is that the advice and suggestions given
by the teacher are too vague and indefinite. The pupil is told
that he lacks organization, that he must develop a better vocabu-
lary, or that he lacks imagination. But how is any child specif-
ically and concretely to know how to improve in these respects?
How may a child know
that such words and phrases as "stature,"
"stately dignity," "florid complexion" might be the most appro-
priate to use when describing certain persons? When the child
learns addition or spelling he knows more precisely what he has to
learn. Instruction in composition can possibly not be particular-
ized as fully, at least not as easily, but that should not prevent as
much particularization as possible.
The chief objection by teachers of English to such a procedure
is that they believe it would kill originality and make mere thought-
less, verbal machines out of their pupils. In answer to this point,
however, we must remember that the most original people in the
world are also the ones who use to the fullest extent the work,
methods, and ideas of others. The most original persons are also
the most imitative persons. Ingersoll said of Shakespeare:
"Of course Shakespeare made use of the work of others, and we might
almost say, of all others. Every writer must use the work of others. The
only question is, how the accomplishments of other minds are used,
whether as a foundation to build higher, or whether stolen to the end
that the thief may make a reputation for himself, without adding to the
great structure of literature.
"Thousands of people have stolen stones from the Coliseum to make
huts for themselves. Thousands of writers have taken the thoughts of

others with which to adorn themselves. These are plagiarists. But
the man who takes the thought of another, adds to it, gives it intensity
and poetic form, throb and life, is in the highest sense original.
"Shakespeare found nearly all of his facts in the writings of others and
was indebted to others for most of the stories of his plays. The question
is not: Who furnished the stone, or who owned the quarry, but who
chiseled the statue?" (P. 645, Modern Eloquence.)
Some of the prominent literary writers have not only pointed

out the importance of conscious imitation in the development of
their own styles of writing, but have described their own conscious
attempts to imitate other great writers. For example, Stevenson

has said this concerning imitation:
"That, like it or not, is the way to learn to write. It was so Keats

learned, and there never was finer temperament for literature than
Keats's; it is so, ifcould trace it out, that all men have learned. Per-
we
haps I hear some one cry out: 'But that is not the way to be original!'
It is not, nor is there any way but to be born so. Nor yet, i you are born
anything in this training that shall clip the wings of your

original, is there
originality. There can be no one more original than Montaigne, neither
could any be more unlike than Cicery; yet no craftsman can fail to see
how much the one in this time tried to imitate the other. Burns is the
very type of a prime force in letters; he was of all men the most imitative.
Shakespeare himself, the imperial, proceeds directly from a school. Nor
is there anything here that could astonish the considerate. Before he
can tell what cadences he truly prefers, the student should have tried all
that are possible; before he can choose and preserve a fitting key of words,
he should long have practiced the literary scales . . . and it is the great
point of these imitations, that there still shines beyond the student's
reach his imitable model."
"Whenever I read a book or passage that particularly pleased me, I
must sit down at once and set myself to imitate that quality of propriety
or conspicuous force or happy distinction in style. I was unsuccessful
and I knew it, but I got some practice in these vain bouts in rhythm, in
harmony, in construction, and in co-orclination of parts. I have thus
played the sedulous ape to Hazlitt, to Lamb, to Wordsworth, to Browne,
to Defoe, to Hawthorne, to Montaigne, to Baudelaire, and to Obermann."
(Stevenson, R. L., Memories and Portraits, p. 55.)
Franklin likewise described his attempt to improve his own

style of writing by
carefully studying a volume of The Spectator
and developing a style similar to it.
LANGUAGE 361
"I read it over and over and was much delighted with it. I thought
the writing was excellent, and wished, if possible, to imitate it. With
that view J took some of the papers and making short hints of the senti-
ments in each sentence, laid them by a. few days, and then, without
looking at the work, tried to complete the papers again, expressing each
hinted sentiment at length, and as fully as it had been expressed before,
in any suitable words that should occur to me. Then I compared my
'Spectator* with the original, discovered some of my faults, and cor-
rected them." To acquire a stock of words and a readiness in recollection
and use of them, he "took some of the tales in the 'Spectator' and turned
them into verse; and after a time when I had pretty well forgotten the
prose, turned them back again." (Quoted by Bolton, F. E., Principles of
Education, p. 421.)
Brander Matthews in a more general way has pointed out the

relative shares of imitation and originality.
"
Consciously or unconsciously every artist is a debtor to the past.
The most original of innovators has made his originality partly out of
himself, partly out of what he has appropriated and absorbed from those
who practiced his art before him. Only a few of his separate contrivances
are his own, and the most he can claim is a patent on the combination.
"... The young artist is a weakling if he openly robs any single one
of his predecessors; he is a dolt if he does not borrow from as many of
them as may have the separate qualities he is striving to combine.
"The arts arc one in reality; and what is true of painting and sculpture
and architecture is true also of literature, of prose and verse. For exam-
ple, there are few men of letters of our time whose prose has been more
praised for its freshness and its individuality than the late Robert Louis
Stevenson; but his was an originality compounded of many samples.
He confessed frankly that he had sat at the feet of the masters, playing
the 'sedulous ape' to a dozen or more, and at last slowly learning how to
be himself. Again, the verse of Dante Gabriel Rossetti has a note of its
own, a note which many younger poets have delighted to echo and re-
echo; but Rossetti told a friend that the exciting cause of his 'Blessed
Damozcl' was the 'Raven' of Edgar Allan Poc, and Poe's own indebted-
ness to Coleridge is obvious, even if it had not been expressly avowed."
(c) Specific attention to frequently recurring grammatical

errors. Specific attention to, and special drill in, the precise func-
tions to be developed have been shown over and over again by
experiment and experience to be most directly efficacious in pro-

ducing improvement. Several valuable studies have been made
in recent years of the most common types of errors in oral and
written language and of the frequency of their occurrence. Char-

ters and Miller ('15) made a study in cooperation with the teachers
in schools in Kansas City, by keeping a record for one week of all
errors in oral language heard by the teachers in and about the
school, and by collecting all errors in the written work of the
pupils for one month. A similar study was made by H. D. Fillers
('17), at Bonham, Texas, based on results from goo pupils in grades
three to eight for written language, and in grades two to eight for
oral language.
A third study has been reported by Sears and Diebel ('16) from
Cincinnati based on oral errors of 1,378 children of grades three to
eight which were heard by the teachers during a period of four
days.
A fourth investigation was made by Starch (unpublished) in
which oral errors of pupils in grades one to eight were collected by
the teachers, and oral errors of university students by a group of
special students,and written errors were collected from some 1,700

themes obtained from eleven high schools. The high schools varied
from very small ones to one having more than 1,000 students. The
total number of oral errors collected was 2,916 from grade pupils
and 1,164 from university students. The total number of written
errors collected from the high school pupils was 2,316, making a
total of 6,396 errors. The results of all four studies are sum-
marized in parallel columns in Table 108.
On the whole, the corresponding parts of all four studies agree
surprisingly well, showing considerable reliability as well as striking
similarity in the types of errors found in different parts of the
country.
LANGUAGE
TABLE ioS
Errors in language
TABLE 108 Continued
These studies show that most of the errors in language are con-
fined to a small number of types. Thus in Charter's tabulation,
71% of all oral errors fall under five types or grammatical .rules,
namely, No. 9, Conlosion of Past Tense and Past Participle, 24%;

LANGUAGE 365
No. 8, Failure of Verb to Agree with its Subject in Number and

Person, 14%; No. 12, Wrong verb, 12%; No. 18, Double Negative,
n%; and No. 20, Syntactical Redunckmce, 10%. The same situa-
tion obtains among written errors; 91% of all errors fall under ten
classes or rules.
There are several rather striking differences between the errors
of written and oral composition. Written composition is much
more apt to get the wrong form of noun or pronoun (16%, 11%)
than oral (2%, i%, o - %) and the same is true of confusion of the
tenses (12%, 19%, vs. 2%, 5%, 3%). On the other hand, oral lan-
guage is much more apt to confuse the past tense and past parti-
ciple (24%, 20%, 19%, vs. 5%, 4%), to use the wrong verb (12%,
21%, 14%, vs. 7%, 4%), and to use the double negative (11%,
14%, 12% vs. i%,o%).
Sears and Diebel also tabulated their material to show the
relative proportions of different types of errors for the various
grades. (Table 109.)
TABLE 109
CLASSIFICATION OF ERRORS
Verbs
X 4
60.0
5 678
GRADES
TOTAL
1. 44. 2 55 .4 54 o 43 "5 48 2 40 9
2. Pronouns 15.9 14.0 6.7 7.7 12.3 18.8 13.5
3. Negatives 11.5 7.1 202 7.3 15.2 14.0 n.6
4. Syntactical redundance. ... 8.0 6.6 11.2 12.6 16.5 9.6 9.7
5. Mispronunciations 14 7 7.8 2 2 4.9 1.7 8.0
6. Prepositions 3.4 3.2 1.8 5.6 4.1 2.6 3.5
7. Adjectives and adverbs. ... 2.0 0.6 2.2 6.6 8.2 4.8 3.3
8. Ambiguous expressions 0.2 ,2
Percentages of errors in each grade due to each class of mistakes. The

forty-one specific errors which Scars and Diebel found most frequent
follow with their respective frequencies:
1. haven't no for haven't any 233

2. seen had saw 180
3. ain't for am not, isn't, aren't 124
4. done 113
5. got, ain't got, haven't got 112
6. 1 and my brother 96
7. kin, jist, git, kitch '.
91
8. ain't for haven't, hasn't 89
9. Frank and me for Frank and 1 80
10. is for are 76
11. them for those 75
12. learn for teach 71
13. can for may . . 60

14. my mother, she 58
15. got for receive, become, grow, is 53
16. that there 38
17. don't for doesn't i 36
18. It was me 36
19. leave for let 34
20. went for gone 32
21. come for came 31
22. never gave 30
23. by my aunt's 28
24. drawcd, throwcd, growed, knowed 27
25. somcpin, for something 25
26. broke for broken 22
27. lay for lie . 21
28. make dinner for prepare, get . ... 21
29. says for said ... . . . . 20

30. all two, both
all .
19
31. rcadin, nothin 18
32. by us for near us 16
33. he docs it like she docs 15
34. why, and, so, at the beginning of sentence or in middle of sentence. ... 15
35. that, which, for who and whose ... 14
36. onct 12
37. in back of 12
38. funny, lots, etc., for queer, many n

39. et for ate n
40. run for ran n
41. set for sit 10
R. I. Johnson obtained samples of written exposition, narration,

and description from 132 high school freshmen and 66 college
freshmen from the Kansas City High School and Junior College in
an effort to determine the persistence of errors in written English.
The high-school freshmen made 2,160 errors in 50,371 words, while
the college freshmen made 787 errors in 32,693 words. Roughly
this was 23 errors per thousand words for the former and 42 errors
per thousand words for the latter. The college freshmen are there-
fore distinctly superior. The following table shows the relative
improvement for the various types of errors. Naturally the raw
results of such an experiment somewhat exaggerated the improve-
ment due to four years of training because of the elimination of
the poorer high school students before reaching college. In the
last column of the table this factor has been compensated for.
LANGUAGE 367
TABLE no
I have computed the correlation of the two series of ranks in

the above tableby the Spearman rank method which yields .26.
This means that there is slight connection between the prevalence
and the persistency of errors, that is, if there is any tendency at all
it is for the more common errors to be also more persistent.
Such studies should prove exceedingly useful in helping the

teacher to devise special exercises and drills designed to eliminate
these errors and to substitute for them correct forms of expression.
(4) Oral versus Written Practice in Composition. The processes
of oral and written composition differ psychologicallyfrom each
other in certain important respects; and the same is even more true
of formal oral composition and everyday speech. Ordinary conver-
sation consists for the most part of short verbal responses each last-
ing but a few seconds, initiated by continuously recurring stimula-
tions from the other individuals engaged. Formal oral composition
on the other hand is a reaction many times as long with but a
single formal impulse at the beginning. With this enormously
increased length of reaction arise all the complex problems of
structure and the accompanying strain on the attention, to say
nothing of the powerful emotional inhibitions of thought and action
resulting from self-consciousness and fear of failure, which prac-
tically do not exist at all in ordinary spontaneous speech.
Again, written composition differs psychologically very much
from oral composition in permitting much more deliberate action.

If a satisfactory thought docs not present itself, there is no imper-
ative necessity for immediate action at all costs. There are no
distractions of the attention from a staring audience. Perhaps
most different of all, written composition presents an opportunity
for revision of structure, choice of material and words which oral
composition cannot permit. Even the vocabulary and sentence

structure to a certain extent.
is different
It has already been pointed out in the chapter on transfer of train-

ing that psychological reactions tend strongly to be limited closely
to the conditions under which training takes place. The moral as
to composition is obvious. Oral composition no less than written
composition requires specific practice.
Probably over nine-tenths of the composition of the average
adult is oral and only a very small fraction is written. And yet
the schools until recently have directed about nine-tenths of all
specific training in composition toward written expression, either
with the belief that training in writing would carry over directly
to oral composition or with no realization at all of the importance
of oral composition. The whole process has disregarded the psy-
chological law of association that associated bonds should be
formed in the order and manner in which they are to be used. If
language is to be spoken, the neural links via the motor speech
centers should be specifically and correctly exercised. It is needless
to point out here the inestimable value of speaking and conversing
in correct, convincing manner. Why should the schools not direct
their efforts far more fully to specific training in oral composition?
The reply might be made that all speaking in general and all re-
citing in classes affords practice in oral composition. But the
trouble is that most of it is done with little or no attention to cor-
rect, well-organized expression. Gross errors, to be sure, are
pointed out to the child; but why should the school not assign a
theme to each pupil in the English class on which he is to give a
3 or 4 minute oral composition in correct, well-planned and thought-
out form? Apropos of this point, a committee of The Illinois As-
sociation of Teachers of English arranged
"A course for the second semester of the ninth grade, which was to be
taught in two ways; one class would have only written exercises; the
other, a combination of two-thirds oral and one-third written. All
classes taking either course were to be given the same written tests at
LANGUAGE
the beginning, at the middle, and at the end of the semester. All the
papers written by each class, including these tests, were to be forwarded
to the committee in charge of the experiment, accompanied
by a report
from the teacher, stating as accurately as possible how much time he
spent in preparation, in conference, and in correcting papers, and also
his opinion as to the results of the experiment.
"The outcome was decidedly favorable to the use of oral composition.
The sections taking the combined course were better at the end of the
semester in thought-vigor, freedom and interest than the others they were
;
no worse in spelling and punctuation and better in handwriting indeed,

the writing sections showed marked degeneration in all matters of me-
chanics. Over half of the 22 schools which carried out the experiment in
full reported greater improvement in the combination sections, while only
two reported less improvement." (Reported by Hosic, '15, pp. 98-99.)
(5) How Much Written Composition is Profitable? Written

composition, even aside from consuming the time that might more
profitably be used for oral composition, is probably overdone in
many English classes. So much required that it becomes a bore
is
with little profit to be derived for the amount of time consumed in

putting down upon paper a few trivial ideas about fictitious and
worthless subjects. Professor Lounsbury and other teachers of
English have begun to doubt the value of such extensive writing.
He says:
"I am by no means disposed to go so far as the historian of New Eng-

land, John Gorham Palfrey, who as I have been told, was wont to ex-
press the desire that an act of Congress should be passed forbidding on
pain of death anyone under twenty-one years of age to write a sentence.
Excess in one direction can not be remedied by excess in the opposite.
Still, none the less am I thoroughly convinced that altogether undue
importance is attached to exercises in English composition, especially

compulsory exercises; that the benefits to be derived from the general
practice in schools is vastly overrated; that the criticism of themes,
even when it is fully competent, is in the majority of cases of little value
to the recipient; that in a large number of instances the criticism is and
must ever be more or less incompetent; and that when the corrections
which are made inefficiently and unintelligently, as is too often the case,
the results are distinctly more harmful than helpful."
William Lyon Phelps, Professor of English Literature at Yale,

has readied a similar conclusion:
"On the subject of required English composition, I am a stout, un-

abashed and thorough sceptic. And although the majority is still against
me, I am in good company. Professor Child read and corrected themes

at Harvard for about forty years: at the end of the time it was his fervent
belief that not only was the work unprofitable to the student, but that
in many cases it was injurious. That it is always injurious to the in-
structor, when it is intcmperatcly indulged, is certain. When I was an
instructor at Harvard, 1 one day met Professor Child in the yard. He
stopped a moment and asked me what kind of work I was doing. I said,
'Reading themes.' He put his hand affectionately on my shoulder, and
remarked with that wonderful smile of his, in which kindness was mingled
with the regret of forty years, 'Don't spoil your youth/ Professor
Wendell, who inherited the bondage under which his predecessor groaned,
has never really believed in the efficacy of the work. Professor Louns-
bury of Yale has given valuable and powerful testimony against it.
Professor Cook and Professor Hecrs two quite different types of men
are in this point in absolute agreement." (P. 117.)
By wayof concrete evidence for his opinion, Professor Phelps

that after requiring only a moderate amount of theme writing
tells
during the freshman and sophomore years he submitted a batch

of compositions written by his juniors to one of the Harvard pro-
fessors. He read them carefully and testified that they were
exactly as good technically as those done by Harvard juniors. He
further says:
"I know of nothing in the world that illustrates more beautifully the
law of diminishing returns than required courses in composition. A
class of students will never under any circumstances write five times as
well by writing five themes as they will by writing one; but the reading
and correcting of five themes require five times the effort on the part of
the body of teachers." (P. 123.)
from the diversity of opinion and practice regarding

It is evident,
the question, that a careful and extensive investigation should be
made to determine if possible the optimum amount of writing to
be required in English courses. The report of the Committee of
the Modern Language Association of America and the National
Council of Teachers of English on the Cost and Labor of English
Teaching (1913), based on returns from 552 English teachers in
93 high schools and 345 English teachers in 96 colleges, states that
"The amount actually written under present conditions averages for
high schools 380 words a week throughout the year, and for colleges 630
words a week. Ideal conditions would slightly increase these averages
to about 430 for high schools and 680 for colleges, and would make pos-
sible equal attention to oral and to written training."
LANGUAGE 371
These facts are interesting as indicative of the present practice;

they are significant as indicative of the consensus of opinions con-
cerning the ideal amount of writing to be done. No one of course
knows what the ideal amount is nor where the region of diminish-
ing returns is located. Opinions do not settle the question. Ex-
perimental and statistical inquiries must be made to determine
the actual results produced by various amounts of writing under
varying conditions and to locate, if possible, the region of dimin-
ishing returns.
(6) Good English in All Oral and Written Work. One of the
great counteracting forces to the influence of the teaching of com-
position in English classes, particularly in high schools and univer-
sities, isthe utter disregard for proper use of language in work
outside of the classes in English. Pupils wear their Sunday clothes
in only one class and go shabby the rest of the time. Instruction
in composition is effective only when it succeeds in making correct,
elegant language a part of the automatic association processes of

thinking. During nine-tenths of the time, the language habits
are being mechanized in a slipshod manner. How can the instruc-
tion in English classes prevail against the iron chains of habit
formed during the rest of the day?
In his classic chapter on habit, James says: "The second maxim
is: Never suffer an exception to occur till the new habit is securely
rooted in your life. Each lapse is like the fall of a ball of string
which one is carefully winding up; a single slip undoes more than
a great many turns will wind again."
Pupils should be required to be as careful in all their oral and
written work as they are in their classes in English. The organiza-
tion of schools should be so modified that the necessary cooperation
between teachers of English and teachers of other subjects could be
made possible. Why should not all written work in all other sub-
jects have to pass muster before the teacher of English? While

all teachers, besides teachers of English, are in theory supposed to
watch carefully over the English used by their pupils, they do as a
matter of fact pay very little attention to it, partly because they
feelthey are too busy teaching the content of their own subjects
to be able to spare the time, and partly because many of them are
not as competent to correct the English of their pupils as they
should be.
A powerful incentive to the pupil to use at all times the best
language that he is capable of could be given by basing his English
mark to the extent of say one-third or one-fourth, upon his work in

the English class, and to the extent of two-thirds or three-fourths
upon the quality and correctness of his English in other classes.
(7) Practice in Expressing Really Important and Personally
Vital Ideas. On the assumption that thought and language are
intimately related, that in fact for practical purposes idea and
word or thought and language are substantially identical. It would
seem to be a fair inference that the exercise in oral and written
composition should be carried on in conjunction with topics con-
cerning which the pupil really has ideas or concerning which, if he
has none, it would be really worth while to acquire ideas. From
a psychological and humanitarian standpoint, it seems almost
a crime to ask a pupil to write several pages of words on a topic
concerning which he has no ideas or in which he has not the slightest
interest, or which has no vital importance to him or to anybody
else. Yet how often is such the case! It is not likely to be very
stimulating to a pupil to be asked to write 150 words with this
phrase as the beginning of the first sentence: "Life to me
seems .... "; to write a thame on that veteran of topics "The
Autobiography"; or on such subjects as "Why Go to Church?"
"Why go to College?" "What Picture Impressed me Most,"
"What I got out of Virgil's Êneid."
Brown and Haggerty incidentally point out in their study that
the pupils wrote distinctly better compositions on certain topics
than on others. The most striking case was the topic of the sixth
week, "How I Earned Some Money" which produced a rise in the
curves to as high a point as was reached even at the end of the
twelve weeks. The topic for the fifth week, "The Pleasures of
Skating (or some other sport)" and for the seventh week, "The
Right Kind of a Chum" produced distinct drops in the curves.
Too much of the practice in theme writing has to do with fictitious
situations and not with vital ideas. The average adult, unless he
enters a profession of which literary work is a necessary part,
rarely has occasion to write about fictitious or far-fetched subjects.
His composition, both oral and written, has to do with problems
which are vital to his life and concerning which he actually has
ideas or concerning which, if he has none, it is very important for
him to acquire ideas. His welfare, his business and professional
success may depend upon the forcefulness of his letter or upon the
convincingness of his interview. Many adults testify to the effect
that they obtained most of their training in organizing and ex-
LANGUAGE 373
pressing their thoughts in connection with their business, profes-

sional,and vocational problems.
Why should not the composition in school be centered around
problems that are real to the pupil qr that will be real to him?
Why should not the written work in other school subjects be
submitted as the work in English composition, and why should not
the correction and development of language be centered around
the expression of thoughts actually necessary in his school work?
The engineering school in a western university severed its instruc-
tion in English from the academic English department because of
the purely formal drill that was provided for its students, and
instead secured the services of a teacher, who was trained both in
engineering and in English, to give instruction in English by using
their writing, their laboratory notes and reports, their language work
in the various courses as the chief basis of instruction in English.
(8) Effect of Differences in Teaching Ability. The differences
of the ability of teachers are probably as striking, if not more so,
in the teaching of English as in any other branch. The differences
in the average abilities of entire classes are very great. The fol-
lowing table gives the median value of the compositions written

on the same topic and under similar circumstances in the high
schools in a certain county in Illinois. Each composition was rated
by three or more judges according to the Hillegas-Thorndike scale.
TABLE in
HIGH NUMBER OF MEDIAN SCORE IN COMPOSITION
SCHOOL PUPILS (HILLEGAS SCALE)
1 6 40.5
2 77 47-4
3 l69 47-7
4 IS 52-8
5 41 52.9
6 113 58-5
7 ... 62.3
8 55 63.0
9 261 63 I
.
10 22 77.2
From this table it appears that the pupils in the best school
wrote compositions of approximately twice as good quality as the
pupils in the worst school. Furthermore there is, so far as these
limited data go, no indication of connection between size of school
and quality of composition. The probability is that the most potent
factor in the situation was the teaching ability of the instructor.
CHAPTER XX
ARITHMETIC
THE PSYCHOLOGICAL PROCESSES INVOLVED IN ARITHMETICAL

OPERATIONS
An analysis of the psychological processes involved in arith-

metical operations may be undertaken in two ways: either we may
analyze the mental activities concerned in a typical arithmetical
operation as performed by an adult or by a practiced pupil, or we
may trace the genetic development and combination of arithmet-
ical concepts and processes in the child. The one would be a
dissection of the finished product as carried out by the trained
individual; the other would be a synthesis of the arithmetical
elements as they arise in the mental growth of the child. Both
methods of attack will lead essentially to the same set of elements.
Let us pursue for the present the first method of approach and let
us take as an illustration a common everyday type of arithmetical
computation. Suppose you purchase at a store, fruit for ioc., a
loaf of bread for ioc., a pound of butter at 4$c., and a bunch of
celery for yc., and give the clerk a dollar bill, how much change
should you receive? What mental processes are either involved
or presupposed in arriving at the answer, 28c., or rather iri the
clerk'smaking the change up from 72 until he reaches one dollar?
An analysis will show at least the following steps: (i) The concepts
of numbers and their meaning, (2) the ability (a) to hear, interpret
and speak (if only by inner speech, which accompanies even the
thinking of the numbers) the sounds for the numbers when the
calculation is carried out mentally, or (b) to write and read the
symbols for the numbers when the calculation is done with paper
and pencil; (3) the previously established mechanical associations
among numbers known as addition, subtraction, multiplication
and division; (4) The fact that the various articles are put to-
gether probably suggests the putting together of the numbers
representijig their value. That is, it acts as a stimulus to arouse
the association process of addition rather than subtraction or
multiplication or division; (5) the occurrence of the successive
374
ARITHMETIC 375
links of the mechanical associations of the numbers leading to the

sum and the continued association of counting by units (pennies)
until 100 is reached.
In brief the steps in the solution of ar representative arithmetical
problem would be as follows:
(1) Number concepts previously acquired.
(2) Ability to read, write or speak symbols for the numbers or
objects to which they refer. The steps involved in these particular
processes have been analyzed in the preceding chapters and hence
need not be repeated here.
(3) Numerous connections previously established between the
various numbers, which are known as the fundamental operations
with numbers, fractions, and so on.
(4) A clue as to the particular connection to be made at succes-
sive points.
(5) The execution of the successive acts with the particular

numbers involved.
(1) The number concept. This is partly acquired by the child
before he enters school. It arises out of, and develops through,
the endless experiences of the child in contact with objects and
repeated occurrences of events. Some investigators of the genesis
of the number idea believe it to arise by means of counting; others
believe it to arise by means of grouping. The probability is that
both factors contribute. What happens is substantially this: The
child deals with various objects and learns through handling them
that they are separate things and that there are several of them.
As he continues to handle them, he established definite meanings
and associations with each one. He furthermore develops the
idea of number or quantity by finding possibly that one or more
may sometimes be missing. He thus acquires a group perception
and gradually widens it to the ability to count, which essentially
consists of associating certain sounds and certain motor speech
processes in speaking the sounds, with successive objects or acts,

probably both, as the successive objects are touched and handled
or at least looked at in turn. These elementary number ideas are
then enlarged rapidly upon entering school by providing more
extensive materials and by associating larger numbers with
them.
(2) This step involves all the detailed and complicated elements
enumerated in the reading and writing processes and as such need
not be considered again as they contain no new elements. All the
other steps are specifically arithmetical processes and require de-

tailed consideration.
(3) The establishment of associations among the numbers.

The processes involved in the four fundamental operations of
adding, subtracting, multiplying and dividing are pure association
processes supplemented by experiences with illustrative material
to show the meaning of the various operations. Ultimately and
fundamentally they are established in the neural and mental
machinery as associative links which, in the trained individual,
become purely automatic. Six plus 2 equals 8, 6 minus 2 equals 4,
6 times 2 equals 12, and 6 divided by 2 equals 3, are pure associa-
tion processes so that when the two numbers with a certain symbol
between them appear and are spoken in succession, the appro-
priate last link is brought up thereby.
(4) The clue as to which process is to take place. The element
in a given situation or problem that suggests which of the four op-
erations at a given point shall take place is the clue which acts as a
stimulus to arouse the appropriate associative reaction. In the
illustration used, the fact of putting the articles together probably
acts as the clue which starts the association of 10 plus 10 plus 45
plus 7 equals 72. Step (4) in this is what we ordinarily
problem
call reasoning. But fundamentally, arithmetical reasoning, as well
as so-called reasoning in general, is essentially a matter of the
perception of differences and similarities and a matter of selective
association processes which are largely automatic in the perfected
stages of arithmetical skill and largely trial and error in the early
stages. The child associates the putting together of objects with
addition; the taking away of objects from a group with subtraction;
the putting together of groups of equal size with multiplication;
and the taking away of groups of equal size with division. This at
firstprobably develops through the multiplication of objects in
these various ways. Something in any given situation or problem
suggests one or the other of the four operations. Through in-
definitely repeated situations, the correct operation is probably
suggested as a matter of association. In a more complex situation
in which the clue does not operate as automatically, the mind pro-
ceeds largely by trial and error by bringing up in turn different
associations until the correct one arises, or until one arises that
satisfiesthe circumstances.
In ordinary terminology, we call this process reasoning. It may
seem as though we were reducing reasoning to mechanical associa-
ARITHMETIC 377
tion which naively we do not consider to be reasoning. Perhaps

in a sense that is really what it amounts to. There probably is no
such thing as reasoning in the sense of forcing one's thought proc-
esses into a given desired direction in a straight and direct line by
sheer force of will. Reasoning, even that of the most original
and inventive type, probably consists fundamentally in starting
with a certain idea, desire, or problem, in short, with a stimulus,
and in waiting for associations to arise and then in following out
in turn by trial and error, one link after another, and in waiting for
each one to bring up its links until a chain of successful links arises
which satisfies the desire or which meets no opposition and which
is then selected. Probably all that voluntary effort will do is to
stimulate possibly a more rapfd arousal of associative links, and to
stir up by virtue of stimulating greater neural activity, such addi-
tional associations as ordinarily do not come up quite so easily.
This is apparently what happens even in the inventive and most
original type of thinking or reasoning. The statements of scientists
and inventors bear witness to the error of the usual belief that
original thinking goes straight to the goal with unerring step. Thus
Jevons remarked:
"In all probability the errors of the great mind exceed in number those
of the less vigorous one. Fertility of imagination and abundance of
guesses at truth are among the first requisites of discovery."
And Faraday said:
"The world little knows how many of the thoughts and theories which
have passed through the mind of a scientific investigator have been
crushed in silence and secrecy by his own severe criticism and adverse
examination; that in the most successful instances not a tenth of the
suggestions, the hopes, the wishes, the preliminary conclusions have
been realized." (Quoted from Lindley.)
Edison has described the invention of the electric lamp as follows:
"During all those years of experimentation and research, I never

once made a discovery. All my work was deductive, and the results I
achieved were those of invention, pure and simple. I would construct
a theory and work on its lines until I found it was untenable. Then
it would be discarded at once and another theory evolved. This was
the only possible way for me to work out the problem. ... I speak
without exaggeration when I say that I have constructed 3,000 different
theories in connection with the electric light, each one ofthem reasonable
and apparently likely to be true. Yet only in two cases did my

experi-
ments prove the truth of my theory. My chief difficulty was in con-
structing the carbon filament. . . .
Every quarter of the globe was
ransacked by my agents, and all sorts of the queerest materials used,
until finally the shred of bamlioo, now utilized by us, was settled upon."
"
(G. C. Lathrop, Talks with Edison," Harpers, Vol. 80, p. 425.)
However, it must not be supposed that our analysis has more

than sketched in its broadest outlines one of the most complex
and difficult mental activities which human beings ordinarily per-
form. Indeed arithmetic is psychologically not a single activity but
a group of fairly distinct activities, some of which are as unrelated
to each other as arithmetic as a whole is to English. Thorndike
found that the average arithmetic grades of the children in a cer-
tain school for a period of two and one-half years gave a correlation
(Pearson) of .39 with the grades in English, and of .36 with those
in geography. On the other hand, Stone tested 500 children on the
four fundamental processes and arithmetical reasoning and ob-
taiced the following correlations:
Addition with subtraction 50

Addition with multiplication 65
Addition with division 56
Subtraction with multiplication 89
Subtraction with division 95
Multiplication with division 95
Arithmetical reasoning with addition 28
Arithmetical reasoning with subtraction 32
Arithmetical reasoning with multiplication 34
Arithmetical reasoning with division 36
These results are especially striking in their low correlation of

the four fundamental processes with arithmetical reasoning. It
is significant also that subtraction, which might be thought to be
closely related to addition, turns out to be much more closely re-

lated to division, the correlations being .50 and .95 respectively.
These results have added significance when it is remembered that
allelements of subtraction or multiplication, for example, which
are normally present in division fundamentals were carefully
eliminated from the scores.
Our numerous previous observations of the specific nature of
psychological activities, however, so far from making us surprised

at Stone's results would rather lead us to expect still greater dif-
ferentiation of the arithmetical processes. Howell found with re-
ARITHMETIC 379
gard to the combination 128=4 and 12-4=8 that knowing

the one by no means assured knowing the other. Table 112 shows
the total number of errors of 300 children in grades three to eight
on eight pairs of complementary subtraction combinations:
TABLE H2
TOTAL ERRORS TOTAL ERRORS
12-4 14 12-7 17
12-8 o 12-5 o
I3-S 8 15-8 17
13-8 15 15-7 o
TT-6 TO II-9 15
11-5 O II-2 O
II-7 22 17-9 12
11-4 5 17-8 i
Here there is evidently a strong tendency for the smaller of two

amounts subtracted to yield a surprisingly greater degree of accu-
racy than its complement.
Now each of these numerous specialized arithmetical activities
has a psychology more or less of its own. Most of these processes
are as yet very imperfectly explored. The psychology of comple-
mentary processes just considered offers an excellent example.

Other important ones are concerned with borrowing, carrying,
and decimal points. Perhaps one of the most significant of all is a
peculiar type of attention, or memory span, such as is involved in
accurately keeping in mind and adding one to the tens at appro-
priate intervals while at the same time adding the units of a long
column of digits; or, in subtracting, in which it is necessary to re-
member when borrowing has taken place so as to make the appro-
priate compensation. Among the specialized aspects of the psy-
chology of arithmetic phenomenon of number preferences.
is the
The United States census reports show that when people are not
certain of their ages they tend strongly to give even numbers and
to avoid odd numbers. Phillips obtained introspections showing
that people like the even numbers but feel uncomfortable at the
thought of odd numbers. Jastrow in an unpublished investigation
had university students estimate the number of steps of two short
flights of stairs which all climbed nearly every day. The avoidance
of the odd numbers and particularly of the prime numbers is very
strikingly shown by Figure 75 where the results of both sets of judg-

ments are combined. The avoidance of a given odd number is
shown by comparing its frequency with the frequency of the num-
bers at each side of it. Eight is an especial favorite.
21
-5=514
3 4 5 6 7 8 9 10 11 12 Hi 14
Tvn n
15 16 17 18 19 20
Estimates
FIG. 75. The number of persons whose estimates of the number of steps
fallon various numbers showing a tendency to avoid odd and prime numbers.
1 2 8 4 5 6 7 8 9 10 11 12 13 14 .15 16 17]
Totals in Addition.
FIG. 76. Showing the relative numbers of addition combinations amounting
to the various totals. Adapted from Table 128.
ARITHMETIC 381
These number preferences are chiefly significant to education in

their genesis. Some indications as to their origin are incidentally
furnished by two studies to be described presently, the results of
which are represented graphically by. Figures 76 and 80. Figure
76 shows very clearly that there is a strong tendency for additions
of digits which result in totals of odd numbers and especially of
prune numbers, to be more difficult and thus to cause the learner

more trouble than even-numbered totals. From quite a different
source, namely, the difficulty of mastering number pictures (Fig-
ure 80), there comes evidence tending toward the same conclusion.
Indeed it is possible that early experiences of a disagreeable nature
involving these odd and prime numbers may have caused the
1
unpleasant feelings reported by Phillips subjects to be associated

with them, and that this unpleasant feeling- tone may have in-
hibited to a certain extent the choice of odd. numbers in Jastrow's
experiment according to the well-known tendency for the un-
pleasant to produce avoiding reactions. Moreover, the special
difficultyencountered with prime numbers suggests that the diffi-
culty of the various totals, apart from the factor of size, is closely
related in its turn to indirect assistance derived from multiplication.
THE MEASUREMENT OF EFFICIENCY IN ARITHMETICAL OPERATIONS
According to our analysis, the measurement of efficiency in arith-

metical operations consists fundamentally in a measurement of
the quickness and correctness with which the various operations
are suggested and carried out under different conditions. Two
types of measuring devices have been worked out. The first type
consists of the well-known Courtis ('10) tests. Courtis prepared at
the outset a series of eight tests now widely used, known as Series A,
to measure the following operations: Addition, subtraction, mul-
tiplication, division, copying figures, reasoning without performance
of the operations, fundamentals consisting of various combinations
of the four fundamental operations, and reasoning, including the
performance of the operations. The general principle of the tests

consists in the selection of units of the material in each test of
approximately the same type and difficulty throughout the test

and of measuring efficiency by the number of operations attempted
or done correctly in a given period of time. For example, the
addition test is composed of a large number of combinations of
single digits, such as 6 plus 3, 7 plus i, etc. The measure of
efficiency is the number of such additions made in one minute.

Each of the other tests is constructed on the same general princi-
ple. The reasoning tests consist of separate problems of approxi-
mately equal length and difficulty, and the measure of efficiency
is the number of problems attempted or solved correctly in a
given number of minutes.

More recently Courtis prepared another set, Series B, which is
confined to the four fundamental operations and contains conse-
quently one test for addition, subtraction, multiplication and
division. The chief difference between the tests in Series B and
the corresponding ones in Scries A is that the problems in the
former contain larger numbers arranged in larger columns so as to
introduce more complex elements.
The second general type of tests is constructed on the principle
of a problems of increasing difficulty. The author ('15)
scale of
prepared a test for measuring ability in solving concrete prob-

lems, designated as Arithmetical Scale A. It is composed of a
series of problems increasing in difficulty by determined steps of
difficulty ranging from o to 15. Ability is measured not in terms

of speed, but in terms of the highest point on the scale at which a
pupil can solve problems correctly. The difficulty of the problem

and the distances between the steps were determined by extensive
experiments with pupils.
Stone ( o8) prepared a set of problems whose difficulty was deter-
}
mined experimentally. The problems are, however, not arranged

in scale form.
Woody ('16) prepared a series of tests on the scale plan for

measuring ability chiefly in the fundamental operations. Each
scale is composed of a series of problems arranged in the order of
increasing difficulty.
Judd and Counts ('16) prepared for the survey of the Cleveland
schools, fifteen sets of arithmetic tests, four for addition, two for
subtraction, three for multiplication, four for division, and two
for operations with fractions. The various sets for any one type
of operation represented successively more and more complex
stages of the operation.
Standards of attainment have been prepared by the authors of
the various tests so that measurements of the abilities of pupils
and schools can be compared directly with the respective norms for
the various grades. These tests have been very useful in getting at
some real facts concerning progress and attainment in arithmetic.
ARITHMETIC 383
Figure 19, Chapter III, shows the xaiige of individual abilities

and the overlapping of abilities in the various grades in a certain
school as measured by the author's arithmetical scale. The facts
shown therein, although indicating enormous ranges of differences
and overlappings, are substantially the same as those found in
other subjects and hence are no longer surprising. Figure 20 ex-
hibits the same facts for addition as measured by the Courtis tests.
Grade
FIG. 77. Difference between boys and girls in solving arithmetical problems
as measured by the author s scale ('15).
Figure 77 shows the difference between the sexes and indicates

that in arithmetical reasoning the boys surpass the girls. This is
one of the few school abilities in which the boys are in the lead.
ECONOMICAL METHODS OF LEARNING ARITHMETICAL OPERATIONS
(i) The Acquisition of the Number Concept. The fundamental

psychological materials which the child must acquire are ideas of
numbers and quantities. He must learn what i, 2, 3, 4, etc., mean.
The practical question is, How may the pupil acquire these number
concepts most economically? A considerable amount of experi-
mental work has been done which bears either directly or indirectly
on the processes by which the child develops definite notions of
the significance of number and quantity. The investigations on
the span of attention or apprehension as measured in terms of the
number of objects that can be grasped simultaneously have chiefly
an indirect bearing. Such researches have been made by Cattell

('86),Messenger ('03), Burnett ('06), Dietze ('85), Warren ('97),
Nanu ('04), Freeman, and others. The general result of these
researches has been to show .that the average person has a span of
four, five or six objects or stimuli presented either simultaneously
or in rapid succession to the sense of vision, hearing or touch.
Investigations bearing more directly upon the development of
number concepts have been carried out by Lay ('98 and '07),
Walsemann, Knilling, and others.
Pestalozzi was probably the first to attack the problem by
devising for the purpose of school exercises his Strichtdbellen (stroke-
1
i ii in mi inn nun mini mum nniim iiimiiiT*
2 O 00 000 0000 00000 000000 0000000 00000000 000000000 0000000000
o oo oo oooooo oooooooo o o o n A [Born]
[Lay]
FIG. 78. Number pictures or arrangements. After Howeli (p. 154).
tables)which were the forerunners of more recent number pic-

tures. These have been devised in two general forms: (a) those
in which the dots are arranged in two parallel horizontal rows
(prepared by Born, Busse, and Behme) and (b) those in which the
dots are arranged vertically (prepared by Hentschd, Beetz, So-
belewsky, and Kaselitz). See Figure 78.
Extensive experiments have been made upon school children by
Lay to determine the relative merits of the various devices and
arrangements of the materials used for teaching apprehension of
the number of objects. Lay investigated such problems as whether
it is better for teaching numbers to present groups of stimuli or
objects simultaneously or successively; whether it is better to

present them in single rows or in double rows, in continuous rows
or in quadrats (groups of four); how many objects children can
ARITHMETIC 385
apprehend at one time; what the effect is of distance between the

objects; and so on.
The chief results have been summarized by Howell in the follow-
ing Table 113, and graph, Figure 79*
a? -Grade
?/-% Mistakes
1JAT
IB 2aT 2B 3AT 3B 4B 6AT
FIG. 79. Curve of error for the number pictures 5 to 12 as a whole. After
Howell ('14, p. 213^.
TABLE 113. After Howell
Comparison of the Born, Beetz, and Lay pictures with one another and with
the Russian Machine in the apprehension of the numbers 5 to 10.
P = chances for mistakes. M = number of mistakes.
The outcome of these experiments is that the quadratic pictures

are superior to other known pictures, particularly to the rows, and
that the Lay and Born pictures are practically equal in merit.
Howell continued experimentation with quadratic number
pictures similar to Lay's. The groups
of dots used on the pictures
ranged in size from three to twelve. Each card was exposed for
5 seconds. The investigation was carried out with pupils in an
elementary school. Howell's results are shown in the curve of
Figure 79. This curve shows a rapid drop in errors from the first
grade to the second, and that pupils reach a high degree of cer-
tainty between the third and fourth grade in apprehending the
numbers. Howell also found that certain numbers are not appro-
FIG. 80. Number of mistakes in perceiving numbers by pupils in various
grades. After Howell ( '14, p. 2 1 7) .
hended as accurately as others, the most difficult ones being seven

and eleven, as shown in Figure 80.
(2) The Operations to be Learned. One of the first important
problems in the economy of learning arithmetic, as in every school
subject, is the determination of the operations that pupils should
really master. Considerable attention has lately been given to
this question with the result that the recent texts and courses of
study have eliminated much material and have decidedly shifted

the emphasis from certain topics to others.
Two methods of attack may be employed in determining the
topics and the material to be learned in arithmetic. According to
the one method, we might obtain the consensus of many expert
opinions as to the topics that should be included and the relative
amount of emphasis on each. According to the other method, we
might proceed to gather a large number of the arithmetical prob-
lems and operations actually involved in the occupations and
ARITHMETIC 387
professions of all classes of people, and to determine on the basis

of such a collection, the sort of material that should be
taught.
Some results have been obtained by both methods.
80 40 60 60 80 90 100
Per cent,
FIG. 81. Percentage of superintendents in 830 cities who favor elimination
of the various topics represented by checked surface and those who favor "less
attention" arc represented by shaded surface. After Jessup and Coffman ('15).
Jessup ('15) and Coffman obtained an expression of opinion

from superintendents of 830 cities with a population of 4,000 or
over as to which topics should be eliminated, which should receive
less emphasis, and which should receive more emphasis than they
do at present. Their results are shown in Figure 81 and Figure 82.
108080406060708090100
Per cent
FIG. 82. Percentage of superintendents in 830 cities who favor giving more
attention to each of these topics. After Jessup and Coffman ( '15).
As a result of these returns Jessup and Coffman recommend the

elimination of the following topics from the elementary course
of study:
"Apothecaries' weight, alligation, aliquot parts, annual interest,

cube root, cases in percentage, compound and complex fractions of more
than two digits, compound proportion, dram, foreign money, folding
ARITHMETIC 389
paper, the long method of greatest common divisor, longitude and time,
least common multiple, metric system, progression, quarter in avoirdu-
pois table, reduction of more than two steps, troy weight, true discount,
unreal fractions."
Furthermore, they recommend greater attention to such topics

as:
"Time saved through the omission of the material mentioned in the

foregoing may be wisely devoted to the study of social, economic and
arithmetical issues involved in such facts as saving and loaning money,
taxation, public expenditure, banking, borrowing, building and loan
associations, investments, bonds and stocks, tax levies, insurance, profits,
public utilities, and the like."
G. M. Wilson ('17) collected 5,036problems from 1,457 persons,

representing practically varieties of occupations and profes-
all
sions. He then classified these problems according to the type of

operation involved, and the number of problems of each type as
shown in Table 114.
TABLE 114. After Wilson ('i 7)
Addition i-Pl 30 Accounts 251

2-PI 706 Addition of Fractions 3
3-P1 748 Amount 1 1
4-P1 193 Area i
Over 4-P1 65 Average Weight 14
Total 1742 Banking 18

Board Measure 12
Cancellation 26
Multiplication i-Pl 1660 Capacity 10
2-PI 904 CircularMeasure i
3-P1 195 Cubic Measure 56

4-P1 17 Debts 56
Over 4-P1 3 Decimals A
Total 2779 Discount 5

Division of Fractions i
Subtraction i-Pl 40 Dry Measure $

2-PI 407 Exchange 5
3-P1 406 Insurance 10
4 -Pl 167 Interest 66
Over 4-Pl 65 Liquid Measure 14
Total 1085
TABLE 114 Continued
Division . .i-Pl 334 Making Change 3

2-n 3i9 Measuring 21
3-Pl 121 Percentage 217
4-Pi 48 Plastering 2
Over. .
.4-Pl 5 Practical Measurement 79
Profit and Loss 16
Total. 39 Proportion 5
Receipts i
Square Measure 27
Fractions. .
2-5 534 Taxes 6
I
Time Measure 13
6-1 o Buying 3 1 .>8
Selling. .646
10 plus
i plus
86
i plus
5 Plus 60
Total .......... 1035

United States Money i-Pl 23
2-Pl 2982
3-Pl 1714
4-Pl 550
Over ........... 4-Pl 247
Total.
"The problems solved in actual usage are brief and simple. They
chiefly require the more fundamental and more easily mastered proc-
esses.
"
In actual usage, few problems of an abstract nature are encountered.
The problems are concrete and relate to business situations. They
require simple reasoning and a decision as to the processes to be em-
ployed.
"The study justifies careful consideration of the following question:
After the development of reasonable speed and accuracy in the funda-
mentals and the mastery of the simple and more useful arithmetical
processes, should the arithmetic work not be centered largely around
those problems which furnish the basis for much business information?"
(Wilson.)
ARITHMETIC 391
W. S. Monroe ('17) has compiled the problems in four text-

books and them according to the types of operation
classified
involved and then compared the frequency of these types of prob-
lems with the number of workers in, the different occupations.
His preliminary report states:
"In the first place, out of a total of 1,023 types of practical problems
found in four text-books, 720, or 71%, occur in occupational activities.
"A study of the frequency with which type problems occur reveals
a significant fact; viz., the frequency ranges from one to 434.
"This wide variation in frequency shows that the authors of our text-
books are far from being in agreement on the type problems of arithmetic.
Only one author out of four has recognized 511 out of 1,023 type problems
and 140 type problems have received the recognition of only two authors
out of four." (Monroe.)
(3) Length of the Class Period. J. M. Rice made an investiga-

tion of efficiency in arithmetic after the general plan of his investi-
gation of spelling. He tested some 6,000 pupils in eighteen dif-
ferent schools in seven cities. His results are exhibited in Table
115, which
"
Gives two averages for each grade as well as for each school as a
whole. Thus, the school at the top shows averages 80.0 and 83.1, and
the one at the bottom, 25.3 and 31.5. The first represents the percentage
of answers which were absolutely correct; the second shows what per
cent of the problems were correct in principle, i. e., the average that
would have been received if no mechanical errors had been made. The
difference represents the percentage of mechanical errors, which, I be-
lieve, in most instances, makes a surprisingly small appearance."
39* EDUCATIONAL PSYCHOLOGY
-
^QO
O M
OO 00 00
co ^N
t* t-
to to O COOO tooO
f ^\O VO*O^^t't>
M
'^COOO O
OtvO t>OO '
O
'
<O **
I
c,
g
* 6 6 o* * cood d* ** >d T- to to 6
-OO OO
t*
t>>vO t^ ^* to to to ^ ^fod
^*vQ ^4" *O co co
t^.
CM \o
tooO H ro tOO O tÔO tOtÔcMMOOw^to IOH
*
r
fl
:s :
:
--, R
. .
^K M W ^, .
i
.
^
s
^ f.-,
\O M O co !> OÔ w w fJO "-inÔ^ *OOO
> to vo M vO 00
N food od M vd o^ ^' M
to to c -*f o to '
! a
ARITHMETIC 393
ssssssssssssssssss
Q
So >O
<. M tO
^.vo co t
M co w M
co co * *> INO vo >o 10 o i
B
O O O to *
I
s \O M M M VOOO M \O O
-
2
a W ONO O O O" O O* O
PO *cO-<cic<o>-irociHi
MMNfO'^>-< r*^< M ^^'- ( - P1fO MMr *5
^^^^^î ^^^^^^^^^^^ 1
OOODOOOOOOOOOOOOOO
With reference to the factor of time in relation to efficiency in

arithmetic, Rice concludes thus:
"A glance at the figures vill us at once that there is no direct

tell
relation between time and does not neces-
result; that special pressure
sarily lead to success, and, conversely, that lack of pressure does not
necessarily mean failure.
"In the first it is interesting to note that the amount of time
place,
devoted to arithmetic in the school that obtained the lowest average
25% was practically the same as it was in the one where the highest
average 80% was obtained. In the former the regular time for arith-
metic in all the grades was forty-five minutes a day, but some additional
time was given. In the latter the time varied in the different, classes,
but averaged fifty-three minutes daily. This shows an extreme variation
in results under the same appropriation of time.
"Looking again toward the bottom of the list, we find three schools
with an average of 36%. In one of these, insufficient pressure might be
suggested as a reason for the unsatisfactory results, only thirty minutes
daily having been devoted to arithmetic. The second school, however,
gave forty-eight, while the third gave seventy-five. This certainly seems
to indicate that a radical defect in the quality of instruction can not be
offset by an increase in quantity,
"If we now turn our from the three schools just mentioned
attention
and direct to three near the top Schools 2, 3 and 4, City I we find
it
the conditions reversed; for while the two schools that gave forty-five
minutes made averages of 64% and 67%, respectively, the school that
gave only twenty-five minutes succeeded in obtaining an average of 69%.
This would appear to indicate that while, on the one hand, nothing is
gained by an increase of time where the instruction in arithmetic is
faulty, on the other hand, nothing is lost by a decrease of time, to a
certain point, where the schools are on the right path in teaching the
subject. Perhaps the most interesting feature of the table is the fact that
the school giving twenty-five minutes a day came out within two of the
top, while the school giving seventy-five minutes daily came out prac-
tically within one of the bottom."
Stone ('08) made a similar investigation, testing some 6,000

pupils in the 6th grade in twenty-six school systems. He reports
results practically identical with those of Rice, namely, that while
the amount of time devoted to arithmetic in different schools
varied from 7% to 22% of
the total school time, yet a comparison
of time expenditure with the efficiency attained showed, according
to his interpretation, that time plays a negligible part.
These results and inferences are interesting and valuable but
ARITHMETIC 395
they cannot be interpreted with absolute assurance. The various

factors cooperating or counteracting are so intricate that a more
careful isolation of the effect of the time element is necessary. In
general, the same criticism made in connection with Rice's and
Cornman's investigations of spelling applies here. Dependable
conclusions could be reached only by an experimental procedure
similar to the one there suggested.
The findings of Rice and Stone probably represent correctly
the situations in the schools examined. A possible explanation of
the fact that the schools giving more time to arithmetic did not
obtain on the whole any higher efficiency than those devoting
less time to it may, perhaps, be sought in the likelihood that the
schools giving longer periods of time may not have worked as
intensively and used their time to as good advantage as the schools
devoting less time.
(4) The Effect of Various Environmental Factors. Both Rice

and Stone massed their results with reference to ascertaining the
effect of such factors as the home environment of the pupils, size
of classes, age of pupils, the time of day of the test, amount of
home-work required of the pupils, method of teaching, teaching
ability, the course of study, the superintendent's training of the
teachers, etc. Rice reports that none of the factors had any influ-
ential part in producing efficiency in arithmetic. The results are
open to the same criticism of complication of factors as were pointed

out previously. It seems quite improbable that these elements
played no part. It is rather a question of more rigorous isolation of
the effect of different factors. Stone, for example, found that the
correlation of excellence in the course of study, as rated by judges,
with efficiency in arithmetical reasoning was .56, and with effi-
ciency in fundamentals .13.
That environmental factors, and perhaps particularly method
and spirit of teaching, do make important differences in the attain-
ments of pupils is shown clearly in such results as those exhibited
in Table 116 which gives the distribution of class averages of the
grades in sixteen different schools as measured by the author's

Arithmetical Scale A.
TABLE 116
Average scores attained in various schools as measured by Arithmetical Scale A

(Starch)
Thus we note that the best eighth grade attained an average of

14.5 as compared with the poorest one which attained an average
of only 9.7. Such differences would not be surprising if they
were the scores of individual pupils. They are, however, the mean
scores of whole classes. It is quite unlikely that the hereditary
differences of the groups as wholes differ so much from one another.
It seems quite probable that the environmental circumstances,
and chief among them the teacher and the attitude of the learner,
were mainly responsible for the ultimate differences in achieve-
ment.
Similar results have been reported by Judd for the fifth and
eighth grades in ninety schools in Cleveland, as measured by his
Test A in simple addition, Figure 83. The best fifth grade made
an average score nearly three times as high as the poorest fifth
grade, and the best eighth grade made a score nearly twice as
high as the poorest eighth grade.
(5) Drill in Fundamental Operations. Various methods of
drill in the fundamental operations have been devised.
Studebaker, Assistant Superintendent of Schools at Des Moines,

has prepared a series of drill cards. The various combinations of
numbers in fundamental operations are given on one side of the
ARITHMETIC 397
card. Below each example there is an opening through the card

in which the pupil may write his answer on the sheet of paper
placed underneath the card. The pupil works as rapidly as he can
within a certain limit of time. Then he turns the card over and
places it again over the sheet of paper containing his answers so
FIG. 83. Median scores of the $th grades and of the 8th grades in 90 schools
in simple addition. After Judd ( 'i 6, p. 1 1 2) .
that they can be seen in the openings of the card and compared
with the correct answers printed on that side of the card.
This plan of drill work has a number of advantages, such as an
incentive to rapid and accurate work, immediate self-checking ot
the answers* and so on.
A considerable number of careful experimental studies on the

influence of drill are now available and without exception they
show drill to be distinctly valuable. Thorndike had nineteen
university students practice adding 48 lo-digit columns of figures
daily for seven days. While the work required on the average less
than an hour in all, there was an improvement of 29% over the
original rate.
J. C. Brown performed two elaborate comparative experiments
to determine whether children under controlled school conditions
more by giving a small part of each
profit class period to drill or
by spending the entire period in ordinary routine work in arith-

metic. In each experiment the children were first tested with the
Stone Arithmetic tests and then divided into two groups of equal
ability on the basis of their performance in the tests. One group
was given the special drill as a part of the regular class work while
the other did the class work as usual. At the conclusion of the
drill,both were tested again by the Stone tests to see which had
made the greater gain. In the first experiment 51 children from
the sixth, seventh, and eighth grades were used. They averaged
thirteen and one-half years of age. Drill on the four fundamental
operations was given to one-half of the group for the first five
minutes of each class period of twenty-five minutes. About half
the drill was oral and half was written. The drill lasted thirty
periods. In the second experiment 222 children were used and the
drill was given for twenty periods. The results of the two experi-
ments are given in parallel columns in Table 117. In each case
section I received the drill and section II received the regular class
work. The pupils did not know that any experiment was in
progress.
ARITHMETIC 399
TABLE 117
In both the experiments there was a decided advantage in using

a part of the recitation period for drill. In the first experiment,
the drilled group gained about twice as much as the undrilled
group; while in the second experiment the drilled group improved
about two and one-half times as much as the undrilled group. The
sixth grade gained the most (35%) and the eighth the least (13.8%).
In order to determine whether group I had gained on fundamentals
at the expense of reasoning, both groups were tested in arithmet-
ical reasoning before and after the drill. Here again the drilled
group did better, making a gain of 6.3%, while the undrilled group
gained only 3.0%. This last factor is interesting in the light of
the small amount of connection between fundamentals and arith-
metical reasoning pointed out above as well as the small amount
of transfer of one arithmetical process to another (Chapter XIV).
Since the improvement in reasoning, which had not been drilled
at all was almost exactly the same proportionately as the processes
which were drilled, it suggests that the drill had a tonic effect
upon the remainder of each recitation period following the drill, to
which much of its value was due.
In order to discover the permanency of the effects of drill, Brown
tested both groups once more after a twelve-weeks vacation and
found that the drilled group was also superior in retention, having
lost.2% while the undrillcd group had lost 2.29%.
The same experiment was repeated by F. M. Phillips. He had
69 children for subjects and gave to one group drill in fundamentals
and in reasoning, both oral and written, for eight weeks. Neither
teachers nor students knew the purpose of the tests. He found
that, "The improvement in fundamentals of the combined drill,
groups was 15% greater than that of the non-drill groups. Inj
reasoning, the drill groups improved 50% more than the non-drill
groups. . . . The greatest gains were made in the sixth grade and
the least in the eighth." Almost all the gain on fundamentals
was in multiplication.
Mary A. Kerr under the direction of Haggerty reported an

experiment carried on for six weeks at Bloomington, Indiana, on
the effects of five minutes of drill in addition at the beginning of
each class period. The drill was begun by adding five three-place
numbers per column, which were gradually increased to nine three-
place numbers per column. Four hundred and twenty-three chil-
dren took the drill. Table n8 shows the average performance on
the Courtis tests, Series B, before the drill began and at its con-
clusion in June and, for comparison, the May scores of the best
twenty Indiana cities for the previous year.
TABLE 118
The decided advance in each grade and the great superiority to

the best twenty Indiana cities in each test bears eloquent testimony
to the value of drill in addition fundamentals.
Supt. Herman Wimmer of Rochelle, Illinois, conducted a series
of comparative experiments on the effects of drill in arithmetic
ARITHMETIC 401
under various conditions. Each experiment lasted six weeks. The

time spent in drill was in all cases subtracted from the regular
class time in arithmetic. The Courtis tests, Series A, were given
before and after each experiment. In-Experiment I, the drill group
was rather miscellaneous, grade 5 being drilled five minutes per
day (probably in fundamentals, though it is impossible to tell
from Wimmer's account), grade 6E five minutes per day, three-
fifths on reasoning and two-fifths on fundamentals, grade 6W
fifteen minutes in a single period per week on reasoning and funda-
mentals in the same ratio as 6E. In Experiment II, two sections
of the sixth grade, which had equal ability as shown by tests, were
drilled five minutes daily, one for speed and the other for accuracy.
In Experiment III, the seventh grade was drilled five minutes
daily on reasoning, while the eighth grade was drilled five minutes
daily on fundamentals. The results of all three experiments are
shown by percentages of gain in Table 119. As in the previous
experiments, drill as such is shown to have a decided value though
much more for reasoning than for fundamentals, the advantage of
the drilled groups being 32.7% and 4.4%, respectively. In Ex-
periment II, drill for speed is seen to have a distinct advantage
over drill for accuracy. Here the gain is considerably more in the
fundamentals than in reasoning. In Experiment III, we find that
the class drilled on reasoning gained very largely in reasoning
alone, while the other class trained in fundamentals gained almost
exclusively in fundamentals. This presents a contrast to Brown's
experiment where, it will be remembered, drill in fundamentals
showed as much gain in reasoning as in fundamentals themselves.
TABLE 119
Wimmer's results
(6) The Optimum Distribution of Drill. A number of experi-

ments have been conducted to determine the most economical
distribution of time, that is, the most economical duration of drill
periods in arithmetic. T. J. Kirby carried out such experiments
on a large scale for both addition and division. He used special
blanks for the practice in both experiments. In addition funda-
mentals there was a beginning and a final test each of fifteen
minutes. Between these periods forty-five minutes of drill were
variously distributed. The following table shows the distribution
of the periods, the median initial ability of each group in examples
correct, and the gross gains as measured by three different methods
of calculation. Seven hundred and thirty-two fourth-grade children
were used as subjects.
TABLE 120. After Kirby

ARITHMETIC 403
When accurate correction had been made for differences in

initial ability, thegains were then in proportion 100, 121, 101, and
146^ respectively. There was no very distinct tendency observ-
able here except that the short periods of two minutes yielded a
distinct advantage over the rest. Unfortunately for the consist-
ency of these results the next shortest period (6 min.) yielded
nearly the least gain of all.
The same general plan was followed by Kirby in the experiment

on division. Six hundred and six children from the second half
of grade three and the first half of grade four were used. The
following table shows in detail the various distributions of the
drill together with the results by each method.
TABLE 121. After Kirby
When inequalities of initial ability had been removed, the gains

were found to be in the proportion of 100, no>, and 177. Thus
we find here very consistent and decided advantage in favor of the
shorter drill periods. Unfortunately in each of these experiments
it is impossible to tell how much of the gain in the shorter practice
periods was due to spontaneous practice outside the class. It is

needless to say, that the children were not permitted to take any
of the practice cards away from the class.
Kirby's experiments as a whole both in addition and division
showed great improvement. Addition with a practice period of
60 minutes yielded an improvement of 48%, while division with
a practice period of 50 minutes yielded an improvement of 75%.
Accuracy was not disturbed in addition, but in division it improved
2.6%.
Kirby also investigated the permanence of the improvement
resulting from the drill. He found that from June to September
fourth-grade children lost 17% of the ability possessed in June and
required 58% as much time to regain the efficiency which they pos-
sessed the preceding year. In division there was a loss of 21%.
It required 60% as much time to recover the preceding year's

efficiency.
Hahn and Thorndike repeated the addition part of Kirby's
experiment. Each grade was divided into two sections, section B
receiving periods exactly half as long as section A. All received
a total of 90 minutes of drill, which was preceded and followed by
a i5-minute test as in Kirby's investigation. Table 122 shows
the distribution of drill for the various groups, the initial ability,
the gain, and the advantage of group A according to two different
methods of scoring the results.
TABLE 122. After Hahn and Thorndike
There is no clearly defined advantage for cither the long or the

short period as was the case with Kirby's experiment on drill in
adding. What little tendency there is, however, is in favor of the
longer periods.
Superintendent Wimmer, in connection with his drill experiments
previously reported, also investigated the problem of economy of
long and short periods of practice. The drill, three-fifths on rea-
soning and two-fifths on fundamentals, was given to one group
for five minutes at the beginning of each class period. The other
group received one 15-minute period per week. Drill lasted for
six weeks. The results are shown in the following table. They
are distinctly in favor of the longer drill period despite the fact that
only three-fifths as much time was spent by this method.
ARITHMETIC 405
TABLE 123
After Wimmer
PROBLKM FAVORABLE DISTRIBUTION OF TIME FOR DRILL
This is particularly the case with reasoning. The rather slight

advantage of the long periods on fundamentals in connection with
the uncertain indications of Hahn and Thorndike's results and
the opposite finding of Kirby suggests that there is little or no
advantage in the distribution of time in arithmetical drill on fun-
damentals but that the longer periods are more favorable for drill
in reasoning.
(?) Special or Economical Methods of Drill. A
number of
special methods of giving drill in arithmetic have been advocated
and used. That of Studebaker has already been noticed. Courtis
has also published practice pads for drill purposes. Flora Wilbur
undertook to determine experimentally the value of this kind of
drill at the Fort Wayne, Indiana, training school. Two classes
of 14 children each were divided each into two groups of equal
ability on the basis of the Courtis tests, Series B. One section
of the fifth grade received four and one-half minutes of drill with
the pads at the beginning of each class period, and one section
of the sixth grade received similar drill for four minutes. The
remaining sections received the regular class work. The experi-
ment lasted from September to May. The results are shown in
percentages of gain in the following table. The drill was clearly
of value in both grades and in all four processes.
4 o6 EDUCATIONAL PSYCHOLOGY
TABLE 124. After Wilbur
Divisionand multiplication profited most and addition least,

as usual in such experiments. The real question remains, how-
is
ever: Is drill with the Courtis practice pad more or less efficient
than drill as ordinarily given?
Kirkpatrick performed two comparative experiments to de-

termine the relative economy of various methods of memorizing
multiplication tables. His subjects were twenty normal school
men, divided into groups of equal ability. As these subjects knew
the ordinary tables, he had them learn the products of 7 multiplied
by all the prime numbers between 17 and 53. One group simply
memorized the table by rote for the first five days, then spent the
periods of the next five days in writing down the answers on a
blank with a card containing the table before them for reference.
The other section spent the periods of all ten days working on
the blank with the table for reference. The time consumed during
the five days of memorizing was about an hour. A test at the end
of the experiment showed that, in a period of two minutes, the
practiced group put down 46.2 answers while the memorizing

group put down 40.9, thus showing a distinct advantage for the
practiced group.
A second experiment was performed with two groups of normal
school students of equal ability, twenty-five in each group. One
group practiced with the keys and blanks as in the first experiment
while the other group spent the same amount of time multiplying
out the products as needed. The experiment extended over eight
ARITHMETIC 407
days. When tested at the end of the experiment, those, using the
key put down 25.4 answers in two minutes, while the computers
wrote down 44.3 examples, showing a decided advantage for the
computation method. He concludes that tables should be learned
by use rather than by memorizing.
Conrad and Alps ('16) investigated the effect of suppressing artic-
ulatory movements upon the effect of drill in rapid adding. They
divided sixty-four high school students into equal groups of equal
ability. The students were then given eight periods of drill in
rapid addition of columns. The pupils in one group were per-
mitted to add in their ordinary way which involved a great deal
of articulation or inner speech. The other group was cautioned
repeatedly and emphatically to "think results only." The former
was called the traditional method and the latter the economical
method. The percentages of gain by the two methods were as
follows:
The traditional method gained in attempts 8.5% and in rights
2 -5% The economical method gained in attempts 34.4% and in
rights 30.9%. This gave an advantage in favor of the economical
method of 25% in attempts and of 33.4% in rights. These results
came out almost startlingly in favor of "thinking results only."
The evil effects of articulation and lip movements have been no-
ticed in connection with reading (page 287). It is probable that
the cause is the same in both cases.
P. B. Ballard investigated the comparative efficiency of the
"equal addition" method and the "decomposition" method in
subtraction.
"In the equal addition method the compensation is made accounts

are squared at the very first number dealt with after the minuend has
been disturbed. In subtracting 37 from 85, after taking 7 from 15 the
disturbed relationship of difference between minuend and subtrahend is
immediately restored by increasing the 3 tens to 4 tens. In the method
of decomposition, however, it is the 8, the second figure dealt with, that
has to be changed to restore the balance. If the minuend figure is zero,
the balancing of accounts is still longer deferred."
Ballard gave tests in fundamentals to 71 English schools of

which 23 had been taught subtraction by the method of equal
addition while the rest had been taught it by the method of de-
composition. While there was little difference in the average ability
of the two groups in the other three fundamental operations,
there was a very striking superiority in the score for subtraction

of the equal addition group. At 13 years of age it amounted
to over 10% and at earlier ages it amounted to over 40%. (See
Figure 84.) Inasmuch as' the decomposition, or less efficient
method, isthe one in general use in this country, it is evident
that this matter deserves careful attention.
oo
00
40
10
84 10i 10i Hi 11} 124 12* 134 13*

Ages
Flo. 84. Showing superiority of teaching subtraction by the "equal addi-
tions" method. After Ballard ('15).
Mead and Sears performed two experiments in comparative

economy of in arithmetic. The first was to compare the
methods
efficiency of the ordinary "take away" method of subtraction
which involves the learning of an entire subtraction table, as
compared with "addition" subtraction which permits the use of
the addition table, thus saving the learning of an entire table.
In the first the formula is "8 minus 2 equals what?", in the second,
"2 plus what equals 8?" Two second-grade classes of approxi-
mately equal median ability as indicated by the Courtis tests were
ARITHMETIC 409
taught subtraction by the respective methods thirty minutes per

day for four months, all other factors being equalized as fully as pos-
sible. Tests given periodically throughout the experiment showed
that at the outset the addition methodl was superior but the "take
away" method gradually overtook it end of the four
until at the
months the "take away" method was superior by 4.5 points which
was nearly one-third of the final median score made by the addition
group. This difference disappeared, however, when both groups
were tested on longer examples.
The second experiment was to compare the multiplicative
method of division. The formula of the first is, "Five into twenty
how many times?", that of the latter, "Five times what equals
twenty?" Two third-grade classes were used as subjects in this
experiment. Other conditions were similar to the experiment on
addition. In this experiment, the final test on combinations
revealed the fact that the multiplicative class stood 4.3 points
above the "into" class, which was about one-fifth of the final
score of the "into" class. This difference disappeared, however,
when the class was tested on longer examples just as that in ad-
dition noted above.
J. A. Drushel investigated the relative efficiency of two methods
of determining the position of the decimal point of the quotient in
the division of decimals. The rule of method A, the older one, is:
"There are as many places in the quotient as those in the dividend
exceed those in the divisor."The rule of method B, the newer of
the two, is: "First render the divisor an integer by multiplying
both dividend and divisor by 10 or some power of ten. Then
proceed as with integral divisors." A
short test in division of
decimals was given to 576 freshmen at Harris Teachers' College.
Of these, 507 had studied division of decimals by method A, while
69 had studied it by method B. The results show that the stu-
dents taught by method A had the very low accuracy of 66% in
placing the points, while those taught by method B had the very
excellent accuracy of99%. If future investigations confirm these
method B should be generally adopted.
results,
(8) Speed vs. Accuracy. Thorndike investigated the relation
between speed and accuracy in simple addition. Six hundred and
seventy-one students were tested apparently on two different occa-
sions in a class experiment in adding columns of nine digits. The
subjects were then grouped according to speed as shown by the
following table:
4io EDUCATIONAL PSYCHOLOGY
TABLE 125. After Thorndike ('15)
With a decrease in the rate of additions there is a steady increase

in the number of errors per 1,000 additions. Thorndike concludes
"
that the sort of individual who is quick in adding is more accurate
also than the one who is slow."
(9) Limits of Attainment. Since there is general agreement
that the fundamental number combinations should become auto-
matic association processes, it is pertinent to ask, how high a
degree of skill should be developed in pupils? This question is
similar to the one discussed in connection with quality of hand-
writing. It is reasonable to maintain that it is probably uneco-
nomical to attempt to develop a degree of speed beyond a certain
point.
How great speed is practically necessary or worth while? Obvi-
ously the school may devote relatively too much attention to the
development of speed in the four fundamental operations at too
great a cost of time. It would, therefore, be important to know
what degree of proficiency is needed for the practical affairs of life.
An investigation of the problem is needed.
(10) Errors. The detection and classification of errors and the
discovery of the frequency with which they occur are highly useful
facts in any school subject because such information will help to
make instruction specific. It will indicate the particular points
at whichdrill should be directed.
Howell made an analysis of the mistakes in division occurring
in the Courtis tests as applied to the pupils in his school. He
found the following rubrics of errors:
ARITHMETIC 4U
" i.
Making the quotient figure the same as the divisor,
(a) When a difference of only one exists between the divisor and
quotient;
(b) When the quotient is commonly used as the divisor of the
given dividend.
"
2. Making some factor (other than the divisor),
commonly used as
the divisor of a given dividend, the quotient figure.
"3. When dividing a digit by itself, making the quotient figure the
same.
"4. When dividing a digit by itself, making the quotient figure zero.
11
5. When dividing by one, making the quotient one.
"6. When the dividend is zero, making the quotient the same as the
divisor.
"7. Pupils whose associations are as yet feeble or become so through

fatigue or distraction arc commonly observed to resort to running up the
table.
"They frequently miss count and get a quotient figure one removed
(say) from the right one.
"8. Making one of the quotient figures the quotient.
"9. Substituting multiplication for division.
"10. Unclassified."
The frequency with which the different types of errors occurred

is shown in Table 126.
TABLE 126. After Howeli
Showing the number of mistakes in the division tables falling into each class
GRADE KINDS OF MISTAKES TOTAL MISTAKES
From this table it appears that certain types of error occur

much more frequently than others. For example, Error No. 4,
"when dividing a digit by itself, making the quotient figure zero,"
occurs 181 times; whereas Error Nos. 3 and 9 occur only 12 times
each. More extensive studies of this sort are much needed. Learn-
ing in any school subject is apt to be more economical the more
specifically the learners' attention may be directed to the particular

processes to be exercised.
A. S. Gist tabulated the errors in subtraction, multiplication,
and division of 812 arithmetic test papers from six schools in
Seattle. Table 127 shows the percentage of different types of errors
separately for each of the three processes and for each of the grades
from 4 to 8. It is noticeable that the greatest difficulty was pre-
sented in subtraction by borrowing, in multiplication by the
tables and in division by the remainder. For the most part the
proportions of the various errors are fairly constant from year to

year.
Besides the prevalence of the general type of errors noted above,
the relative difficulty of the various elementary combinations is
a matter of much practical importance. C. L. Phelps attempted
to determine this for the addition combinations by finding the
percentage of errors made on each of the 55 addition combinations
of the Courtis tests in 5,950 papers made by repeatedly testing
238 eighth-grade children. These results are shown in detail in
Table 128. There is a clear relation between the per cent of errors
and the size of the total resulting from the combination. The
writer computed the correlation between the two by Spearman's
rank method and found a coefficient of .57. The relation is shown
in detail by the curve of Figure 76. For the most part the odd
and especially the prime numbers are more difficult than the even
numbers adjacent to them. While there is seen to be a fairly
steady rise up to 10, there is an abrupt increase at the beginning
of the teens which continues as far as the investigation extends.
TABLE 127
The per cent of each type of error in the examples of subtraction, multiplica-
tion, and division respectively from the 812 papers from six schools in
Seattle. (After Gist ('17).)
GRADE 4TH STH 6tH 7TH STB
Subtraction:
Borrowing 54 56 52 51 55
Combinations 36 38 45 44 41
Omissions 2 i 2 3 i
Reversions i 2 tf o o
7 o, o, etc 5 3 }A o o
Left-hand digit , o o o o 2
ARITHMETIC 413
TABLE 127 Continued
4TH STH 6TH 7TH 8TH
Multiplication:
Tables 79 73 73 77 75
Addition 18 20 22 19 20
Cipher in multiplier 1.5 6 5 4 5
Division:
Remainder too large 34 39 27 19 10
Multiplication 22 15 19 37 33
Subtraction TT 14 18 25 23
Last remainder o, and o in dividend ... 7 15 19 7 n
Multiplicand larger than dividend 7 4 i r r
Failure to bring down all of dividend 7 4 3 o 6

Failure to bring down correct digit .... 2 r 4 4 6
Failure to place all
Cipher in quotient, as
of quotient in quo
90898
. .
37847
7 i i 3 3
A somewhat similar investigation was carried out by Holloway.

He tabulated the number of errors made on each of the addition
combinations by 1,065 third-grade children. His results are given
in Table 128 in the order of the number of errors, parallel to those
of Phelps. While there is considerable agreement between the two
studies as to the relative difficulty of the various combinations,
there are also rather striking disagreements. How much these
differences are due to the stress which had been put upon the
various combinations in the teaching of the children cannot be
determined. The latter defect is much less likely in Holloway's
results because of his much wider range of subjects.
Holloway also tabulated the errors in the multiplication com-
binations in the test papers of 1,215 third-grade children. They
'
are given in detail in Table 129.

TABLE 128
Table showing the relative difficulty of the various addition combinations for
the third and eighth grades respectively
HOLLOWAY PHELPS
19 .62
19 -45
18 .67
17 .54
17 2.22
17 .27
16 .50
15 1. 1 6
15 -72
15 i 30
13 -86
13 i 09
13 .88
9 .07
9 -37
8 .12
8 1.46
.62
52
39
.28
25
24
.24
-IS
.14
.05
ARITHMETIC 41$
TABLE 129
Table showing the order of difficulty as determined by the number of errors

of 1,215 third grade children at end of year. After Phelps.
11 X ii .735 11X9 '181 5X4 55

12 X ii .. 655 7X5 lSl 6X2 . .
50
11 X 10 638.. 9X3... .169 5 X 3- 46
12 X 10 542 9X5.. 168 ii X 2. . .
46
12 X8 460 . ii X 8 .
167 i X i. .
41
9 X 7 455 8 X 3 -I5I 9 X 2. .
39
12 X 7--- 438 ii X 6... 144 10X3-. 38
8 X7 435 . 6 X 5 138 7 X 2 .
38
12X12... 425 11X7-.-. 137 5X5- 34
9 X8.... 422 8X5- --I37 4X2 32
12 X9-- 417 6X4-.. 133 10X4- 3i
9X 6 390 6X6 129 10X2.. 31
8X8.. 361 11X5 "3 HXI..-3 1
12 X 6. .. 361 6 X 3 .. 102 4 X i. 31 . .
8 X 63... 42 ii X 3--.. 99 3 X i.. 28

9 X 4- 292 10 X 9- 94 5 X 2. 26 . .
7 X 6.... 285 10 X 7 86 3 X 3--- 25

12 X 5----27I ioX8....8s 9X1. ...22
7 X 7 268 12 X2 81 3 X 2. 21 . .
9 X 9. .-
263 10 X6 79 9X1 21
12 X 4 ...250 4 X 4-..- 78 6 X i... 21
10 X 10.. ..241 4X3-- --76 12 XL. ..20
8 X 4- ---235 7 X3-." 7i 5 X i... 20
7X 4.... 192 10X5 5S 2XL... 20
12 X 3----i*3 8 X 2 "" S8 2X2.. .18
8 XL 18
10 X i 12
CHAPTER XXI
HISTORY
PSYCHOLOGICAL PROCESSES IN LEARNING HISTORY

It more difficult, at least more uncertain, to make an analysis
is
of the psychological processes concerned in the learning of history

than it is in case of most of the subjects treated thus far, for the
reason that teachers are not as fully agreed as to what is to be
learned in history. In general there are two extreme views: One
would hold that the learning of history means the learning of the
main facts names, dates and events of the human race; while
the other would hold that history means the acquisition of ability
to interpret the significance of human events. In their most ex-
treme forms the two views would be a memorizing of isolated
facts versus an interpretation of facts with little emphasis on
facts. Probably no one holds either of these extreme positions and
the distinction is useful only for analytical purposes since the
psychological processes involved in learning history would be
quite different in these two modes of approaching the subject.
Those who stress the interpretational aspect of history would
stress the conception that the purpose of history in the public
schools is training for citizenship or the development of patriotism.
Let us assume for our present analysis that the obvious aim of
what is to be learned or acquired in history is facts of the events
of human beings and the connections among these facts a view
to which probably every one could agree. What are the psycho-
logical functions concerned in acquiring and connecting historical

facts? Let us take as a concrete instance a given historical event
and see what mental operations are necessary to grasp, interpret
and remember it. Let us take the statement that Columbus dis-
covered America in 1492. The psychological processes involved
or assumed in learning and grasping this statement would be
substantially as follows: To begin with, it would involve all the
steps enumerated in the analysis of the reading process, or of the
process involved in understanding spoken language when the facts
are heard instead of read, since practically all history is learned
416
HISTORY 417
through reading. The chief difference would be in the emphasis

and elaboration of some of the steps. Taking the factors as enu-
merated in Chapter XVI, the main difference between ordinary
reading and learning history would be in step (6): "The establish-
ing or arousal of association processes whereby the incoming im-
pulses are interpreted." These specific processes of association
and interpretation over and above ordinary reading, necessary for
grasping a historical statement are as follows:
(1) A mental picturing or conceiving of the persons, actions,
localities and objects, concerned in the event.
(2) A mental picturing or conceiving of points and locations in

time.
(3) Processes (i) and (2) applied to events preceding the one in
question.
(4) Processes (i) and (2) applied to events succeeding the one
in question.
(5) A judgment concerning the internal motives of the persons

and the external conditions that led to the event.
(6) A judgment concerning the effect of the event upon the
motives of the persons and environmental conditions involved in
succeeding events.
(7) A remembering of the mental processes in steps (i) to (6),
in so far as a permanent memory of them is considered important.
Steps (i) to (4) use primarily the imagination, steps (5) and (6)
judgment and reasoning, and step (7) memory.
Thus the grasping, interpreting, and remembering of the state-
ment that Columbus discovered America in 1492 would mean, (i)
An imagining or conceiving of Columbus as a person, his asso-
ciates, ships, water, the voyage and landing on new soil; (2) an
imagining or conceiving of the time so as to give a notion of how

long ago 1492 was; (3) and (4) a similar procedure with events
coming before and after this particular one, such as Columbus's
interview at the court of Queen Isabella of Spain, the need for
another route to India, succeeding voyages, settlement of the new
country, and the like; (5) and (6) judgments concerning the effect
of the preceding events in bringing about the particular event
under consideration, and judgments concerning the effect of the
latter in bringing about later events; and (7) a repetition of the
learning of the fact with its connections to fix it in mind.

(
MEASUREMENT or ATTAINMENT IN HISTORY

The difficulties that beset any endeavor to devise an objective
and generally acceptable method of measuring attainment in
history are very great for the reason that historians differ very
widely in the selection, emphasis and interpretation of facts and
in the manner of stating the facts; and also for the reason that
teachers as well as texts differ in the relative emphasis on the learn-
ing and remembering of facts as compared with their interpreta-
tion. The author has prepared a plan with the aim of meeting
these difficulties as far as possible so as to secure a test that could
be used fairly wherever American history is taught; it is necessary
to exercise much care in selecting the right material for testing
purposes. The scheme finally carried out was as follows: Five
widely used text-books in American History were carefully com-
pared and all facts and interpretations given in all five were selected
and formulated into statements or sentences. This gave a total
of 278 statements a remarkably small body of facts common to
five texts. These statements were then made into a mutilated
text or completion test. Certain important words or phrases were
omitted which are to be supplied by the pupils doing the test. The
entire 278 statements would be too long as a single test. Hence
they were split up into four parallel sets, each containing 69 or 70

statements, by taking for the first set, statements numbers (i),
(5), etc.; forthe second set, numbers (2), (6), etc.; these four tests
may be used interchangeably at different times in testing a class.
Direct comparisons and measurements of progress can thereby
be made. The score of a pupil is the number of omitted parts
correctly supplied. The following statements serve to illustrate
the nature of the resulting test. They are the first ten statements
of the first test.
-discovered America in 1492.

2. John Cabot exploring for the in 1497 landed
on the coast and claimed the country for
3. around the globe in 1519-1521.

sailed
4. discovered the Mississippi River in 1541.
5. Two expeditions sent out by to settle Vir-
ginia IA ^585 and 1587 respectively, failed.
6. was governor of Virginia after Delaware
left.
HISTORY 419
7. in service of the Dutch East India Com-

pany, explored the river in 1609.
8. was Governor of the Dominion of New Eng-
land, which was composed of (i) -
(2)
9. John Winthrop came to America in 1630 and settled
10. New Hampshire was founded in-
The following are the average scores for the ends of the different
grades obtained from approximately 2,000 pupils:
Grade 6 7 8 H. S.
Scores 7 20 38 38
Individual differences and overlapping of successive grades as

shown in Figure 22 are enormously wide. In a certain eighth grade
composed of thirty-six pupils, the best pupil made a score of 102

and the poorest a score of 4. The differences among various schools
are indicated in Table 130. Very wide differences exist, which are
probably due chiefly to differences in methods of teaching. Thus
the best eighth grade made a score of 66 and the lowest one a
score of 19. Even in the same school system the differences among
schools may be very wide. For example, in city N, the best eighth
grade averaged 52 and the poorest 19. Another striking observa-
tion is the fact that the average attainment of the high school
pupils isno better than that of the eighth grade. Apparently the
pupils relearn in the high-school course in American history about
as much as they forgot during the intervening two or three years
since they left the eighth grade. This does not necessarily mean
that American history in the high school is useless since the re-
learning will help to guard against further loss. More extensive
tests are needed on this point.
Striking sex differences in knowledge of history have been re-

vealed by such a test as the one here described. The median scores
for boys and girls in the writer's test were as follows:
BOYS GIRLS
NUMBER MEDIAN SCORE NUMBER MEDIAN SCORE
High School 47 4i 73 3$
8th grade 288 45 352 31
7th grade 94 24 101 J7
Bell and McCollum applied a history test to 1,500 pupils and found
similar differences. The boys
in the elementary schools did 28%
better and in the high schools better than the girls. This
31%
superiority on the part of the boys may be due to their greater
interest in battles which in turn may be due to the greater strength
of the fighting instinct.
TABLE 130
Scores in the writer's American History Test, Series A

GRADE 7 s II. S.
City A 66 35
"
B 30
"
C 3; 4S
"
I), School i 22
* 4
2 59
"
"3 42
K 17 33
"
*' - 40 -
65
-
40
48
33
39
-
41
29
33
19
~ -,,
3 52
"
4 ... 17 50
"5 -'9
ECONOMIC METHODS IN THE LEARNING AND TEACHING OF HISTORY
Experimental work that has been done up to the present time in

the psychology and pedagogy of school subjects has been confined
almost entirely to the subjects thus far considered. Yet the prob-
lems and factors entering into such a subject as history are exceed-
much worth while and for the most part as
ingly intricate arid as
capable of experimental determination as most of the problems in
the other subjects. The discussion of factors and conditions
affecting the most economic procedure in history will, therefore,
HISTORY 421
have to be limited to a few suggestions and beginnings in experi-

mental work. The problem for the future will consist in determin-
ing the factors which promote or retard the elements enumerated
in the first section of this chapter. Economic procedure in learning
history resolves itself into discovering the most favorable means
of, and a measurement of their actual effects in, grasping, imagin-
ing, judging and remembering the important events of the human
race.
(1) By what means may the imagining or conceiving of a given

event be brought about most effectively? Numerous devices are
employed to assist the imagination, such as pictures, dramatiza-
tions, pageants, etc. These are probably used with profit but no
one has ever determined to what extent they actually contribute,
or whether they contribute at all to the better understanding of
the event, or how much time devoted to dramatization, for ex-
ample, is worth the returns it may bring. Experimental work
should obviously be undertaken. Dramas and pageants may be
easily overdone and may often deal with unimportant phases of
the persons or events concerned.
(2) By what method may the imagining or conceiving of a given
event in time with respect to other events be accomplished? Ori-
entation in time is a very complex psychological process and prob-
ably develops rather gradually through the years of a child's
experience. It probably develops from the immediate perceptions
of changes in the child's environment to the gradual extension to
longer historical periods which are not directly perceived but are
thought of in symbolic form. Thus the writer pictures different
periods and points in history hi spatial terms by imagining a hori-
zontal line about three feet in length extending from a point, which
represents the present, toward the left, that is back to the past.
The discovery of America is located about four inches to the left
from the beginning point, the birth of Christ about a foot and a
half to the left, and so forth for other approximate locations in
time. As a mere opinion, the writer believes that it would be
advantageous to introduce the pupil to the study of history by
giving a bird's-eye view over long stretches of time, as this would
probably aid the imagination in conceiving time. Historians
usually object rather strenuously to this method of introducing or
teaching history. Psychologically, it would seem easier to imagine
long periods of time and the relative location of events in them by
viewing all history pretty much at a glance, and by giving then
more and more detailed consideration to each period. But this

again a problem requiring experimental determination.
is
(3) How may judgments about the personal motives of histori-

cal figures and causal relations among events be best developed?
Nobody knows. All we can say is to encourage the making of

such judgments and interpretations according to the best insight
of the pupil and then to check them up with those of competent
historians.
(4) What most effective methods of remembering his-

are the
torical events? In spite of the extensive experimental work in the
field of memory, there is very little in the way of concrete advice
that can be given to a pupil to assist him at this point. The fol-
lowing, partly general and partly specific, suggestions may be
given. These have been corroborated by experimental data and
have been stated in Chapter XII on How to Study, and in Chapter
XVI on Reading, and will, therefore, be only mentioned here.
a. Thoroughly understand the facts you wish to remember.
b. Systematize the facts to be remembered.
c. Look for the essentials.
d. Recall, after every paragraph or two, the essential ideas read.
e. At longer intervals, re-think or review the essential ideas
again.
f. Develop your own special means, associative links, or schemes
for remembering certain facts. Systems of memory such as that
developed by Loisette for remembering, for example, the names of
the Presidents of the United States, consist in establishing certain
associations of similarity in sound between certain parts of the
successive names, as shown in the following illustration:
George WashingTON In. "Ton" and "John" make a fairly good

JOHN Adams In. i>y sound.
JOHN Adams In. "John" and "Thorn" (the "h" is silent

THOMas Jefferson in bothnames) make an IN. by sound,
imperfect but adequate if noticed.
Thomas JefferSON In. Both names terminating with the same

James MadiSON syllable, "son," makes a clear case of In.
by sound and spelling.
JAMES Madison In. This pair of names furnishes an example

JAMES Monroe of perfect In. by sound and spelling in
the Christian names.
HISTORY 423
James MONroe In. "Mon" and "John" give us a good In.

JOHN Q. Adams by sound.
JOHNQ. Adams In. "Jack" is a nickname for John a case

Andrew JACKson ofSynonymous In.
Etc. (Loisctte, p. 26.)
The main objection to such a plan is its artificially. The chief

advantage is that it does draw specific attention to the facts to be
remembered; but everyone can develop his own special links or
clues for retaining facts with which he has difficulty. These are
likely to be more natural, more serviceable and more permanent
than artificial ones forced upon the learner from the outside. The
main point is that each one should attempt to establish such clues
which will usually result in discovering useful associative links and
at the same time force attention upon the facts to be retained.
(5) Another very important problem in the economy of learning
history is the question of essential material. What should the
child really be expected to master? What facts, names and dates
should he actually learn? What interpretations should he be led
to make andacquire? In general there has been a distinct shift
from regarding history as a chronicle of wars to regarding history
as a tracing of the development of political, industrial and social
institutions. In recent years, various committees have been at
work to decide* upon a body of minimum essentials. Thus the
committees of Iowa and Minnesota have made the following sug-
gestions as summarized by Betts ('17):
"Wars. Limit the study of wars to their remote and immediate

causes; their general geography; resources and problems of nations in-
volved; general plan of military operations; a few critical battles; im-
portant leaders; what the war settled, and the after effects; cost in men
and treasure. This plan will reduce the war phase of history study by
more than half.
"Eliminate the detailed study of battles except: Battle of Quebec;
Lexington and Concord; Bunker Hill; Saratoga; Yorktown; Lake Erie;
Merrimac and Monitor; Gettysburg; Vicksburg; Manila.
"Dates. Limit the memorizing of dates to events of central impor-
tance like the following: 1492, the discovery of America; 1607, settlement
of Jamestown; 1619, slavery introduced; 1620, Pilgrims land at Plymouth;
1643, confederation of colonies; 1775, Lexington, Concord and Bunker
Hill; 1776, Declaration of Independence; 1781, Cornwallis surrenders;
1789, First Congress; 1793, Whitney's cotton gin; 1803, Louisiana Pur*
424 EDUCATI**!- PSYCHOLOGY
chase; 1807, Fulton's steamboat; 1812, war with England; 1820, Missouri
Compromise; 1823, Monroe Doctrine; 1826, first railroad; 1844, first
telegraph; 1846, sewing-machine invented; 1845, first reaper; 1846-
1848, secession and Civil War; 1863, Emancipation
Mexican War; 1861,
Proclamation; Gettysburg, Vicksburg; 1866, Atlantic cable; 1876, first
telephone, 1878, electric light invented; 1898, war with Spain; 1903,
firstwireless across Atlantic; 1914, world war in Europe.
"Other omissions. Detailed provisions of various tariff acts (but the
meaning of tariff should be understood); details of political campaigns
except Jefferson's, Jackson's, Lincoln's and any current campaign in
progress; critical study of political party principles (but give broad
distinctions between chief rival parties) ; financial panics except those of
7, 1873, 1893." (Pp. 271 and 272.)
Bagley ('15) has been engaged on working out a possible scien-

tific plan for determining the relative emphasis upon various
portions of history by discovering the frequency of reference to
persons and events made in current magazines and newspapers.
He concludes that such a study is suggestive but doubts whether
it
may serve as a final criterion for determining the amount of
time and emphasis to be given to various phases of history.
Bagley and Rugg ('16) made a study of twenty- three text-books,
published between 1865 and 1912, by comparing the amount of
space given to various topics in each book, by determining the
shift in emphasis in the course of this period of time as measured
by the space given to different topics, and by listing the topics and
names included in all, or in a certain fraction of these texts.
"(i) In so far as can be determined from the materials presented in

the text-books, elementary American history as taught in the 7th and 8th
grades has been and still is predominantly political and military history.
"
(2) Within the past fifty years, the emphasis upon military affairs
as measured by the proportion of space devoted to wars has declined. In
was formerly
general, battles and campaigns are treated less in detail than
the rule, while proportionately more space is devoted to the causes and
the results of the wars. The lessening emphasis upon details of the wars
is first noticed in some of the text-books published between 1881 and
1888, and the tendency has been general and decided since that time.
"
(3) The later books give a perceptibly heavier emphasis to the facts
of economic and industrial development than do the earlier books, al-
though political development still constitutes the essential core of ele-
mentary historical instruction.
"
(4) As regards the treatment of specific eras or epochs, the principal
increases in emphasis are to be noted in connection with: (a) the period
HISTORY 425
1783-1812 (especially in the treatment of the so-called 'critical period'

between the close of the Revolution and the adoption of the Constitu-
tion) (b) the non-military affairs of the period 1812-1861; and (c) Euro-
;
pean events preceding and during the periods of discovery, exploration,

and settlement.
"
(8) Numerous changes have taken place in the construction of ele-
mentary text-books in history during the past fifty years. The more
important of these are: (a) a movement toward a simpler 'style' with
larger emphasis upon clear statements of causal relationships: (b) the
*
introduction and development of the problem* as a method of teaching
history, and a consequent encouragement of 'judgment* as contrasted
with rote memory, of rational as contrasted with verbatim mastery;
(c) a marked decline in the employment of imaginative pictures as illus-
trations and an increase in the use of pictures that represent sincere at-
tempts to portray actual conditions; (d) a marked decline in the use of
anecdotal materials; (e) a larger and wider use of maps." (Pp. 56 and
57.)
Horn conducted an investigation after the manner of

('17)
Bagley's magazine-newspaper method by checking through
twenty-seven recent books on current industrial, political and
social problems, in order to ascertain the facts, persons and dates
referred to and the frequency of reference. His general impression
of this inquiry is stated thus:
"This investigation has not attempted to answer the question as to

the complete content of the course of study in history. Neither does it
assert that the purpose of history is to throw light on modern social
problems, or that this is even one of the chief purposes of studying history.
Without regard to what the aims of teaching history are, this investiga-
tion has been carried on to examine into the implications of one par-
ticular assertion: namely, that history should render pupils more in-
telligent with regard to modern conditions, problems and activities. If
one assumes (i) that this is the function of history, (2) that the method of
research here followed is satisfactory, and (3) that sufficient data have
been collected, then there seems to be no escape from the conclusion that
the present elementary and high-school courses of study in history are
in very serious need of reconstruction." (Horn, '17, p. 171.)
CHAPTER XXII
MARKS AS MEASURES OF SCHOOL WORK
Importance of Marks. In order to determine the fruitfulne&J
or wastefulness of methods of learning and teaching school sub-
jects, it is necessary to evaluate the achievements of pupils as
accurately as possible. Furthermore, the successful operation of
a school demands an accounting of the work of its pupils.
Marks have been the universal measures of school work. So
many problems in the management of a school credit, failure,
promotion, retardation, elimination, graduation, honors, recom-
mendations for positions, indeed the entire scholastic machinery
of a school hinge upon the assignment of marks that it is highly
imperative to examine in detail the value, accuracy and reliability

of marks as well as to ascertain the possibility of some sort of
standardization of marks.
Variationsamong Teachers and Schools in the Distribution of
Marks. The manner in which marks are distributed to pupils
varies enormously from teacher to teacher and from school to
school. No one realized the seriousness of the situation until
specific tabulations and comparisons were made.
Meyer published marks assigned by 40
the distribution of the
different professors at the University of Missouri to their students
during a period of five years as exhibited in Table 131.
426
MARKS AS MEASURES OF SCHOOL WORK 427
TABLE 131. After Meyer ('08)
Distribution of the marks of 40 teachers in the University of Missouri for a

period of five years. The numbers are the percentages receiving the
various grades.
TOTAL No.
TEACHERS A B C F OF STUDENTS
Philosophy 55 33 10 2 623
Latin 1 52 42 6 o 130
Sociology 52 30 13 5 958
Mathematics 1 40 31 16 13 208
Economics 39 37 19 5 461
Greek 39 26 24 1 1 287
Latin II 36 40 19 5 577
French 36 29 25 10 295
Political Science .... 34 30 27 9 592
Mathematics IT ... .32 29 23 i^ 145
German 1 30 29 20 1 1
580
Psychology I 30 36 24 10 907
German II 20 38 25 u 94*
Elocution 20 Or 19 o 917
Geology 22 48 22 S 293
History 1 14 53 27 6 779
Zoology 1 21 45 28 6 479
Psychology II 19 47 29 5 238
History of Art 25 40 30 5 685
Bacteriology 20 45 31 4 263
Freehand Drawing . . . 18 47 25 10 506
Chemistry 1 23 40 31 6 205
English 1 21 41 30 8 964
Astronomy 13 40 33 5 22 5
History II n 51 33 5 8o6
Zoology II 24 37 31 8 250
German III 22 37 28 13 44i
Chemistry II 9 48 43 o 21
Education 18 38 35 9 266
Mathematics III 19 36 26 19 182
Mathematics IV 25 29 36 10 380
Physiology 20 33 40 7 426
Anatomy 19 34 3<> Ir 544
Mathematics V 16 34 35 *5 209
Engineering 1 13 36 42 9 813
Mechanical Drawing. , 18 29 41 12 538
Mechanics 18 26 42 14 495
Engineering II 16 26 46 12 826
Chemistry HI i u 60 28 1903
English II 9 ** 35 28 1,098
Similar tabulations have been published for Harvard University

by Foster, Table 133; for the University of Wisconsin by Dearborn,
Table 134; and for Cornell University by Finkelstein. Finkelstein
('13) has shown the effect of the personal equation in the marks
assigned to the same students in a year course which was in charge
of one teacher in the first semester and of another in the second
semester.
TABLE 132. After Finkelstein
No. OF
STUDENTS PER CENT RECEIVING THE VARIOUS MARKS
0-39 40-44 45-49 50-54 55-59 60-64 65-69
ist semester ... 263 .4 .4 2.3 45 1.9 13 7 n.8
and semester .. 257 1.2 .8 1.5 12.1 12.8
No. OP
STUDENTS
70-74 75-79 80-84 85-89 90-94 95-100 Exempt
semester, con. ..
i st 263 16.7 15.6 20.2 6.4 5.7 .4 12.5
2nd semester, con. .. 257 10.5 13.6 98 33.9 3.8 37.7
The instructor in the first semester exempted from the final

examination 12.5% of the class, while the instructor in the second
semester exempted 37.7%. The latter obviously graded very
much higher than the former.
MARKS AS MEASURES O? SCHOOL WORK 429
TABLE 133. Alter Foster ('n).
Harvard College. Distribution of 8,969 grades. Elementary Courses
Averages ............ 7 20 42 3 213

TABLE 134
Percentages of grades assigned by 45 individual instructors in the University of

Wisconsin to freshmen and sophomores. After Dearborn ('10}
German:
35- 26.3 21. 12.3 5-2
36.. 12. 49.1 24. II. I 3-7
37 .. 34-3 2.9 2.9 67
3... 11.4 22.9 3- 2 61
39- -
17-3 37-3 29 6.6 9 3 75
40. .
17.9 2b 17.1 4.0 123
14-7 29-5 33 18 9 3- 1 95
42. 12 3 27.4 *1 2 r
.9
43- 29 o 30.1 14 4 3 93
44- 21 6 42.0 12 5 2.27 88
45- 22.4 1-7
JL
T.
10
70 75 80 86 90 96 100
FIG. 85. Distribution of all grades in a high school. After Gray ( '13, p. 66).
These tables agree showing extremely wide differences among

in
teachers in the manner In the tabulation for the
of giving marks.
University of Missouri, one professor assigned the grade A to 55%
of his students, and the grade F to only 2%, while another profes-
sor assigned the grade to onlyA i%

of his students and the grade
F to 28%. At Harvard, one professor gave the grade to 35% A
and the grade E
to i%, while another professor gave to and A i%
E to 32% of his students..
The situation in high schools is substantially the same. Gray
tabulated the marks assigned by all the teachers in eight high
Ln
T l
u
nu
70 75 80 85 90 86
i
100
FIG. 86. Distribution of all grades in another high school. After Gray (*i3,p. 67).
schools. The two of these schools are shown in
distributions of
Figures 85 and 86. The one grades high and the other grades low.
In one school the great mass of the pupils receive 85 to 100; in the
other they receive 85 to 70.
Table 135 shows the distribution of the grades by the different
teachers in a high school of about 150 pupils.
TABLE 135
Distribution of 8,490 grades by the different teachers in a high school during

four years. (From a private report by Superintendent J. F. Waddell,
Evansville, Wisconsin.)
74 75-8o 81-86 87-92 93-100

English (1913-1914, 1914-1915).. 33% 39% 13% 12% 3%
English (1915-1916) 15 27 27 25 6
Latin and German 5 15 13 33 34
Mathematics (1913-1914) 22 31 18 21 8
Mathematics (1914-1915) 23 27 24 13 13
Mathematics (1915-1916) 5 12 15 28 40
History (1913-1914) n 25 22 33 9
History (1914-1915; 1915-1916). . 10 18 25 30 17
Science (1913-1914) ir 36 25 26 2
Science (1914-1915; 1915-1916) . . 10 33 24 19 14
Domestic Science o 12 27 51 10
Distribution of grades for the year after the above tabulation was made known
to the teachers. Extreme variations are considerably reduced.
English (1916-1917) 6% 32% 27% 27^ 6%

Latin and German (1916-1917) .. 9 25 22 29 15
Mathematics (1916-1917) 5 20 28 27 20
History (1916-1917) 2 17 37 34 10
Science (1916-1917) 5 26 33 22 14
Domestic Science (1916-1917) ... o 10 40 41 9
Variation among Teachers in the Evaluation of the Same School

Products. A more and crucial method of examining the
direct
variations of teachers' marks than the tabulation of the grades as
distributed by different teachers is to measure experimentally the
differences in the values assigned by different teachers to the same
pieces of work.
Starch and Elliott ('12 and '13) made a series of investigations
in which two examination papers in first-year-high-school
final
English were graded by 142 English teachers in as many high

schools, one final examination paper in geometry was graded by
118 teachers of mathematics, and one final examination paper in
American history was graded by 70 teachers of history. The
variations in these marks are shown in Figures 87, 88, 89 and 90.
The differences are astounding; the marks for any given paper run
practically over the entire range of the percentage
scale ordinarily
used. The marks of the first English paper run all the way from
64 to 98, of the second English* paper from 50 to 98, of the geometry

paper from 28 to 92, and of the history paper from 43 to 90.
65 70 75 80 85 90 85
FIG. 87. Distribution of the marks assigned by 142 English teachers to a

final examination paper in high-school freshman English. After Starch and
Elliott ('12).
75 80 85 90 95
50 60 65 70
FIG. 88. Distribution of the works assigned by 142 english teachers to an-
other final examination paper in high-school freshman English. After Starch
and Elliott ('12).
28 53 55 60 65 S 55 5 % 3)
FIG. 89. Distribution of marks assigned by 114 mathematics teachers to a

final examination paper in geometry. After Starch and Elliott ( '13).
i
60 60 70 80 90
FIG. 90. Distribution of marks assigned by 70 history teachers to a final

lamination paper in American history. After Starch and Elliott ('13).
This investigation has established two conclusions: first, that

teachers differ enormously in evaluating the same pieces of work
in terms of the ordinary percentage scale; and second, that they
differ as much in one subject as in .another. They disagree as
much in evaluating a paper in mathematics as in English or his-
tory. Apparently mathematical papers are not marked with
mathematical precision any more than any other papers are.
The author made a further investigation by having ten final
papers in freshman English in the University of Wisconsin graded
by ten instructors of freshman English. The marks are shown in
Marks assigned by ten instructors to ten final examination papers in English.
Av. 79.4 81.4 79.8 73.7 65.5 78.7 83.8 85.1 79.7 80.2 78.7 5.3 .074
The variations shown in this table are practically as large as

those found in the previous inquiry. It was thought that the wide
range of marks shown in the first study might be due to the fact
that the teachers were in different schools. However, Table 136
shows that teachers in the same department differ almost as much.
Less extensive results obtained by having various members of a
department grade the same paper show that as much variation
exists in other subjects as in English.
Causes of Variation. Why do teachers differ so much in estimat-
ing the worth of a given product and in the distribution of marks
to groups of pupils? Four possible factors may be mentioned:
(i) Differences in the standard of severity or leniency in different
schools; (2) differences in the standard of severity or leniency of
different teachers; (3) differences in credit or penalty assigned by
different teachers to any given fact or error in a piece of work; and
(4) minuteness of the discrimination between successive steps of
merit or quality in a given scale of qualities. How potent is each

factor in producing the total variation in evaluating a given paper?
The first thought that occurs in regard to the wide range of
marks for the same papers 2,3 shown in Figures 87 to 90, was that
it must be due to the fact that these teachers were situated in
different schools with different standards and ideals. It turns out,

however, that factor one relatively insignificant. If we compare
is
the mean marks of the ten English papers assigned

variation of the
by ten instructors in the same department with the mean variation
of the marks of the two English papers assigned by teachers in
different schools we can determine the part played by factor one.
The mean marks is 5.3 and of the
variation of the former set of
latter is 5.4. Hence the mean variation of the marks assigned by
teachers in the same department is only o.i less than the mean
variation of marks assigned by teachers in different schools.
The potency of factor two may be ascertained from the data in
Table 136. The general average of all the grades assigned by the
ten instructors to the ten papers is 78.7. If we compare the aver-
age of all the marks given by any one instructor with the general
average we obtain a measure of his particular standard of severity
or leniency. Thus instructor 5 graded on the average 13.2 points
lower and instructor 8 graded 6.4 points higher than the general
average of all the teachers. If now we raise or lower each instruct-
or's gradesby as many points as the average of his grades is below
or above the general average, we find that the mean variation of
these weighted marks is 4.3. This mean variation is only i.o
point smaller than the mean variation of the original unweighted
marks. Hence factor two accounts for a relatively small share of
the total variation. Factors three and four must then account for
the remaining mean variation of 4.3. The strength of factor four
can be determined experimentally by having the same teachers
re-grade their own papers without knowledge or, their former
marks. The author carried out such an
experiment\and obtained
the results exhibited in Table 137.
The mean variation of these" marks, comparing the first with the
second for each paper, is 2.2 points. A part of this variation,
however, is due to the slight shift in standard on the part of each
teacher from one grading to the next. By applying the same
process of weighting explained in connection with factor two, the
mean variation drops from 2.2 to 1.75 points. Hence 1.75 repre-
sents the amount of variation contributed by factor four to the
total mean variation and by subtraction we find that factor three
contributes 2.55 points. The four factors therefore contribute
the following amounts:
Factor one 10 points

Factor two i . oo "
Factor three 2
"
.
55
Factor four i
"
.
75
"
Total mean variation 5 40.
It is obvious then that factors three and four are the most im-
portant ones in producing the large differences of values assigned

by teachers to a given piece of school work.
How Large Should the Units of a Marking Scale Be? The
answer to this question depends primarily upon the fineness of the
discriminations of successive degrees of quality in terms of the
scale used, and, secondarily, upon the convenience of using a given
scale of marks. The smallness of distinguishable shades of quality
of anything can be determined by ascertaining the amount of
difference in terms of a given scale that can be discriminated in
the long run by the judges.
A general principle that has been followed commonly in psycho-
logical measurements, in which the units depend upon the discrimi-
nation of judges, has been to regard a difference in amount of the
thing in question which can be distinguished correctly by 75% of

the judges or judgments concerning it as the smallest psychological
unit that can be used with reasonable certainty. For example, if
we take ten shades of blue and ask 100 judges to arrange them in
the order of blueness from left to right, we would regard that dif-
ference in blueness between any two successive shades which
75 of the judges agree in perceiving the one bluer than the other as
the least that can be distinguished with a fair degree of confidence.
The 75% point is chosen because it is midway between pure
chance and absolute certainty. If only one of two possible judg-

ments may be made, that is, if a given shade is either more or less
MARKS AS MEASURES O* SCHOOL WORK 439
blue than another, then 50% of the judgments would be correct

by pure chance guesses. If 100% of the judgments are correct,
it means that the difference is so large that it can be recognized
correctly every time, and the amount of difference may range

anywhere from just large enough to be always recognized up to
rn infinite difference.
According to this principle, how large would the steps be on
the marking scale? For example, if we took two papers, a and 6,
in a given subject which differed in quality just enough so that
three-fourths of the examiners or teachers would consider b better
than a, how large would this difference be, say on the usual 100
percentage scale?Data for answering this question with approx-
imate accuracy are found in Figures 87 to 90. The probable
error or median deviation of the marks given by the teachers to
the papers represented in these four figures are 4.0, 4.8, 7.8 and 7.1
respectively, with an average of 6.4. By definition, two times the
probable error includes the middle half of the measures or marks.
For example, median is 80.2 and the probable
in Figure 88 the
error is 4.8, the middle 71 of the 142 marks lie between 75
that is,
and 85. Obviously one-fourth or 35 of the marks lie above 85.

Consequently so far as this particular paper is concerned the next
better paper would have to be 4.8 points better so that three-
fourths of the examiners would consider it better.
Now the average probable error of the four sets of marks is 6.4.
Hence the difference between two papers in general must be approx-
imately 6.4 points so that three-fourths of the examiners would
consider one better than another. On this principle then the step
on the 100 percentage scale, with 70 as the usual passing grade,
turns out to be approximately 7 points. This would produce a
scale of steps as follows: 70-76, 77-84, 85-92, and 93-100. That
is, the marking scale would have five steps, failure and four passing
steps above 70. which may be designated as excellent, good, fair,

poor, and failure, or perhaps preferably by the symbols A, B, C, D,
and E. Such would be the size of the steps so that three-fourths
of the examiners of a given set of papers would agree in distinguish-
ing between the qualities of the papers.
However, any individual teacher agrees with himself more
closely in re-grading a set of papers than he agrees with other
teachers, as indicated in Table 137. This table shows that the
probable error or median deviation of a given teacher's mark"- in
re-grading his own papers is approximately 2 points. By the saîe
reasoning the amount of difference in quality between two papers

would have to be 2 points in order that an individual teacher would
consider one paper better than another in three out of four in-
dependent markings. Hence the marking scale for an individual
teacher, who grades papers from his own viewpoint and compares
them only with his own judgments, could have each step in a five-
step scale subdivided into three smaller steps of about 2 points each
by using the plus and minus. That is 70 to 76 would become
70-72 or D- 73-74 D, and 75-76 D+, and so on.
,
Whether a marking scale such as the 100 percentage scale or

fine
a coarser five-step scale should be used is largely a matter of con-
venience and personal habit. The advantage of a coarse scale is
perhaps that it avoids giving the pupil the impression that the
evaluation of a piece of work is more accurate than it actually is.

The advantage of a fine scale is that it probably encourages the
examiner in making as fine distinctions as possible. In practice
a fine scale can probably be used as readily and as quickly as a
coarse one if the teacher is accustomed to using a fine scale. A
person may use as fine a scale as he wishes provided he recognizes
the amount of the probable error in terms of the units of that par-
ticular scale. In terms of a 100 percentage scale the probable error
is about 6 or 7 points; in terms of the five-step scale it is about
one step which is 6 or 7 times as large as a point on the percentage

scale. The absolute amount of variation is substantially the same
on the two scales.
How Should Marks be Distributed to Groups of Pupils? If a
five-step scale is used, what percentages of pupils should in the
long run receive each of the five marks? The answer to this ques-
tion that I advocate is that the marks of large numbers of unse-
lected pupils should be distributed approximately in conformity

with the normal distribution or probability curve. Three lines of
evidence for this position may be presented, the last two of which
are fundamentally based upon the first:
First, mental and physical traits, when measured in large num-
bers of individuals, are distributed in a manner which yields a
distribution surface very nearly identical with that of the proba-
bility curve. Concrete evidence for this statement has been pre-
sented in Chapter III, Figures 7 to 10, to which the reader should
turn. It seems reasonable to infer that abilities in school subjects
are very probably distributed in the same manner as other mental
traits.
Iii the second place, when abilities in school

subjects are meas-
ured by objective methods, they are found to be distributed in
very close conformity to the probability curve. Concrete evidence
for this is presented in Figures 16 to 27, Chapter III.
Thus, for example, the scores of 662 seventh grade pupils in the
author's geography test shows the following distribution when the
total range of the base line is divided into five equal sections:
Scores 0-27 28-54 55-8i 82-108 109-135

% of pupils ft
24 37 24 9
This is a remarkably close conformity to the theoretical dis-

tribution proposed on the following pages.
In the third place, the distribution of marks assigned by many
teachers to large numbers of students conforms fairly closely to
the normal distribution curve. When the marks of many teachers
are combined, the idiosyncrasies of individual teachers tend to be
counterbalanced. Tables 138 to 142, and Figures 91, 92, 93,
and 94, show the distributions of marks in various institutions and
the extent to -which they differ from the theoretical probability
curve.
TABLE 138
Distribution of grades in the College of Letters and Science, University of
Wisconsin, for the years 1907, 1910 to 1915. From the reports of Presi-
dent . A. Birge.
CONDITIONED Ex- No. OP
INCOMPLETE & FAILED POOR FAIR GOOD CELLENT GRADES
Elementary Course.. 3.6 9.3 15.3 33.2 29.4 9.2 42,557
Advanced Courses... 3.2 3.5 7.9 30.9 41.8 12.7 39,302
TABLE 139
Distribution of grades at Cornell University for the years 1902, 1903 and 1911.
Adapted from Finkelstein ('13, p. 22), to give the distributions for a five-
point scale, 60 being the passing grade.
NUMBER OF GRADES
0-59 60-69 70-79 80-89 00-100
9.2 22.5 30.0 27.2 II. I 20,348
TABLE 140
Distribution of all grades for two academic years at Harvard College. After
Foster ('n, p. 262)
TABLE 141
Distribution of grades at the University of Missouri. After Foster, p. 289

NUMBER
OF GRADES
24,979
TABLE 142
Average and the elementary courses for

percentages for Cornell, Missouri,
Harvard and Wisconsin. These percentages do not total 100 because the
incomplete grades for Wisconsin and Missouri are not included.
E D C B A NUMBLR OF GRADES
8.7 17.9 38.0 24.5 8.2 96,853
If we grant that marks in the long run should be assigned ac-
cording to the normal distribution curve, what percentage of

pupils should receive each of the five steps of the marking scale?
If the base line of the probability curve in Figure 15, Chapter III,
isdivided into five equal divisions, then the area above the various
divisions would comprise the following percentages of the total
1
area:
A, Excellent, or 93-100
= 7%
B, Good, or 85- 92 = 24%
C, Fair, or 77- 84 = 38%
D, Poor, or 70- 76 = 24%
E, Failure, or 69
= 7%
Figures 91 to 94 indicate how closely distributions of the marks

at Wisconsin, Cornell, Harvard, and Missouri run parallel to the
theoretical curve. The only difference is a slight skewing to the
right. Not quite as many D's are assigned and very slightly more
E's and A's are assigned than the theoretical distribution would
demand. Thus the marks as actually assigned by hundreds of
1
The ends
of the probability curve would reach the base line only at infinity. Hence
an arbitrary point of termination must be selected. This has been placed at a point
3.65 P. E. values from the median. This point has been selected because it yields 7%
for the E and A surfaces which is approximately the percentage of pupils receiving these
grades in many institutions.
teachers lo thousands of students furnish impressive support for

the theory of the probability distribution of grades.
Certain objections, however, both of a theoretical and a practi-
cal kind,must be considered. In the first place, the soundness of
Con. & Failed Poor Fair Good Excellent
FIG. 91. -Distribution of 42,557 grades, broken line, in elementary courses

in the College of Letters and Science of the University of Wisconsin. The con-
tinuous line is the theoretical distribution. After a report by President E. A.
Birge.
the theory rests on the supposition that the pupils are unselected,
chance specimens of mankind as a whole. This supposition, of
course, never obtains absolutely for any group of human beings
brought together anywhere. The very reason that brings any
group together at the same time selects them. Pupils in school are
0-59 60-69 70-79 80-89 90-100

FIG. 92. Distribution of 20,348 grades at Cornell University. After Fin-
kelstein ('13).
not random samplings of human beings of their respective ages

the less so as one goes up the educational ladder. The tendency
is that every rung of the ladder selects on the whole slightly better
and better specimens. The fact, however, seems to be that the

selection which does take place is not of the sort that materially
modifies the form of the distribution curve but rather tends to
contract its base. The selection that does take place is not an
abrupt cutting off, but a gradual slicing off along a large share of
the distribution surface.
E D C B A
FIG. 93. Distribution of 8,969 grades in elementary courses at Harvard Uni-
'
versity. After Foster ( 1 1) .
The writer undertook to ascertain the actual elimination of

university students as it really takes place on the basis of the records
of 476 freshmen tabulated by Dearborn. It was found that the
E D c B A
FIG. 94. Distribution of 24,979 grades at the University of Missouri. After
Meyer ('08).
following percentages of students dropped out of the University

in the various grades of scholarship at the end of the freshmen and
sophomore years:
TABLE 143
CONDITIONED
& FAILED POOR FAIR GOOD EXCELLENT
Percentage of students of each
grade dropped during fresh-
man year 100 52
Percentage of those remaining
in each grade, dropped during
sophomore year , , 45
MAKKS AS MEASURES OK SCHOOL WORK 44 j
This table reads that all students whose average grade was
"conditioned" or "failed" dropped out during the freshman year;
52% of those whose grade was "poor" dropped out during the
freshman year and 45% of those remaining whose grade was
"poor" dropped out during the sophomore year, etc. It is obvious,
therefore, that there is elimination from all classes of scholarship
with the exception of the highest from which there is very little or
no loss. The general effect of the actual elimination upon the
distribution curve is to shift the left end of the curve toward the
right and to change the general form of it only slightly as indicated

in Figure 95.
The outcome of this evidence is that the distribution of the
well as of the high
grades for the freshman year of the college as
66-68 69-72 73-76 77-80 81-64 85-88 89-92 93-96 97-100
FIG. 95. The continuous line shows the theoretical distribution of the marks
of students. The upper broken line represents the change in this curve due to
the dropping out of students during the freshman year. The lower broken line
represents the change in the curve due to the elimination during the sophomore
year. After Starch ('13).
school should conform quite closely to the theoretical distribution

curve and that slight shifts to the right should be made for the
successive four years. It may seem curious to recommend that
after the elimination of the successive years of the high school, the
distribution to be followed in the freshman year of the college
should be approximately normal again. The explanation is that
the standards of the college are somewhat higher than those of the
should eliminate all of
high school; so that, even if the high school
the poorest 7% of its pupils, the next poorest 7%, who are able to
to meet the de-
complete the high school, are likely to be unable
mands of the college.
The second objection urged by teachers against the adoption of
the theoretical distribution of grades here recommended is that it
would be unfair to lay down a rule that 7% of the pupils should
be failed How do we know; possibly by good teaching all pupil^
may reach a sufficiently high attainment to be passed in the course.

The answer to this statement is that the effects even of the best
teaching will so rarely raise the attainments of pupils sufficiently
high so that none of the pupils would fall below the passing grade;
and furthermore, in the interests of reasonably high standards of
scholarship the attainments of approximately 7% of large num-
bers of pupils will very probably not merit a passing grade. There
should be doubly good reasons for passing all students or for failing
considerably less than 7%. Many of the cases of "good teaching"
or "unusual classes" prove to be spurious when it is possible to
check them up by outside means.
A third point is not so much an objection as a question of practi-
cal use of the principle of the distribution of grades; namely, in
how large classes or groups of pupils should we expect fairly close
conformity, and how close conformity should be expected? The
answer to the question which I shall give, on the basis of experience
in attempting to observe the principle in the assignment of grades,
is that for groups of students of 100 or more quite close conformity
should be expected. By quite close conformity I mean a deviation

of not more than about 25% above or below the number of grades
that theoretically should fall on a given step of the five-division
scale. For example, the theoretical distribution demands that
7% of the pupils should receive the grade of A or Excellent. For
groups of 100 or more pupils, this percentage should ordinarily
not be lower than 5 nor higher than 9; the percentage of B's should
ordinarily not run lower than 18 nor higher than 30; the percentage
of C's should not run higher than 48, nor lower than 28, etc. The
larger the number of pupils concerned, the closer the conformity
should be. For groups smaller than 100 a wider latitude should be
permissible whenever there is genuine reason for wider deviation. I
advocate conformity to the theoretical distribution within the limits
ofcommon sense with as much deviation as may seem permissible
lor good cause.However, really genuine reasons for large devia-
tions, even with classes as small as 25 pupils, unless obviously
selected by special cause, is much rarer than teachers ordinarily
believe.
In support of this contention, the author ('15) reported an
experiment in which twenty-four compositions written by sixth
and seventh grade pupils were graded by 23 teachers according to
the usual percentage method with 70 as a passing grade. After
the papers had thus been graded, the teachers were requested to
grade them according to a five point scale and give the grade of E
to two papers, D to from four to six papers, C to from eight to ten
papers, B to from four to six papers, and A 'to two papers. Even
if the teachers felt, for example, that there were no papers good
enough to receive the grade of A, they were to select the two best
ones and call them A. The outcome was that those teachers who
in their original grades differed most from the combined judgment
of all the teachers were forced to comply more closely to the actual
average marks as given in the first grading. One teacher marked
the highest paper 85 in the original grading, and objected to giving
it a grade of A in the forced distribution on the ground that no
paper in the lot was good enough to receive so high a grade, and
yet the average of the marks given by all the teachers to this
paper was 92.9, the best paper in the entire group.
The theory of the probability distribution of marks should be
observed with sense and reason and not in a purely mechanical
manner. A blind, unintelligent observance of the principle is
bound to lead to injustice, particularly with small classes. In one
such case which came to the author's attention it led to the giving
of a mediocre grade to a pupil of very high ability.
A fourth point frequently raised by teachers to justify unusually
high or low marks is that the particular class in question is an
unusually good one or poor one. Such a claim ought to be allowed
only if it can be justified by good evidence. There are, of course,
differences in classes, but these are almost never as great as we are
inclined to believe. Large differences between successive classes
in the same subject are for the most part illusory for the reason
that the judgment of an individual teacher is more likely to deviate
from a correct estimate than the average ability of a group devi-
ates from the average of other groups. The teacher who says to
each succeeding class that this is the best class he has ever had in
this subject would possess, if this judgment were correct, a magic
power for elevating the intellectual level of human beings.

The feeling on the part of teachers that a given class is an un-
usually good or poor class is quite often due to one or two unusu-
ally good or poor individuals whose impression upon the mind of
the teacher is outstanding, rather than to a higher or lower level
of the class as a whole.
As concrete evidence of the extent to which a teacher may err
in such opinions and of the manner in which the opinion of the
teacher may he checked up, the curves in Figure 06 are presented.
The continuous curve shows tie distribution of the grades of a

teacher in Latin and German. When her attention was called to
the predominance of high marks, she claimed that her pupils were
exceptionally good. The broken curve shows the extent to which
her claim was unfounded since it shows that, according to their
abilities in other subjects, they were an average group.
How May Variation in the Assignment of Grades be Reduced.
i. a common sense compliance in the distribution of marks
By
with the normal distribution or probability curve.
(a) To this end the administrator of a school should tabulate
at stated intervals the marks assigned by each teacher and exhibit
the tabulation to the teachers. This in itself will usually lead
without request or compulsion to a very considerable correction
ee 75 81 87 93 100
FIG. 96. The continuous line shows the distribution of the marks of a teacher
of Latin and German in a high school. The broken line shows the distribution
of themarks of the same pupils in their other subjects. After an unpublished
report of Supt. J. F. Waddell, Evansville, Wisconsin.
of aberrations on the part of those teachers who deviate most

widely. At the University of Missouri the adoption of a plan of
distribution in conformity with the normal curve reduced the
irregularity in grading in the ratio of five to two.
(b) The teacher himself will find it useful to tabulate at frequent
intervals the distribution of his grades. In making out the marks
of a set of papers and particularly in making out the final grades
for a course, the author has followed for several years the practice
of plotting a distribution of the grades as tentatively made out.
If the assignment is decidedly abnormal in having
considerably
too many or too few of the different grades, a shift is made of
borderline cases, unless there is an obvious reason to the
contrary,
to obtain a reasonably normal distribution. Every teacher feels
that there is a considerable number of cases concerning which he
is in doubt as to whether they should have the one or the other
MARKS AS MEASURES OF SCPIOOL WORK 449
grade. For example, if the tentative list of grades contains too

many or too few A's, the lowest A's may be shifted to B's or the
highest B's may be shifted to A's.
2. Variability and uncertainty in grades may be reduced by
adopting particularly in departments containing several teachers,

a plan of giving certain weights or penalties for certain types of
errors or defects. This should be done by departmental conference
so as to secure a consensus of judgments on the various types of
errors and amounts of penalties. Much could be done in this
direction toward greater uniformity in methods of grading. If
organizations of teachers would take this matter up, much of the

chaos which is now present in methods of grading could be re-
duced to order.
At the present time A's or B's obtained from different teachers
often mean quite different things. By observing the points here
suggested they would mean more
nearly the same thing. Evalua-
tion of achievement terms of judgment depends obviously upon
in
the judge. Marks as such will at best depend upon the examiner.
They will probably always have to be used. More impersonal and
objective methods for determining achievement in school work
are being developed at the present time. To what extent these
educational measuring devices will be able to replace the usual
examinations and grades will depend upon their future develop-
ment.
BIBLIOGRAPHY '
Abbott, E. E.: 1909. On the Analysis of the Memory Consciousness in

Orthography. Psychological Review Monograph, n: No. i, 127-
158.
Arai, T.: 1912. Mental Fatigue. Columbia University Contributions
to Education, No. 54.
Ash, I. E.: 1914. Fatigue and Its Effects upon Control. Archives of
Psychology, No. 31.
Ayres, L. P.: 1912. A Scale for Measuring the Quality of Handwriting
of School Children. Russell Sage Foundation, New York City.
Ayres, L. P.: 1913. The Spelling Vocabularies of Personal and Business
Letters. Russell Sage Foundation, New York City.
Ayres, L. P.: 1915. A Measuring Scale for Ability in Spelling. Russell
Sage Foundation, New York City.
Bagley, W. C.: 1911.

Educational Values.
Bagley, W. C.: 1915. The Determination of Minimum Essentials in
Elementary Geography and History. Fourteenth Yearbook of the
National Society for the Study of Education, 131-146.
Bagley, W. C., and Rugg, II. O.: 1916. The Content of American
History as taught in the Seventh and Eighth Grades: An Analysis of
Typical School Text-books. University of Illinois Bulletin.
Bair, J. II.: 1902. The Practice Curve. Psychological Review Mono-
graph, No. 19.
Baldwin, B. T.: 1914. United States Bureau of Education, Bulletin
No. 581.
Billiard, P. B.: 1915. Norms of Performance in Fundamental Processes
of Arithmetic, Journal of Pedagogy, 3:9-20.
Ballou, F. W.: 1914. Scales for the Measurement of English Comix>si-
tion. Harvard University.
Batson, W. H.: 1916. Acquisition of Skill. Psychological Review
Monographs, 21, No. 3.
Bean, C. II.: 1912. The Curve of Forgetting. Archives of Psychology,

No. 21.
Bennett, C. J. C.: 1907. Formal Discipline. Columbia University.

Bennett, N. F.: 1916. The Correlations Between Different Memories.
Journal of Experimental Psychology, i :404~4i8.
Bergstrom, J. A. : 1894. The Relation of the Interference to the Practice
Effect of an Association. American Journal of Psychology, 6 1433-
442.
45i
4$2 BIBLIOGRAPHY
Belts, G. H.: 1909. The Distribution and Functions of Mental Imagery.

Columbia University Contributions to Education, No. 26.
Betts, G. H.: 1917. Class-Room Method and Management.
Binet, H., and Simon, T. 1905. (a) Sur la n&essite" d'etabler un diagnos-
:
tic scientific dcs 6tats inferieurs de Intelligence. Annee Psycho-

logique, 11:163-190. (b) Methodes nouvelles pour Ic diagnostic du
niveau intellectuel des anormaux. Ibid., 191-244. (c) Application
des me'thodes nouvelles an diagnostic du niveau intellectuel chez
enfants normaux et anormaux d'hospice et d'Scole primaire. Ibid.,
245-336.
Boas, F.: 1896-1897. Growth of American Children. Rejxjrt of United
States Commissioner of Education, 11:1555.
Bolton, T. L.: 1902. Ueber die Bezichungen zwischcn Ermudung,
Roumsinn der Haut and Muskelleistung. Psychologische Arbeiten,
4:175-234.
Book, W. F.: 1008. The Psychology of Skill: with Special Reference to
itsAcquisition in Typewriting. University of Montana, Publica-
tions in Psychology; Bulletin No. 53, Psychological Series No. i.
Borst, M.: 1904. Recherches exp'tlles sur 1'educabilite ct la fidclite du
tcmoignoge. Archives de Psychologic, 3:233-314.
Breslich, E. R.: 1912. Teaching High School Pupils to Study. School
Review, 20:505-515.
Briggs, T. H.: 1913. Formal English Grammar as a Discipline. Teach-
ers' College Record, 14, No. 41.
Briggs, T. H.: '13. The Right Way to Read. Education, 33:550-562.

Brown, H. A.: 1914. Measurement of Efficiency of Instruction in Read-
ing. Elementary School Teacher, 14:477-400.
Brown, H. A. 1916. The Measurement of Ability to Read. Department
:
of Public Instruction, New Hampshire, Bulletin No. i.

Brown, J. C.: 1913. An Investigation on the Value of Drill Work in the
Fundamental Operations in Arithmetic. Journal of Educ. Psych.,
2:81-88; 3 061-570.
Brown, M. D., and Haggerty, M. E.: 1917. The Measurement of Im-
provement in English Composition. The English Journal, 6 151 5-527.
Browne, C. E.: 1006. The Psychology of the Simple Arithmetical
Processes. American Journal of Psychology, 17:1-37.
Bryan, W. L., and Harter, N. 1897-1899. Studies in the Physiology and
:
Psychology of the Telegraphic Language. Psychological Review,

4:27-53; 6:348-375.
Buckingham, B. R.: 1913. Spelling Ability: Its Measurement and
Distribution. Teachers' College Record.
Burk, C. F.: 1900. The Collecting Instinct. Pedagogical Seminary,
7:179-207.
Burnett, C. J.: 1906. The Estimation of Number. Harvard Psycholog-
ical Studies, 2:349-404.
BIBLIOGRAPHY 453
Burt, C.: 1909. Experimental Tests of General Intelligence. British

Journal of Psychology, 3:94-177.
Carter, M. H.: 1909. The Conservation of the Defective Child, Mc-

Clure's Magazine, 33:160.
Cattell, J. M.: 1886. Inertia of the Eye and Brain. Brain 8:294-309.
Cattcll, J. M.: 1886. Ueber die Zeit des Erkennens, u. s. w. Philos.
Studien, 3:94-127.
Cattell, J. M.: 1905. Examinations, Grades and Credits. Popular
Science Monthly, 66:367-378.
Cattcll, J. M.: 1906. A
Statistical Study of American Men of Science,
III. The Distribution of American Men of Science. Science, New
Series, 24:723-742.
Charters, W. .W., andMiller, Edith: 1915. A
Course of Study in Gram-
mar. University of Missouri, Education Bulletin, No. 9.
Colvin, S. S., and Myers, E. J.: 1909. The Development of Imagination
in School Children and the Relation between Ideational Types and
the Retentivity of Material Appealing to Various Sense Depart-
ments. Psychological Review Monographs, n: No. x, 85-126.
Conrad, H. E., and Arps, G. F.: 1916. An Experimental Study in Eco-
nomical Learning. American Journal of Psychology, 27:507-529.
Cook, W. A.: 1912. Shall we Teach Spelling by Rule? Journal of Educa*
tional Psychology, 3:316:325.
Cook, W. A., and O'Shea, M. V.: 1914. The Child and His Spelling.
Coover, J. E.: 1916. Formal Discipline from the Standpoint of Experi-
mental Psychology. Psychological Review Monographs, 20: No. 3.
Coover, J. E., and Angell, F.: 1907. General Practice Effect of Special
Exercise. American Journal of Psychology, 18:327-340.
Cornell, W. S.: 1008. The Relation of Physical to Mental Defect in
School Children. Psychological Clinic, i '.233.
Cornman, O. P.: 1902. Spelling in the Elementary School.
Counts, G. S.: 1917. Arithmetic Tests and Studies in the Psychology
of Arithmetic. Suppl. Educ. Monographs, i : No. 4. University of
Chicago.
Courtis, S. A.: 1910. Manual of Instruction. 82 Eliot Street, Detroit,
Michigan.
Dallam, M. T. : 1917. Is the Study of Latin Advantageous to the Study

of English? 'Educational Review, 54:500-503.
Davidson, P. E.: 1914. The Recapitulation Theory and Human In-
fancy. (Columbia University.)
Davis, W. W.: 1898-1900. Researches in Cross-Education. Studies
from the Yale Psychological Laboratory, 6:5, 6, and 8, 54-108.
Dearborn, W. F.: 1906. The Psychology of Reading. Archives of Psy-
chology, No. 4.
4S4 BIBLIOGRAPHY
Dearborn, W. F.: 1909. The General Effects of Special Practice in

Memory. Psychological Bulletin, 6:44.
Dearborn, W. F.: 1909. The Relative Standing of Pupils in the High
School and the University. Bulletin of the University of Wiscon-
sin, No. 312. ,
Dearborn, W. F.: 1910. Experiments in Learning. Journal of Educa-

tional Psychology, 1:373-388.
Dearborn, W. F.: 1910. School and University Grades. Bulletin of the
University of Wisconsin, No. 368.
De Candolle, A.: 1873. (Second Edition, 1885.) Histoire des sciences
et des savants depuis deux siecles, etc.
Delabarre, E. B.: 1898. A
Method of Recording Eye Movements.
American Journal of Psychology, 9.
Dietzc, G.: 1885. Ueber den Bcwusstseinsumfang.
Dodge, R.: 1900. Visual Perception During Eye-movements. Psy-
chological Review, 7 1454-465.
Dougherty, M. L.: 1913. Report of the Binet-Simoii Tests Given to
483 Children in the Public Schools of Kansas City, Kansas. Journal
of Educational Psychology, 4:338-352.
Downey, J. E.: 1908. Control Processes in Modified Handwriting: An
Experimental Study. Psychological Review Monographs, 9, No. i.
Downey, J. E.: 1910. Judgments on the Sex of Handwriting. Psy-
chological Review, 17:205-216.
Drushel, J. A.: 191-. A
Study of the Amount of Arithmetic at the Com-
mand of High School Graduates who have had no Arithmetic in
their High School Course. El. School Journal, 17:657-661.
Earle, E. L.: 1903. The Inheritance of the Ability to Learn to Spell.

Columbia Contrib. to Phil., Psy., and Ed. 2, No. 2:41-44.
Ebbinghaus, H.: 1885. Ueber das Gedachtniss.
Ebbinghaus, H.: 1897. Ueber eine neue Methode zur Prtifung geistiger
Fahigkeiten und ihre anwendung bei Schulkindern. Zeitschrift ftir
Psychologic, 13:401-459.
Ebert, E., and Meumann, E. : 1905. Ueber einige Grundfragen der Psy-
chologic des Gedachtnisses. Archiv ftir gesamte Psychologic, 4:1-32.
Eldridge, R. C. 191 1. Six Thousand Common English Words. Niagara
:
Falls, New York.

Erdmann, B., and Dodge, R.: 1908. Psychologische Untersuchungen
tiber das Lesen, auf Experimenteller Grundlage. Halle.
Fillers, H. D.: 1917. Oral and Written Errors in Grammar. Educa-

tional Review, 54:458-470.
Finkelstein, I. E.: 1913. The Marking System in Theory and Practice.
Foster, F. M.: 1917. The Results of a Recent Spelling Test at the
University of Iowa. School and Society, 5:506-508.
BIBLIOGRAPHY 455
Foster,W. T.: 1911. Administration of the College Curriculum.

Fracker, G. C.: 1908. On the Transference of Training in Memory.
Psychological Review Monographs, No. 38:56-102.
Freeman, F. N.: 190-. Untersuchungen iiber den Aufmerksamkeitsum-
fang und die Zahlanffassung. Padagogische-Psychologische Ar-
beiten, 1:88-168.
Freeman, F. N.: 1914. Practical Studies of Handwriting. Elementary
School Teacher, 14:167-179.
Freeman, F. N.: 1914. The Teaching of Handwriting.
Freeman, F. N.: 1915. Handwriting. Fourteenth Yearbook of the
National Society for the Study of Education.
Freeman, F. N.: 1916. The Psychology of the Common Branches.
Friedrich, J^: 1897. Untersuchungen iiber die Einflusse der Arbeits-
dauer und der Arbeitspausen auf die geistige Leistungsfahigkeit der
Schulkinder. Zeitschrift fiir Psychologic, 13:1-53.
Fulton, M. J.: 1914. An Experiment in Teaching Spelling. Ped. Sem.,
21:287-289.
Galton, F. : 1869 and 1892. Hereditary Genius: An Inquiry into its Laws
and Consequences.
Galton, F.: 1883. Human Faculty and its Development.
Inquiries into
Gesell, A. L.: 1906. Accuracy in Handwriting, as Related to School
Intelligence and Sex. American Journal of Psychology, 17 1394-405.
Gilbert, J. A.: 1894. Researches on the Mental and Physical Develop-
ment of School Children. Studies from the Yale Psychological
Laboratory, 11:40-100.
and Fracker, G. C.: 1897. The Effects of Practice in Re-
Gilbert, J. A.,
action and Discrimination for Sound upon the Time of Reaction and
Discrimination for other Forms of Stimuli. University of Iowa
Studies in Psychology, 1 162-76.
Gist, A. S.: 1917. Errors in the Fundamentals of Arithmetic. School
and Society, 7:175-177.
Goddard, H. H.: 1912. The Kallikak Family.
Goddard, H. H.: 1914. Feeble-mindedness: Its Causes and Conse-
quences.
Graves, S. M.: 1917. A Study in Handwriting. J. of Educ. Psych.,
7:483-494.
Gray, C. T.: 1913. Variations in the Grades of High School Pupils.
Gray, C. T.: 1915. A
Score Card for the Measurement of Handwriting.
Bulletin of the University of Texas, No. 37.
Gray, W. S.: 1917. Studies of Elementary School Reading Through
Standardized Tests. Suppl. Educ. Monog., i, No. i.
Gulick, L. H., and Ayres, L. P.: 1908. (Second Edition, 1913.) Medical
Inspection of Schools.
Guyer, M. F.: 1916. Being Well-Born.
456 BIBLIOGRAPHY
Haggerty, M. Scales for Reading Vocabulary of Primary

E.: 1917.
Elementary School Journal, 1 7 104-1 1 5.
Children. :
Hahn, W. H., and Thorndike, E. L.: 1914. Some Results in Addition

under School Conditions. Journal of Educ. Psych., 5:65-84.
Haines, T. H.: 1916. Mental Measurements of the Blind. Psych.
Rev. Monog., 21, No. i.
âll, G. S.:1883. Published first in Princeton Review, 2:240-272.
"
Then reprinted in Aspect of Child Life and Education," 1907.
Hall, G. S.: 1904. Adolescence.
Hall, G. S.: 1908. Youth.
Hall-Quest, A. L.: 1917. Supervised Study.
Hamilton, F. M.: 1907. The Perceptual Factors in Reading. Archives
of Psychology, No. 9.
Harris, J. H. and Anderson, H. W.: 1916. Measuring Primary Reading
in Dubuque Schools (Iowa).
Harris, L. H.: 1915. A
Study in the Relation of Latin to English Com-
position. School and Society, 2:251-252.
Heck, W. H.: 1913. A Study of Mental Fatigue.
Heck, W. H. 1913. A Second Study of Mental Fatigue in Relation to the
:
Daily School Program. Psychological Clinic, 7 :29~34.

Henmon, V. A. C.: 1910. Sex Differences and Variability in Color Per-
ception. University of Colorado Studies, 7, No. 4.
Henmon, V. A. C.: 1912. The Relation between Mode of Presentation
and Retention. Psychological Review, 19:79-96.
Hewins, N. P. 1916. The Doctrine of Formal Discipline in the Light of
:
Experimental Investigation.
Hillegas, M. B. 191 2. A Scale for the Measurement of Quality in English
:
Composition by Young People. Teachers' College Record, 13, No. 4.

Hoge, Mildred, and Stocking, Ruth, J.: 1912. A Note on the Relative
Value of Punishment and Reward as Motives. Journal of Animal
Behavior, II, 43.
Holiingworth, H, L.: 1912. Correlation of Abilities as Affected by Prac-
tice; Journal of Educational Psychology, Jan., 1912.
Holiingworth, H. L.: 1914. Individual Differences, During and After
Practice. Psychological Review, January, 1914.
Holloway, H. V.: 1917. The Relative Difficulty of the Tables, Tren-
ton, N. J.
Horn, E.: 1917. Possible Defects in the Present Content of American
History as Taught in the Schools. Sixteenth Yearbook of the
National Society for the Study of Education.
Hosic, J. F. 191 5. The Essentials of Composition and Grammar. Four-
:
teenth Yearbook of the National Society for the Study of Education.

Houser, J. D.: 1917. An Investigation of the Writing Vocabularies of
Representatives of an Economic Class. Elem. School Journal,
17:708-718.
BIBLIOGRAPHY 457
Howell, H. B.: 1914. A Fundamental Study in the Pedagogy of Arith-

metic.
Hoyt, F. S. : 1906. Studies in the Teaching of English Grammar. Teach-
ers'College Record, 7, No. 5.
Huey, E. B.: 1898 and 1900. Experiments in the Physiology and Psy-
chology of Reading. American Journal of Psychology, 9:575-586;
11:283-302; 12:292-312.
Huey, E. B.: 1910. The Psychology and Pedagogy of Reading.
Hull, C. L.: 1919. Quantitative Aspects of the Evolution of Concepts.
Psychological Review Monograph Supplements, 1919.
Hull, C. L., and Montgomery, R. B.: 1919. An Experimental Investi-
gation of Certain Alleged Relations between Character and Hand-
writing. Psych. Rev., 26:63-74.
Jacoby, P.: 1881. (Second Edition, 1904.) fitudes sur la selection dous
ses rapports avec I'heYeclitS chez Thomme.
James, W.: 1890. Principles of Psychology.
Javal, E.: 1878 and 1879. Sur la Physiologic de la Lecture. Annoles d'
Oculistique.
Javal, E.: 1879. Conditions de la lecture facile. Comptes rendus de la
Societie de Biologic, page 8.
Jessup,W. A.: 1915. Current Practices and Standards in Arithmetic.
Fourteenth Yearbook of the National Society for the Study of
Education, 14:116-130.
Johnson, R. I. 1917. The Persistency of Errors in English Composition.
:
School Review, 25:555-580.

Jones, W. F.: 1913. Concrete Examination of the Material of English
Spelling.
Jones, W. F.: 1915. Spelling Book.
Jost, A.: 1897. Die Assoziationsfestigkeit in ihrer Abhangigkeit von der
Verteilung der Wiederholungen. Zeitschrift ftir Psychologic, 14:
436-472.
Judd, C. H., and McAllister, C. N., and Steele, W. M. : 1005. Analysis of
Reaction Movements. Psychological Review Monographs, 7:141-184.
Judd, C. H.: 1908. The Relation of Special Training to General In-
telligence. Educational Review, 36:28-42.
Judd, C. H.: 1915. The Psychology of High School Subjects.
Judd, C. H.: 1916. Measuring the Work of the Public Schools.
Judd, C. H. (Edited by) : 1916. School Survey, Grand Rapids, Michigan.
Kelley, T. L.: 1914. Educational Guidance. Columbia University.

Kelly, F. J.: 1915. The Kansas Silent Reading Test. Bulletin of State
Normal School, Emporia, Kansas.
Kerr, Mary A.: 1916. The Effects of Six Weeks of Daily Drill in Addi
tion. Indiana University Studies.
453 BIBLIOGRAPHY
King, I.: 1917. Comparison of Ra?j>id and Slow Readers in Comprehen-

School and Society, 4:330-334.
sion.
Kirby, T. J.: 1913. The Results of Practice under School Conditions.

Columbia University.
Kirkpatrick, E. A.: 1914. An Experiment in Memorizing vs. Incidental
Learning. Journal of Educ. Psych., 5:405-413.
Kline, L. W.: 1909. Some Experimental Evidence on the Doctrine of
Formal Discipline. Bulletin of State Normal School, Duluth, Minn.
Kraepelin, E., and Hoch, A. : 1895. Ueber die Wirkung der Theebestand-
theile auf korperliche und geistige Arbeit. Psychologische Arbeiten,
1:378-488.
Kraepelin, E., and Hylan, J. P.: 1902. Ueber die Wirkung kurzer
Arbcitszciten. Psychologische Arbeiten, 4:454-494.
Kraepelin, E., and Rivers, W. H. R.: 1896. Ueber Ermtidung und Er-
holung. Psychologische Arbeiten, i :378-488.
Kraepelin, E.: 1902. Die Arbeitscurve. Philsophische Studien, 19:459-
507-
Kraepelin, E.: 1903. Ueber Ermudungsmessungen. Archive ftir die
gesamte Psychologic, i 19-30.
Kuhlmann, F. 191 2. Revision of the Binet-Simon System for Measuring
:
the Intelligence of Children. Journal of Psycho-Asthenics, Mono-

graph Supplements, i, No. i.
Laser, H.: 1894. Ueber geistige Ermtidung beim Schulunterricht.

Zeitschrift fur Schulgesundheitspflege, 7:2-22.
Lay, W. A.: 1898. Fuhrer durch den ersten Rechenunterricht. Wies-
baden.
Lay, W. A.: 1907. Fuhrer durch den Rechenunterricht der Unterstufe.
Laudolt: 1891. Nouvelles recherches sur la physiologic des mouvements
des yeux. Archives d'Ophthalmologie, 2:385-395.
Lindley, E. H.: 1897. A Study of Puzzles. American Journal of Psy-
chology, 8:431-493.
Llewelyn, E. J.: 1916. Reading in the Mt. Vernon Schools. Elementary
School Journal, 17:123-127.
Lodge, G.: 1915. Chapter X. Principles of Secondary Education.
Edited by Paul Monroe.
Loisette, A. 1 899. Assimilative Memory.
:
Lowell, A. L.: 1910. College Studies and the Professional School.

Harvard Graduates' Magazine, Dec., 1910; also Educational Re-
view, Oct., 1911.
Lueba, J. H., and Hyde, W. 1905.
: An Experiment in Learning to Make
Hand Movements. Psychological Review, 12:351-369.
Lyon, D. O.: 1913. Relation of Length of Material to Time Taken for
Learning and the Optimum Distribution of Time. Journal of Edu-

cational Psychology, 5:155-163.
BIBLIOGRAPHY 459
Matthews, B.: The Study of Fiction in Counsel upon the Reading of

Books by Henry VanDyke.
Mead, C. D.: 1915. Silent versus Oral Reading with 100 Sixth Grade
Children. Journal of Educational Psychology, 6:345-348.
Mead, C. D.: 1917. Results of Silenfversus Oral Reading. Journal of
Educational Psychology, 8:367-368.
Mead, C. D., and Scars, Isabel. 1916: Additive, Subtraction and Mul-
tiplicative Division Tested. Journal of Educ. Psych., 7:261-270.
Messenger, J. F.: 1903. The Perception of Number. Psychological
Review Monographs, 5:123-144.
Meumann, E. :
1914. (First Edition, 1907.) Vorlcsungen zur EinfUrhung
in die Experimentelle Padagogik und ihre Psychologischen Grund-
lagen. Vols. I, II, and III.
Meyer, M.: 1908. The Grading of Students. Science, 28:243-250.
Minnick, J. H.: 1913. An Experiment in the Supervised Study of
Mathematics. School Review, 21:670-675.
Monroe, W. S.: 1917. A Preliminary Report of an Investigation of the
Economy of Time in Arithmetic. Sixteenth Yearbook of the Na-
tional Society for the Study of Education, 1917:111-127.
Mosso, A.: Fatigue.
Munn, A. F.: 1909. The Curve of Learning. In Studies in Development
and Learning. Edited by E. A. Kirkpatrick, Archives of Psychology,
No. 12.
Nanu, H. A.: 1904. Zur Psychologic der Zahl-Auffassung. Wurzberg.

Nicholson, F. W.: 1915. Success in College and in After Life. School
Oberholtzer, E. E.: 1914. Testing the Efficiency of Reading in the

Grades. Elementary School Jo urnal, 1 5 13 1 3-3 2 2 .
Partridge, E. A.: 1915. High-School Latin A New Phase of an Old

Subject. Classical Journal, 10:404-412.
Pearson, K.: 1904. On the Laws of Inheritance in Man. II. On the
Inheritance of the Mental and Moral Characters in Man, etc.
Biometrika, 3:131-190.
Pearson, K. : 1907. On the Relationship of Intelligence to Size and Shape
of Head, and to Other Physical and Mental Characters. Biometrika,
5:105-146.
Perkins, A. S.: 1914. Latin as a Vocational Study in the Commercial
Course. Classical Journal, 10:7-16.
Peters, C. C.: 1917. Influence of Speed Drills on Silent Reading. Jour-
nal of Educational Psychology, 8:350-366.
Peterson, H. A.: 1912. Note on a Retrial of Professor James' Experi-
ment on Memory Training. Psych. Rev., 19:491-492.
460 BIBLIOGRAPHY
Phelps, C. L.: 1913. A Study or Errors in Tests of Adding Ability.

Elem. School Journal, 14:29-39.
Phelps, W. L.: Teaching in School and College.
Phillips, F. M.: 1913. Value of Daily Drill in Arithmetic. Journal of
Educ. Psych., 4:159.
Pintner, R.: 1913. Oral and Silent Reading of Fourth Grade Pupils.
Journal of Educational Psychology, 4:333-337.
Pintner, R., and Paterson, D. G.: 1916. Psych. Rev. Monog., 20: No. 4.
Pintner, R., and Paterson, D. G.: 1917. A Scale of Performance Tests.
Proctor, M.: 1917. Supervised Study on the Pacific Coast. School
Pryor, H. C: 1915. Spelling. Fourteenth Yearbook of the National
Society for the Study of Education, 1915:78-89.
Puffer: 1912. The Boy and His Gang.
Pyle, W. H. :
1913. Economical Learning. Psychological Bulletin, 10:73.
Radassawljewitsch, P. R.: 1907. Das Behalten und Vergessen bei

Kindern und Erwachsenen nach experimentcllen Untersuchungen.
Rice, J. M.: 1897. The Futility of the Spelling Grind. The Forum, 23:
163-172, 409-419-
Rice, J. M.: 1902. Educational Research: A Test in Arithmetic. The
Forum, 34:281-297; Causes of Success and Failure in Arithmetic.
The Forum, 34:437-452.
Riley, J. L.: 1008. The Springfield Tests, 1846-1905, 1906.
Ritter, C.: 1900. Ermudungsmessungen. Zeitschrift fur Psychologic,
etc., 24:401-444-
Ruediger, 1907. The Field of Distinct Vision. Archives of Psychology,
No. 5.
Ruger, II. A.: 1910. The Psychology of Efficiency. Archives of Psy-

chology, No. 15.
Rugg, H. O.: 1916. The Experimental Determination of Mental Dis-
cipline in School Studies.
Sackett, L. W. Measuring a
: School System by the Buckingham Spelling
School and Society, 2:860-864, 894-898.
Scale.
Schmidt, W. A.: 1917. An Experimental Study in the Psychology of
Reading. Suppl. Educational Monographs (Chicago), i, No. 2.
Schuster and Elderton: 1907. Inheritance of Ability. Publications of
Eugenics Laboratory, London.
Schuyler, W., and Swift, E. J.: 1907. The Learning Process. Psy-
chological Bulletin, 4:307-310.
Scripture, E. W., Smith, T. L., and Brown, E. M.: 1894. On the Educa-
tion of Muscular Control and Power. Studies from the Yale Psy-
chological Laboratory, 2:114-119.
Sears, J. B.: 1915. Spelling Efficiency in the Oakland Schools. School
and Society, 2:531-537, 569-574-
BIBLIOGRAPHY 461
Scars, Isabel, and Diebel, Amelia: igr6. A

Study of Common Mistakes
in Pupils' Oral English. El. School Journal, 17:44-54.
Seashore, C. E. : 1 901 Suggestions for Tests on School Children. Educa-

.
tional Review, 22:69-82.

Seashore, C. E., and Kent, G. H.: 1965. Periodicity and Progressive
Change in Continuous Mental Work. Psychological Review
r
T
'nographs.
Sikorski, J.: 1879. Sur les effects de la lassitude provoquee par Ics
travoux intcllectuels chez cnfants le Tage scollaire. Annales d'hy-
giene publique, 2 1458-464.
Simpson, B. R.: 1912. Correlations of Mental Abilities. Columbia
University.
Sleight, W. G.: 1911. Memory and Formal Training. British Journal
of Psychology, 4:386-457.
Smledey, F.: 1900-1901. Rept. Dcpt. Child Study and Pedagogic In-
vestigation, No. 3.
Smith, F. O
191 2. : A
Rational Basis for Determining Fitness for College
Entrance. University of Iowa Studies in Education, N. S. 51.
Sneed, C. M., and Whipplc, G. M.: 1909. An Examination of the Eyes,
Ears, and Throats of Children in the Public Schools of Jefferson
City, Missouri. Psychological Clinic, 2 :234~238.
Spillman, W. J.: 1909. The Country Boy. Science, 29:739-741 ; 30:405-
407.
Starch, D.: 1910. A Demonstration of the Trial and Error Method of
Learning. Psychological Bulletin, 7:20-23.
Starch, D.: 1911. Experiments in Educational Psychology, 1911.
Revised, 1917.
Starch, D. 191 1: Unconscious Imitation in Handwriting. Psychological
.
Review, 18:223-228.
Starch, D.: 1911. The Transfer of Training in Arithmetical Operations.
1

Starch, D 191 2
. : Periods of Work in Learning. Journal of Educat ional
.
Psychology, 3:209-213.
Starch, D. 1913. The Measurement of Handwriting.
:
Journal of Educa-
tional Psychology, 4:445-464.
Starch, D.: 1913. Reliability and Distribution of Grades. Science
38:630-636.
Starch, D.: 1913. Correlation Among Abilities in School Studies. Jour-
nal of Educational Psychology, 3:415-418.
Starch, D.: 1915. The Measurement of Efficiency in Reading. Journal
Starch, D.: 1915. The Measurement of Efficiency in Handwriting.
Starch, D.: 1915. The Measurement of Efficiency in Spelling. Journal
462 BIBLIOGRAPHY
Starch, D.: 1915. The Measurement of Ability in English Grammar.

Journal of Educational Psychology, 6.
Starch, D.: 1915. Can. the Variability of Marks be Reduced? School
Starch, 0.11915. The Inheritance of Abilities m school Studies. School
Starch, D.: 1915. Some Experimental Data on the Value of Studying
Foreign Languages. School Review, 23:697-703.
Starch, D.: 1916. The Measurement of Ability in Arithmetic. Journal
of Educational Psychology, April, 1916.
Starch, D.: 1917. The Estimated Value of School Subjects. School and
Society, 5:59-60.
Starch, D.: 1917. Further ExperimentalData on the Value of Studying
Foreign Languages. School Review, 25:243-248.
Starch, D.: 1917. Similarity between Brothers and Sisters in Mental
Traits. Psychological Review, 24:235-238.
Starch, D., and Ash, I. E.: 1917. The Mental Work Curve. Psycholog-
ical
Review, 24:391-402.
Starch, D., and Elliott, E. C.: 1912. The Reliability of Grading High
School Work in English. School Review, 20:442-457.
Starch, D., and Elliott, E. C. 1913. The Reliability of Grading Work in
:
Mathematics. School Review, 21:254-259.

Starch, D.,and Elliott, E. C.: 1913. The Reliability of Grading Work in
History. School Review, 21:676-681.
Stone, C. W.: 1908. Arithmetical Abilities and Some Factors Deter-
mining Them. Columbia Contributions to PMucation, No. 10.
Studebaker, W.: 1916. Economy Practice Exercises in Arithmetic.
J.
Scott, Forcsman & Company.
Suzzallo, H. 1910. The Teaching of Spelling.
:
Suzzallo, H., and Pearson, H. C.: 1913. Comparative Experimental

Teaching in Spelling. Teachers* College Record, 13: No. i.
Swift, E. J.: 1908. The Mind in the Making.
Swift, E. J.: 1912. Youth and the Race.
Taussig, A. E. 1909. The Prevalence of Visual and Aural Defects among

:
the Public School Children of St. Louis County, Missouri. The

Psychological Clinic, 3:140-160.
Terman, L. M.: 1916. The Measurement of Intelligence.
Tcrman, L. M.: 1917. The Intelligence Quotient of Francis Galton in
Childhood. American Journal of Psychology, 28:210.
Thompson (Woolley), H. B.. 1903. Psychological Norms in Men and
Women. University of Chicago Contributions to Philosophy,
IV, No. i.
Thorndike, E. L.: 1900. Mental Fatigue, Psychological Review, 7:466-
482, 547-579-
BIBLIOGRAPHY 403
Thorndike, E. L.: 1905. Measurements of Twins. Archives of Phil.

Psychology and Scientific Methods, No. i.
Thorndike, E. L.: 1910. Handwriting. Teachers' College Record, n:
No. 2.
Thorndike, E. L.: 1910. Practice in the Case of Addition. American

Journal of Psychology, 21:483-486.'
Thorndike, E. L.: IQII. Mental Fatigue. Journal of Educational
Psychology, 12:61-80.
Thorndike, E. L.: 1912. Education.
Thorndike, E. L.: 1912. The Permanence of Interests and Their Rela-
tion to Abilities, Popular Science Monthly, Nov., 1912.
Thorndike, E. L.: 1912. The Curve of Work. Psychological Review,
19:165-194.
Thorndike, E. J^. :
1914. Educational Psychology, Vol. I. The Original
Nature of Man. Vol. II. The Psychology of Learning. Vol. III.
Mental Work and Fatigue and Individual Differences and Their
Causes.
Thorndike, E. L.: 1914. The Measurement of Ability in Reading.
Teachers' College Record, 15, No. 4.
Thorndike, E. L.: 1915. The Disciplinary Values of Studies: A Census
of Opinions. Education, 35:278-286.
Thorndike, E. L.: 1915. The Relation between Speed and Accuracy in
Addition. Jour, of Educ. Psych., 5:537.
Thorndike, E. L.: 1917. The Curve of Work and the Curve of Satisfy-
ingness, Journal of Applied Psychology, i :265~267.
Trabue, M. R.: 1916. Completion-test Language Scales. Columbia
University Contributions to Education, No. 77.
Uhl, W. T.: 1916. Use of the Results of Reading Tests as Basis for
Planning Remedial Work. Elementary School Journal, 17:266-
275-
Waldo, K. D.: 1914. Tests in Reading in the Sycamore Schools, Ele-

mentary School Journal, 15:251-268.
Wallin, J. E. W. 1911 Spelling Efficiency in Relation to Age, Grade and
: .
Sex, and the Question of Transfer.

Walsemann, H. J.: 1907. Anschauungslehre der Rechenkunst.
Warren, H. C.: 1897. The Reaction Time of Counting. Psychological
Review, 4:569-591.
Webb, L. W.: 1917. Transfer of Training and Retroaction. Psycholog-
ical Review Monographs, No. 104. 24:
Wells, F. L. 1 908. Normal Performance in the Tapping Test. American
:

Wells, F. L.: 1912. The Relation of Practice to Individual Differences.
American Journal of Psychology, 23:75-88.
464 BIBLIOGRAPHY
West, A. F. (Editor) 191 7. : Value of the Classics, Princeton University

Press.
Whipple, G. M. 1910. Manual of Mental and Physical Tests.

:
Whipple, G. M.: 1916. How to Study Effectively. Public School

Publishing Co., Bloomington, III.
Whitley, M. T.: 1911. An Empirical Study of Certain Tests for In-
dividual Differences. Archives of Psychology, No. 19.
1
Wilbur, Flo ra 1916. Experiments with Courtis Practice Pads. Indiana

:
University Studies.
Wilcox, M. J.: 1917. Does the Study of High School Latin improve
High School English? School and Society, 6:58-60.
Wilson, G. M.: 1917. A Survey of the Social and Business Use of Arith-
metic. Sixteenth Yearbook of the National Society for the Study
of Education, 1917:128-142.
Wimmer, H.: 1916. An Experimental Study of the Effects of Drill in
Arithmetical Processes under Varying Conditions. Indiana Univer-
sity Studies.
Winch, W. H.: 1904. Immediate Memory in School Children. British
Winch, W. H.: 1906. Immediate Memory in School Children: Auditory*
British Journal of Psychology, 2:52-57.
Winch, W. II. :
1908. The Transfer of Improvement in Memory in School
Children: I. British Journal of Psychology, 2:284-293.
Winch, W. H.: 1910. The Transfer of Improvement in Memory in
School Children: H. British Journal of Psychology, 3:386-405.
Winch, W. H.: 1910. Accuracy in School Children. Does Improvement

in Numerical Accuracy Transfer? Journal of Educational Psy-
chology, 1:557-589.
Winch, W. H.: 1911. Further Work in Numerical Accuracy. Journal of
Educational Psychology, 2:262-271.
Winship, A. E.: 1900. Jukes-Edwards: A Study in Education and
Heredity.
Woodworth, R. S.: 1899. The Accuracy of Voluntary Movement. Psy-
chological Review Monographs, No. 13.
Woodworth, R. S., and Thorndike, E. L.: 1901. The Influence of Im-
provement in One Mental Function upon Other Functions. Psy-
chological Review, 8:247-261; 384-395; 553-564-
Woods, F. A.: 1906. Mental and Moral Heredity in Royalty.
Woody, C.: 1916. Measurements of Some Achievements in Arithmetic.
Columbia University Contributions to Education, No. 80.
Yerkes, R. M., Bridges, J. M., and Hardwick, R. S.: 1915. A Point-

Scale for Measuring Mental Ability.
INDEX
Abbott, E. E., 167. Auditory defects, frequency, 128.
Aiken, 265. Aussage tests, 133.
Aldine system of reading, 290, 291, Ayres, scale in writing, 35; in spelling,
293, 294. 324-
Ames, W. R., 263, 264.
Anderson, H. W., 290. ft W. C., 224, 247, 424, 425.
Angell, 198, 200. Bair, J. H., 161, 198.
\
Apperception, doctrine of, 139; eval- Baldwin, B. T., 59, 60.
uation of, 140. Ballard, P. B., 407, 408.
Arai, T., 173, 174. Ballou, F. W., 353.
Arithmetic; steps involved in, 374*".; Barr, Martin W., 80.
number concept, 375; establish- Batavia plan, 42.
ment of associations, 376; clue as to Balsonf W. H., 146, 147, 150, 161.
process required, 376f.; number Beacon system of reading, 290, 291,
preferences, 379f.; methods of meas- 292, 293.
uring efficiency in, 38 if.; economic Bean, C. II., 157.
methods in learning, 383^; acquiring Beetz, 384, 385.
number concept, 383^; number pic- Bergstrom, J. A., 200.
tures, 384f.; operations to be learned, Betts, G. II., 167, 423.
386f.; length of class period, 391; Biglow, R. P., 218.
certain environmental factors, 3951.; Binet, H., 99, 300.
optimum distribution of
drill, 396f.; Binet-Simon Tests^ 69f.; 99f.; Ter-
drill, 402f.; special methods of drill, maji's revised scale, ioof., 107; nec-
405f.; speed versus accuracy, 4O9f.; essary qualifications for tester;
limits of attainment, 410; errors, measurements resulting from use of,
41 of.; relative difficulty of combina- io7f.
tions, 414. Birge, E. A., 441, 443-
Arithmetic tests: Boas, F., 18.
Courtis, 20, 35, 37, 89, 165, 166, 381, Bolton, T. L., 171.
400, 408, 410, 412. Book, W. F., 144, 145, 146, 150, 157,
Starch, 382, 383, 39& 161.
Stone, 382. Born, 384, 385, 386.
Woody, 382. Born, Busse, and Belune, 384.
Judd and Counts, 382. Borst, M., 133, 136.
Arithmetical reasoning, 20; Fig. 4, Breslich, E. R., 189.
Fig. 5, 21; Fig. I9,'3$. Briggs, T. H., 225, 226.
Arps, G. F., 407. Brown, H. A., 274, 275, 289.
Association, see Learning, Spelling, Brown, J. C., 398.
Correlation. Brown, M. D., 353, 354, 372.
Attention, largely controlled by in- Brown, J. Stanley, 189.
stinct, 13; control of in studying, Brinckerhoff, Morris and Thorndike,
180. 55-
466 INDEX
Bryan, W. L., and Harter, N., 142 Curve of learning, 141 ;

characteristics
143, 150, 151, 153, 158. yf, i4if.; initial rise in, i43f.; pla-
Buckingham, B. R., 325., teaus, i5of.
Burk, C. F., 22.
Burnett, C. J., 384. Dnllam, M. Ther~i., 246.
Burns, 55. Duvidson, P. E., 24.
Burt, C., 53, 54, 59, no. Davis, W. W., 210.
Dearborn, W. F., 60, 154, 161, 162,
Cambridge plan, 44. 177, 201, 202, 206, 207, 262, 265,
Capacities, definition of, 10; variation 266, 267, 268, 269, 428, 430, 444-
in, see Individual Differences. De Bruin, L. C., 328.
Carter, M. H., 123. De Candolle, A., 92.
Cattell, J.M., 92, 384. Dexter, E. G., 178.
Charters, W. W., 357, 363- Dexter, Emily S., 84, 85.
Charters, W. W., and Miller, Edith, Diebel, Amelia, 362, 363, 365.
362. Dietze, G., 384.
Classical versus non-classical students, Distribution curve, 28, 30, and as-
236, 237!. signment of marks, 443f.; memory
Cliborne, J. H., 121. ability, 26; A-test, 27; cancellation,
Coffman, 387, 388. 27; association, 28; chest measure-
Cohn, 125, 127. ments, 30; height of women, 31;
Color-blindness, 124, 126. head girth of boys, 31; of tossings
Colvin, S. S., 168. of pennies, 32; and marks, 443-448.
Commenius, 224. Distribution of mental abilities, 29.
Conrad, H. E., 407. Distribution of practice, i53f.
Cook, W. A., 333, 334. Dodge, R., 266.
Coover, J. E., 196, 198, 200, 206, 208, Dougherty, M. L., 109.
209. Downey, J. E., 298, 300.
Cornell, W. S., 129. Drill in writing, 3i9f.; in spelling,
Cornman, O. P., 337, 338, 339, 340, 339f.; in arithmetic, 396f.
395- Drushel, J. A., 409.
Correlation: problem stated, 49; co- Dugdale, R. G., 77.
efficient of, 49f.; among specific Duquid, 290.
mental abilities, 5of.; among abil- Durr, 124.
school subjects, 54f.; conclu-
ities in
sion concerning, 54, 57f.; between Earle, E. L., 82.
special capacities and general intel- Ebbinghaus, H., 69, 153, 157, 171.
ligence, between mental and
59; Ebbinghaus test, no.
physical traits, 59; between early Ebert, E., and Meumann, E., 161,
and later mental abilities, 6of. 201, 202.
Coubal, L. J., 89, 164. Edison, T. A., 377.
Counts, G. S., 382. Education: Definition of, i; funda-
Courtis, S. A., 274, 275, 286, 382, meriTaTproblems of, 2.
406. E. C., 45, 433, 434-
Elliott,
Courtis tests, 20, 35, 37, 89, 165, 166, Ellis, 92.
381, 400, 408, 410, 412. English: methods of measuring effi-
Craig, Helen, 43. ciency in, 35 2f.; results of measure-
Cross education, 21 of. ments in, 353f.; economic methods
fculture epochs theory, 25. of acquiring skill in, 354^; acquisi-
tion of ideas, 355; of words and
INDEX; 467
forms, 355; grammar and correct Gessell, A. L., 300.

English, 356!.; imitation in acquir- Gilbert, J. A., 67, 71; and Fracker,
ing, 357; specific attention to errors G. C., 194, 195-
in grammar, 361!.; oral versus writ- Gilman, Charles, 122, 123, 114.
ten practice, 367!.; good English in Gist, A. S.,'4i2.
all classes, 371!.; types o'f topics Goddard, H. H., 77, 78, 79, 100, 107,
which should be given, 372; effects '130.
of differing teaching ability, 373; Gormley, 241.
and Latin, see Latin. Graves, S. M., 306, 307.
Environment, see Environmental In- Gray, C. T., 321, 431, 432.
fluence. Gray, W. S., 265, 268, 274, 275, 294.
Environmental influence on different Griesbach, 170.
original abilities, 88; experiments Gross, 299.
concerning, 88f.j conclusions con- Gulick, L. H., and Ayres, L. P., 127,
cerning, gi. 128, 130.
Erdmann, B., 2J56. . Guyer, M. F., 77, 79.
Extra- work plan, 43.
Haggcrty, M. E., 275, 353, 354, 372,
Family resemblances, see Mental 400.
Heredity. Hahn, W. H., 404, 405.
Faraday, 377. Haines, T. H., 107.
Far-sightedness, 123. Hall l< G._Stanley, 17, 19, 23, 135, 265.
Fatigue, methods of studying, 170; Hall-Quest, A. L., 189, 190.
""'"experiments on, 17 if.; and school Hamilton, F. M., 270.
work, 175; see also Progress of Handwriting: steps involved in, 297;
Learning. sex differences in, 299^, 303^; cor-
Feeblemindedness, inheritance of, see relation of with other traits, 300;
Mental Heredity. measurement of efficiency in, 301 f.;
Fillers, H. D., 362, 363. economic procedure in learning to
Finkelstein, I. E., 428, 440, 443. write, 304^; perception of forms in,
Fordyce, 274, 275. 304f.; length of practice periods,
Forgetting, is6f. 307f.; standard of proficiency in,
T^râ^jd^scipljneji 217, 252. 3iof.; relation of speed to quality
Foster, F. M., 235, 236, 325. in,3iif.; methodsof teaching, 3 14!".;
Foster, J. W., 218. factors affecting execution of move-
Foster, W. T., 177, 428, 4291 44i, 442, ments in, 3i7f.; movement drills,
444. 319; correct form in, 320; analysis

Fracker, G. C., 161, 194, 195, 202, of imperfections in, 32of.
203, 212. Handwriting scales, Thorndike, 301,

Franklin, 360. 303, 310.
Freeman, F. N., 282, 299, 306, 308, Ayres, 301, 303, 310, 312, 313, 316.
309,310,312,313,317,318,321,384- Starch, 301, 303.
Friedrich, J., 171, 172. Freeman, 302.
Fulton, M. J., 342. Palmer, 302, 321.
Zaner, 302, 321.
Galton, Sir Francis, 74, 85, 86, 105. Gray, 321.
General intelligence and special ca- Harris, J. H., 200.
pacities, 109; results of Simpson's Harris, L. H., 234, 235.
experiment on, 1 10; Hurt's study of, Heck, W. H., 172, 173.
ixof.; conclusion concerning, inf. Height and age, Fig. i, 18.
468 INDEX
Hendricks, 294. Individual instruction plan, 42, 44.

Henmon, V. A. C., 70, 71, 84, 164^ Infallibility of instincts, 15!.
168, 169, 281. Ingersoll, L. R., 351.
Hentschel, Butz, Sobelewsky and Instinct, and motivation, 12; collect-
Kaselitz, 384. ing instinct and age, Fig. 6, 22.
Herbart, 139. Instincts, defined, 9; and reflexes, 9;
heredity, see Mental Heredity.

Jid capacities, 10; classified, iof.;
HewTRs, N. P., 247, 249. )veremphasis of in education, nf.;
Hicks, W. E., 391. die in education, nf; sudden de-
Hillegas scale, 35, 39, 40, 41, 245. velopment of, i9f.; unrevivability
History: psychological steps involved of, 2 if.
in, 416!.; measurement of attain- Instincts, based upon dy-
theories
ment in, 4i8f.; economic methods lamic theory, 14, isf.; transited-
in learning, 42of.; suggestions for less, 14, 17; recapitulation, 14,
method of study, 42 2f.; essential Z2f.
material in, 423^ Intelligence quotient,. 104; how de-

History tests: termined, iosf.
Starch, 418. Interest in relation to learning, see
Bell and McColIum, 420. Progress of Learning.
Hoge, Mildred and Stocking, Ruth J., Interests, permanency of, 62.
14.
Hollingworth, H. L., 52. Jacoby, P., 92.
Holloway, H. V., 413, 414. James, W., 21, 23, 138, 139, 193, 2x1.
Horace Mann system of reading, 290, 212, 213, 214, 371.
291, 292, 293. Jastrow, J., 379, 381.
Horn, E., 425. Javal, E., 266.

Hosic, J. F., 369. Jessup, W. A., 387, 388.
Houser, J. D., 330. Jevons, 377.
Howe system in reading, 294. Johnson, R. L, 366.
Howell, H. B., 378, 384, 335, 386, Jonathan Edwards family, 77, 80.
410, 411. Jones, W. F., 346.
How to study, 176; waste in studying, Jost, A., 153.
176; value of study, i77f.; types of Mjj^CUH., 199, 213, 214, 215, 264,
studying, 179; problems in study- 285, 286, 306, 313, 314, 3*5> 3*6 ;
ing; see also Studying. 396, 397-

Hoyt, F. S., 226, 227. Jukes family, 77, 80.
Huey, E. B., 266, 270, 281.
Hull, C. L., 148, 149, 263, 264, 301. Kallikak family, 77f.; diagram of de*
Humphrey, 240. scendants, 78.
Hypermetropia, 123. Kansas silent reading test, 275.
Kelly, F. 274, 275.
J.,
Imagery^ i66f. Kelley, T. L., 61, 62.

Ulfflvltttfal differences, 26; quantita- Kent, G. H., 174, 175.
tive nature of, 26; means of graphic Kerr, Mary A., 400.
representation, 28; range of, 28f.; King, Miss, 172.
in reading, 33; in writing, 34; in King, Irving, 235, 285, 286.
spelling, 34; in arithmetical reason- Kirby, T. J., 155, 402, 403, 404, 405.
ing, 35; in addition, 37; in English, Kircher, H. W., 280.
38, 40, 41, 42; in geography, 39; Kirkpatrick, E. A., 406.
in drawing, 40. Kline, L. W., 161, 197.
INDEX 469
Knight, 346. .
Marks: importance of, 426; variations
Knilling, 384. in distribution of, 426f.; variations
Kraepelin, E., 291,. in evaluation of same school subject,
Kuhlmann, F., 100, 108. 433f.; causes of variations, 435!.;
Language, psychological processes in- evaluation of factors involved, 438.;
volved in, 349!. -~ size of units on marking scale, 438f .;
how distribute marks, 44of.; objec-
Landolt, 266. tions to use of distribution curve
Language scales: in assigning marks, 443 f.; methods
Starch, 352, 353. of reducing variations in grades,
Hilligas-Thorndike, 353, 373. 448f.
Harvard-Newton, 353. Matthews, Brander, 361.
Trabue, 353. Measurement of mental capacities, 97;
Laser, H., 171. value of, 97f.; methods of, 99, 109.
Lathrop, G. C., 378. Mead, C. D., 287, 288, 289, 408.
Latin: general value of, 229; effect of Memorizing ability, regular increase
study of modern language, 23 if.; during school life, igi.
and scholarship, 232; effect on Eng- Memory, see Correlation, Transfer of
lish, 233^, 24if., 246f., 356; on rhet- Training, Learning.
oric, 235; and college honors, 236f.; Mental heredity; problem of, 73;
and original capacity, 238f. methods of studying, 73; views of,
Lay, W. A., 384, 385, 36. 74; Galton's study of, 74f.; Wood's
Learning: problems in rate and prog- study of in royalty, 75; in various
ress of, 141; curve of, 141; early low grade families, 77f.; in the
progress in, 143; various curves of, Jonathan Edwards family, 77; its
142-149; analytic types of, 148; effect on degeneracy and crime, 79;
distribution of practice in, iS3f.; on feeblemindedness, 78, 79f.; gen-
plateaus in, 142, 149, i5of.; see also eral interpretation, 95f.
Progress in Learning and Curve of Messenger, J. F., 384.

Learning. Methods of teaching, their relation to
Learning process: analyzed, iisf.; psychology, 3.
common and special elements in, Meumann, E., 2oif., 299.
118; general versus special, 119; Meyer, M., 426, 427, 444.
problems concerning, ii9f. Mill, John Stuart, 97.
Length of practice periods, see Prog- Minnick, J. H., 190.
ress of Learning. Monitorial group plan, 42, 43.
Liddle, Carrie W., 200. Monroe, Paul, 230.
Lincoln, 30. Monroe, W. S., 275, 391.
Lindley, E. H., 377. Mosso, A., 171.
Llewelyn, 289. Motivation in learning, 165.
Lodge, G., 229, 230. Mueller, A. D., 273.
Loisette, A., 422. Munn, A. F., 154.
Lounsbury, 369. Musical discrimination, 130.
Loveland, 190. Myers, E. J., 168.
Lueba, J. H., and Hyde, W., 153- Myopia, 123.
Lyon, D. O., 156. Myth, Creation of Woman, 6$L
McAllister, 264. Nanu, H. A., 384.

McGuire, Margaret F., 122. Nature's infallibility, see Infallibility
Magneff, 157. of Instincts.
INDEX
Near-sightedness, 123 i58f.; specific versus general prac-

Neff, 218. tice, i59f.; relation to school work,
JNewton, 95. i62f.; practice with knowledge, 163;
interest in improvement, i64f.;
Oberholtzer, E. E., 287. imagery in, i66f.; conclusion on
Observation, accuracy of, i32f.; in imagery, 169; fatigue, i69f., 75.
children, 133; reasons for inaccu- Promotion, by groups, 44, 45; by sub-
racy in, 134; range of, i34f.; how im- jects, 46; further suggestions for,
prove, 136; effect of practice on, 46f.
137- Pryor, H. C., 343, 347;
Odin, 92. Psychology and teaching, 3, 258f.
Original abilities, affected by differing Psychology of learning, and methods
environments, 92f.; Cattell's inves- of teaching, 3; need for more ex-
tigation of scientific men, 92; French tended and exact studies in, 3f.;
men of letters, 93; efficiency in analysis of problems of, nsf.; state-
school subjects, 93; birth places of ment of problems of, ii9f.; in school
eminent men, 94. subjects, 257; problems of in school
O'Shea, Harriet, 281. subjects, 260.
O'Shea, M. V., 327. PuebloJPlan, 42.
Overlapping, 36f.; extent of, 38; im-
portance of, 39; remedies for, 4if.; Pyle, W. H., 154.
percentages of as basis for com-
parison, 66, 69. Radossawljewitch, P. R., 157.
Rate of tapping, Fig. 2, 19.
Parker, F. W., 55, 59. Rational method in reading, 294.

"
Partridge, E. A., 233, 234. feea3Tng scales, Gray, 274, 275,
Paterson, D. G., 107, 130. 291.
Pearson, K.,*6i, 80, 82, 345, 346, 347, Kelly, 274, 275.
378. Starch, 274, 291, 292.
Perkins, A. 240, 242, 243.
S., Courtis, 274, 275, 280.
Peters, C. C., 282. Brown, 274, 275.
Permanency of interests, 62.
"
Fordyce, 274, 275, 280.
Kansas silent reading, 275, 280.
Phelps, C. L., 412, 413, 414. Reading, steps involved in, 261; re-
Phelps, William Lyon, 369, 370. ception of stimuli, 26 2f.; size of field
Phillips, F. M., 379, 381, 400. of distinct vision, 264; range of at-
Physical defects and school work, tention, 265^; eye movements, 266f.;
i3of. transmission of nerve impulses to
Pintner, R., 107, 130, 289. visual center, 269; arousal of asso-
Plateaus, see Learning. ciation processes, 269^; transmis-
Tlato, 250. sion of nerve impulse from visual

Poellman, 77. center, 27if.; measurement of effi-
Point scale tests, 107. ciency in reading, 274f.; results of
Probability curve, 33. measurements, 27$f.; economic pro-
Probability, integral, 32. cedure in learning to read, 278f.;
Proctor, M., 188. improvement in reading ability,
Progress of learning, factors in: length 281 f., 294f.; relation of speed to com-
.fend distribution of work periods, prehension, 283^; relation between
?i53f- 156; forgetting, i56f., and oral and silent reading, 287^; phon-
*
school work, 158; concentration, ics, 290; comparison of teaching
INDEX 47*
methods in reading, 20,of.; Beacon ments, 8yf.; young and old twins
system, 290-295; Aldine system, compared, 87.
290-294; Horace Mann system, 290, Simon, T., 99.
293; Ward method, 294; Howe Simpson, B. R., 53, 59, 109.
system, 294; Rational, method, Sleight, W. G., 161, 203, 204, 205,
294. , 206.
Reception of stimuli, problems, 132. Sloane, William, 218.

Rejall, 157, 161. Smedley, F., 18, 20, 129.
Report, unreliability of, i32f.; sources Smith, A. G., 55.
of error in, 134. Smith, F. O., 61.
Rice, J. M., 92, 93, 94, 334, 335, Spearman, 367, 412.
337, 338, 340, 39*i 392, 393, 394, Specific topics for educational psy-
395- chology, brief outline, 4f.
Rickhard, 190. Spelling, lists of common words in:
Riley, J. L., 348. Eldridgc, 327, 328, 329, 356.
Ritter, C., 171. Ayres, 327, 328, 356.
Rogers, A. C., 79. Jones, 327, 328, 331, 356.
Royal families, heredity in, see Men- Cook and O'Shea, 327, 328, 329,
tal Heredity. 356.
Ruediger, 224, 264, 271. Starch, 327, 328, 329, 356.
Ruger, H. A., 148, 209. Boston minimum, 331, 333.
Rugg, H. 0., 50, 250, 251, 424 Stockton, Santa Cruz, Chicago
Speller, 331.
Sackett, L. W., 311, 325- Nicholson, Chico, 333.
Schmidt, W. A., 268, 269. Spelling: steps involved, 322f.;
Schocklow, 199. methods of measuring efficiency in,
Schuster and Elderton, 84. 323f.; results from use of scales,
Schuyler, W., 158. 325f.; economic methods in learning
Scripture, E. W., Smith, T. L., and to spell, 326f.; determination of com-
Brown, E. M., 210. mon words in, 327^; proper place-
Search, P. W., 42. ment of words, 33of.; influence of
Sears, Isabel, 408. rules in,33!".; length of class period,
Sears, J. B., 325, 362, 365. 334f.; methods of teaching, 338f.;

Seashore, C. E., 130, 174, 175. effect of drill in, 339f.; waste in
Sensory defects: their effects in teaching of, 340; laws of association

general, 120; on school work, in, 343; centering attention on order
I29f. of letters in, 343; personal incentive
Sex differences: educational signifi- to effort, 343f.; special attention to
cance of, 63; popular vs. scientific parts of words, 344; writing
difficult
view of, 63fquantitative nature of,
.
;
the words, 345; context versus
64; methods of comparing, 65, 69, column spelling, 345; teaching
summary of, 68f.; in range ot abil- homonyms, 34sf.; class versus inde-
ities, 70, 72. pendent study, 346; grouping of
Sikorski, J., 171. words, 347; imagery in, 347f.; pres-
Similarities between brothers and sis- ent day compared with past effi-
ters, 80; in special mental traits, ciency in, 348.
8if.; in abilities in school subjects, Spillman, W. J., 94.
82f.; in university work, 84. Squire, Carrie R., 224.
Similarities of twins, 85; Gallon's in- Stanford revision of Binet-Simon
vestigation, 86; Thorndike's experi- tests, 69, 70, loo.
472 INDEX
Steele, 264. Transfer of training in school sub-

Stevens, W. J., 45. jects; opinions concerning, 2i7f.;
Stevenson, J. A., 51, 52. specific estimates of values of school
Stevenson, R. L., 360. studies, 209f.; table giving estimated
Stimuli, reception of, 132; interpreta- values Q! studies, 221; in arithmetic,
tion of, 138. 222; in grammar, 224!".; in foreign
St. Louis plan, 44, 45. languages, 229^; in Latin, 233 f.; in
Stone, C. W., 378, 394, 395- science, 247f.; in geometry, 25of.;
Strabismus, 124. general interpretation concerning,
Studebaker, J. W., 396, 405. 25 2f.
Studley and Ware, 33. Twins; similarity of, 85; experimental
Studying, waste in, 1 76; value of, 1 77f .; work with, 86f.
.types of, 179; problems in, 179; Typewriting, 145^

control of attention in, 180; diffi-
culties in, 180; suggestions for, i83f.; Van Landegend, E., 89.
Whipple's rules for, i86f.; whole and Varying abilities in school subjects;
part method in, i8sf. See also How range, 33, 36; general causes, 33.
to Study. Visual defects, 120, 121; types of, 123;
Supernormal child, need for differen- causes of, i24f.; frequency of, I25f.;
tiated training for, see Overlap- relative frequency in colored and
ping. white children, 125; reasons for
Supervised study, 42, 43, 44, i88f. variations in reports on, 126; in-
'Surface of frequency, 28. crease in higher grades, 127;
Suzzalo, H., 345, 346, 347- methods of avoiding, I27f.
Swift, E. J., 144, 15; i5 2 * 158, 161,
230, 231. Waddell, J. F., 433-
Wagner, 347.
Taussig, A. E., 125, 126, 127, 128. Waldo, K. D., 282, 283, 294.
Telegraphy, i42f. Wallin, J. E. W., 325, 34i, 342,
Terman, L. M., 69, 100, 105, 107, 345-
108. Walsemann, H. J., 384.
Thomas, 64. Ward method in reading, 294.
Thorndike, E. L., i, 10, 16, 40, 50, 55, Warren, H. C., 384.
56, 62, 66, 67, 68, 71, 72, 85, 86, 87, Washington, 358.
89, 148, 150, 155, 157, 161, 165, 172, Waste in education due to lack of
173, 175, 195, 197, 212, 213, 222, exact data on various learning proc-
275, 300* 301, 307, 3o8, 309, 311, esses, 3f.
378, 404, 405, 409, 4io. Webb, L. W., 199, 200.

Trabue, M. R., 353- Wells, F. L., 160, 161.
Transference of training: problems, Whipple, G. M., 50, 68, 124, 125,
191; experimental technique, 192; 127, 128, 129, 133, 136, 186, 265,
criticism of , 2 1 1 f . ; in memory, James 286.
. experiment, i93f.; other experi- Whitley, M. T., 89, 156, 161.
ments, 201-208; in reaction time, Whole and part methods in learning,
I94f .; in perception and discrimina- see Studying.
tion, i9$f.; in sensori-motor associa- Wilbur, Flora, 405, 406.

tion, i98f.; in attention, 2o8f.;
in in- Wilcox, M. J., 238, 239, 240.
genuity, 209!.; summary concerning, Wiley, Harvey, 218.
212; methods by which transfer Wilmarth, Alfred, 79.
takes place, 21 if. Wilson, O. M., 389, 390.
INDEX 473
Wimmer, Herman, 400, 401, 402, Woolley, Helen Thompson, 66, 67.
404. Word lists, see Spelling.
Winch, W. H., 161, 207. Writing, see Handwriting.
Witmer, 122.
Woods, F. A., 75, 76. Yerkes, R! M., 100; and Bridges, 107,
Woodworth, R. S., 195, 197, 210,
212. Zero family, 77.
Woody, C., 382.
This book was distributed courtesy of:
For your own Unlimited Reading and FREE eBooks today, visit:
http://www.Free-eBooks.net
Share this eBook with anyone and everyone automatically by selecting any of the
options below:
To show your appreciation to the author and help others have

wonderful reading experiences and find helpful information too,
we'd be very grateful if you'd kindly
post your comments for this book here.
COPYRIGHT INFORMATION
Free-eBooks.net respects the intellectual property of others. When a book's copyright owner submits their work to Free-eBooks.net, they are granting us permission to distribute such material. Unless
otherwise stated in this book, this permission is not passed onto others. As such, redistributing this book without the copyright owner's permission can constitute copyright infringement. If you
believe that your work has been used in a manner that constitutes copyright infringement, please follow our Notice and Procedure for Making Claims of Copyright Infringement as seen in our Terms
of Service here:
http://www.free-ebooks.net/tos.html

SDVASDFV

Uploaded by

Copyright:

Available Formats

SDVASDFV

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

SDVASDFV

Uploaded by

Copyright:

Available Formats

u <OU_1 64297 >m

OSMANIA UNIVERSITY LIBRARY

MACMILLAN & CO., LIMITED

BY THE MACMIL1AN COMPANY.

Set up and elcclrolyped. Published JUH

November, 1921; October, 19*2; April, OctoL

viewpoint in discussing these problems. The result of

problems. In the second place, a great deal of experimental

chology and pedagogy of learning.

psychology of more of the high school subjects; but this is impossi-

city of definite or substantial material in this field. It was, how-

ever, included because the topic is exceedingly important in school

PART I. THE NATIVE EQUIPMENT OF HUMAN BEINGS

PART II. THE PSYCHOLOGY OF LEARNING: A. IN GENERAL

PART III. THE PSYCHOLOGY OF LEARNING: B. OF SCHOOL SUBJECTS

PROBLEMS AND SCOPE OF EDUCATIONAL PSYCHOLOGY

What is Education? The problems and the scope of educational

to acquire skill, both motor and intellectual, in the use of his

changes may be brought about?

brought about by these agencies. This latter phase of the problem

economical methods of learning. Accordingly, the field of edu-

may designate as:

method, it follows that the most economical way of teaching read-

following: It has been found by recent tests and measurements

that some schools obtain

time we were able to save 36 minutes a day for eight years, we

I. The native equipment of human beings.

(a) In single traits.

2. At different times of life in the same individual.

II. The Psychology of Learning.

2. The measurement of ability and progress in learning

Reflexes, Instincts, and Capacities. The equipment with

ness and degree of complexity. AnUnstinct/niay be defined from

the action. In the case of acquired responses, such as habits,

The difference between reflexes and instincts is largely a differ-

actions involving the use of large groups of muscles or, in many

1. Food getting and protective responses.

II. Responses to behavior of other human beings.

23. Self-conscious behavior.

Relation of Education to Native Endowment. The inherited

Perhaps the most important role of instinct in education lies

itscontrol of the attention processes. Owing to the peculiarities

potency over others in attracting the attention and interest of the

multiplication table, a skillful teacher may on one day appeal to

into a rhyme or song. Or the tendency to play will be utilized by

reaching speculations and theories with regard to the nature and

Critique of the Dynamic Theory of Instincts. To work with

They are so deep-seated in the human psychophysicai organism

child manifests a tendency to draw, the school should take ad-

necessary precursor or a necessary accompaniment of useful tenden-

the whole, of being prophecies of similar interests and intellectual

ticularly useful at the present time. We would be better off if in

"The imperfections and misleadings of original nature are in fact

That these primitive tendencies persist with great strength is

"In education, don't cut off the tadpole's tail."

The Theory of the Transitoriness of Instincts. This concep-

FIG. i. Height of boys and girls measured in centimeters, based on measure-

FIG. 2. Rate of tapping. Number of taps made with right hand in 30

to cross this bridge successfully, there is sometimes almost a break of

The correctness of the theory of sudden appearance is primarily