Fes Manual
Fes Manual
Fes Manual
MANUA L
FOR
~-0mp;_efilili_~~~t.an11!]JNJJJi~-_
=e-o~hesion
&p;r__=.ess1;o.n
-fiii/e~-• nilenae
Mo=mJ®-r_Jema~Lan
0Iganflsxit1on
~e~=-tfJnra1 rnr,1 1:~fftaHo~
Copyrights © 1998. All rights reserved by Mis PSY-COM SERVICES, B-4, 8012, S,.J.Enclavc,
New De/hi-110 029 (INDIA). ~ 26106433, Fax: 26175191, email: slantee2000@yahoo.com
_J
Jts
imp orta nce
Fam ily Env iron men t:
5. This group has resour ces that are used to adjust in life.
. f Tl~e Family Environmc !nl ~-i <.;alc-! (rt.::S) w;-I5 dvvuloped os a mc:,:m:; lo gut
~n orination about Ille fo111ily environme nt in .:i r<.1pid · obJ·ective ond s1ondard-
Ized · ·
manner.. It Is a l.>ricf and non-stres sful seal<➔, applicable to all but the
1
owest education al levels. It is appropria te for use with ages of 1 O years and
above, throughou t adulthood . The scale can be used on young children or
even low literate adults who can read and comprehe nd simple English. The
scale gives an accurate appraisal of family environme nt. This scale is easily
administe red individual ly or to large groups at one time . It can even be used
as self-admin istering test. It includes 98 statement s and number of items for
each dimension are divided equally. Each statemen t has two possible an-
swers .
Competit ive Framewo rk (Cf): This dimension measures the extent to which ac-
tivities (such as school and work) are cast into an achievem ent oriented or
competitiv e framework . High score on this dimension means that the family
members are high on competitiv eness and achievem ent orientation . They give
importanc e to grades in school, or success one achieves at work and other
areas of life. However, low score on dimension Cf means that the family mem-
- bers are low on competiti ve~ss and need for achievem ent is also low. They
don't really work too hard or hardly worry about job promotion s, school grades
etc.
Cohesion (Co): This dimension measures the degree of commitme nt, help and
support family members provide for one another. High score on this dimensio n
mean·s that the family members support each other and they have a strong
feeling of togethern ~ss. They usually get along well and have time to pay
attention to everyone in the family. Low score on Co, on the other hand, means
that the family members openly express anger, show aggressio n in their be-
haviour. There is always conflict among family members and they often criti-
cize each other.
Expressiv eness (Ex): Expressiv eness measures the extent to which family mem-
bers _are encourag ed to act openly and to express their feelings directly. High
score on dimension Ex means that the family members are free to say anything
they want to, feelings of disagreem ent or disapprov al can be freely expresse d
at home. They usually indulge in spontane ous discussio ns without any fear or
hesitation of hurting each others feeling. However, low score on l;:x means
family members display difficulty in expressin g their feelings and thoughts to
one another. Important family matters are not discussed openly, and they are
always cautious about what they say to each other.
4
t to w.l1icl1 f r.1111ily mc:n, ..
. • · me~surcs ti 1C, ex l l '·' ll • .. · Hi(lll score 011
Independe nce (In): This dimension , u cir 0 wn deC 1 ~'°11 ,_.
"' ·
bers are assertive, self-suffic ient 011d mak_c 1d to be indcpcn~u 11~, lli~y u,,u~•II~
::, •.•
In means that family members arc cncourc:lg e . ,. Ive tl1e11 p1oblcms tliL!m
do things on their own they are given freedorn 10 .;;, 0 embers arc 11ot re:.:illy
'
se'Jves. However, low score that family r11 • · , t
on In means •c: I ittlc or no privocy il
encourag ed to speak up for themselve s a nd th ere 1_'' . ost of tile decision s
home. There is usually one family member who makes Ill
for them .
. ree of emphasis given
Moral Orientatio n (Mo): This dimension measures th ~ deg n dimensio n Mo
to et_hical, moral and religious issues a~d values. High 5 ~~re O They have high
means that family members give high importanc e to religion. H r low
ethical and moral values and are encourag ed to follow them. oweve_ ·
score on dimension Mo means that family members attac h l ·ttl I or no ,mpor-
e
tance. to religion, and show little or no ethical and moral values.
Organizat ion (Or): This dimension measures the degree of importanc e .0 ~ ~~ear
organizat ion and structure in planning family activities and res~on~1 b1l,ti~s,
-and extent to which set rules and procedure s are used to run family life. High
score on this dimension means that family members place high emphasis on
rules, there activities are carefully planned. Each individual 's duty in the family
is clearly defined, and they are usually neat and orderly. Low score, on the
other hand, means that the family rules are pretty flexible in their hou5-eho ld
and activities are not really planned carefully. There duties are not' well de-
fined and they are usually dirty and indisciplin ed in their daily activities.
Recreatio nal Orientatio n (Ro): This dimension measures the extent of participa -
tion and interest in social, recreation al, political, intellectua l and cultural activi-
ties. The high score on Ro means that family members have varied interests in
various recreation al activities and that they spare time for there hobbies, cul-
tural activities and/or intellectua l discussio ns. They go out very often to see
plays, attend lectures, watch sports events or just go out for fun. However, low
score· on Ro means that family members are really not very much interested in
arts-and culture, politics, or other inte1rectua l activities. They spend most of
there free time watching T.V. or listening to radio. They rarely go out to watch
plays, attend lectures, watch sports events or go out just for fun.
5
Tab /e-1 Brie f Des crip tion of Din iens ions Mea
sur ed by FES
Dim ens i<;> n
Dcs crlp tlo1 1
The limit of 50% has been found to be statistically sound when this scale
was given to the subjects under two condition: (1) 'anonymity', where the test
was said to have no connection whatsoever with their family life and they were
told that the results will be used for research purposes only, and (2) 'strongly
motivated' conditions, where subjects believed that their family life depended
on their responses and it will be used to sort out his/her family problems, if
any. Out of the 28 questions which showed greatest distortion were correlated
with the distortion of the total test. Here also it was found that even in 'anonym- _
ity' condition subjects do not show 'total honesty' probably because one tends
always to give the best possible impression of themselves (called faking
good), or tries to hide his/her bad points (faking bad}. For this purpose sepa-
rate norms were prepared for 'Vi' scores. We suggest following interpretation
for 'Vi' score for test users.
7
Adn 1in istr atio n
8
Scoring
2 . Examiner should also ensure that each answer has one and only one
answer.
3 . Once this is done, place the transparent. stencil scoring key over the · answer
sheet. Be sure that . it is aligned properly with the outer box printed on the
body of the answer sheet and the scoring key.
5. Each item in this scale measures some dimension as indicated in the key.
6. Add the scores horizontally for each dimension and write it in the space
provided for that dimension on the right hand side of the answer sheet.
7. For dimension 'Vi' use 'Vi Score Key'. Compare answers through
corresponding boxes, if the answer is same, give a score of 1 . If the answer
is different, no score is given. Add the score and note it down in the bottom
of the answer sheet.
8. Convert these raw scores to sten scores by procedure mentioned later in the
norms.
Norma tive Samp le
. T_he FES has only UU ilc:rns, wltic:I1 covt-! 1 s c: v c.· n fa111ily e11viro1111 Hrnl
dimensio ns . The detciiled tfoscriptio n with their psycllolog icc1I meaning II as
already been discussed earlier. Final 98 ilcrns lie.WC been selected from a tolc1I
pool of more than 200 items tlwt had been tested and refined in programm atic
studies . Each item in the scale has two answer choices from which the subject
may select one answer . All the items were carefully screened and several
criteria were considere d in the final item select.,ion .
Normativ e data on the FES were collected for more than 725 families.
The sample includes families from more than 15 locations all over India .
Sample also included various types of families like joint families, nuclear
families, single-pa rent, families drawn from minority groups, and families of all
age groups (newly-m arried families, families with preschoo l and adolesce nt
childr:en, families ..whose children have left home, and families compose d of
older: retired adults) , with atleast 2 or more members in each family. Table-3
provides demogra phic details of these families with geograph ical locations
(mainly towns and cities).
. . . hnve no need to
ammer wr 11 ,,.
lications of FES, the ex to stnndord ::;co,c-> .
In many research apbpt . d with the scoring key . dividual stands
convert the raw score o arne know how an 111 f .
Commonly however the test user will want to . tables presented here are( or
in relation ' to the reference
' popu Ia t'on
1 •
The norm The term , sten • comes ro111
the conversion of raw scores into the st en scores. s witli a ten point range
the 'standard ten', i.e. sten scales are ~ta nd a rd ~f~~:ndard score points, from
and they are distributed over ten equal intervals
1 to 1'0.
u have decided
Use table 4 for parents and table 5 for children . . Onc:t:~traight forward.
on the norm table to be used, the use of _norm ~ables rs _q~• the raw score of
For example to convert raw score of dimension _Cf, frn on the top or
dimension Cf in the row Cf and read the corresponding s_ten s~ore Please
bottom column of 'Sten' scores . Do likewise for other dimensions a Iso. ..
note that the values within the table (i.e ., in the body of t_he ta~le) are . ra~
scores" - values which you obtained with the help of stencil .. scoring key,: an
the values on the top and bottom columns are corresponding sten scores ·
To use norm tables the test user must note that the sten scores derived
from these tables are n-stens or normalised stens, i.e. they do not result from a
linear transformation of the raw score distribution, but instead from an area
transformation of the raw scores designed to produce a more normal
distribution. For test users who wish to convert group mean raw .score profile to
n-sten (standard deviation sten), the means and standard deviations of the
norm ·group are given at the bottom of the table.
Please note that the sten of 4-7 indicates average score, sten of 8-10
indicates high and extremely high score and sten of 1-3 indicates low and
extremely low scores.
11
Table 4: Norms for General Population (Parents)
Ste n Score::.;
Dimension 1· 2 3 4 :, G 7 8 9 10 Mc:[111 SO
l~c1w Scores
Cf 9 10 11 12 7.26 2 .38
1- 2 3 4-5 6 7 8
10 11 - 12 - 7.03 2 .89
Co 1 2 3 -4 5 6 7-8 9
10 11 12 7.57 2 . 37
Ex 1- 3 4 5 6 7 8 9
11 12 9.12 1 .74
In 1- 5 6 7 8 9 10
9 10 11-12 6.91 2 .91
Mo 1 2 3 4-5 6 7-8
10 11 12 7.01 2 .71
Or 1 2-3 4 5 6 7-8 9
11 12 7 .64 2.91
Ro 1-2 3 4 5-6 7 8 9-10
9 10 11 12-14 7.50 1.80
Vi 1- 2 3 4 5- 6 7 8
3 4 5 6 7 8 9 10 Mean so
Dimension 1 2
Sten Scores
Sten Scores
DimeA-Sion 1 2 3 4 5 6 7 8 9 10 Mean so
Raw Scores
Cf 1-3 4 5 6 7 8 9 10 11 12 7 .75 2.18
Co 1 2 3-4 5 6 7-8 9 10 11-12 6 .92 2.91
Ex 1 2 3 4 $ 0 7 8 9-10 11-12 5.98 2.34
In 1-4 5 6 7 8 9 10 11 12 8.49 1 .98
Mo 1 2 3-4 5 6 7-8 9 10-11 12 5 .78 2.98
Or 1 2 3-4 5 6 7 8-9 10 11 12 6.86 2.78
Ro 1 2-3 4 5 6-7 8 9 10-11 12 7 .32 2.80
Vi 1-2 3 4 5-6 7 8 9 10 1 1 12-14 7 .50 1.80
Dimension 1 2 3 4 5 6 7 8 9 10 Mean SD
Sten Scores
: Few psychometric criteria were kept in mind while selecting items for the
final form of the FES: (1) The overall item split should be as close to 50-50 as
possible to avoid items characteristic only of unusual families, (2) Items should
correJate more highly with their own-subscale than with any other (all the or
final 98 items met this criterion), (3) The subscales should have tow to
moderate inter-correlatio ns, and (5) Each item (and each subscale) should
discriminate among families . Each of these criteria was met in the .final form of
the FES.
Reliabil ity of FES
For calculating split-half reliability, a trial run of the present scale was
conducted on more than 300 individuals . The scale was divided into two parts
using odd-even method and care was taken to ensure that each half contained
more or less the $ame number of questions.
14
Tab le 7: lnt erc orr e/a tio ns
.. . . --- - Or Ro
- ··· Mo
Ex In
Dime nsion Cf Co
--- -- ·- -· -- ---- - - •·•·. .0!)
.7G ..10
Cf .11 -. 05 -.01
.01 .21
.40 .28 .20
Co
-.01 .31
.24 .01
Ex
-.05 .09 .24
In
.27 .04
Mo
.12
Or
Ro
Validit y of FES
measure
. i·t.r
The basic .definition 0· r vn • 111 .. ,1 .1I 1s
1 1ly ,s
.
. ..111 111<kx of how well . --1 ~c.:il~
h" 1
s w lat it purports to measure:. 111 tile case of an ability tost 01 c.111
a~ ,eve~ent test, this can be do1w directly and more easily . Gut tllis is riot so
simple wrth personali ty test or family environm ent scrile .
. Now, the question here is, what was FES designed to measure ? As
~ent1one d earlier the FES was conceived primarily as a device to measure 7
independ ent family environm ent dimension s that resulted from a set of items
adm!niste red on normal and clinical groups . The items in this final form are the
survivors from several thousand s of items originally tried, and constitute only
those which continue to have significan t validity against the dimensio ns. The
factor analyses verified the existence and structure of these 7 independ ent
family environm ent dimension s. More formally, to understan d the concept of
validity used here we must know the backgroun d theory of factorial validity .
In the FES, thes~ functional unities are identified by analyzing the inter-
correlatio ns among a number of separate, re1atively- restricted measures , which
identify the common factors for obtained coefficie nts. Factor analysis ,
therefore, is intended to reduce the number of variables , or test categorie s,
needed to represent an individua l's traits for specified purposes . Factorial
validity here, is therefore , determine d by the weights (called, "loadings ") .
Contribute d to the total-test scores by each of qerived scores. It is determine d
by the relative independ ence of one another (low intercorre lations). Thus, to
be true to its design each dimension should have one "importan t", loading -
on the dimension that it was intended to measure - and 6 "unimpor tant"
loading . These factorial validity coefficien ts are shown in Table-8 below.
Co .83 Or .79
Ex .77 Ro .80
In .82
16