Canadian Association of Applied Linguistics/Canadian Linguistics Association

Joint Plenary Talk - May 30, 2005, London, Canada

Integrative motivation and second language acquisition
Robert C. Gardner
Department of Psychology
The University of Western Ontario
London, Canada
Abstract
This talk discusses integrative motivation from the perspective of the socio-educational
model of second language acquisition, and will focus on six topics. The first concerns the concept
of motivation which can be shown to comprise at least 10 cognitive, affective and conative
components. The second topic is the socio-educational model of second language acquisition. It
is presented in two forms. One is the fundamental model which proposes that two major factors,
ability and motivation, interact with language acquisition contexts to produce linguistic and nonlinguistic outcomes. It is assumed that the educational setting and the cultural context influence
motivation. The other shows the relationships among ability, attitude, motivation, language anxiety
and language achievement variables. The third topic is the Attitude Motivation Test Battery
(AMTB) in which the components of the model and their assessment are presented. The fourth topic
focusses on a brief review of some empirical findings based on our research.
The rest of the discussion deals with two issues that I believe are important for this area of
research. Topic five concerns the distinction between motivation and motivating. In our research,
we have demonstrated how the issue of motivating can be studied based on our model. For example,
we have found that, predictably, attitudes toward the learning situation vary significantly among
classes at the end of the academic year in university classes in French in Canada, while significant
class differences in integrativeness, motivation and attitudes toward the learning situation were
obtained for children learning English in Spain. We also found that over the course of the year,
some significant changes were obtained which varied on the basis of the grade obtained at the end
of the year.
The sixth topic deals with the generalizability of the socio-educational model and the AMTB.
There has been the view expressed that they may not be appropriate to settings where the focus is
on foreign language learning. Our research in four countries shows that the scales of the AMTB
produce internal consistency reliability coefficients and correlations of the major constructs with
final grades in English comparable to those obtained in our Canadian studies.
In this talk, I have attempted to review the concept of integrative motivation. It is my view
that the concept has been defined in many different, but related, ways by some researchers but that
from the perspective of the socio-educational model of second language acquisition, it has a very
definite and predictable meaning which is applicable equally to situations involving second and
foreign language acquisition.

Integrative motivation and second language acquisition1

Robert C. Gardner
Department of Psychology
The University of Western Ontario
London, Canada
When I began preparing this talk, I googled the internet for the term integrative motivation,
and much to my surprise I obtained 591 hits. This may not be a particularly large number in the
grand scheme of things, but I was impressed. There is a fair amount of interest in the topic. Reading
some of the files, however, I found that many people have different conceptualizations of the term.
Most of them overlap the concept of integrative motivation that I have been discussing for a number
of years but not completely. For example, I found one that defined integrative motivation as when
students want to learn a language to become part of a speech community. In another, I read that
integrative motivation is characterised by the learners positive attitudes toward the target language
group and the desire to integrate into the target language community. These two definitions
include some of the elements of integrative motivation as we typically refer to it, but not all. In
1985, I discussed the concepts of motivation and integrative orientation, and then concluded that
The concept of the integrative motive includes not only the orientation but also the motivation (i.e.,
attitudes toward learning the language, plus desire plus motivational intensity) and a number of other
attitude variables involving the other language community, out-groups in general and the language
learning context (Gardner, 1985, p.54). To me, this characterization is much broader than those
often given by other researchers.
I also dont believe that integrative motivation to learn French, for example, is characterized
by statements like Studying French is important to me because it will enable me to better
understand and appreciate French Canadian art and literature or that Studying French can be
important for me because it will allow me to meet and converse with more and varied people.
Admittedly, these are items from our 1985 Integrative Orientation Scale, but they reflect an
orientation, not necessarily a motivation. Individuals might well agree whole-heartedly with these
types of items, but may or may not be motivated to learn the language.
In 1991, Crookes and Schmidt called for what they termed a reopening of the research
agenda. Personally, I am all in favour of new approaches and research agendas, and was at the time.
Unfortunately, in calling for this new agenda, I believe they misrepresented what they termed the
old one. They referred to the initial article by Gardner and Lambert (1959), implying that we
distinguished between integrative and instrumental motivation. But we didnt even use those terms
in that article. They stated Motivation is identified primarily with the learners orientation toward
the goal of learning a second language. Integrative motivation is identified with positive attitudes
toward the target language group and the potential for integrating into that group, or at least an
interest in meeting and interacting with members of the target language group (Crookes & Schmidt,
pp 471-472). This, it should be noted is their definition, not mine. In fact, they even noted that it
wasnt mine. They state What is not noticed or commented upon, however, is that the integrative
motive in Gardners more recent model (Gardner, 1985, 1988) is no longer equivalent to attitudes
toward the target language community and is not equivalent to a score on the integrative orientation
subscale of the AMTB or any other subscale of the AMTB (p. 475). But, the simple truth is, it

3
never was any of these.
The Crookes and Schmidt (1991) article was useful in that it urged individual researchers
to seek other ways of conceptualizing motivation; it did a disservice, however, in suggesting that
researchers should simply rule out the socio-educational approach. As we shall see, the socioeducational model is a paradigm that is completely compatible with many of the new research
agendas that have developed.
My intention today, therefore, is to begin by discussing the concept of motivation in general,
to outline the many attributes of the motivated individual, and to show the implications of this for
second language acquisition and for the related research. Following this, I will outline the socioeducational model of second language acquisition, and show how it provides a general research
paradigm that can be used to better understand the role of motivation in second language acquisition.
One feature of our research is the Attitude/Motivation Test Battery (AMTB), and I plan to discuss
it in a bit of detail. I do not mean to suggest that this is the only measure of affective variables in
second language acquisition research - - an inspection of the research literature reveals that there are
many tests of attitudes, motivation, anxiety, and orientations, as well as related measures of
personality attributes, learning strategies, and the like. The AMTB was developed, however, to fit
directly into the socio-educational model of second language acquisition, and thus serves as a
convenient way to refer to specific aspects of the model.
I then plan to focus on two issues in this area that I feel are particularly important. One is
on the distinction between motivation and motivating. The other deals with the often-stated view
that the socio-educational model and the AMTB are appropriate to the Canadian context but not
necessarily to other contexts.
Thus, by way of introduction, my plan is to discuss:
1.
The meaning of motivation
2.
The socio-educational model of second language acquisition
3.
The Attitude motivation Test Battery (AMTB)
4.
Theoretical Generalizations and Empirical Findings Related to the Model
5.
Motivation vs motivating
6.
The generalizability of the socio-educational model and the AMTB
The Meaning of Motivation
Motivation is multifaceted, and it has been defined in many different ways by different
researchers in psychology and other scientific disciplines. Kleinginna and Kleinginna (1981)
present 102 statements referring to the concept. A simple definition is, therefore, not possible.
Nonetheless, one referenced by Crookes and Schmidt can be used to give the general flavour of what
motivation is. Keller (1983) states Motivation refers to the choices people make as to what
experiences or goals they will approach or avoid, and the degree of effort they will exert in this
respect (p.389). Most of the important characteristics of the motivated individual are alluded to
in that definition, as a review of the literature will reveal. Motivated individuals display many
characteristics. They are goal-directed. But many of us have goals - - things we would like to have,

4
but may never achieve because we lack many other characteristics of the motivated individual that
would aid in our obtaining those goals. Motivated individuals express effort in attaining the goal,
they show persistence, and they attend to the tasks necessary to achieve the goals. They have a
strong desire to attain their goal, and they enjoy the activities necessary to achieve their goal. They
are aroused in seeking their goals, they have expectancies about their successes and failures, and
when they are achieving some degree of success they demonstrate self-efficacy; they are selfconfident about their achievements. Finally, they have reasons for their behaviour, and these reasons
are often called motives. As you can see some of these characteristics reflect cognitions, some
reflect affect, and some reflect behaviours or behavioural intentions. Motivation to learn a second
language is not a simple construct. It cannot be measured by one scale; perhaps the whole range of
motivation cannot be assessed by even three or four scales. It definitely cannot be assessed by
asking individuals to give reasons for why they think learning a language is important to them.
The point I am trying to make here is that motivation is a very broad-based construct. It has
cognitive, affective and conative characteristics, and the motivated individual demonstrates all
facets. A reason is not motivation. One can want to learn a language for reasons that might reflect
an integrative orientation, but unless this is accompanied by other features of motivation it is not
motivation. Similarly, we can want to learn for reasons that might be classified as instrumental, but
without the motivational features, this doesnt reflect motivation. If one is motivated, he/she has
reasons (motives) for engaging in the relevant activities, persists in the activities, attends to the
tasks, shows desire to achieve the goal, enjoys the activities, etc...
There are a large number of studies, often claiming to be studying motivation, that have
presented individual language learners with reasons for studying a second language. Some have
shown correlations of these reasons with measures of achievement. Some have not. Some have
shown, often by factor analytic means, that the reasons can be grouped in various ways (Travel,
Educational, Dominance, Friendship, etc...), claiming to have identified other orientations. In my
opinion, these other orientations can be further classified as being in the integrative or instrumental
camp, but this is a minor matter. I am not trying to be critical of these studies. They serve a very
useful role in identifying how reasons group together, which might well reflect cultural and even
contextual differences. What I am saying is that, these studies may or may not be studying
motivation. Without some indication of the relationship between the choice of reasons and the other
features of the motivated individual, we will never know. Without an association with the other
attributes that characterize the motivated individual, a reason is just a reason, not a motive.
The Socio-educational model of second language acquisition
Before describing the socio-educational model of second language acquisition, I would like
to outline what I consider to be the fundamental model of language learning. This is shown in
schematic fashion in Figure 1.

5
Figure 1
The Fundamental Model
Ability
Educational
Setting
Motivation
Cultural
Context

Formal
Contexts

Informal
Contexts

Linguistic
Outcomes

Non-linguistic
Outcomes

As indicated in the figure, the model proposes that there are two primary individual
difference variables involved in language learning, viz., ability and motivation. It is proposed that,
other things being equal, the student with higher levels of ability (both intelligence and language
aptitude) will tend to be more successful at learning the language than students less endowed.
Similarly, other things being equal, students with higher levels of motivation will do better than
students with lower levels because they will expend more effort, will be more attentive, will be more
persistent, will enjoy the experience more, will want more to learn the material, will be goal
directed, will display optimal levels of arousal, will have expectancies, and will be more selfconfident with their performance. These two factors are expected to be relatively independent
because some students high in ability may be high or low in motivation for any host of reasons, and
vice versa. This is not a particularly novel observation. Many educational psychologists such as
Carroll (1963), Bruner, (1966), and Glazer (1976) have proposed that ability and motivation are two
important factors associated with achievement in school.
In the model, both ability and motivation are seen to be involved in both formal and informal
language learning contexts. The formal context refers to any situation where instruction is carried
out (typically, the language classroom), and informal contexts are any other situation where the
language can be used or experienced, etc., (i.e., listening to the radio, language clubs, the street,
etc.). It is further assumed that whereas both ability and motivation would be equally involved in
formal contexts, motivation would be more involved than ability in informal contexts simply
because motivation would tend to determine whether or not the individual even takes part in the
informal contexts. This is indicated by the dashed arrow linking Ability to Informal Contexts. In
the model, both contexts are shown to lead to both linguistic and non-linguistic outcomes. Finally,
it is shown that the educational situation and the cultural context are shown to have influences on
motivation, but not ability.
1.
2.
3.

Clearly, language acquisition involves a number of factors, such as, for example:
Quality of Instruction - - Teacher, Curriculum, and Lesson Plans
Opportunities to Use the Language
Socio-cultural milieu and Expectations

6
4.
5.
6.
7.

Student Ability - - Scholastic (Intelligence) and Language Aptitude

Student Affect - attitudes, motivation, anxiety
Personality Variables
Learning Strategies

Inspection of these variables will reveal that they can be grouped into two classes. The first
three refer to environmental characteristics, while the last four are characteristics of the student.
Obviously environmental variables play important roles but, equally obviously, much of the
environmental influences are mediated by the individual. For example, one can have a high quality
of instruction, but the actual success of this instruction depends upon how it is received by the
student. Some students will profit from the high quality of instruction more than others, and the
extent of these differences are dependent on characteristics of the students.
The socio-educational model is concerned primarily with motivation and factors that support
it, and it assumes that other variables such as personality, strategy use, etc., can be explained in
terms of the motivation construct. The socio-educational model is shown schematically in Figure
2.
Figure 2
The Socio-educational Model
Ability
Attitudes to
Learning Situation
Motivation

Language
Achievement

Integrativeness

Instrumentality

Language
Anxiety

In the socio-educational model, it is proposed that the individuals motivation to learn a

second language is related to two classes of variables. One is Attitudes toward the Learning
Situation. Clearly, the nature of the learning situation will influence a students level of motivation.
An interesting, devoted skilled teacher with a good command of the language, an exciting
curriculum, carefully constructed lesson plans, and meaningful evaluation procedures will promote
higher levels of motivation, other things being equal, than a teacher lacking in some of these
attributes.
The other variable seen to be important in influencing motivation is Integrativeness. This

7
construct seems to give a lot of people a hard time, because they have difficulty discriminating
between it, integrative orientation, integrative motivation, and an integrative motive. Perhaps the
label, Integrativeness, is confusing, but the concept itself is not. It derives historically from the
construct of identification used by Mowrer (1950) to explain the motivation a child has to learn the
language of the parents. He proposed that because of the reinforcements and drive reducing
behaviour of the parents early in life, the child is motivated to adopt features of the parents which
could serve as rewarding and drive fulfilling substitutes when the parent is not present. One aspect
of this is verbal behaviour. The child learns that by making sounds similar to the parents he/she can
achieve some level of comfort in their absence. This leads to the gradual adoption of many of the
parental characteristics, resulting in the childs identification with the parents. Hence, identification
serves as the motivation to learn the first language.
The concept of integrativeness was postulated as an extension of this concept. Gardner and
Lambert (1972) proposed that a similar type of process could be important for second language
acquisition. Obviously, it was not identification in the same sense as Mowrer proposed it, but we
argued that because of their cultural background, early home experiences, child rearing
characteristics, etc., some individual learners would be more open to other ethnic, cultural, and
linguistic groups than others, and this openness could influence their motivation to learn the other
language. Perhaps it is the case that some individuals do learn another language in order to integrate
with another culture (in fact, Lambert, 1955 describes such a case), but the majority do not. We
never meant integrativeness (or integrative orientation, or integrative motive) to mean one wanted
to become a member of the other cultural community, but rather an individuals openness to taking
on characteristics of another cultural/linguistic group. Individuals for whom their own ethnolinguistic heritage is a major part of their sense of identity would be low in integrativeness; those
for whom their ethnicity is not a major component, and who are interested in other cultural
communities would be high in integrativeness. This, it should be noted is not a new interpretation
of the concept. It has been central to our research and the socio-educational model from the
beginning. We have consistently proposed that the words, pronunciations, grammar, and the like
are salient characteristics of another cultural community, and thus the individuals openness to other
cultures (i.e., integrativeness) will influence his/her motivation to learn the language.
Other researchers have proposed similar constructs more appropriate to their contexts,
labelling them differently. For example, Yashima (2002) and Yashima, Zenuke-Nishide, and
Shimizu (2004) have demonstrated that a construct labelled International Posture influences
motivation in structural equation models linking a series of attitude and motivation variables to
second language achievement and/or willingness to communicate and frequency of communication
in the second language. In those studies, international posture is defined as a general attitude
toward the international community that influences English learning and communication among
Japanese learners (cf., Yashima, 2002, pp. 62-63) and is measured in terms of intercultural
friendship, approach tendencies (to interact with foreigners), interest in international activities, and
interest in foreign affairs. In short, it also reflects an openness to other cultures.
Recently, Czizr and Drnyei (2005) have proposed a redefinition of integrativeness
focussing on a cognitive representation. Czizr and Drnyei state Our proposed interpretation

8
equates integrativeness with the Ideal L2 Self, referring to the L2-specific dimension of the learners
ideal self (p. 30). Personally, I believe that this might confuse things considerably; it certainly will
make communication about integrativeness difficult. The concept of integrativeness as initially
proposed refers to an affective dimension and it is historically linked to the concept of identification.
It is quite possible that individuals who are high in integrativeness may have different perceptions
of their self and their ideal self, particularly as they relate to the second language but it would seem
better to use a different label. In any event, research will be needed to establish whether there is any
relation between the two conceptualizations.
Each of these characterizations recognize the role played by personal socially relevant
factors in influencing motivation, though their nature is modified by the cultural milieu in which the
studies are conducted. Another study by Kraemer (1993) also considers a different socially relevant
construct. Her study involved Israeli students learning Arabic. She demonstrated that a socially
relevant construct, identified as Social/Political Attitudes, had an indirect effect on motivation, and
these attitudes reflected close social distance, equal civil rights, and optimism about peace in the
future. Like integrativeness in the socio-educational model, these reflect an openness to the other
community. The point is that integrativeness represents a socially relevant, as opposed to an
educationally relevant construct.
The socio-educational model also postulates that other variables are implicated in second
language learning. For example, it is clear that Language anxiety plays a role in language learning,
though the role can be complex. Anxiety can have motivational properties suggesting that it might
well facilitate achievement. It also has debilitating components that interfere with learning and
production, however, and probably because of these, language anxiety is generally negatively related
to achievement as well as to self-confidence with the language. This negative relationship can be
interpreted as indicating that high levels of language anxiety interfere with language achievement,
or that low levels of achievement cause individuals to be anxious in situations where they are to use
their language.
Another variable that can be implicated in second language achievement is an instrumental
orientation, or more generally instrumentality. In many situations, individuals might well want to
learn a language for purely practical reasons, and to the extent that this orientation is related to
achievement it is reasonable to expect that the relationship would be mediated by motivation, as was
the case with integrativeness.
The diagram is intended to represent the assumed relationships among these constructs. The
bidirectional arrows linking Integrativeness to Attitudes toward the Learning Situation and also to
Instrumentality are meant to indicate that the two pairs of constructs are expected to be positively
correlated with one another. Individuals with high levels of integrativeness would be expected to
view the language learning situation positively, other things being equal and vice versa. And,
individuals who are high in integrativeness would also be expected to be high in instrumentality.
As has been well documented, there does tend to be positive relationships between these two classes
of orientations. There is no reason to expect them to be independent of one another.

9
The diagram also shows uni-directional arrows linking Attitudes toward the Learning
Situation, Integrativeness, and potentially Instrumentality to Motivation. This is meant to imply that
motivation is supported by these constructs. Levels of motivation are influenced and maintained
by Attitudes toward the Learning Situation and Integrativeness, and where relevant, Instrumentality.
Unidirectional arrows also link Ability and Motivation to Language achievement. This is meant to
indicate that in the socio-educational model, it is assumed that these are the two major variables
responsible for individual differences in achievement in the language learning context. As stated
earlier, it is assumed that individual differences in ability (both intelligence and language aptitude)
account for some of the differences in achievement, while differences in motivation account for
other differences. A student can achieve high levels of achievement based on ability or motivation,
but since they are relatively independent, the individual who is high in both ability and motivation
will be even more successful, other things being equal.
The model indicates two directional arrows from language anxiety and achievement. This
is meant to indicate that individuals experiencing high levels of language anxiety will tend to do
more poorly on the measures of achievement, and that individuals who lack facility in the language
will also tend to be more anxious in situations where they are called upon to use their language.
The model does not formally refer to environmental characteristics, though obviously they
are an integral part of the model. Undoubtedly, quality instruction begets quality learning, other
things being equal. How the material is presented, immediate reinforcements, clearly identified
objectives - - all will promote learning. Opportunities to use the language reinforce and strengthen
what is learned and thus can promote learning. The socio-cultural milieu offers role models,
scenarios, situations and expectations that can foster language achievement. And, low quality
instruction, few opportunities to use the language, and a socio-cultural milieu that opposes bilingual
development, or simply ignores second language acquisition because of all the other issues in life
can inhibit achievement in the other language. But these environmental factors operate on and
interact with the individual. In the end, it is how the individual responds that leads to the ultimate
results.
Nor does the model refer specifically to personality or learning strategies, even though it
recognizes that these variables can play a role in second language achievement. Personality
variables such as extroversion, introversion, sociability, and the like influence an individuals
tendency to respond in situations, and the use of learning strategies can influence learning. It seems
likely that their influences, however, would be indirect, acting through motivation, ability, or
language anxiety.
The model does permit many generalizations about the learning context, and suggests many
hypotheses that could be tested. Before discussing some of them, let us consider the measures that
correspond to the socio-educational model.
The Attitude Motivation Test Battery (AMTB)
The AMTB was developed to measure the various components of the socio-educational
model of second language acquisition. There are 11 subtests, nine with 10 items each, and two with
4 items. The major conceptual variables and the subtests designed to assess them are as follows:

10
Attitudes toward the Learning Situation.
Attitudes toward the learning situation refer to affective reactions to any aspect of the class
and could be assessed in terms of class atmosphere, the quality of the materials, availability of
materials, the curriculum, the teacher, etc. In the AMTB, these attitudes are assessed in terms of the
participants evaluation of the teacher and the course, though it is recognized that other factors could
be included. The two measures (with abbreviations used in Figure 3 in brackets) are:
Language Teacher - Evaluation (Teach)
Language Course - Evaluation (Course)
Integrativeness.
Integrativeness can be assessed in many ways, and represents group-related affective
reactions. As measured in the AMTB it involves the individuals orientation to language learning
that focusses on communication with members of the other language group, a general interest in
foreign groups, especially through their language, and favourable attitudes toward the target
language group. That is, it reflects an openness to other cultures in general, and the target culture
in particular. Individuals who are high in integrativeness do not focus on their own ethno-linguistic
community as part of their own identity, but instead are willing and able to take on features of
another language group (if only just the language) as part of their own behavioural repertoire.
Individual differences in integrativeness probably are formed from the interplay of cultural factors,
family beliefs and attitudes, child-rearing characteristics, and possibly even genetic predispositions.
Three measures from the AMTB are:
Integrative orientation (IO)
Interest in Foreign languages (IFL)
Attitudes toward the Language community (ALC)
Motivation.
There are many facets of motivation that can be examined but I believe that the fundamentals
are best identified by three measures that assess effort and persistence, the desire to learn the
language, and affective reactions to learning the language. Any one, in and of itself, does not
properly encompass the many features of the motivated individual, but it is felt that these three do
an adequate job in this regard. In the socio-educational model, it is assumed that attitudes toward
the learning situation and integrativeness serve as the major supports for motivation, though it is
recognized that under some circumstances instrumentality could also serve the same purpose. Other
factors such as, for example, scholastic motivation or even some personality characteristics could
also have an influence on levels of motivation, but it is unlikely that the effects would be as long
standing. The three scales in the AMTB used to assess motivation are:
Motivational Intensity (MI)
Desire to Learn the Language (DESIRE)
Attitudes toward learning the language (ALL)
Language Anxiety.
Anxiety about the language could be aroused in many situations (i.e., interpersonal
communication, language drills, examinations, etc.). Such anxiety could result from more general
forms of anxiety such as trait anxiety, previous unnerving experiences in language classes, or

11
because of concern about deficiencies in language knowledge and skill. That is, language anxiety
could have deleterious effects on learning, and inadequate skill could give rise to feelings of anxiety.
For the purposes of the socio-educational model, we distinguished between two broad situations,
the language class, and contexts outside of the classroom situation where the language might be
used. Two measures are employed:
Language Class Anxiety (CLASS)
Language Use Anxiety (USE)
Instrumentality.
The notion of instrumentality refers to conditions where the language is being studied for
practical or utilitarian purposes. Like integrativeness, there could be many causes for such feelings
varying from the cultural setting to idiosyncratic experiences of the individual. To date, the only
measure that we have found appropriate is the following scale:
Instrumental Orientation (INST)
Our research has used these measures in various ways, depending on the purpose of the
study. Thus, we might focus on:
1.

2.

3.

Individual variables. That is, we investigate the correlations among these variables.
Generally, the variables are interrelated, even though their factor structure changes
slightly depending on the other variables in the matrix. In one study where we used
multiple measures of each of these variables, we obtained the five factors
corresponding to these five categories (cf., Gardner & MacIntyre, 1993). In later
studies, we were more interested in the functional relationships among the variables
and tested structural equation models comparable to the one presented in Figure 3.
In most cases, the measurement models are well defined, and the major paths are as
predicted.
Aggregate measures of the constructs. In other studies, we focussed more on
aggregate measures reflecting the major components. We thus calculated total scores
for each one. For example, the score for motivation is the aggregate of motivational
intensity plus desire to learn the language plus attitudes toward learning the
language.
Aggregates of the constructs. In some other studies, we computed Integrative motive
scores by aggregating Integrativeness, Attitudes toward the Learning Situation, and
Motivation. In still other studies, we computed AMI (Attitude Motivation Index)
scores based on the sum of Integrativeness, Attitudes toward the Learning Situation,
Motivation, and Instrumental Orientation minus Language Anxiety.

These measures link to the socio-educational model as shown in Figure 3.

12
Figure 3
The Socio-educational Model with Indicator Variables
TEACH

CLASS

Attitudes to
Learning
Situation
Integrativeness
IO

MI

DESIRE

ALL

Motivation
Language
Achievement

ALC

IFL

Instrumentality
INST

Language
Anxiety
CLASS

USE

Theoretical Generalizations and Empirical Findings Related to the Model

The model permits a number of generalizations relevant to understanding the process of
second language acquisition. One major one is that it helps to clarify the notion of the integrative
motive. As postulated, an individual who has a high degree of integrativeness, has a favourable
evaluation of the language learning situation, and is highly motivated to learn the language can be
said to be integratively motivated. That is, it seems reasonable to conclude that such individuals are
willing or interested in coming closer psychologically to the target language community. Obviously,
not everyone studying a language is integratively motivated. Thus, some individuals may not be
particularly high on integrativeness, but still positively evaluate the class and be motivated to learn.
We would anticipate that they too would be successful in learning the language, though they may
not evidence the same degree of devotion as someone who is integratively motivated. If, for
example, their motivation was based largely on pragmatic reasons for learning the language, one
might expect that they may become less motivated once they achieve their specific goal (cf., Gardner
& MacIntyre, 1991). One example of this might well be language programs that are offered to
workers for job enhancement. One could well imagine individuals motivated by these job related
inducements could do very well in the course, but once they have achieved their objective make little
use of their language skills.
Rather than treating the socio-educational model as one describing associations among the
major variables, it can be used to test specific predictions about the role of these variables in
language learning. Thus, the model would predict that by and large motivation would be more
highly related to achievement in the second language than would Integrativeness or Attitudes toward

13
the Learning Situation. This has been well documented in the literature. For example, Masgoret
and Gardner (2003) conducted a meta analysis of 75 samples of data (involving 10,489 participants),
and investigated the relationships of the three major constructs and the two orientations, to measures
of grades, objective measures, and self ratings of proficiency in the language.
Not all of the correlations were significant, of course, but the majority were. For example,
the mean correlation of motivation with grades was .37, with individual values ranging from .03 to
.55. In all, 96% of the correlations were significant. There was also considerable variability in the
correlations of the other measures with grades, but there correlations were much lower. The mean
correlations were .24 for both Integrativeness and Attitudes Toward the Learning situation, .20 for
Integrative Orientation, and .16 for Instrumental Orientation. That is, these other measures are not
as highly correlated with measures of achievement as motivation even though the majority of the
correlations were significant. Often, however, these measures are highly correlated with motivation
indicating that they serve a supporting function for motivation. The operative variable in the
relationships, however, is motivation.
The socio-educational model has generated many other predictions related to second
language acquisition. Some of the major findings are:
1.
2.
3.
4.
5.
6.
7.
8.

Integrative motivation is predictive of classroom behaviour in three grades levels of

highschool French instruction (Gliksman, Gardner, & Smythe, 1982).
Students who drop out of language study once it is no longer required are lower in
motivation, integrativeness, and attitudes toward the learning situation (Gardner, 1983).
Structural Equation models support the socio-educational model as described (Gardner,
1985).
Faster learning of English/French pairs is related to integrative motivation (Gardner,
Lalonde, & Moorcroft, 1985).
Integrative and instrumental motivation are both related to the learning of English/French
pairs (Gardner & MacIntyre, 1991).
Integrativeness, attitudes toward the learning situation, and motivation are factorially distinct
(Gardner & MacIntyre, 1993).
The socio-educational model can incorporate other motivational variables in structural
equations (Tremblay & Gardner, 1995).
Attitudes and motivation influence state motivation which influences the rate of learning of
English/Hebrew word pairs (Tremblay, Goldberg, & Gardner, 1995).

In summary, there is ample evidence that the socio-educational model provides a

parsimonious interpretation of second language acquisition, and that the AMTB provides measures
that can be used to investigate various aspects of the language learning process.
I would like to turn attention now to two issues that are common to this area of research and
which bear on the construct of integrative motivation. These two issues are:
(a) the distinction between motivation and motivating, and
(b) the generalizability or lack of it of the socio-educational model and the AMTB.

14
Motivation vs Motivating
A distinction has been made between motivation and motivating, implying that they are
different in many important ways. Often motivation has been viewed as very much like a trait (cf.,
Drnyei, 2001). This is not the view held by Gardner, though it is true that motivation is seen as a
characteristic of the individual. Those making the distinction view motivating as something that can
be done by someone else, notably a teacher. That is, we speak of the student as being motivated (or
not), and the teacher as being good at motivating the students (or not). The importance of studying
how to motivate students was one of the new agenda items that Crookes and Schmidt (1991)
recommended.
There is no doubt that a teacher can provide an environment that is conducive to learning by
demonstrating expertise in the material, being supportive and reinforcing, being well organized,
having interesting and exciting lesson plans, encouraging cooperation in the classroom, and being
consistent in evaluating students, etc. The learning environment is very important and can certainly
aid students. It seems very reasonable to hypothesize that some things can motivate the student to
attend to a lesson, to learn the material well, and/or to do well in a language class, and it would be
extremely beneficial to conduct research onto these factors. It may even be the case that someone
could motivate a student to learn and use the language outside of the classroom and beyond, but
more research is needed on this as well. It is certainly a prediction of the socio-educational model
that this latter form of motivating may be more difficult than one might anticipate, because of the
role that attitudinal factors like integrativeness play in influencing an individuals level of
motivation. But this is an open empirical question.
It is reasonable to predict that some teachers are better at teaching and motivating than
others, and thus if we were to investigate attitudes and motivation in different classes, we should
find differences, particularly toward the end of the year. Moreover, the socio-educational model
would predict that if teachers can motivate students, this should be evidenced in measures of
attitudes toward the learning situation and perhaps motivation. But on the assumption that
integrativeness is more group-related, they would be less susceptible to actions of the teachers.
I was able to test both of these assumptions using data from three different studies we
conducted recently. One was done in Canada with university students studying French (Gardner,
Masgoret, Tennant, & Mihic, 2004), and two were conducted in Spain, one with elementary school
children and one with secondary school students studying English (Gardner & Bernaus, 2004). For
the study conducted in Canada, I analyzed the data for the aggregate scores for the five concepts
discussed earlier asking if there was significant variation among the 9 classes on either of these five
measures. In this analysis, I found significant effects on Attitudes toward the Learning Situation on
measures taken toward the end of the year. Thus, there is some evidence that class experiences can
have an effect, particularly on attitudes toward the learning situation. Class differences are not
reflected in the other measures, however.
For the Spanish students, there was significant variation toward the end of the academic year
among the 9 classes for the elementary school students and among the 8 classes for the secondary
school students in motivation, integrativeness, and attitudes toward the learning situation. For the

15
secondary school students, there was also significant variation in instrumental orientation. Thus,
there is clear indication that the dynamics of the classroom and perhaps the characteristics of the
surrounding community influenced many of the attitudinal and motivational characteristics of these
younger students.
These results are intriguing. Clearly, there are differences between the results for the
Canadian sample and for the Spanish ones, which might be attributable to differences in the cultural
settings. There is also a difference in the ages of the students, and this might well be a contributing
factor that warrants further study. What is clear in all three samples is that the attitudes toward the
learning situation vary from class to class, indicating that clearly here the teacher and the
surroundings can have an influence on the students. This in itself warrants further study to
determine what features of the classroom environment account for the differences. These result
indicate that teachers can make a difference, and possibly with younger children these differences
might generalize to both integrativeness and motivation.
One finding that is fairly common in the literature is that over the course of an academic
year, and even longer, students levels of motivation and attitudes tend to decline (see, for example,
Gardner, 1985). This would seem to have implications concerning whether or not teachers can
motivate students. If motivation and attitudes continually decline over the course of the year, it
throws into question the motivating capabilities of the teachers concerned. It is nonetheless possible
that some students might be motivated and some not, and it is reasonable to expect that how much
would be reflected in their final grades in the course. The model would predict that individuals who
obtained good grades would show higher levels of motivation and possibly more favourable attitudes
(or perhaps less decline) than students obtaining lower grades.
I also investigated the effects of the year in class on students scores on the five aggregate
variables as a function of their final grades in the class. In both studies, the students were tested at
the beginning of the year, and again toward the end. For the Canadian sample, I found significant
decreases over time for Motivation, Attitudes toward the Learning Situation, and Language Anxiety.
Moreover, the changes for Motivation and Attitudes toward the Learning Situation were moderated
by the final grade the students obtained in the class. The results for the two interactions are shown
in the Figures 4 and 5.

16
Figure 4
Fall and Spring Motivation scores as a function of Final Grades - Canadian Sample

Motivation Scores
6.0

Mean Motivation

5.8
5.6
5.4
5.2

Grade Obtained

5.0

<B

4.8

4.6

Fall

Spring

Time of Testing

Figure 5
Fall and Spring Learning Situation Attitudes as a Function of Final Grades - Canadian Sample

Attitudes Toward the Learning Situation

Attitudes Toward Learning Situation

5.4

5.2

5.0

4.8

Grade Obtained

4.6

<B
4.4

4.2

Fall

Spring

Time of Testing

17
As can be seen, for motivation, there were very slight decreases for students who obtained
A grades, slightly greater ones for those obtaining B grades and even larger ones for those obtaining
less than B grades. Similar results were obtained for attitudes toward the learning situation.
I conducted a similar analysis for the elementary and secondary students in Spain (Gardner
& Bernaus, 2004) and obtained similar results for both groups of students. Both groups showed
significant declines in integrativeness and motivation, and significant increases in language anxiety
over the course, while the elementary students also demonstrated a significant increase in attitudes
toward the learning situation and the secondary students showed a significant decline in instrumental
orientation. Of greater relevance to this discussion, however, the interaction between Time of
testing and Final Grade (defined as top, bottom, and middle third) was significant for the secondary
students. This interaction is shown in Figure 6.
Figure 6
Fall and Spring Learning Situation Attitudes as a Function of Final Grades - Spanish Sample

Attitudes toward the Learning Situation

Secondary School Students
Learning Situation Attitudes

4.8
4.6
4.4
4.2
4.0

Grade Group

3.8
Bottom Third

3.6

Middle Third

3.4

Top Third

3.2
Fall

Spring

Time of Testing

As can be seen, these results are very similar to those obtained on the Canadian sample,
though in this case the top students actually showed a slight increase in favourable attitudes.
Nonetheless, the students in the middle third of the final grade range showed a slight decrease, while
those in the lower third of the class showed a larger decrease in favourable attitudes.
These results again demonstrate that what takes place in the classroom can have a decided
influence on attitudes toward the learning situation, and possibly (at least in the case of the Canadian

18
sample) on motivation. There is no indication in these data that final performance moderates
integrativeness in any of the samples. An interesting next step in research might be to investigate
the effects of specific strategies teachers use to motivate students, and determine what effects they
have on students attitudes and motivation.
The Generalizability of the socio-educational model and the AMTB.
A statement that is often made in the literature is that the AMTB is appropriate for Canada
because it is a bilingual country and hence the research is concerned with second language learning,
but that it is not appropriate for foreign language learning. Reasons given for this are either that the
language is not readily available (Oxford, 1996) and/or that it lacks political importance in the
community (Drnyei, 2001). Some researchers, therefore, feel that the AMTB is not appropriate
for their settings.
We decided to test this hypothesis by conducting research in a number of countries. To date,
we have completed studies in four countries, Croatia, Poland, Romania, and Spain. I would like to
extend my sincere appreciation to my colleagues in these settings for their invaluable assistance and
cooperation. They are Merc Bernaus, Universitat Autnoma de Barcelona, Espaa, Jelena
Mihaljevi Djigunovi, Sveuilite u Zagrebu Hrvatska, Croatia, Gabriela S. Matei, Consultant
independent, Timioara, Romnia, and Anna Murkowska, Uniwersytet Warsawski, Polska.
The results obtained indicate that the AMTB is clearly appropriate to these four countries,
and that the results typically obtained in Canada are obtained there as well. One indication of the
comparability of these results to those obtained in Canada are the Cronbach measures of internal
consistency reliability. These values are presented in Table 1, along with the sample sizes indicated
in brackets along with the class type (i.e., n = 166 in the elementary sample from Croatia). As can
be seen, the reliability coefficients are invariably high for each age group in each country. The
median reliabilities range from .79 to .88. Closer inspection will reveal that except for Romania,
the reliability coefficients tend to increase with age. Overall 35 of the 48 reliability coefficients are
higher (or equal) for the older students than the younger ones. Such results would be expected since
their attitudes would tend to be more fully developed.

19
Table 1
Internal Consistency Reliabilities
Croatia

Poland

Romania

Spain

PS 6
(166)

SS 2
(153)

KL 1
(216)

KL 3
(194)

lev 6
(158)

lev 8
(155)

eso 2
(232)

eso 4
(199)

Atts. To English People

Int. Foreign Languages
Integrative Orientation
Motivational Intensity
Desire
Atts to Learn English
English Teacher
English Course
English Class Anxiety
English Use Anxiety
Instrumental Orientation
Parental Encouragement

.73
.80
.65
.78
.79
.90
.90
.89
.88
.88
.59
.77

.79
.76
.55
.80
,81
,90
.92
.89
.91
.92
.61
.81

.84
.63
.83
.68
.79
.86
.89
.88
.85
.83
.69
.75

.86
.67
.72
.59
.81
.88
.89
.86
.88
.89
.76
.71

.84
.79
.81
.64
.83
.85
.91
.89
.76
.78
.62
.83

.79
.73
.64
.68
.79
.80
.93
.89
.83
.78
.62
.81

.80
.79
.81
.73
.83
.87
.90
.86
.86
.81
.60
.85

.82
.77
.80
.77
.90
.92
.93
.90
.87
.85
.72
.89

Medians

.80

.81

.83

.84

.82

.79

.81

.88

The second indication of the applicability of the results to these countries are the correlations
of the measures with grades obtained in English at the end of the year. These results are presented
in Table 2.

20
Table 2
Correlations between Aggregate Measures of the AMTB and Grades in Four Countries
Younger Students
Measures

Croatia

Poland

Romania

Spain

ALS
INT
MOT
ANX
INST
PE

.26**
.28**
.37**
-.65**
.08
-.23**

.34**
.39**
.45**
-.30**
.36**
.12

-.06
.22**
.25**
-.34**
.17*
.03

.14
.32**
.40**
-.39**
.22*
.12

Older Students
Measures
ALS
INT
MOT
ANX
INST
PE

.18*
.20*
.35**
-.39**
.07
.04

.12
.30**
.40**
-.50**
.15
.23**

.17*
.35**
.28**
-.46**
.16
.15

.18*
.37**
.49**
-.38**
.21**
.04

As can be seen, all of the correlations involving integrativeness, motivation, and language
anxiety are significant and in the expected direction. In fact, 35 of the 48 correlations are
significant, though one (involving parental encouragement for the young Croatian students) is
contrary to expectations. Of the 13 correlations that are not significant, 6 involve the measure of
parental encouragement, 3 are concerned with the measure of attitudes toward the learning situation,
and 4 are associated with instrumental orientation. In short, in these countries, the best predictors
of final grades are language anxiety, motivation and integrativeness, roughly in that order. These
results are very consistent with those obtained in Canada.
Conclusions
In summary then, the major points that I have tried to make today are that:
1. Motivation is a multifaceted concept, involving cognitive, affective and behavioural
components. Its essence cannot be captured by only one aspect. Thus, orientations in and of
themselves do not necessarily reflect motivation. Reasons for doing something may indicate
motivation and maybe not. Unless the reasons are accompanied by the many other features of
motivation, the reason is just a reason, not a motive.
2. Integrative motivation refers to a constellation of attributes. It is not something that some
people have and others dont. What our research has indicated clearly, I believe, is that if an
individual is highly motivated to learn an other language, has an open and accepting approach to

21
other cultural groups and/or a strong emotional interest in the target language group, and has a
positive evaluation of the learning situation, then we might describe that person as being
integratively motivated to learn the language. We would probably also find that the person is very
successful in learning and using the language.
3. The socio-educational model is one parsimonious way of accounting for individual
differences in second language acquisition. It has served us well in suggesting hypotheses that could
be tested, and it helps to organize research findings.
Finally, I have tried to demonstrate how this conceptualization of the motivation to learn a
second language fits in well and could greatly improve movements to identify ways of motivating
students. Presumably, the aim is to motivate them to develop a high level of proficiency in the
language and not just to enjoy the language class and evaluate the learning situation positively. This
might have an influence on motivation, as the socio-educational model suggests, but research is
needed to test this hypothesis further.
And finally, finally, our initial findings suggest that if one pays attention to adapting the full
AMTB to other cultural settings, the results obtained will be consistent with those obtained mostly
in Canada and provide further support for the validity of the socio-educational model.
References
Bruner, J. S. (1966). Toward a Theory of Instruction. New York: Horton.
Carroll, J. B. (1963). A model of school learning. Teachers College Record, 64, 723-733.
Crookes, G., & Schmidt, R. W. (1991). Motivation: Reopening the research agenda. Language
Learning, 41, 469-512.
Czizr, K. & Drnyei, Z. (2005). The internal structure of language learning motivation and its
relationship with language choice and learning effort. Modern Language Journal, 89, 19-35.
Drnyei, Z. (2001). Teaching and researching motivation. Essex: Longman.
Gardner, R. C. (1983). Learning another language: A true social psychological experiment.
Journal of Language and Social Psychology, 2, 219-239.
Gardner, R. C. (1985). Social psychology and second language learning: The role of attitudes and
motivation. London: Edward Arnold Publishers.
Gardner, R. C. (1988). The socio-educational model of second language learning: Assumptions,
findings and issues. Language Learning, 38, 101-126.
Gardner, R. C. & Bernaus, M. (2004). The applicability of the Attitude/Motivation Test Battery to
EFL students in spain. Unpublished manuscript, University of Western Ontario.
Gardner, R. C., Lalonde, R. N., & Moorcroft, R. (1985). The role of attitudes and motivation in
second language learning: Correlational and experimental considerations. Language
Learning, 35, 207-227.
Gardner, R. C., & Lambert, W. E. (1959). Motivational variables in second language acquisition.
Canadian Journal of Psychology, 13, 266-272.
Gardner, R. C., & Lambert, W. E. (1972). Attitudes and motivation in second languagelearning.
Rowley, Mass.: Newbury House.
Gardner, R. C., & MacIntyre, P. D. (1991). An instrumental motivation in language study: Who

22
says it isn't effective? Studies in Second Language Acquisition, 13, 57-72.
Gardner, R. C., & MacIntyre, P. D. (1993). On the measurement of affective variables in second
language learning. Language Learning, 43, 157-194.
Gardner, R. C., Masgoret, A.-M., Tennant, J. & Mihic, L. (2004). Integrative motivation: Changes
during a year-long intermediate-level language course. Language Learning, 54, 1-34.
Glazer, R. (1976). Components of a psychology of instruction: Toward a science of design. Review
of Educational Research, 46, 1-24.
Gliksman, L., Gardner, R. C., & Smythe, P. C. (1982). The role of the integrative motive on
students' participation in the French classroom. Canadian Modern Language Review, 38,
625-647.
Keller, J. M. (1983). Motivational design of instruction. In C. M. Reigeluth (Ed.), Instructional
Design Theories and Models (pp 386-433). Hillsdale, NJ: Erlbaum.
Kleinginna, P.R. & Kleinginna, A.M. (1981). A categorized list of motivational definitions with a
suggestion for a consensual definition. Motivation and Emotion, 5, 263-291.
Kraemer, R. (1993). Social psychological factors related to the study of Arabic among Israeli
highschool students. Studies in Second Language Acquisition, 15, 83-105.
Lambert, W. E. (1955). Measurement of the linguistic dominance of bilinguals. Journal of
Abnormal and Social Psychology, 50, 197-200.
Masgoret, A.-M. & Gardner, R. C. (2003). Attitudes, motivation, and second language learning:
A meta-analysis of studies conducted by Gardner and Associates. Language Learning, 53,
123-163.
Mowrer, O. H. (1950). Learning Theory and Personality Dynamics. N.Y.:Ronald.
Oxford, R. L. (1996). New pathways of language learning motivation. In R. L. Oxford (Ed.),
Language Learning Motivation: Pathways to the New Century (Technical Report No. 11,
Honolulu: University of Hawaii Second Language Teaching and Curriculum Center, 1-8.
Tremblay, P. F., & Gardner, R. C. (1995). Expanding the motivation construct in language
learning. Modern Language Journal, 79, 505-518.
Tremblay, P. F., Goldberg, M. P., & Gardner, R. C. (1995). Trait and state motivation and the
acquisition of Hebrew vocabulary. Canadian Journal of Behavioural Science, 27, 356-370.
Yashima, T. (2002). Willingness to communicate in a second language: The Japanese EFL Context.
The Modern Language Journal, 86, 54-56.
Yashima, T., Zenuke-Nishide, L., & Shimizu, K. (2004). The influence of attitudes and affect on
willingness to communicate and second language communication. Language Learning, 54,
119-152.
Endnote
1. A joint plenary address presented at the meetings of the Canadian Association of Applied
Linguistics and the Canadian Linguistics Association on May 30, 2005 at The University of
Western Ontario, London, Canada. Preparation of the manuscript was facilitated by a grant to
R.C. Gardner from the Social Sciences and Humanities Research Council of Canada (Grant No.
410-2002-0810).