|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
INTERPLAY AMONG ANXIETY, MOTIVATION, AND AUTONOMY IN SECOND
LANGUAGE LEARNERS OF FRENCH: A QUANTITATIVE AND QUALITATIVE
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
I would like to express my sincere gratitude to Dr. Theresa Antes, Dr. Gillian Lord,
Dr. Candace Harper and Dr. Virginia LoCastro for having served on my committee, and
for providing me with invaluable guidance, feedback and encouragement.
My heartfelt thanks go to the instructors and students of the two French classes
observed. It would have been difficult to conduct this study without their willingness and
cooperation throughout the data collection process.
The encouragement and patience extended to me by my dear family and friends
were much appreciated. I wish so much that my mother had lived just a year longer to
witness the completion of this dissertation.
TABLE OF CONTENTS
A C K N O W L E D G M E N T S .................................................................................................. ii
LIST OF TABLES ................ ......... ....................... ....... vi
ABSTRACT .............. ............................................. vii
1 IN TR OD U CTION ............................................... .. ......................... ..
Purpose of the Study ............... ..................................................3
R research Q u estion s........... .............................................................. .. ...... ..... ..3
Q u antitativ e ............................................... .......................... 3
Q u alitativ e ................................................. ......................... 3
Significance of the Study .............................................................................. ...... .4
O organization of the Study .......................................................... ..............4
D efinitions of T erm s U sed.......................................... ....................................... 5
Affective Factors ................... ......... ................. .................5
A n x iety ....................................................... 5
M motivation ................................................................. 5
A u to n o m y .................................................................. .............................. . 5
2 LITERATURE REVIEW ...........................................................6
Individual D differences ....................................................... 6
A anxiety ............................................................. .8
M otiv atio n ...........................................................17
The A nxiety-M motivation Interface .............................................................................. 25
A utonom y ................................................................26
The Motivation-Autonomy Interface ................................................................ 29
The Anxiety-Motivation-Autonomy Interface ........................................... 32
3 M ETH OD OLO G Y .............................................................35
P a rtic ip a n ts ..................................................................................................... 3 5
P rocedu res ......................................................................................................... 39
In strum ents ....................................................... 42
In-house F rench T est ........4....... ...... ........................................ .............. ..42
The Foreign Language Classroom Anxiety Scale (FLCAS)..............................42
The Attitude/Motivation Test Battery (AMTB) ...............................................44
The Questionnaire on Autonom y ............................................. ............... 44
Class O observations ...................................... ................... ..... .... 45
Journals ...................................................................................................... 46
Student Interview s .............................................. .. ...... .. ............ 47
Teacher Interview s ........................................ ................. ..... .... 47
Quantitative Data............... ... .. ...................... .........48
Q u alitativ e D ata .......................................................50
N o te s ..................................................................................................................... 5 1
4 RESULTS AND DISCU SSION .......................................... ........................... 52
T y p e o f A n x iety .................................................................. ................................5 2
Research Question 1 ..................................................... ............ 54
R research Question 2 ................. ............... ............ .. .... .. .......... 58
Correlation between Anxiety and Motivation................. ............................59
Correlation between Anxiety and Autonomy.....................................................60
Correlation between Motivation and Autonomy ..............................................63
Research Question 3 ..................................................... ............ 65
Research Question 4 ..................................................... ............ 70
Research Question 5 ..................................................... ............ 74
Research Question 6 ..................................................... ............ 76
Research Question 7 ..................................................... ............ 85
T est A n x iety ................................................................8 8
Extensive M material ............................................................ .... ......... 88
Fast Pace of Class .................. ..................................... ................. 89
L language R equirem ent............................................... ............................. 89
Communication with Instructor..................... ..... ........................... 89
D dissatisfaction w ith C lass......................................................... ............... 90
5 C O N C L U SIO N ......... ......................................................................... ........ .. ..... .. 93
S u m m a ry ..................................................................................................9 3
Pedagogical Im plications ............................................................................ .94
A nxiety and M otiv ation ............................................................ .....................94
A nxiety and A utonom y ............................................... ............................ 95
M otivation and A utonom y ........................................... ........................... 95
Anxiety, Motivation and Class Grades.....................................................96
Teacher Effect ............................................ ................ ............ 97
Possible Reasons for Lack of Motivation.........................................................97
Lim stations of the Study ......................................................... .............. 97
F utu re research ..............................................................9 8
A PARTICIPANT BACKGROUND QUESTIONNAIRE ..........................................100
B IN-HOUSE FRENCH TEST ........................................................ ............. 102
C QUESTIONNAIRE ON ANXIETY.................................... ......................... 106
D QUESTIONNAIRE ON MOTIVATION..............................................................109
E QUESTIONNAIRE ON AUTONOMY ............... ...........................................113
F T O PIC S FO R JO U R N A L .......................................................................... .... .... 117
G QUESTIONS FOR STUDENT INTERVIEWS......................................................119
H QUESTIONS FOR TEACHER INTERVIEWS ............... ................... ...........120
I OVERVIEW OF PARTICIPANTS WHO EXPERIENCED HIGH LEVELS OF
ANXIETY, MOTIVATION, AND AUTONOMY...............................................121
B IO G R A PH IC A L SK ETCH ................................................................ .....................129
LIST OF TABLES
3.1 Relevant demography of students (total N=32) ....................................... .......... 38
3.2 The range of points for the three categories on each measure...............................42
4.1 Student classification based on scores received on questionnaires (n=32) .............55
4.3 The correlations among participants' levels of the three variables..........................59
4.4 Students who were highly anxious and/or highly autonomous ............................61
4.5 Paired samples t-test results showing change only in motivation levels..................66
4.6 Group differences in motivation scores ....................................... ............... 67
4.7 Correlation of changes from test to retest ........................................ ............... 69
4.8 Changes in one factor showing corresponding changes in others............................70
4.9 Summary of correlations between the French test scores/class grades and
affective factors ............. ..... ....... ................. ................ 71
4.10 Correlation between anxiety scores of moderately anxious participants and their
F ren ch test scores ................................................... ................ 7 3
4.11 Correlation between anxiety scores of moderately anxious participants and their
final class grades ......................... ......... .. .. ..... .. ............. 74
4.12 Group-wise mean scores on the three questionnaires during Week 5......................75
4.13 Group-wise mean scores on each questionnaire during Week 13..........................75
4.14 Students who are highly motivated and highly autonomous during Test 1
a n d /o r T e st 2 .............................................................................................................7 6
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
INTERPLAY AMONG ANXIETY, MOTIVATION, AND AUTONOMY IN SECOND
LANGUAGE LEARNERS OF FRENCH: A QUANTITATIVE AND QUALITATIVE
Chair: Theresa A. Antes
Major Department: Linguistics
Research in second language learning has shown a close connection between
anxiety and motivation as well as between motivation and autonomy. This study goes
one step further in examining the interactions among these three affective factors,
namely, anxiety, motivation and autonomy, and their influence on class performance.
Both quantitative and qualitative data were collected over the course of the semester.
The 32 participants, who were first-year French students at the University of Florida,
completed four instruments during Weeks 5 and 13 of the semester. The instruments
used were an in-house French test, the Foreign Language Classroom Anxiety Scale, the
Attitude/Motivation Test Battery, and The Roles of Learners and Teachers, which is a
questionnaire on autonomy. The students were observed in their French classes at least
once a week. Eleven students wrote journal entries and six participated in an interview.
The results showed that, in general, approximately 22-25% of the participants in this
study were highly anxious, 53-59% were highly motivated, while 13-16% were highly
autonomous. Of the three affective factors, there was a correlation only between
motivation and autonomy. Further, the participants' levels of anxiety and autonomy were
stable from test to re-test, but there was a statistically significant decrease in their levels
of motivation. This change in the participants' levels of motivation did not correlate with
a corresponding significant change in the other two factors, but results show that there are
possibilities of changes occurring in both anxiety and autonomy. Only anxiety had a
negative correlation with students' class performance. Results also showed that students
who are autonomous are also motivated, but students who are motivated are not always
Research in the field of second language acquisition acknowledges that both
cognitive and affective factors within a language learner affect the level of success that
will be achieved in learning a second language (L2). Earlier research concentrated on
discovering the effect of cognitive factors on language learning, but since the 1960s,
there have been more investigations on determining the effect of affective factors, such as
anxiety and motivation (Gardner, 1985; Skehan, 1989). Affective factors are defined as
"those that deal with the emotional reactions and motivations of the learner" (Scovel,
Most studies have found an inverse relationship between anxiety and language
achievement of L2 learners, but in some cases it has been observed that anxiety
encourages a student to work harder, resulting in better class performance (Horwitz,
Horwitz & Cope, 1986; Phillips, 1992; Trylong, 1987). With regards to motivation,
evidence shows that motivated students perform better in the classroom than those who
are unmotivated. MacIntyre and Gardner (1991) are of the opinion that both anxiety and
motivation influence language learning and are good predictors of success. Their data
suggest that the two variables have an inverse relationship, such that the higher the levels
of anxiety experienced by the learners, the less motivated they tend to be. Consequently,
these students put less effort into their learning process, which often results in lower class
grades. The same studies show, however, that moderate levels of anxiety can act as an
incentive and motivate the learners to work harder, potentially resulting in higher grades.
More recently, researchers have examined the relationship between motivation and
autonomy. Dickinson (1995) posits that when learners are actively and independently
involved in their own learning, their motivation levels increase and they learn more
effectively. Spratt, Humphreys and Chan (2002) speculate on whether autonomy
precedes motivation or motivation precedes autonomy in a language learner. In other
words, they question whether students should first be motivated in order to develop and
show signs of being actively and independently involved in their learning, or whether
they should first be autonomous, which will then influence their motivation levels to
increase. While most investigations reveal that motivation and autonomy are distinct
factors, views differ on whether students need to first be motivated or autonomous in
order to be successful language learners.
Recognizing that second language research has found a close relationship between
motivation and anxiety, and also between motivation and autonomy, one wonders how
these three affective factors interact together and what effect they would have together on
students' class grades. Anxiety and motivation are considered to be good predictors of
success, and learners' levels of motivation are said to rise when they are actively and
independently involved in their own learning. Does this mean that teachers will have to
concentrate on encouraging students to be autonomous in order to increase their
motivation to learn more effectively, rather than trying to motivate them to follow the
established curriculum? What relationship do anxiety and autonomy have? Can anxious
students be autonomous? What happens to the correlation with other factors when there
is a change in one of them? These are the questions the present study seeks to answer.
Purpose of the Study
This study seeks to observe the levels of anxiety, motivation, and autonomy in
beginning French students, and to determine if these affective factors are stable over an
eight-week period. Correlation tests are run to examine how these three affective factors
interact with each other and what effect they have on participants' class performance. An
attempt is also made to ascertain the degree to which motivation and autonomy are
distinct factors. Finally, a comparison is made between the quantitative results obtained
from two intact classes taught by different instructors in order to investigate possible
differences and, if any, reasons for these.
This study employs both quantitative and qualitative data. Accordingly, the
research questions are divided into two sections:
1. To what extent, if any, are the participants anxious, motivated, and autonomous?
2. To what extent, if any, is there a correlation among the participants' stated levels of
anxiety, motivation, and autonomy?
3. To what extent, if any, are the participants' levels of anxiety, motivation, and
autonomy stable over time?
4. What is the correlation between the three affective factors and students'
performance, as measured by the French test and the final class grade?
5. In what ways, if any, do the results obtained from the participants in the two intact
classes, taught by different instructors, differ?
6. Are motivation and autonomy distinct factors?
7. What are some possible reasons for a change, if any, in the levels of anxiety,
motivation, and autonomy experienced by the participants from test to retest?
Significance of the Study
Earlier studies have explored the relationship between the two affective factors of
anxiety and motivation and their effect on class performance. Other studies have also
investigated the role of autonomy in language learning, but no known study thus far has
investigated the combined interactions and influence of anxiety, motivation, and
autonomy on class performance or language achievement. Furthermore, unlike most
previous studies, which relied on one type of data, this study used data collection
procedures that were both quantitative and qualitative in nature in the hope that a more
complete picture of the interaction among the three affective variables can be obtained.
The participants completed questionnaires on anxiety, motivation, and autonomy towards
the beginning as well as the end of the semester, and were observed in class at least once
a week throughout the semester. They also submitted fourjournal entries on specific
topics given to them at intervals, and at the end of the semester were interviewed about
their experiences in the French course.
Organization of the Study
This dissertation is divided into five chapters. After this introductory chapter,
Chapter 2, provides an overview of the research carried out in the area of individual
differences of second language learners, and focuses on the three affective factors under
consideration in this study. Chapter 3 describes the methodology used in this study. A
detailed account of the instruments used and both the quantitative and qualitative data
collection procedures are described. The Results and Discussion chapter, Chapter 4,
gives the results of the tests run on the quantitative data collected and discusses these
results, along with describing the qualitative data collected. Finally, Chapter 5 brings the
dissertation to a close by summarizing results, suggesting pedagogical implications,
discussing the limitations of the study, and outlining avenues for possible future research.
Definitions of Terms Used
Affective factors are defined as "those that deal with the emotional reactions and
motivations of the learner" (Scovel, 1978:131).
Anxiety is defined as "the feeling of tension and apprehension specifically
associated with second language contexts, including speaking, listening, and learning"
(MacIntyre & Gardner, 1994:284).
"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 that respect" (Keller,
Autonomy is "the ability to take charge of one's learning ... To take charge of
one's learning is to have, and to hold, the responsibility for all the decisions concerning
all aspects of this learning" (Holec, 1981:3).
A substantial amount of research done in second or foreign language learning has
revealed that the individual differences found in learners are caused by both cognitive
and affective factors. As early as the 1920s, researchers (e.g., Henmon, 1929) first began
to investigate cognitive factors such as language learning aptitude, learning strategies and
intelligence; however, in the last four decades, second language acquisition researchers
(e.g., Gardner et al., 1976; Horwitz et al., 1986) have become more aware of the fact that
researching individual differences that are affective in nature, like motivation, anxiety
and self-confidence, is just as important as researching the cognitive variables like
language learning aptitude, learning strategies, and intelligence. They have realized that
these two sets of factors work together to influence both the process and the outcome of
language acquisition. Believing that each language learner is unique and works with a
distinct combination of cognitive and affective variables that determine the process of
second language acquisition, Gardner, Day and MacIntyre (1992) hypothesize that "there
are probably as many factors that might account for individual differences in achievement
in a second language as there are individuals" (p. 212). This is probably why much
second language research has focused on individual differences in recent years.
Schumann (1994) strives to shed some light on the connection between these two
sets of variables, the affective and cognitive, by citing Mishkin and Appenzeller (1987),
who explain that the amygdale, a part of the temporal lobe in the brain, "assesses the
emotional significance and motivational relevance of stimuli; this appraisal then
influences attention and memory" (p. 233). We thus learn that linguistic input, which is a
form of stimulus, is first evaluated for its emotional significance and motivational
relevance to the learner before it can be processed by the brain. This evaluation
determines whether or not the linguistic input is attended to and stored in memory.
Considerable research in the area of second language learning reveals that emotions
play an important role in language acquisition (e.g., Horwitz, 2001). Tomkins (1970)
states that human beings are always experiencing some sort of emotion in varying
degrees, and strong emotion can disrupt cognitive and physiological processes. This
could account for the fact that some language learners perform better when they
experience positive emotions such as motivation and enthusiasm, or perform poorly when
they have negative emotions such as anxiety or low self esteem. MacIntyre and Gardner
(1994b) concur when they claim, "Some of the strongest correlations between affective
variables and achievement measures involve anxiety" (p. 284). Several studies have
shown that anxiety causes cognitive interference, resulting in significant negative
correlations between language anxiety and course grades (e.g., Gardner, Moorcroft, &
MacIntyre, 1987; Phillips, 1992; Trylong, 1987).
As a result of the growing awareness that emotions play an important role in
language learning, researchers, since the 1970s, have increasingly focused on the
affective variables of a language learner, such as anxiety (e.g., Horwitz, Horwitz & Cope,
1986; Onwuegbuzie, Bailey & Daley, 2000; Saito & Samimy, 1996) and motivation (e.g.,
Dornyei & Schmidt, 2001; Gardner, Tremblay & Masgoret, 1997). In the last two
decades researchers have also turned their attention to the variable of autonomy (e.g.,
Benson, 2001; Little, 1991; Littlewood, 1996). They have found that both anxiety and
motivation are good predictors of success in language learning, and that motivation and
autonomy share a relationship. For these reasons, this study strives to investigate and
shed some light on the interplay among the three affective factors of anxiety, motivation
and autonomy in college students learning French as a foreign language.
In 1978, Scovel examined the relationship between anxiety and second language
achievement and concluded that anxiety is one of the most predominant variables
identified in language learning tasks. He observed that different studies used a variety of
instruments to measure anxiety and came up with different results. He believed that their
results would be less conflicting if the researchers specified the type of anxiety (e.g., state
anxiety, trait anxiety, and test anxiety) they were measuring which would have made it
easier to compare the results. Much later, Young (1994) provided a comprehensive
overview of the major research carried out on language anxiety since the 1970s, and
tabulated the findings of various researchers, although the results were often
contradictory and a consensus had yet to be reached. Some studies showed a negative
correlation between language anxiety and language performance (Aida, 1994; Coulombe,
2000; Saito & Samimy, 1996), while other studies showed no effect or a positive
correlation between the two (Chastain, 1975; Kleinmann, 1977). In one study, some of
the least proficient students scored the highest on an anxiety scale, whereas others scored
the lowest (Backman, 1976). One of the reasons Young (1994) offers for these
inconsistent results is that "many of the studies had different goals, objectives,
definitions, and conceptual schemata, rendering comparisons difficult" (p. 4). The
second reason she offers is the lack of a reliable and valid instrument for measuring
foreign language anxiety.
Observing that researchers had "neither adequately defined foreign language
anxiety nor described its specific effects on foreign language learning," Horwitz et al.
(1986:125) attempted to fill this gap by conceptualizing foreign language anxiety as a
distinct variable, so that both language students and teachers would be able to recognize
the symptoms and consequences. They developed the FLCAS as a standard measure to
determine levels of foreign language anxiety. Young (1994) explains that with the
introduction of a valid and reliable measuring instrument, the Foreign Language
Classroom Anxiety Scale (FLCAS) (Horwitz et al., 1986), research on language anxiety
began to consistently show a moderate negative correlation between anxiety and
language achievement. This observation supported Scovel's view that researchers were
getting conflicting results in their studies because they were using a variety of
instruments to measure different types of anxiety.
Early researchers did not try to define foreign language anxiety because of its
complex, and multidimensional nature (Young, 1991). But later, MacIntyre and Gardner
(1994b) defined language anxiety as "the feeling of tension and apprehension specifically
associated with second language contexts, including speaking, listening, and learning"
Some researchers have found that language learners who experienced a higher level
of anxiety were more likely to be motivated to perform better in class and receive higher
grades than students with a lower level of anxiety. Alpert and Haber (1960) called this
type of language anxiety "facilitating." Other research (e.g., Trylong, 1987) found that
anxiety has the opposite effect on a language learner, where higher levels of anxiety
result in lower grades and is termed "debilitating." Scovel (1978) explores the likelihood
that both faces of anxiety, are at work together whenever a person performs any activity.
One part within an individual stimulates and stirs, while the other quells and balances,
resulting in an optimal mental state that motivates the learner to perform the task well. It
seems that having the right amount of anxiety, rather than none at all, might be beneficial
to language learning. The anxiety one feels at the possibility of failure may serve to
motivate a language learner to strive harder and achieve success. This not-so-unusual
phenomenon is elucidated by Eysenck (1979) who posits that "the extent to which
anxiety either facilitates or impairs performance is determined by the extent to which
high-anxiety subjects compensate for reduced processing effectiveness by enhanced
effort" (p. 365). Thus, it is widely accepted that anxiety can influence both the amount of
effort a language learner invests in a task and the quality of performance.
Other distinctions that researchers have made in their quest to discover the different
types of anxiety are "trait anxiety" and "state anxiety." Spielberger (1983) defines trait
anxiety as an individual's likelihood of becoming anxious in any situation, whereas state
anxiety is apprehension which is experienced only in some specific situations, for
example, in a language classroom. MacIntyre and Gardner (1989) define trait anxiety as
a stable propensity to be anxious in a number of different situations, while referring to
state anxiety as an unpleasant emotional condition or temporary state which one
experiences at a particular moment in time. In other words, trait anxiety is considered to
be a permanent personality feature, while state anxiety depends on the manner in which
an individual reacts to the circumstances she or he is in.
Yet another type of anxiety that researchers have investigated is the situation-
specific anxiety in which the "respondents are tested for their anxiety reactions in a well-
defined situation such as public speaking, writing examinations" (Maclntyre & Gardner,
1991:90). In these studies, the "respondents are required to make attributions of anxiety
to particular sources.... By testing more detailed hypotheses, the process by which a
given situation generates anxiety can be examined" (p. 91) and the anxiety-creating
sources are identified.
Researchers have not only identified different kinds of anxieties, but they have
also documented the observable changes that occur in individuals who experience
anxiety. Physiological signs of anxiety include perspiration, sweaty palms, dry mouth,
muscle contractions and tension, and an increase in heart rate (Chastain 1975; Gardner et
al.1985; Steinberg and Horwitz 1986). Behavioral signs of anxiety in an academic
setting include avoiding class, not completing assignments, and a preoccupation with the
performance of other students in class (Bailey 1983; Horwitz et al. 1986; Young 1992).
It is possible that language teachers may consider such behavior as disruptive and thus
miss an opportunity to identify and help their anxious students. Language anxiety even
results in learners experiencing a "distortion of sounds, inability to produce the intonation
and rhythm of the language, 'freezing up' when called on to perform, and forgetting
words or phrases just learned or simply refusing to speak and remaining silent" (Young
Teachers in most classrooms have observed students who have exhibited one or
more of the signs of anxiety while engaged in different language tasks, and at different
levels of instruction. They have also probably wondered what exactly happens when a
student is anxious in the language classroom that makes it impossible for her/him to
benefit from instruction. In order to find answers, researchers have examined the effect
of language anxiety on different tasks. As expected, a negative relationship was found
between foreign language anxiety and course grades of students in high school (Gardner
et al., 1987) as well as in college (Trylong, 1987). Language anxiety has a negative
impact on performance in oral examinations (Phillips, 1992), and on production of
vocabulary (Gardner et al., 1987). It was also found that students with the highest levels
of foreign language anxiety tended to report that (1) they spend too much time on some
subjects and not enough time on others, (2) they frequently do not get enough sleep and
feel sluggish in class or when studying, (3) they do not take breaks to avoid becoming too
tired while studying, and (4) they have trouble settling down to work and do not begin to
study as soon as they sit down (Bailey, Onwuegbuzie & Daley, 2000). Horwitz et al.
(1986) observed that anxious students studied harder to compensate for their lack of
confidence, had difficulty retrieving items from memory during exams, and had a fear of
making mistakes, leading to silence instead of participation. Clearly, anxiety has a
negative effect on language learners and it is not surprising that researchers claim that
anxiety is one of the best predictors of second language achievement (Ehrman & Oxford,
1995; MacIntyre & Gardner, 1994a, 1994b) because of the negative correlation between
anxiety and performance. Teachers work very hard to ensure that language learners are
successful in their classes. If teachers are able to identify their anxious students, they
might be able to modify their teaching styles and strategies to help the anxious students
feel more comfortable in class and thus have the opportunity to learn better.
One wonders what it is that happens to an anxious student's cognitive ability, to
cause poor class performance. As early as 1979, Eysenck hypothesized that an anxious
person's cognitive abilities are used not only to complete a task but also to deal with the
emotional aspects of accomplishing the task, doing neither full justice, thus resulting in
poor performance. He further postulated that an anxious student studies much harder
than a less anxious student to compensate for the cognitive interference. When the task is
simple, the anxiety experienced by a learner is facilitative. But when the task is difficult,
the anxiety experienced interferes with the cognitive abilities and is debilitative.
Maclntyre (1995) elaborates on this hypothesis:
... a demand to answer a question in a second language class may cause a student
to become anxious; anxiety leads to worry and rumination. Cognitive performance is
diminished because of the divided attention and therefore performance suffers, leading to
negative self-evaluations and more self-deprecating cognition which further impairs
performance. (p. 92)
He further claims that this relation between anxiety and performance is relatively
strong evidence in favor of considering anxiety as one of the major factors that creates
individual differences among second language learners.
One of the first studies that sought to observe and analyze anxiety in students at
different levels of proficiency was by Saito and Samimy (1996). Their study comprised
257 students in the beginning, intermediate, and advanced levels of Japanese courses at
the University of Texas at Austin. The goals of their study were (1) to find the
relationship between language anxiety and language performance at the three different
instructional levels, and (2) to find the variables that significantly correlated with
Japanese class anxiety at each of these three levels. The variables included in their
analyses were gender, year in college, length of time in Japan, time spent studying for the
class, language class anxiety, language class risk-taking, language class sociability,
strength of motivation, attitude toward the Japanese class, concern for grade, and final
The researchers found that "year in college" was the best predictor of students'
final grades only at the beginning level. "In other words, for college students in the
beginning Japanese classes, those who began (studying Japanese) at an upper level, such
as seniors or graduate students, were predicted by the model to receive higher grades,
while those who had just started college, such as freshmen, were predicted to receive
lower grades" (Saito & Samimy, 1996:245). An explanation provided by the authors for
this occurrence is that the students with high achievement scores are the most determined
to protect their grade point average. It is possible that the senior students had already
mastered some learning strategies in college and were better able to cope with tests and
assignments, resulting in better grades than the freshmen and juniors. It is also possible
that the seniors were more concerned about their grades than the freshmen, and so were
more determined to protect their grade point average.
On comparing the mean scores for language anxiety of the students at the
beginning, intermediate and advanced levels, the researchers discovered that the
advanced students scored the highest, the intermediate students scored the lowest and the
beginners scored in between the other two levels. These results contrasted with those
obtained by Gardner et al. (1977), who discovered that language beginners had the
highest mean scores for anxiety and the advanced students had the lowest mean scores,
indicating that foreign language anxiety "decreases as proficiency and training increase"
(p. 251). Similar results were obtained by MacIntyre and Gardner (1991), who claimed
that experience and proficiency are inversely proportionate to anxiety levels in a language
learner. Differing from both these stances, Bailey et al. (1998) investigated the anxiety
reported by 253 college students from a variety of disciplinary backgrounds enrolled in
either Spanish, French, or German classes. Their analysis indicated no difference in
anxiety about foreign languages among students in these three classes.
Saito and Samimy provide us with a rationale for the conflicting results they
obtained in their study in comparison with the results of Gardner et al. (1977). They
claim that foreign language anxiety is high in beginners of more commonly taught
languages, such as Spanish and French, because the learners are still in the process of
developing successful learning strategies. However, foreign language anxiety is high in
the advanced students in less commonly taught languages, such as Japanese, because at
this level more emphasis is placed on reading and writing rather than speaking and
listening, which is a challenge to the students. As a result, they do not have as much time
to practice their conversational skills and lack the confidence to speak the language,
which in turn, raises their anxiety levels. These findings suggest that the levels of anxiety
that language learners experience at the beginning, intermediate, and advanced stages of
language learning, depend on the language being learnt.
Onwuegbuzie et al. (1999) conducted a study with 210 university students in the
United States to investigate which factors out of 26 possible, correlated with language
anxiety. They included:
direct measures of self-perception (e.g., perceived intellectual ability, perceived
scholastic competence, perceived self-worth, and expected final course average for
current language course) as well as measures of constructs that are manifestations of self-
perceptions (i.e. social interdependence and study habits) (p. 221)
The participants represented 43 different degree programs offered at a mid-southern
university, and were enrolled in French, Spanish, German and Japanese classes at the
beginning, intermediate and advanced levels of instruction. The majority of the students
had never traveled outside the United States; however, they took this language course as
a requirement toward their degree. Most of the participants had previously studied a
foreign language, either in high school or college and the final grades they expected
ranged from 68 to 100, with a mean of 87.1.
The researchers found seven variables that accounted for 40% of the total variance
in foreign language anxiety. Those were: a student's expectation of her or his
achievement in a foreign language course (18% of variance); the number of foreign
countries visited (5%); perceived scholastic competence (5%); perceived self-worth
(5%); age (4% of variance); number of high school foreign language courses taken
(1-2%) and academic achievement (1-2%). The researchers did not find any significant
differences in the levels of foreign language anxiety between students enrolled in
beginning, intermediate or advanced classes, but they did find a significant difference in
anxiety levels with "year of study" as an independent variable. Foreign language anxiety
appeared to increase linearly, which means that in the same language classroom,
freshmen reported the lowest levels of foreign language anxiety and seniors reported the
The researchers do not give any information about the students' final grades in that
language classroom, but they do suggest some reasons for the high anxiety level in the
senior students. A higher proportion of seniors had already taken at least one college-
level foreign language course, which was not the case with the freshmen or sophomores.
It may be possible that many of the seniors had formed negative attitudes about language
classes through previous experience with learning languages in high school or college,
which resulted in their higher levels of anxiety or probably led to their decision to delay
taking a language course until it could not be put off any longer. The fact that the seniors
in this study reported higher levels of foreign language anxiety is consistent with the
results obtained by Crook (1979) and Hunt (1989) in their research. Finally, the
advanced level students in this study had a higher grade point average than the beginning
or intermediate level students, which might also account for higher anxiety levels in the
language classroom. Karabenick and Knapp (1988) found that students with high
academic achievement avoided asking for help because they perceived it as a sign of
failure, which in turn, increased their anxiety levels. High achievers were also anxious
about taking foreign language courses because they felt their grade point average would
be adversely affected and were most determined to protect their grade point average. All
these studies on language anxiety reveal that anxiety is a complex factor which is
manifested in many different forms.
The term "motivation" is used and is easily understood in everyday conversations,
yet, like anxiety, it has been difficult for researchers to agree on a definition or model for
motivation as it relates to language learning. Motivation is considered to be one of the
most important factors determining success in second or foreign language acquisition and
language teachers are constantly striving to find ways to motivate their students. Its role
in language learning has been examined in relation with other affective factors, such as
language anxiety and language aptitude. In 1959, Gardner and Lambert demonstrated
that motivation and language aptitude were both related to achievement in second
language learning. This study was then replicated in many different contexts and
languages and it is now generally believed that individual differences in motivation are
independent of individual differences in language aptitude.
Until recently, Gardner's (1985, 1988, 1997) socio-educational model of SLA,
which has undergone many changes and reformulations, dominated research in
motivation. Gardner (1996) refers to motivation as the driving force in any situation and
distinguishes between two distinct perspectives. One is that motivation is an internal
attribute of an individual and the other is that it is an external attribute, which means that
a person can be motivated by an external trigger or force, for example, when a teacher
attempts to motivate a student. Gardner (1996) says, "you can't motivate a rock" (p. 25),
which suggests that a student must already have the potential to be motivated by an
external force, for example, a teacher, and that motivation cannot be created out of thin
A great deal of emphasis has been placed on the learner's aim or motive in learning
a second language. If the learner's goal in learning a target language is to become a part
of the target group, then the learner is said to have integrative motivation, which involves
positive attitudes toward the L2 community and a desire to integrate into the community.
But, if the learner's aim is to learn a second or foreign language for financial or other
benefits, for example, to get ajob, then the learner is said to have instrumental
Gardner's model introduces the notion of "integrative motive," and shows that it is
comprised of three variables: (1) integrativeness, which is the person's desire to interact
with members of the target language; (2) attitudes toward the learning situation; and (3)
motivation, which comprises the person's attitude toward learning, desire to learn, and
the effort invested. Of these three, Gardner posits that it is "motivation" that influences
second language achievement, while the other two variables support motivation:
Motivation to learn the second language is viewed as comprising three elements.
First, the motivated individual expends effort to learn the language....there is a
persistent and consistent attempt to learn the material.... Second, the motivated
individual wants to achieve the goal. Such an individual will express a strong
desire to learn the language, ... Third, the motivated individual will enjoy the task
of learning the language. Such an individual will say that it is fun, a challenge, and
enjoyable, even though at times enthusiasm may be less than at other times
Gardner drives home the point that a truly motivated language learner will posses
all three of these elements: an effort to learn, a desire to learn the language, and positive
feelings while engaged in learning tasks. "Thus, Gardner's concept of motivation
provides for behavioral, cognitive, and affective components" (MacIntyre, et al.,
In addition to integrative and instrumental motivation, Ely (1986) adds a third type:
the need to fulfill a language requirement. Identifying this type of motivation is practical
as more and more students are learning a foreign language to fulfill a language
requirement at their university these days. Their motivation levels could be measured by
the amount of effort they put into their learning because, according to Keller (1983),
"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 that respect" (p. 389). This
definition is one that language teachers might easily relate to as they work to motivate
their students in the classroom. Gardner and Smythe (1975, 1981) developed the
Attitude/Motivation Test Battery (AMTB) to assess the motivation levels within language
learners. This proved to be a very reliable instrument and has been widely used for many
decades by various researchers.
Gardner's socio-educational theory was widely accepted until the mid-1990s,
when researchers began to feel that the framework provided by the model was too limited
and needed to be studied from different perspectives. They believed that the distinction
between integrative and instrumental motivation in Gardner's model was emphasized too
much, without any reference to other variables that played an important role in language
acquisition, for example, self-efficacy, expectancy, attributions, and locus of control
(Domyei, 1990, 1994; Oxford & Shearin, 1994). Crookes and Schmidt (1991) felt that
researchers investigating motivation should focus less on the reasons for learning
language and more on defining motivation as teachers and educators would, namely,
noting that a motivated student is one who "becomes productively engaged in learning
tasks, and sustains that engagement, without the need for continual encouragement or
direction" (p. 480). They believed that not only the concept, but also the definition of
motivation needed to be developed so that all teachers and educators could relate to and
accept the ideas proposed by the researchers.
Since then, several models and constructs have been proposed by researchers to
define and describe the role of motivation in second language acquisition, each of them
trying to add dimensions that reflect the attitudes and behavior patterns of motivated
students in a language classroom and to expand the socio-educational model.
MacIntyre et al. (2001) believed that it would be unproductive to develop new models or
constructs that covered the same basic concepts as did the socio-educational model.
Therefore, they decided to examine the level of empirical similarity between the variables
addressed in four different models of motivation: (1) Gardner's socio-educational model,
which was designed for language learning situations, (2) Pintrich's (1991) Motivated
Strategies for Learning Questionnaire (MSLQ), whose variables were taken from the
social-learning literature on motivation as applied to academic learning contexts, (3)
Kuhl's (1994b) Action Control Scale (ASC-90), which was used in psycho-physiological
studies of motivation; and (4) McCroskey's (1992) Willingness to Communicate (WTC)
scale from the literature on native language communication. We observe that Gardner's
socio-educational model was designed for language learning situations, whereas the other
three models were designed to be used in other contexts but could also be applied to
In the second model, Pintrich et al. (1991) examine general academic motivation
from an expectancy-value perspective, in which "expectancy" refers to the behavioral
outcomes one would expect, and "value" refers to the desirability of those outcomes. The
authors believe that the higher the value placed on a particular task, the more motivated a
person will be to complete the task. This reasoning is also extended to learning
languages. Pintrich et al. claim that the two-part process of learning involves the
assimilation of new information and the restructuring of internalized knowledge, called
accommodation. They believe that the process of restructuring internalized knowledge
requires much more effort than assimilating new knowledge, and whether a language
learner would take on the task of accommodation or not would depend on how motivated
the learner is. The authors developed the Motivated Strategies for Learning
Questionnaire (MSLQ) to measure motivation levels in language learners. These
motivation levels or scores will indicate the likelihood of a language learner exerting the
effort to restructure their internalized knowledge, which is a part of the learning process.
Kuhl (1994a), on the other hand, does not believe that learners will automatically
be more motivated in a high value task because there are many learners who continue to
engage in tasks that are not very productive or rewarding even though they have a choice
of tackling more rewarding tasks. Kuhl's model of motivation focuses on a theory of
action-versus-state orientation. He claims that action-oriented people prefer to be active
and are unwilling to be passive. State-oriented people, on the other hand, are more likely
to let things happen without any effort or intervention on their part, and tend to
contemplate on past and present feelings without taking any action to change their
affective state. This model, called the Action Control Scale (ACS-90) evaluates a
person's ability to initiate and maintain levels of behavior. It was initially developed to
assess differences in personality which affected a person's ability to maintain the drive to
start or complete a task (MacIntyre et al., 2001), and was then extended to include the
affective variable of motivation in language learning (Kuhl, 1994b).
The fourth model MacIntyre et al. examined was the Willingness to Communicate
(WTC) scale by McCroskey (1992), which consists of four communication contexts:
public speaking, talking in meetings, talking in small groups, and talking in dyads.
McCroskey explains that there are differences in the way individuals are motivated or
willing to initiate communication when they are free to do so. He believes that the two
most important factors that influence willingness to communicate are communication
competence and communication apprehension. Communication competence is the ability
to convey verbal or written information, whereas communication apprehension is defined
as the level of fear one experiences when thinking of how one's communication with
others will turn out (MacIntyre et al., 2001). Communication apprehension can have a
substantial, negative impact on communication competence (Rubin, 1990) and the WTC
scale would reflect a learner's level of motivation. This situation seems to parallel the
impact anxiety has on student performance.
As stated above, MacIntyre et al. (2001) examined these four models to test for the
degree of overlap among the constructs and found that there is considerable overlap
among the various concepts addressed by these constructs. For example, Gardner's
instrumental orientation and integrative motive significantly correlated with Pintrich's
extrinsic goal orientation and expectancy-value, respectively. MacIntyre et al. (2001)
observe "It might be surprising that there is such a high degree of overlap, but it should
be noted that the Gardner model has always covered a great deal of conceptual ground"
Even though Dornyei (1994) agreed that researchers would find it difficult to
surpass the work done by Gardner and his associates, who "established scientific research
procedures and introduced standardized assessment techniques and instruments, thus
setting high research standards and bringing L2 motivation research to maturity" (p. 273),
he believed that Gardner's social psychological approach was too influential and also
limited. Dornyei (1998) sought to expand on this approach by developing "a more
pragmatic, education-centred approach to motivation research, which would be consistent
with the perceptions of practicing teachers and, thus, be more directly relevant to
classroom application" (p. 204). He identified three dimensions of motivation in order to
outline a comprehensive and general framework of motivation that he feels would be
more relevant to the L2 classroom setting. The first level of his construct of motivation is
the "Language Level," where the focus is on orientations and motives related to the L2.
This level includes different aspects of the second language such as the culture it
conveys, the community it is spoken in, and the potential usefulness of being proficient in
the language. This part of the framework is similar to Gardner's integrative and
instrumental motivational subsystem. The second level of Dornyei's framework is the
"Learner Level," which involves both the affective and cognitive abilities of a language
learner. Individual differences, such as self-confidence and the need for achievement, are
considered in this level. In talking about self-confidence, other factors, such as language
anxiety and perceived L2 competence, are also included. The third level of the construct
is the "Learning Situation Level," which is made up of intrinsic and extrinsic motives and
three different motivational conditions: course-specific, teacher-specific and group-
specific motivational components. Dornyei's model of L2 motivation does seem to be
more education-centered and thus more consistent with the perceptions of practicing
teachers than Gardner's socio-educational model. Teachers are aware of the three levels
Dirnyei introduces, understand the demands they have to meet in each of these areas, and
are prepared to put in efforts to motivate their students. This model seems to focus
directly on the most important aspects of second language learning: the language, the
learner, and the learning environment; but its effectiveness in explaining and measuring
motivation in language learners needs to be researched and established.
Second language research has shown that motivation is multi-faceted. Some of the
new themes and approaches Dornyei (2001) presents are: social motivation, task
motivation, a neurobiological explanation of motivation, and motivation from a process-
oriented perspective. Further, Oxford and Shearin (1994) explore whether motivations
differ among learners of second versus foreign languages and believe that there are
chances that these two motivations work differently because in a second language
learning situation, the learners are exposed to the L2 inside as well as outside the
classroom, which is not the case in a foreign language learning situation. Integrative
motivation is an important part of the socio-educational model, but it may not have as
much meaning in foreign language situations, where the students may not come in direct
contact with the L2 community.
The social psychological approach to motivation in second language learning
emphasizes integrative motivation because this approach is concerned with the individual
in the context of a group. Undoubtedly, there is an urgent need to expand the theory of
motivation beyond Gardner's widely accepted model and to try to formulate a more
education-centered construct which teachers will be able to relate to and implement
successfully in their classrooms, particularly in foreign language classrooms.
The Anxiety-Motivation Interface
As has been stated, each language learner works with a unique combination of
cognitive and affective variables, which influences the process of second language
acquisition. Among the affective factors researched, anxiety and motivation, by
themselves or together, have received the most attention. Schumann (1994) interprets
Krashen's (1981) "affective filter" hypothesis as relating to these two factors as well:
"when motivation is lacking, anxiety is high, and self-esteem is low, the filter is up and
input will not become intake (i.e., input will not be processed so as to produce learning)"
(p. 233). He makes a very clear connection between these two affective variables.
Gardner et al. (1992) posit that "anxiety and motivation are two separate dimensions with
overlapping behavioral consequences (that are) correlated yet distinguishable"
(p. 212). The causal sequence cannot be established and probably differs from learner to
learner. It may be that a highly motivated language learner experiences low or no anxiety
levels and an unmotivated learner is very anxious.
Researchers who have investigated affective variables in language learners believe
that both anxiety and motivation can predict language proficiency. Maclntyre and
Gardner (1991) are of the opinion that "at the earliest stages of language learning,
motivation and language aptitude are the dominant factors in determining success ...
(and) anxiety plays a negligible role in proficiency" (p. 110), but at the later stages of
language learning, anxiety plays a more crucial role in determining success. They
believe that if a language learner's experiences are negative, then foreign language
anxiety may become a regular occurrence and influence student performance.
The third affective variable being investigated in this study is autonomy. The most
frequently cited definition of autonomy in the context of foreign language learning is that
autonomy is "the ability to take charge of one's learning" (Holec, 1981:3). Holec further
To take charge of one 's learning is to have, and to hold, the responsibility for all
the decisions concerning all aspects of this learning,
i.e.:--determining the objectives;
-- defining the contents and progressions;
-- selecting methods and techniques to be used;
-- monitoring the procedure of acquisition properly speaking (rhythm, time,
-- evaluating what has been acquired (p. 3)
Holec focuses here on the learning process, and the factors a language learner would need
to control, for effective learning to take place.
Little (1991), on the other hand, does not believe that autonomy is essentially a
matter of having control of and organizing one's learning process. He feels that
autonomy is a capacity for detachment, critical reflection, decision-making,
and independent action.... The capacity for autonomy will be displayed both in
the way the learner learns and in the way he or she transfers what has been
learned to wider contexts (p. 4).
Little believes that autonomous learning occurs not only in the confines of a classroom
but spills over into every other area of life. This view of autonomy encompasses the
capacity and the ability to be in control of one's cognitive faculties, and also to be able to
use this capacity for language learning and beyond.
Little (1991) strives to dispel some of the misconceptions people have about
autonomy in language learning. He states that some people think an autonomous person
makes the decision to learn without a teacher or requires a teacher to give up all control
over the student's learning. Little believes that this is not true because "we are social
beings (and) our independence is always balanced by dependence ... our capacity for self-
instruction probably develops out of our experience of learning in interaction with others"
Another misconception some people have is that "autonomy is something teachers
do to their learners" (p. 3). Little believes that autonomy cannot be developed in learners
through any set curriculum. This is probably why researchers are of the opinion that
autonomy is a "capacity" that learners have. Little reminds us that autonomy is not a
single behavior that can easily be described but that it takes different forms depending on
individual factors such as age and perceptions of learning needs. Finally, he states that it
is possible for learners to be highly autonomous in some areas and non-autonomous in
others, so autonomy is not "a steady state achieved by certain learners" (p. 4).
According to Littlewood (1996), an autonomous person is "one who has an
independent capacity to make and carry out the choices which govern his or her actions.
This capacity depends on two main components: ability and willingness" (p. 428).
Littlewood believes that an autonomous learner would need to have both characteristics,
ability and willingness, in order to be autonomous, since it is possible for a person to be
willing and not have the ability to make informed decisions, or vice versa. He further
explains the terms "ability" and "willingness." Littlewood claims that in order to have
the ability to become autonomous, a person should have both the "knowledge" of all the
options available to choose from and the necessary "skills" to make appropriate choices.
He also believes that the willingness of a person to become autonomous depends on
having both the "motivation" and the "confidence" to be responsible for the choices.
When discussing autonomy in second language learning, researchers focus on the
ability or the capacity of an individual to make choices independently, and furthermore,
to assume responsibility for those choices. It is believed that possessing this
characteristic will help not only in making informed decisions leading to success in the
language classroom, but that this trait will help the individual in all walks of life, even
outside the classroom.
Not many studies conducted thus far have focused on collecting both qualitative
and quantitative data on autonomy. Rivers (2001) analyzes self-directed language
learning behaviors of adult third-language learners based on qualitative data gathered
from 11 learners of Georgian and Kazakh at the University of Maryland and concludes:
The accurate use of metacognitive, affective, and social strategies to control the
language learning process and the learning environment is the hallmark of self-
directed language learning. In order for such learning to occur, learners must be
able to determine accurately what their needs are, and they must have the freedom
to take action to meet those needs. In the absence of either accurate self-
assessment or genuine autonomy, self-directed language learning will not occur. (p.
With the limited amount of research that has been carried out on autonomy, it is
difficult to understand whether the capacity to be autonomous language learners and
control learning comes naturally or if it needs to be acquired through training. There is
also a need to understand what conditions can help students develop autonomy in
language learning and what factors will hinder its development. Teachers may be able to
observe the behaviors of students in the classroom and gauge if a student is becoming
more autonomous or not, but it would be beneficial to have a global system of measuring
autonomy (Benson, 2001).
The Motivation-Autonomy Interface
While researchers generally argue that motivation and autonomy are separate
constructs, there is a definite interface between them. Keller (1983) believes that a
motivated person can choose to attempt a particular task and decide on the degree of
effort that will be exerted, but if the learner does not have the knowledge or the skills to
accomplish the task, no amount of motivation is going to result in the successful
accomplishment of the task. Additionally, there is research (e.g., Wang and Palincsar,
1989) that shows that when learners take responsibility for their own learning and make
independent choices to facilitate their goals, their motivation increases and they are better
able to achieve their goals.
Deci and Ryan's (1985) self-determination theory is based on a distinction between
intrinsic and extrinsic motivation and we can see here a relationship with autonomy as
well. They state that people who are intrinsically motivated engage in activities because
they are interested in the activity for its own sake, and find it enjoyable or satisfying to
engage in it. People who are extrinsically motivated expect rewards which are unrelated
to the activity itself or engage in a task to avoid punishment. The authors believe that
these types of motivations are not categorically different, but lie on a continuum of self-
determination. They believe that learners who are intrinsically motivated will become
more effective learners and that this learning will become even more productive when the
learner is self-determined and has some control over the learning process. Their
attribution theory of motivation is concerned with learners' perceptions of the reasons for
success and failure in learning. Learners who believe they are responsible for their own
success or failure will take greater responsibility for their learning. Hence, these self-
determination and attribution theories provide a very strong association between
motivation and autonomy.
Assuming that there is a distinct connection between motivation and autonomy, the
question that language teachers may be confronted with is how their students can be
motivated and become autonomous learners. Should teachers focus on motivating
students so that they can ultimately become autonomous learners? Or should they train
students to become autonomous learners so that they will be motivated to be successful
language learners? In other words, does motivation precede autonomy, or does autonomy
Spratt, Humphreys, and Chan (2002) research the question of whether autonomy or
motivation comes first and find that some research supports the view that motivation
precedes autonomy while other research points to autonomy being a precondition for
motivation to occur. In their own study, the authors found that "motivation is a key
factor that influences the extent to which learners are ready to learn autonomously" (p.
245). Other researchers who hold similar views are Spolsky (1989), who claims that
learners will be willing to invest the required time in learning a language only if they are
motivated, and Skehan (1989), who feels that there is more evidence supporting the fact
that motivation causes successful learning, rather than the other way round. These
studies suggest that motivation generates autonomy.
Further, Spratt et al. (2002) posit that motivation may play an inhibiting or enabling
role in relation to autonomy and could operate in different directions. Borrowing the
terminology from the literature on anxiety, can we then state that autonomy is either
facilitative or debilitative depending on the other factors within the language learning
situation? The present study seeks to find the correlations between motivation and
autonomy. It is possible that further research with different populations in a variety of
learning situations may provide some answers to this question. One of the aims of this
study is to determine if there is a correlation between motivation and autonomy and to
examine if they are distinct factors.
Research also provides evidence to support the view that autonomy needs to be
present to trigger motivation in language learners. Deci and Ryan's (1985) theories show
that learners are intrinsically motivated if they are self-determined and have control over
their language learning process. Dickinson (1995) asserts that for learners to be
successful and to be motivated, they need to take responsibility for their own learning.
Dornyei and Csizer (1998) advise teachers to motivate their students by promoting
learner autonomy. These studies imply that autonomy precedes motivation in a language
Motivation is not a stable factor and can undergo changes moment by moment in
terms of type and intensity, depending on the learner's experiences. It is possible for a
learner to be motivated in some areas of study and not in others, or at some times and not
at others. Motivation has been regarded and researched as an affective variable for
decades. If there is such a strong connection between motivation and autonomy, is
autonomy an affective factor too? Or is it a stable, cognitive factor? Different
researchers present different viewpoints about autonomy. Nunan (1996) claims that there
are degrees of autonomy and that it is not an absolute concept. Little (1990) believes that
autonomy is not a steady state achieved by learners, but a process. Thanasoulas (2000)
posits that one does not become autonomous, but rather one only works towards
autonomy. The underlying belief seems to be that autonomy is a construct that is
continually changing. It is possible for language learners to be taught how to be
autonomous learners, but the degree to which they will succeed differs. Many factors,
such as the personality traits of the learner, her/his goal for language learning, and the
cultural context in which learning occurs, will determine the extent to which a learner
will become autonomous. Both motivation and autonomy are dynamic, multifaceted
variables and it is very likely that we see a different facet each time there is a change in
the balance of factors, such as individual variables and learning situations. It will be
worthwhile to investigate the interactions between different combinations of factors in a
variety of learning situations with different populations, to examine if any patterns can be
observed that will provide insights into these affective factors.
The Anxiety-Motivation-Autonomy Interface
While trying to understand the interplay between anxiety, motivation, and
autonomy, one wonders what role anxiety plays with regards to both motivation and
autonomy. The relationship between anxiety and motivation has been well documented.
Gardner and MacIntyre (1993) suggest that anxiety and motivation have a reciprocal
relationship and that each variable influences the other. They investigated whether
anxiety caused poor performance or if poor performance caused anxiety and found that
anxiety leads to poor performance (MacIntyre & Gardner, 1994b). In describing the role
played by affective factors in second language acquisition, Schumann (1994) states,
"When motivation is lacking, anxiety is high ... the filter is up and input will not become
intake ... When motivation and self-confidence are high and anxiety is low, the filter
will be down and the relevant input will be acquired" (pp. 232, 233). An explanation for
this negative correlation is offered by MacIntyre and Gardner (1989), who say that
performance levels in the second language drop because anxiety uses up the cognitive
resources and attention that could have been used to perform better in the second
Research on learning strategies provides some evidence that successful language
learners do take measures to control their language anxiety (e.g., Naiman, Frohlich, Stern,
& Todesco, 1978). We are reminded by Oxford (1990) that "good language learners are
often those who know how to control their emotions and attitudes about learning"
(p. 140), which suggests that successful language learners control such emotions as
anxiety. In light of the fact that autonomous learners take charge of their learning and
control their emotions by using anxiety lowering strategies, encouraging themselves and
staying focused on the task at hand, there also seems to be a connection between anxiety
Researchers have also found a relation between motivation and autonomy. They
are in mutual agreement that both motivation and autonomy are distinct factors but have a
definite interface between them. Some researchers feel that students need to be
motivated before they can be autonomous, while others feel that autonomy has to precede
motivation in a student for more successful language learning.
There is a need for researchers to investigate all three of these individual
differences using qualitative data (e.g., interviews and journals) to complement the
quantitative data collected, in order to provide a more complete picture and deeper
insights into the roles played by these factors, individually and collectively, in language
learning. In this study, both quantitative and qualitative data have been collected and the
three affective factors of anxiety, motivation, and autonomy have been examined.
The present study seeks to observe the extent to which the participants are anxious,
motivated, and/or autonomy, the correlation, if any, among these factors, the extent to
which these factors are stable over an eight-week period, and to ascertain whether there is
any correlation between these factors and students' class performance. The results
obtained through quantitative analysis are then corroborated with the qualitative data
collected through journals, interviews and class observations. This study further
investigates the correlation between motivation and autonomy to examine whether
motivation and autonomy are distinct factors.
The participants in this study were enrolled in two second semester French classes
(FRE 1131) at the University of Florida. Out of a total of 32 participants, 23 (72%) were
native speakers of English while 9 (28%) were bilinguals and had a native-like
proficiency of English. Six of these nine bilinguals were fluent in Spanish, one student
was fluent in Chinese, another student spoke an Indian language, and the last one spoke
Haitian Creole. Participation was voluntary and data were collected throughout the
semester from these 32 students from two intact classes: 15 students from one class and
17 students from the other. These classes were selected because they had a higher
enrollment than some of the other classes.
Students at the University of Florida are placed in specific French classes
depending on their prior experience with foreign languages, if any. If the students have
not taken foreign language classes in high school, they are placed in FRE 1130, which is
the first semester French class. After their first semester, students are enrolled in FRE
1131, the second semester French class. If the students have taken any previous French
classes, however, they take the SAT II standardized test to determine their placement in
either FRE 1130 or FRE 1115. Both are first semester French classes and they follow the
same syllabus, but FRE 1130 meets five times a week and FRE 1115 meets only three
times a week.
The students in this study had already completed their first semester of French and
were in the second semester of their language study. Each of these classes had an
enrollment of about 20 to 25, so the students from two classes were recruited in order to
obtain an adequate sample size. At the beginning of the semester a total of 44 students
were enrolled in these two classes, but during the course of the semester, eight students
dropped the class, and two students were dropped from the study because they did not
complete all the required components of the study. Finally, another two students were
eliminated from the study because it was determined that they were trait anxious rather
than state anxious, which could potentially impact the anxiety results and skew the
All the participants were undergraduate students ranging in age from 18 to 22
years. From one intact class 13 females and two males participated, while 14 females
and three males participated from the other class. The participants completed a
background questionnaire (see Appendix A) in which they were asked for information
such as their age, their native language, their year in school, and the length of time, if
any, they had spent in French-speaking countries. Research done by Saito and Samimy
(1996) and Onwuegbuzie et al. (1999) clearly supports the view that there is a correlation
between the "year of study" and language anxiety experienced by language learners.
Onwuegbuzie, Bailey, and Christine (2000) found that students with high levels of
foreign language anxiety tended to be older, "had never visited a foreign country, had not
taken any high school foreign language courses, (and) had low expectations of their
overall average for their current language course" (pp. 88, 89). Aida (1994), observed
that first-year students who had been to a country where the target language was spoken
showed significantly lower anxiety in the language class than students who had never
visited or lived in countries where the target language was spoken. The adverse effect of
anxiety on performance in language class has been well documented. In their study,
Onwuegbuzie et al. (1999) discovered that students' expectations of their overall
achievement in foreign language courses was the biggest predictor of foreign language
anxiety. This finding is in harmony with MacIntyre and Gardner's (1991) claim that the
negative expectations students have about their language courses result in "worry and
emotionality." (p. 110). Since research shows that these factors influence or predict
success in language learning, the participants in this study were asked to share that
information with the researcher2.
All of the 32 participants in this study had had some experience of learning French
in high school. Some of them had also taken Spanish classes. However, only 20 had
visited foreign countries, which included England, Germany, France, Spain, India, China,
Portugal, Hong Kong, Chile and Venezuela. Of these 20 participants, only 10 had visited
French-speaking countries for varying lengths of time ranging from a day to about three
weeks on vacation. None of the students had spent any time studying in these countries.
A summary of the relevant demographic information of the participants in this study is
provided in Table 1.
Table 3.1: Relevant demography of students (total N=32)
Number of students Percentage
Female students 27 84%
Male students 5 16%
between 18-22 32 100 %
Year in school
Freshman 7 22 %
Sophomore 10 31 %
Junior 7 22 %
Senior 8 25 %
Foreign countries visited
Yes 19 59 %
No 13 41%
French-speaking countries 10 31 %
Previous foreign language experience
Yes 32 100 %
No 0 0 %
Expected letter grade in class
A 10 31 %
B 19 59 %
C 3 10%
Among the participants, seven students were freshmen, 10 were sophomores,
seven were juniors and eight were seniors. As far as grade expectations were concerned,
10 participants expected to get an 'A' in their French class, 19 students expected a 'B,'
and 3 students expected a 'C.' These students pursued a diversified range of majors:
English, Public Relations, Political Science, Education, Finance, Psychology, Journalism,
Telecommunications, Music, Microbiology, Agriculture, History, Anthropology,
Mathematics, Quantitative Science, Business, Plant Pathology, Pre-Med, Advertising and
Permission was obtained from the University of Florida Institutional Review
Board to conduct this study. Permission was also obtained from the instructors of the two
second-semester French course to visit their classes to collect the data. The researcher
explained the entire project and the time frame for data collection to the instructors, and
asked for time in class to administer the French test and three questionnaires on anxiety,
motivation, and autonomy. Then, the researcher met the classes two days before the
administration of the first instrument to introduce the semester-long project to the
students and enlist their help. They were not told that the affective factors of anxiety,
motivation, and autonomy were being researched. The participants were assured that
their identity would not be revealed and that their decision to participate (or not) would
not affect their class grade, and were then asked to sign a consent form. This precaution
was taken not only because the law requires it but also with the intention of making the
students feel comfortable with the entire procedure, to give them a chance to ask any
questions, and to reduce any anxiety that could be caused by having to deal with the
During Week 5 of the semester, the four instruments: an in-house French test and
questionnaires on anxiety, motivation, and autonomy, were administered to the students.
Data were collected during Week 5, and not earlier in the semester, because it was hoped
that by this time the students would have had a chance to overcome any initial discomfort
and anxiety they might have experienced due to being in a new class with new classmates
and instructor. It was hoped that any anxiety they felt during Week 5 would be directly
related to their language learning experiences in general, as opposed to specific new-
During Week 13 of the semester the same four instruments were administered to
the students a second time. Final and oral examinations were to be held during Week 18
of the semester, so this administration was scheduled to occur before any final exam
anxiety might appear. The in-house French test and the three questionnaires were
completed at the beginning of the class period so that the students would not be tempted
to rush through the questionnaires in order to leave class early. Additional data were also
collected out of class. The students were given specific journal topics and were invited to
submit journal entries. They were also invited to participate in one-on-one interviews on
a volunteer basis. These additional data will be discussed in more detail below.
In order to decrease the tendency of over-relying on the results obtained from the
quantitative data that were collected, and to increase the reliability and validity of the
results, qualitative data that were obtained from the journal entries and the interviews
with students and instructors was examined for corroboration with the questionnaire
results. Inconsistencies in the different data sets were identified to gain a better
understanding of the interplay between anxiety, motivation, and autonomy among the
participants in this study.
The participants were first grouped into four categories on the basis of the scores
they received on each of the three questionnaires. These four categories were labeled
'low,' 'fairly low,' 'fairly high' and 'high,' which meant that the participants experienced
low, fairly low, fairly high or high levels of anxiety, motivation, or autonomy. This
method of grouping students proved to be problematic because the number of participants
who fell into the "high" category was too low to investigate or describe. One way to get
a larger group would have been to combine the number of students who were included in
the "high" as well as the "fairly high" categories. But, if that were done, the range of
emotions experienced by this group would have been very broad, ranging from the center
to the highest point on the scale, in which case this group would not have represented
students who had high levels of anxiety, motivation, or autonomy. The participants were
then grouped into three equal categories of 'high,' 'moderate,' and low' based on their
scores on each of the three questionnaires. For example, the total possible score on the
anxiety scale was 165. All the participants who obtained scores between 1 and 55 were
considered to have little or no language anxiety and were placed in the "low" category.
Students who scored between 56 and 110 were considered to experience moderate levels
of anxiety and placed in the "moderate" category, while those who scored between 111
and 165 were considered to be highly anxious.
The participants were divided into three categories on the other scales as well.
The total possible score on the Attitude/Motivation Test Battery (AMTB) was 375. The
participants who obtained scores between 1 and 125 were believed to have little or no
motivation; the students who scored between 126 and 250 were considered to be
moderately motivated while those who scored between 251 and 375 were considered to
be highly motivated.
The total possible score on the questionnaire measuring autonomy was 300.
The participants who obtained a score between 1 and 100 were believed to have low
levels of autonomy. The students who scored between 101 and 200 were considered to
be moderately autonomous while those who scored between 201 and 300 were
considered to be highly autonomous. Having these three groups for each measure made
it possible to identify students who were highly anxious, motivated and/or autonomous.
Table 3.2 gives the range of points for the three categories on each measure.
Table 3.2: The range of points for the three categories on each measure.
Low Moderate High
Anxiety 1 55 56 110 111 -165
Motivation 1 125 126 250 251 -375
Autonomy 1 100 101 -200 201 -300
Trait Anxiety 1 12 13 24 25 35
In-house French Test
An in-house French grammar and vocabulary test (see Appendix B) was
developed by the Assistant Director of the French program for this study to measure the
performance of the students in their French class. The test includes some material that
the students covered during the previous semester in French and other material that the
students were expected to cover during the semester they were currently enrolled in. This
meant that during the first administration of the test, the students were not expected to
know all the answers. The test was constructed in this way to control for any potential
ceiling effect among more advanced students.
The Foreign Language Classroom Anxiety Scale (FLCAS)
The questionnaire used to measure the levels of anxiety in foreign language
learners is the Foreign Language Classroom Anxiety Scale (FLCAS) (see Appendix C),
developed by Horwitz, Horwitz, and Cope in 1986. The FLCAS is a 33-item instrument
that determines the degree to which students feel anxious during language classes by
assessing their communication apprehension, test anxiety, and fear of negative evaluation
in the foreign language classroom. Each item is a statement followed by a five-point
Likert response scale, with which the participants indicate the degree to which they agree
or disagree with each of the items. Items on this scale are both positively and negatively
worded; for example, one item might be, "I start to panic when I have to speak without
preparation in French class," and another item may be, "I feel confident when I speak in
French class". When the negatively worded items are scored, the scale is reversed. The
total possible score ranges from 33 to 165, with the higher scores indicating higher levels
of foreign language anxiety. Validity and reliability studies have shown that the scale is
both valid and reliable (Horwitz et al., 1986, Horwitz, 1991) with an alpha coefficient of
.93 and an eight-week test-retest coefficient of .83. Also, Aida's (1994) study of 96
students in a second-year Japanese course used the same instrument and reported a
Cronbach's alpha coefficient of .94. Alpha coefficient ranges in value from 0 to 1 and
shows how well a set of items measure a single unidimensional construct. The higher the
score, the more reliable the generated scale is. A reliability coefficient of .70 is
considered as "acceptable" in the literature.
To determine if the anxiety experienced by the participants in this study was state
anxiety rather than trait anxiety, seven extra Likert response questions were added to the
33-item FLCAS, making it a 40-item questionnaire. These items were adapted from the
General Anxiety Scale Items by Spielberger (1972), and comprise numbers 34 to 40 on
the questionnaire. The validity and reliability scores of the FLCAS were not altered
because the scores obtained on the seven additional questions were computed separately.
As was noted above, the data from these additional items resulted in the elimination of
two students from the study.
The Attitude/Motivation Test Battery (AMTB)
The questionnaire used to measure levels of motivation was the modified
Attitude/Motivation Test Battery (AMTB), originally developed by Gardner (1985) and
revised by Gardner, Tremblay and Masgoret in 1997 (see Appendix D). This instrument
investigates such factors as attitude toward learning French, desire to learn French, and
motivational intensity in learning French. Like the FLCAS, each item here is also
followed by a five-point Likert response scale for participants to indicate the degree to
which they agree or disagree with the statements. Again, some of the items on this scale
are also both positively and negatively worded. For example, one item might be, "When I
have a problem understanding something we are learning in my French class, I always
ask the instructor for help," and another item may be, "I don't bother checking my
corrected assignments in my French courses." Validity and reliability studies have
shown that the scale is both valid and reliable, with coefficients of .91 and .89
respectively, and a six-week test-retest coefficient of .79.
The Questionnaire on Autonomy
The students' levels of autonomy in learning French were determined by using the
questionnaire formulated by Spratt, Humphreys and Chan (2002) (Appendix E.) This
questionnaire was strongly influenced by Holec's (1981) definition of autonomy, and the
researchers attempted to incorporate the notions of "ability" and "responsibility" in the
five areas of their questionnaire, which aims to:
assess students' readiness for learner autonomy in language learning by
examining their views of their responsibilities and those of their teachers, their
confidence in their ability to operate autonomously .... It also investigated their
actual practice of autonomous learning in the form of both outside and inside
class activities. (p. 245)
Some of the questions required the students to tabulate two answers: to what extent
they thought it was the teacher's responsibility or their own responsibility "to make sure I
make progress outside class" or "to decide the objectives of my French course." The
terms "Not at all", "A little", "Some", "Mainly" or "Completely" were used on the five-
point Likert response scale in this questionnaire. This questionnaire on autonomy was
compiled, piloted, amended and then administered to 508 participants by Spratt,
Humphreys and Chan at the Hong Kong Polytechnic University in 2002. No information
on the reliability and validity of this questionnaire was included in the report.
In an effort to get a complete picture of the participants' apprehensions,
motivations, initiative, and the confidence they had throughout the semester, the
researcher observed both the French classes each week. The researcher chose to observe
the classes in person rather than to videotape the sessions in light of the findings by
Maclntyre and Gardner (1994a), which indicate that the increased anxiety caused by a
video camera in the classroom can impair students' performance.
All the participants were observed in their French classes at least once a week
throughout the semester. The class observations provided insights into the participants'
levels of anxiety, motivation, and autonomy through their behavior and attitudes in class.
Anxiety is exhibited in different ways, for example, through physiological and behavioral
signs discussed by researchers (Chastain, 1975; Horwitz et al., 1986). Students who are
enthusiastic or confident in class, or those who take the initiative to enhance their
learning in class, reflect their levels of motivation and autonomy. When necessary, at the
end of class, the instructor was consulted about certain situations or occurrences in class.
At other times, and throughout the semester, the students were asked for their views and
opinions about their French class. These conversations took place outside of class and
normal class activities were never interrupted.
These class observations were excellent opportunities for the participants and the
researcher to establish a rapport, which was of utmost importance during the end-of-term
interviews with the students. The researcher was also able to observe the classroom
behavior of the participants which helped, to some extent, in understanding the thoughts
expressed by the students in the questionnaires, journals and interviews. Also, having
interacted with the researcher during the administration of the questionnaires, engaging in
small talk before class began, and observing that the instructor had no access to their
responses, either written or verbal, the participants seemed to be comfortable,
forthcoming, and honest while volunteering information during the journal writing
exercise and interviews.
The participants were asked to submit four journal entries of at least one typed
page in length in which they answered specific questions, in English, regarding their
experiences learning French. These journal entries were written outside of class. About
ten students (29%) volunteered to write all four journal entries and were each given an
honorarium of $10 for their efforts. Specific dates were set to submit these journal
entries directly to the researcher and were spaced at least ten days apart. All the students
who volunteered to write journal entries completed all four entries. Each of the topics
(see Appendix F) given to the students targeted information about one of the variables or
constructs being investigated in this study. The first journal topic aimed at getting the
students to reflect on their foreign language learning experience in its entirety. The
second journal topic focused on anxiety, the third on motivation, and the fourth topic
targeted information on autonomy.
All the students who submitted journal entries were invited to participate in an
interview with the researcher. From the ten students who submitted journal entries, seven
students agreed to be interviewed, representing 22% of the total participant population.
Each participant was given an additional honorarium of $10 for participating in the
The students were interviewed, in English, out of class. The primary purpose of the
interview was to explore the participants' points of view, feelings, and perspectives about
learning French in general and their classroom experiences in particular. Some open-
ended questions were prepared to ask during the interview (see Appendix G), if needed,
but mostly the students were allowed to talk freely about their experiences and their
opinions about how anxious, motivated, or autonomous they were throughout the
semester. Some of the questions that were asked in the interview were, "What thoughts
go through your mind as you enter your French classroom?" and "If you needed to, how
would you go about designing a syllabus for your French class?" These semi-structured
interviews were conducted to complement the information obtained from the
questionnaires and to gain a better understanding of what the participants' numerical
The instructors of the two classes were also interviewed after the end of the
semester and submission of grades. Some open-ended questions to ask during the
interview were planned out, but mostly the instructors were allowed to freely talk about
their perspectives on their teaching and their students. Some of the questions (see
Appendix H) that were asked in the interview were, "What are your expectations of the
students, in terms of how they can learn best?" and "What are some ways in which you
will recognize that a student is anxious in your class?" These semi-structured interviews
helped to understand the teachers' perspectives of the importance of particular class
activities as opposed to the students' perspective of what should be done in class to
enhance language learning. The teachers' opinions about the levels of anxiety,
motivation, and autonomy in their students also served to substantiate the results obtained
through other data collection procedures.
Some of the research questions are answered quantitatively by using statistical tests
to evaluate the significance of the data collected, while others are assessed through
observation of the data collected. In this first section, questions that are answered
quantitatively are discussed.
Research Question 1 strives to identify the participants who are anxious, motivated,
and autonomous. This was done by categorizing the students into three groups of "high,"
"moderate," and "low," based on their scores on each measure. The participants whose
scores fell into the "high" category were considered to be anxious, motivated, and/or
Research Question 2 aimed at finding out the correlations between the participants'
levels of anxiety, motivation, and autonomy. A Pearson's Product Moment Correlation
Coefficient test was run on the scores the participants obtained on the three
questionnaires measuring anxiety, motivation, and autonomy to determine if there were
any correlations among them both at the time of the first administration of the
questionnaires and at the second administration. The questionnaires measuring the three
affective variables were administered twice to the participants, during Weeks 5 and 13 of
Research Question 3 investigated whether the affective factors within the
participants had undergone any changes from Time 1 to Time 2 or if they remained
stable. This was done by running a Paired-Samples T-Test on the difference in scores
obtained on each of those variables (e.g., anxiety score during second administration
minus anxiety score during first administration). The mean scores obtained by the
participants on different variables during each administration were also examined to
observe if there were any significant changes.
In order to answer Research Question 4, which sought to find the correlation
between the three affective factors and students' performance, a Linear Regression test
was run with the scores on anxiety, motivation, and autonomy as the independent
variables and class performance as measured by both the in-house French test and
students' final class grades, as the dependent variable. The in-house French grammar and
vocabulary test was administered two times during the semester, but the final class grades
represented the grades the students obtained over the duration of the semester in a wide
variety of areas: written homework, workbook activities, compositions, chapter exams,
oral proficiency tests, attendance, preparation, and participation.
As mentioned earlier, the participants in this study were from two different French
classes. Until this point the participants have been analyzed as a whole unit, but with
Research Question 5 the focus turns to the two classes as separate data sets. Here, an
attempt is made to identify any differences in the results obtained from the participants in
the classes taught by different instructors. An Independent Samples T Test was run to
compare the overall performance of the 2 sets of participants.
In order to answer Research Question 5, both the quantitative data from the scores
on the questionnaires and qualitative data collected through journals and interviews were
examined to determine if motivation and autonomy were distinct factors. If the results
of Research Question 3 showed that the participants' emotions had changed from Time 1
to Time 2, Research Question 7 sought to examine some possible reasons for the change.
The qualitative data in this study were collected through different modes: class
observations, journal entries, and interviews with students and instructors.
Research Question 6 aimed at determining whether motivation and autonomy are
distinct factors. Students who were highly motivated and highly autonomous were
identified and an attempt was made to see if any patterns emerged.
Research Question 7 discusses some possible reasons for a change in the affective
factors experienced by the participants from test to retest. This information was obtained
from the journal entries submitted by the students and the interviews. The qualitative
measures reveal much information about the three affective variables under investigation,
for example, the areas of language learning which cause anxiety, activities that motivate
students to work harder in and out of class, and students' thoughts about the instructors'
responsibility to teach them effectively as opposed to their own responsibility.
The next chapter presents the results of the statistical tests that were run and
discusses both the quantitative and qualitative findings.
1. According to Hatch and Lazaraton (1991), the minimum number of students
required to get a normal distribution is 30, so even with these eliminations, this
study meets that basic requirement, with a final number of 32.
2. The reason for obtaining these details from the students was to have more complete
information about the language learner even though it goes far beyond the scope of
RESULTS AND DISCUSSION
Quantitative data were collected during weeks 5 and 13 of the semester and the
participant scores from the three questionnaires on anxiety, motivation, and autonomy,
along with the scores from the in-house French pre- and post- tests were recorded.
Research questions 1 through 5, which are quantitative in nature, are answered in this
chapter by running a number of statistical tests to ascertain if the results were statistically
significant and if they were indicative of a trend within the sample.
Qualitative data were collected throughout the semester in the form of journal
entries, individual interviews with both students and instructors, and classroom
observations. Research questions 6 and 7, which are qualitative in nature, are also
discussed in this chapter by providing information obtained from the participants through
their journals and interviews.
In all, 34 participants completed all the tests and questionnaires, which were part of
the data collection procedures in this study. The scores obtained by the participants on
each questionnaire indicated the degree to which the participants were anxious, motivated
and/or autonomous. Recall that the procedure used to identify students who experienced
higher than average levels of anxiety, motivation, and autonomy, has been described in
Type of Anxiety
In order to increase the chances that research on anxiety and language achievement
obtained comparable results, Scovel (1978) suggested that language researchers should
be specific about the type of anxiety they were measuring. Accordingly, it is noted that
the present study focused on students who experienced foreign language anxiety or
anxiety that is felt only in specific situations, also called 'state anxiety,' and not trait
anxiety, which is a non-situation-specific personality trait. So, to determine the degree of
foreign language anxiety, it was important to ascertain that the students did indeed
experience state anxiety rather than trait anxiety. This was done by adding seven extra
questions measuring trait anxiety to the FLCAS. These seven questions were adapted
from Spielberger's (1983) State-Trait Anxiety Inventory which, according to Horwitz
(1986), has a low but significant correlation with the FLCAS at a correlation coefficient
of .29 (p=.002). Each item is a statement followed by a five-point Likert response scale.
The added questions are:
1. I get just as nervous in my other classes as I do in my French class.
2. I am usually calm and not easily upset.
3. I am a high-strung person.
4. I avoid meeting new people even though they are English speakers.
5. I am satisfied with the progress I am making in all my classes, except French.
6. I go through periods in which I lose sleep over worry.
7. It doesn't bother me when I am not in control of a situation.
The total possible score on these seven questions is 35. Students who obtained a
score between 1 and 12 were considered to have low levels of trait anxiety, while
students who scored between 13 and 24 were considered to be moderately trait anxious.
To qualify as being highly trait anxious, a participant had to score between 25 and 35
points on the trait anxiety scale. These three categories of 'low,' 'moderate,' and 'high,'
were devised the same way as for the other measures used in this study.
The scores obtained by the participants on the trait anxiety part of the questionnaire
were closely examined to identify students who were highly trait anxious. Out of a total
of 34 participants, two were identified as being highly trait anxious. The first highly trait
anxious participant, A15, scored 30 points during both administrations of the trait anxiety
scale, whereas the second highly trait anxious participant, B04, had a score of 27 during
the first administration of the test and a score of 25 during the second administration.
An interesting observation to be made is that the scores A15 received on the state
anxiety part of the questionnaire identified her as also being highly state anxious. Her
scores for the two administrations were 131 and 137, respectively. However, such was
not the case with B04, who scored 89 points during both administrations of the state
anxiety scale. These scores indicate that A15 experienced both trait and state anxiety,
while B04 experienced only trait anxiety. Understanding that trait and state anxieties
could have a different influence on a language learner and that the interplay between trait
anxiety, motivation, and autonomy could possibly be different from the interplay between
state anxiety, motivation, and autonomy, a decision was made to eliminate both these
participants from the study. As a result, a total of 32 participants were left in this study.
The above-mentioned scores indicate that it is possible for a language learner to be trait
anxious yet not state anxious, so it is advisable for researchers to take note of Scovel's
suggestion and ensure that they measure the appropriate variables in their studies.
The results obtained for each of the research questions are discussed in turn below,
and the type of anxiety measured is state anxiety.
Research Question 1
To what extent, if any, are the participants anxious, motivated, and
In order to determine to what degree the participants were anxious, motivated and
autonomous, the following sub-questions of Research Question 1 were considered:
a. To what extent, if any, are the participants anxious?
b. To what extent, if any, are the participants motivated?
c. To what extent, if any, are the participants autonomous?
The scores received by the participants at Time 1 and Time 2 on each of the three
questionnaires were examined. Table 4.1 provides information on the number of students
who were categorized as having high, moderate, or low levels of anxiety, motivation,
and/or autonomy, and Appendix I gives an overview of all the participants who
experienced high levels of anxiety, motivation, and antonomy.
Table 4.1: Student classification based on scores received on questionnaires (n=32)
High Moderate Low
Number of Number of Number of
students students students
First administration 8 (25%) 23 (72%) 1 (3%)
Anxiety Second administration 7 (22%) 23 (72%) 2 (6%)
First administration 19 (59%) 13 (41%) 0
Motivation Second administration 17 (53%) 15 (47%) 0
First administration 5 (16%) 27 (84%) 0
Autonomy Second administration 4 (13%) 28 (87%) 0
The results are discussed below.
Students were identified as being anxious if their scores fell into the 'high'
category, which meant their scores ranged from 111 to 165 on the anxiety scale.
Emphasis was placed on this category because research has consistently shown that high
levels of anxiety have a negative effect on the performance of language learners, whereas
moderate levels of anxiety can either be facilitative or debilitative, either motivating
students to work harder to reach their goals, or impeding their progress in their language
study program. MacIntyre (1995) calls this an inverted "U" relationship between anxiety
and performance. The topmost segment of the inverted "U" divides the range between
low and moderate anxiety, which may be facilitating, from the range between moderate
and high anxiety, which may be debilitating to a language learner.
As shown in Table 4.1, during Week 5 of the semester, eight students (25%), out of
a total of 32 participants, experienced high levels of anxiety with scores ranging from 111
to 134, and during Week 13 of the semester, seven participants (22%) experienced high
levels of anxiety, with scores ranging from 112 to 126. Table 4.2 gives the details of the
students who were anxious at some point in the semester.
Table 4.2: Students who were anxious during Time 1 and Time 2
Student Time 1 Time 2 Increase Decrease
Points Points in Points in Points
All 116 120 4 (2.4%) Anxious Time 1 and 2
A20 116 117 1 (0.6%) Anxious Time 1 and 2
B18 134 126 8 (4.8%) Anxious Time 1 and 2
B22 127 124 3 (1.8%) Anxious Time 1 and 2
A08 119 99 20 (12%) Anxious Time 1
B12 122 102 20 (12%) Anxious Time 1
B14 112 107 5 (3%) Anxious Time 1
B17 111 110 1 (0.6%) Anxious Time 1
A03 110 123 13 (7.9%) Anxious Time 2
A18 97 113 16 (9.7%) Anxious Time 2
B15 107 112 5 (3%) Anxious Time 2
Only four participants (Al A20, B18 and B22) out of eight, were highly anxious
during both administrations of the questionnaire on anxiety with a maximum of 4.8%
change in their scores. Other four students (A08, B12, B14, and B17), who were anxious
at the beginning of the semester were not anxious at the end of the semester. Their
decrease in scores ranged from 0.6-12%. Yet another three students (A03, A18, and B15)
who were not anxious at the beginning of the semester were anxious at the end of the
semester, and their anxiety scores increased by a maximum of 9.7%. These results show
that a total of 11 students in this study were highly anxious at some point in their French
class during the course of the semester, which is about 34% of the participants.
In order to find out if the participants in this study were motivated, their scores on
the AMTB were examined. Students who scored between 251 and 375 were considered
highly motivated. The results showed that 19 students (59%) were highly motivated
during the first administration of the questionnaire, with scores ranging from 257 to 305,
and 17 (53%) of those participants remained highly motivated during the second
administration, with scores ranging from 259 to 299. It is encouraging to observe that
more than half the participants started off their studies motivated and continued this trend
until the end of the semester. Thirteen (41%) participants were moderately motivated
during the first administration, but during the second administration, 15 (47%)
participants were moderately motivated. In other words, approximately the same subset
of students were moderately motivated during both the first and second administrations of
the motivation questionnaire. However, two students, A17 and A20, who were highly
motivated during Week 5 were only moderately motivated during Week 13 of the
semester. None of the participants had "low" levels of motivation at any time during the
Students who scored between 201 and 300 on the autonomy scale were considered
to be highly autonomous. With regards to the number of participants who were
autonomous, the data show that only five (16%) of the participants, A03, Al B10, B13,
and B17, were highly autonomous during the first administration of the questionnaire,
with scores ranging from 201 to 249, and only four (13%) participants, A03, A08, B10,
and B17, were highly autonomous during the second administration, with scores ranging
from 203 to 222. Three of the participants were highly autonomous during both
administrations. Two of the students who were highly autonomous during Week 5 were
only moderately autonomous during Week 13 of the semester, whereas one student who
was moderately autonomous during Week 5 was highly autonomous during Week 13. In
comparison with the number of students who were highly motivated, fewer in this group
were highly autonomous. Most of the students were moderately autonomous, with
twenty-seven (84%) students falling into this category during the first administration, and
twenty-eight (87%) during the second administration. None of the students fell into the
category of "low" autonomy.
Overall, the results show that about 34% of the participants in this study were
highly anxious at some point during the semester, 59% of them were highly motivated,
but only 19% of them were highly autonomous.
Research Question 2
To what extent, if any, is there a correlation among the participants' stated
levels of anxiety, motivation, and autonomy?
In order to determine if there was a correlation among the three affective factors
under consideration in this study, a Pearson's Product Moment Correlation Coefficient
test was run. Correlation coefficients tell us how well two or more variables are related
to each other. They can also be used to determine how much variability in one factor is
explainable by variation in the other factors.
The correlations were examined for both administrations of the three
questionnaires. The results showed that there were no correlations between anxiety and
motivation, or between anxiety and autonomy, but there was a statistically significant
moderate correlation between motivation and autonomy with a coefficient of .623
(p=<.001) at Time 1 and .624 (p=<.001) at Time 2. Table 4.3 summarizes these results.
Table 4.3: The correlations among participants' levels of the three variables
First administration Second administration
anxiety and motivation .052 -.148
Sig .776 .417
anxiety and autonomy .111 .030
Sig .545 .871
motivation and autonomy .623 .624
Sig .000* .000*
Correlation between Anxiety and Motivation
In considering the participant scores in this study, the results show that there is no
correlation between anxiety and motivation. These results are contrary to those of
Gardner, Day & MacIntyre (1992), who examined the effects of both motivation and
anxiety on computerized vocabulary acquisition and proposed that anxiety and
motivation "are two separate dimensions with overlapping behavioral consequences ...
correlated yet distinguishable" (p. 212).
Gardner's (1985) Attitude/Motivation Test Battery included some questions on
French class anxiety and French use anxiety. These questions were identified and the
scores of each participant on the extracted anxiety questions were correlated with the
scores on the FLCAS to ensure that the anxiety questions, from both scales, were
consistent and measured the same affective factor. The results show that the correlation
between the anxiety scale and the anxiety questions from the motivation scale was high,
at .801 and .849 during Test 1 and Test 2, respectively and p=<.001 at both times,
suggesting that the type of anxiety measured was the same.
In an effort to observe the true motivation scores of the participants, the AMTB
questions on anxiety and autonomy were disregarded. It was interesting to note that
when a correlation test was run between the anxiety and new motivation scores, a
statistically significant moderate inverse correlation was found with a correlation
coefficient of -.482 (p=.005) during the second administration of the instruments. These
results are now similar to those obtained by Gardner, Day & MacIntyre (1992) who
believe that there is a correlation between anxiety and motivation. This change in results
suggests that the AMTB questions on anxiety and autonomy have an impact on the
As mentioned before, eight students were highly anxious during Test 1; of these
eight, two were freshmen, two were sophomores, two were juniors, and two were seniors.
During Test 2, out of the seven students who were highly anxious, two were sophomores,
three were juniors, and two were seniors. Interestingly, none of the freshmen were
anxious at the end of the semester, which is in keeping with the results reported by
Onwuegbuzie et al. (1999), who posited that foreign language anxiety increased linearly,
meaning that the freshmen reported the lowest levels of anxiety while the seniors
reported the highest levels. As noted earlier, more than half of the students were
motivated during both Test 1 and Test 2.
Correlation between Anxiety and Autonomy
With regards to the relationship between anxiety and autonomy, again the data show that
there is little or no correlation between the two, with a correlation coefficient of. 111
(p=.545) during Test 1, and a correlation coefficient of .030 (p=.871) during Test 2.
Table 4.4 gives a summary of all the students who were highly anxious and/or highly
autonomous during both administrations of the questionnaires.
Table 4.4: Students who were highly anxious and/or highly autonomous (represented by
'X')during Test 1 and Test 2
Participant Highly Anxious Students Highly Autonomous Students
Week 5 Week 13 Week 5 Week 13
A03 X X X
A08 X X
All X X X
A20 X X
B10 X X
B17 X X X
B18 X X
B22 X X
There is no apparent pattern in the levels of anxiety and autonomy experienced by
the participants. Two students, B10 and B13, who were highly autonomous, did not
experience high anxiety levels in their French class, whereas two other students, A03 and
B17, were highly autonomous during both Test 1 and Test 2, and were also highly
anxious during one of the administrations of the anxiety scale. B 18 and B22, were not
highly autonomous but were highly anxious. Thus, in this sample, students were highly
autonomous but not anxious; highly anxious but not autonomous, and also both highly
autonomous and highly anxious at the same time. These results suggest that experiencing
high levels of anxiety in a language classroom does not indicate whether a student will or
will not take charge of his or her learning. Neither can it be said that a highly
autonomous student will not experience any language anxiety in the classroom.
A closer look at Table 4.4 shows that four students who were highly anxious during
Test 1 were not as anxious during Test 2, and three other students who were not highly
anxious at Test 1 were anxious at Test 2. Also, two students who were highly
autonomous at Test 1 changed categories and were not as autonomous at Test 2. Finally,
A08, who was not highly autonomous at Test 1 was so at Test 2. These results show that
the participants' levels of anxiety and autonomy changed within the eight-week period.
Participants A03, A18, and B15 were not highly anxious during week 5 but they
were anxious during Week 13. Class observations reveal that both A03 and A18 would
come early to class, sit at the back and not engage in any conversations with their
classmates. They would attempt to answer any questions the teacher asked them. Their
class grades for the first half of the semester were lower than they were toward the end of
the semester. It is possible that these students became anxious about their grades and put
in greater efforts, resulting in improved grades at the end of the semester. This
information is corroborated from the qualitative data collected from A18. She said that
she was not highly anxious at the beginning of the semester because she was familiar
with the course material and had heard that the instructor was good. During the course of
the semester, she said that she was frustrated with the tediousness of the workbook and
did not do her daily assignments. She was anxious in class whenever she had not done
the reading for that day and felt tense at the thought of the instructor calling on her to
answer a question. Though she had no intentions of continuing her language study, she
wanted to get a good grade in class, so she put in extra efforts to reach her goal.
B 15's behavior in class was very different from the previous two participants. She
would usually come late to class, sit next to one of her friends, and chat and play
throughout the French class. Her grade halfway through the semester bottomed out to an
overall average of 42%. She was not anxious at Time 1 but was anxious at Time 2,
suggesting that it was possible that she began to be anxious about her grade and put in
efforts to improve her grade; this is corroborated by her final class grade, which came up
These results show that there is no correlation between anxiety and autonomy, and
that the degree to which the participants are anxious or autonomous can change over the
course of a semester.
Correlation between Motivation and Autonomy
In examining if there is a relationship between motivation and autonomy, the data
show that there is a reasonably high correlation between these two affective factors, with
a coefficient of .623 during the first administration and .629 during the second
administration. Both these correlations are statistically significant at p=<.001.
Language learning research has established that there is a strong connection
between motivation and autonomy. Describing motivated behavior is a challenging task
because it inevitably includes some characteristics of autonomous behavior. Neither is it
easy to describe autonomous behavior and claim that it cannot also be labeled as
motivated behavior. Hence, it is not surprising to find that some questions in the AMTB
seemed to overlap with autonomy. The following questions focused on the efforts the
students put into their language learning, suggesting that they were taking charge of their
learning and showing some autonomous behavior.
1. I keep up to date with French by working on it almost every day.
2. If it were up to me, I would spend all my time learning French.
3. When I have a problem understanding something we are learning in my French
class, I always ask the instructor to help.
4. I really work hard to learn French.
5. When I am studying French, I ignore distractions and stick to the job at hand.
Again, these questions were identified and the scores of each participant on the
extracted questions were correlated with the instrument on autonomy. This time, it was
found that the correlation between the autonomy scale and the autonomy questions from
the motivation scale had a moderately high correlation at .633 and .689 during the first
and second administrations, respectively and p=<.001. This process ensured that the
autonomy questions, from both scales, were consistent and measured the same affective
These moderately high correlations between the autonomy questions on both the
scales, and also the anxiety questions on the AMTB and FLCAS, reinforce what
researchers have said about motivation and anxiety, and motivation and autonomy being
factors that are closely connected, yet are distinct and distinguishable as different
affective factors in a language learner. The relationship between motivation and
autonomy will be further examined while discussing Research Question 6.
To further examine the correlation between motivation and autonomy, a correlation
test between the scores on motivation and autonomy was run a second time; this time
with the new motivation scores. The results were similar to those obtained the first time
showing statistically significant moderate correlations between motivation and autonomy
with correlation coefficients of .603 (p=<.000) and .534 (p=.002) during the first and
second administrations, respectively.
An investigation into the correlations among the levels of anxiety, motivation, and
autonomy observed in the participants of this study show that there is a moderately high
correlation between motivation and autonomy. When a correlation test was run between
anxiety and motivation using the AMTB in its entirety, there was no correlation between
the two factors, but when the questions on anxiety and autonomy in the AMTB were
disregarded, a negative correlation was found between anxiety and motivation. These
results support those of Gardner, Day & MacIntyre (1992) who found a correlation
between anxiety and motivation.
Research Question 3
To what extent, if any, are the participants' levels of anxiety, motivation and
autonomy stable over time?
The sub-questions that were considered to discover if there were any changes in the
participants' stated levels of anxiety, motivation, and autonomy were:
a. To what extent, if any, does the participants' anxiety change from test to retest?
b. To what extent, if any, does the participants' motivation change from test to retest?
c. To what extent, if any, does the participants' autonomy change from test to retest?
d. If a change has occurred in one of the affective factors, is there a corresponding
change in the other two factors?
The questionnaires measuring anxiety, motivation and autonomy were administered
twice to the participants, during Weeks 5 and 13 of the semester. A series of paired
samples t-tests were run to investigate the stability of the levels of emotions experienced
by the participants. The results in Table 4.5 show that there was no change in the
participants' levels of anxiety and autonomy from test to retest, but that there was a
significant decrease (p=.038) in their motivation levels.
Table 4.5: Paired samples t-test results showing change only in motivation levels
Mean t (2-tailed) coefficient Sig.
First Anxiety 93.44 .770 .447 .789 .000*
Second Anxiety 91.63
First Motivation 253.75 2.165 .038* .905 .000*
Second Motivation 248.88
First Autonomy 177.84 -.429 .671 .718 .000*
Second Autonomy 179.19
There was a high correlation between the first and second administrations of the
instruments measuring the three variables, suggesting that the instruments were reliable.
With regards to the possible changes in the levels of anxiety felt by the participants,
the mean anxiety score was 93.44 during Week 5 and 91.63 during Week 13, a negligible
decrease of 1.81 points, which was statistically non-significant at p=.447. In other words,
there was no significant overall change in the participants' stated levels of anxiety from
test to retest.
When the participant scores on the motivation scale were examined, it was found
that during the first administration, the mean score for the participants' level of
motivation was 253.75 while during the second administration, the mean score for
motivation was 248.88, a decrease of 4.87 points. This change was statistically
significant with p=.038, so an overall decrease in participants' motivation levels is
observed from Week 5 to Week 13 of the semester.
These results were puzzling because the decrease in motivation levels seemed to
contradict earlier results which showed a high correlation between the first and second
administrations of the motivation scale, with a correlation coefficient of .905 (p=<.001).
Therefore, each class was then examined separately to better understand the changes in
the levels of motivation. An independent samples t-test was run to compare the scores
obtained by the participants in each of the two groups. On the first administration, the
mean motivation score of the first group was 242.27 points, while the second group
scored 263.88 points, a difference of 21.61 points, which was statistically significant at
p=.032. This meant that the second group was significantly more motivated than the first
group during Week 5 of the semester. Table 4.6 gives the mean motivation scores of the
participants during both administrations.
Table 4.6: Group differences in motivation scores
Group Mean Sig
First administration 1 242.27
2 263.88 .032*
Second administration 1 239.00
2 257.59 .074
During the second administration of the motivation scale, the mean motivation
scores of the second group remained higher than the scores of the first group, but this
time the difference of 18.59 points was not statistically significant (p=.074), which meant
that there was no difference in the motivation levels of the two groups during Week 13 of
the semester. Looking at the mean scores of each group separately, we observe that the
motivation levels of group 1 decreased by 3.27 points while the motivation levels of
group 2 decreased by 6.29 points, which is almost twice the decrease experienced by the
first group. A paired samples t-test showed that the decrease in motivation in group 1
was not statistically significant (p=.461), but the decrease in motivation levels in group 2
was statistically significant (p=.006). It is only when we look at the motivation levels of
both the classes separately that we understand that the overall decrease in motivation
levels as a group come mainly from the scores of one class alone. Thus far, the results
show that there was no statistically significant change in the participants' stated levels of
anxiety, but that there was a statistically significant decrease in their levels of motivation
from Time 1 to Time 2. The possible reasons for this decrease in motivation levels will
be discussed in the section concerning research question 7.
Looking at the participant scores on the autonomy scale, it was observed that the
highest individual score at Time 1 was 249 points, which dropped to 222 points at Time
2, but the lowest individual score was not very different during the two administrations.
This could have suggested a possible decrease in levels of autonomy, but interestingly,
the mean score rose from 177.84 during the first administration to 179.19 during the
second administration. However, this increase was statistically non-significant (p=.671),
so again we understand that as a whole there was no change in the participants' stated
levels of autonomy from test to retest.
The next question to explore is whether a change in one of the affective factors
experienced by the participants, corresponds to a change in the other two factors. It was
observed that the motivation levels within the participants, as a whole group, had
decreased from Time 1 to Time 2. To determine if there were any corresponding changes
in the other two factors, the difference between the two scores obtained during the first
and second administrations was computed for each variable, and a Pearson's Product
Moment Correlations Coefficient test was run to find the correlation among these three
differences. Table 4.7 shows the correlation of changes from Test 1 to Test 2.
Table 4.7: Correlation of changes from test to retest
Correlation between anxiety and motivation -.330 .065
Correlation between anxiety and autonomy .065 .726
Correlation between motivation and autonomy .335 .061
The results showed a low negative correlation between motivation and anxiety at
-.330 (p=.065), which though statistically insignificant was suggestive of a trend. There
was a low positive correlation between motivation and autonomy at .335 (p=.061), but
again, this change was merely suggestive and not statistically significant. In other words,
when there is a significant change in the motivation levels in the participants from Time 1
to Time 2, there is no statistically significant corresponding change in the levels of
anxiety and autonomy. These were the results when motivation levels were considered in
the whole sample. But when the motivation levels of each group were considered
separately to identify if a change in motivation showed a corresponding change in anxiety
and autonomy, a moderate negative correlation of -.613 (p=.015) was found between
anxiety and motivation in Group 1. This suggests that a decrease in motivation was
accompanied by an increase in anxiety within this group. Table 4.8 shows the
corresponding changes that occur when one of the factors changes.
Table 4.8: Changes in one factor showing corresponding changes in others
Anxiety Motivation -.613 .015*
Group 1 Anxiety Autonomy .142 .612
Motivation Autonomy .376 .167
Anxiety Motivation .011 .967
Group 2 Anxiety Autonomy -.171 .513
Motivation Autonomy .206 .427
Earlier results indicated that the anxiety experienced by the participants is more
likely debilitative rather than facilitative, hence it is not very surprising to observe the
trend towards an inverse relationship between anxiety and motivation.
Research Question 4
What is the correlation between the three affective factors and students'
performance, as measured by the French test and the final class grade?
In order to determine if there is a correlation between the French test scores and the
scores obtained by the participants on the questionnaires measuring anxiety, motivation
and autonomy, correlation tests were run between the scores of the participants on each
of the measures and the French test scores.
Multiple Linear Regression tests were run on different data sets using SPSS 12.0.
The first time, correlations between the first administrations of the in-house French test
scores and the scores on the questionnaires on anxiety, motivation, and autonomy were
measured. The second time, correlations between the second administrations of the
French test scores and scores on the anxiety, motivation, and autonomy questionnaires
were measured. In order to find if there is a correlation between the final class grades
and the scores obtained by the participants on the questionnaires measuring anxiety,
motivation, and autonomy, the same procedure was followed as with the French test
scores. Table 4.9 tabulates the results obtained.
Table 4.9: Summary of correlations between the French test scores/class grades and
Anxiety Motivation Autonomy
First administration of the three
questionnaires & proficiency test
Correlation coefficient -.625 .224 -.145
Sig .000* .232 .437
Second administration of the three
questionnaires & proficiency test
Correlation coefficient -.399 .107 -.007
Sig .031* .637 .975
First administration of the three
questionnaires & final class grade
Correlation coefficient -.402 .271 ,207
Sig .016* .186 .311
Second administration of the three
questionnaires & final class grade
Correlation coefficient -.415 .088 .312
Sig .015* .671 .136
The results show that there were statistically significant moderate negative
correlations between the French test scores and first anxiety scores at -.625 (p=<.001),
and a low negative correlation between the French test scores and second anxiety scores
at -.399 (p=.031), but that there were no correlations between the first French test scores
and the scores on motivation or autonomy, either during the first or second
Linear regression tests were also run between the final class grades and the scores
on both administrations of the questionnaires on anxiety, motivation and autonomy.
Results show that moderate negative correlations were found between the final class
grade and the first anxiety score at -.402 (p=.016), and between the final class grade and
the second anxiety score at -.415 (p=.015), but again, there were no correlations between
the final class grades and the scores on either the first or second administrations of the
questionnaires on motivation or autonomy.
The results of all four linear regression tests consistently show that there is a
statistically significant negative correlation between anxiety and student performance as
measured by the French test and final class grade, but no significant correlations between
motivation and autonomy, and student performance. These findings suggest that in this
sample the higher the levels of anxiety experienced by a participant, the lower his/her
performance in class. It does not seem, however, that the students' performance
improves with higher levels of either motivation or autonomy because the students' levels
of motivation and autonomy did not have a statistically significant effect on either their
French test scores or their final class grade.
Earlier studies conducted with university students had similar results. Horwitz's
(1986) study had a significant moderate negative correlation between foreign language
anxiety and the final class grades. MacIntyre and Gardner (1989) also found significant
negative correlations between French Class Anxiety and performance on a vocabulary
learning task. Aida (1994) found a significant negative correlation between FLCAS
scores and final grades among American second-year Japanese students, and Coulombe
(2000) found a small but significant correlation between FLCAS scores and final grades
in eleven French classes ranging from beginning to advanced. Thus it is observed that
regardless of the foreign language being studied or the level the students are in university,
the negative correlation between foreign language anxiety and class performance is
Thus far, the anxiety levels of all the participants was discussed as a whole, but
since moderate levels of anxiety can either be facilitative or debilitative, a decision was
made to examine a subset of the participants. An attempt was made to determine whether
the anxiety experienced by the moderately anxious students in this study was facilitative.
23 students were moderately anxious during each of the two administrations of the
anxiety questionnaire. 18 of these 23 participants were moderately anxious during both
Test 1 and Test 2.
The scores of the twenty-three (72%) moderately anxious participants were
correlated with their overall class performance as measured by the French test scores.
The results showed that during the first administration of the anxiety scale, there was an
inverse relationship between anxiety and proficiency scores which was statistically
significant at p=.043. This meant that during the first administration the anxiety felt by
the participants had a negative impact on their French test scores. However, the results
were no longer statistically significant during the second administration, even though a
negative correlation between moderate levels of anxiety and class performance was
observed. These results suggest that the moderately anxious students in this study also
experienced debilitative, rather than facilitative anxiety. Table 4.10 shows the correlation
between the anxiety scores of the moderately anxious participants and their French test
Table 4.10: Correlation between anxiety scores of moderately anxious participants and
their French test scores
Correlation Coefficient Sig
First administration .426 .043*
Second administration .361 .090
Since the inverse relationship between anxiety and the French test scores during the
second administration was not significant at p=.090, a correlation test was run between
the scores obtained by the moderately anxious participants during both administrations of
the anxiety scale and their final class grade, in place of the in-house French test scores, to
see if the results would differ in any way. The scores from the second administration of
the anxiety scale and the final class grades were obtained at the end of the semester,
whereas the in-house French test was given at the beginning and the end of semester.
This time, the inverse relationship between anxiety scores and class grades was
statistically significant at p=.027. These results are shown in Table 4.11.
Table 4.11: Correlation between anxiety scores of moderately anxious participants and
their final class grades
Correlation Coefficient Sig
First administration -.609 .007*
Second administration -.461 .027*
It is interesting to note that there is an inverse correlation even between the first
anxiety scores and the final class grades, which is statistically significant at p=.007.
These results may indicate that the final grades of the students were more reliable and
valid than the scores obtained from the in-house French test that was created for the
purpose of this study. Despite these limitations, it is observed that when the results are
statistically significant, the anxiety experienced by the moderately anxious students is
debilitative rather than facilitative.
Research Question 5
In what ways, if any, do the results obtained from the participants in the two
intact classes, taught by different instructors, differ?
An independent samples t-test was run to compare the overall performance of the
participants in the two classes. The scores obtained during the first administration of the
questionnaires on anxiety, motivation and autonomy were first computed. As noted
earlier, it was found that Group 1 had a mean score of 242.27 on the motivation scale,
while Group 2 had a mean score of 263.88 (a difference of 21.61 points), a significant
difference with p=.032, meaning that the second group was more motivated than the first
group during Week 5 of the semester. As far as the levels of anxiety and autonomy were
concerned, it was found that there were no significant differences between the two
groups. These results are summarized in Table 4.12.
Table 4.12: Group-wise mean scores on the three questionnaires during Week 5
Group 1 Group 2 Sig
Anxiety mean score 92.20 94.53 .751
Motivation mean score 242.27 263.88 .032*
Autonomy mean score 171.33 183.59 .178
A second independent samples t-test was run to compare the overall performance of
the two groups during the second administration of the instruments. Table 4.13 provides
the participant mean scores on the three questionnaires during Week 13.
Table 4.13: Group-wise mean scores on each questionnaire during Week 13
Group 1 Group 2 Sig
Anxiety mean score 94.33 89.24 .496
Motivation mean score 239.00 257.59 .074
Autonomy mean score 176.80 181.29 .504
This time, the higher levels of motivation in Group 2 were no longer statistically
significant during Week 13 of the semester, but may be suggestive of a trend. In other
words, it is possible that Group 2 experienced marginal levels of motivation and that a
small difference in levels would indicate that these high levels were statistically
Group 2 had higher levels of motivation during week 5 than during Week 13 of the
semester. Although they were not as motivated at the end of the semester as they were
toward the beginning, they were still marginally, but not significantly more motivated
than the participants in Group 1. These results show that the motivation levels within
each group did not change a great deal. Some possible reasons for the marginal decrease
in motivation levels within the sample will be discussed later in this chapter.
Research Question 6
Are motivation and autonomy distinct factors?
A correlation test showed that there is a moderate correlation between participant
scores on motivation and autonomy. Test 1 had a correlation coefficient of .623
(p=<.001), while Test 2 had a coefficient of .624 (p=<.001). In order to find out if
motivation and autonomy are distinct factors, it may be helpful to examine the scores of
participants who are both motivated and autonomous during Test 1 and Test 2 and
observe if any pattern emerges from the data. Table 4.14 summarizes this information.
Table 4.14: Students who are highly motivated and highly autonomous during Test 1
and/or Test 2
Participant TEST 1 TEST 2
Motivated Autonomous Motivated Autonomous
A03 X X X X
A05 X X
A08 X X X
All X X X
A12 X X
B01 X X
B02 X X
B06 X X
Table 4-14: Continued
Participant TEST 1 TEST 2
Motivated Autonomous Motivated Autonomous
B08 X X
B10 X X X X
B11 X X
B12 X X
B13 X X X
B17 X X X X
B18 X X
B19 X X
B21 X X
As mentioned in the discussion of research question 2, during the first
administration of the instruments in this study, 19 students (59%) were highly motivated
but only 5 of them (16%) were also highly autonomous, meaning 14 students (74%) were
motivated but not autonomous. During the second administration, 17 students (53%)
were highly motivated and only 4 (13%) were highly autonomous. In other words, 13
students (76%) were motivated but not autonomous. The data indicates that it is possible
for the students in this sample to be highly motivated, but not highly autonomous. None
of the students who were highly autonomous lacked high motivation, which implies that
to be highly autonomous a student needs to be highly motivated. Three students (50%)
were highly motivated during both administrations of the motivation scale, but were
highly autonomous during only the first or second administrations of the autonomy scale.
More students are motivated than autonomous in this sample, which implies that there are
greater chances of students being motivated than autonomous in a given classroom.
The three students who were both highly motivated and highly autonomous during
both tests are A03, B10, and B17. Two students, All and B13, were autonomous during
Test 1 but not Test 2. Interestingly, A08 was not autonomous at the beginning of the
semester, but was so toward the end of the semester. Results shows that two students,
A17 and A20, were highly motivated during Test 1 but were only moderately motivated
during Test 2, which implies that they lost some of their motivation. This is always
assuming that the participants answered the questionnaires honestly and reliably. Their
responses on the AMTB show that they had lost some confidence in using French both
inside and outside the classroom. They even began to think that most foreign languages
sounded crude and harsh, and that they would prefer to spend more time on courses other
than French. It seems that their attitude toward learning French was not as positive as it
was during Test 1.
A17 did not write any journal entries, nor was she interviewed, hence her responses
on the questionnaires cannot be corroborated. A20, however, expressed her feelings in
the journal entries she submitted. Mid-way through the semester, she said that the French
class was getting progressively harder as they were learning more every day, and earlier
concepts were being built upon. She felt that most of the material was "foreign" to her
and that she was easily confused with the many concepts that were presented in class.
Her main reason for taking the French class was to fulfill the language requirement.
Her overall feelings about learning French do not seem very positive at this point in the
semester because she says, "I honestly do not feel that motivated to learn French because
of all the tedious work that is involved in it. Five days of foreign language class per
week is a lot for a college student with many other classes to handle." She wished the
instructor would "slow down the pace of the class as much as possible in order not to
confuse or discourage students."
These results indicate that participants who are motivated and/or autonomous at
one point in the semester may or may not remain this way at another point in the
semester. They may either gain or lose that characteristic at any stage of their language
learning experience indicating that the state of being motivated or autonomous is not a
constant. These results concur with the views of researchers (e.g. Holec, 1981; Little,
1991) who posit that autonomy is not a steady state that learners achieve once and for all,
and that it can fluctuate at different stages of the language learning experience.
It should be noted that while trying to formulate questions that target motivation, it
is possible for some facets of autonomy to be included in those questions, because of the
high correlation between these two affective factors. The AMTB included some
questions on autonomy, for example, and the students had to indicate whether they kept
up to date with French by working on it almost every day, whether they wanted to learn
French so well that it would become second nature to them, whether they ignored the
feedback they received in French class, and whether they ignored distractions when
studying French. The responses of the students who were only motivated, and those who
were both motivated and autonomous were examined. The results did not show any
significant difference in the way that these two groups of students responded to those
questions on autonomy, which indicates, once again, that there is a strong interface
between motivation and autonomy.
In examining the questionnaire on autonomy, however, there was a difference in
the way that students who were motivated but not autonomous answered questions from
those who were both motivated and autonomous. The autonomous students were
confident that they would be good at choosing learning objectives and materials,
evaluating their own learning, identifying their weaknesses, and deciding what they
should learn next in the French class. Such was not the case with those who were only
motivated. So, how do we distinguish motivation and autonomy? It is observed that the
autonomous students have the ability to take charge of their learning and to make
responsible decisions about all aspects of their learning. Characteristics such as working
hard on French almost every day, wanting to learn as much French as possible, asking the
instructor for help, and sticking to the job at hand while ignoring distractions, overlap and
are found in motivated as well as autonomous students, but the previously mentioned
characteristics can be found only in autonomous students. The responses of the students,
described below, show that both motivation and autonomy are highly correlated, but they
are not one and the same factor with different names or labels. As was mentioned earlier,
clearly identifying and distinguishing between the affective factors of motivation and
autonomy is a challenging task. It seems that in assessing students' levels of motivation,
the focus is on the degree of effort the students put into their language learning, whereas
in assessing students' levels of autonomy, the focus is on their ability to determine
language learning objectives, to select methods and techniques that will help meet their
objectives, and to be able to evaluate what they have learnt (Holec, 1981).
The information obtained from students' journals and interviews was examined to
better understand how motivation and autonomy are correlated, yet distinct. First of all,
two students who are both highly motivated and highly autonomous, but have different
perspectives of their language learning experiences, are discussed. Next, other
participants who are motivated, but not autonomous are discussed to ascertain whether
the qualitative data corroborates the results from the questionnaires, showing that
motivation and autonomy are distinct factors.
B10 and B17 were both highly motivated and highly autonomous during both
administrations of the test. B10 enjoyed the class, liked her instructor and thought he
sincerely wanted them to understand the language. She appreciated the fact that her
instructor would talk in French "all the time." Outside of class she reported that she
practiced speaking French with her friends and enjoyed watching French movies. B10
was also an independent learner and showed signs of being responsible for her own
learning. In one of her journal entries, she writes:
I keep up with the listening manual and workbook. I read before each lesson and I
show up to every class. I study for the tests and do all the review exercises and I
think I am doing really well in class. ...I follow all the instructions and have top
marks in the class. I feel that in this class I am in control of my own learning but
by following the guidelines set for me.
These remarks show that this participant is both highly motivated and highly
autonomous. She confidently states that she is in control of her own learning, yet
behaves differently from what one would expect from an autonomous student. She has
full faith in the instructor, the syllabus and the structure of the class and feels she controls
her own learning, not by creating a new or different plan of study, but by following the
guidelines set for her because they produced the desired results in her learning and the
curriculum was working for her. One might assume that being as highly motivated as she
was, if the curriculum had not worked for her, she might have actively found ways to
modify it so that here learning would be effective. We observe this participant's
motivation in the effort she exerts to achieve her goals, and her autonomy in her
identification of the course of action that will give her the desired results. Both the
qualitative and quantitative data on this participant show that B10 was highly motivated
and highly autonomous in her French class.
According to the scores on the questionnaires, B17 was also highly motivated and
highly autonomous during both administrations of the test, but the data from the journals
and interview give a preliminary impression that B17 was neither motivated nor
autonomous in class. Though B17 had the same instructor as B10, she had a very
different perspective of him. In her journal, she states that she lost her motivation to
study French because her instructor is "very bad at teaching .... This makes me very
frustrated and unmotivated to study." Yet her scores on the motivation scale indicate that
she was highly motivated during both Weeks 5 and 13 of the semester. It is possible that
B17 was unmotivated to study, but that she was still motivated to learn French, as is
observed in the efforts she puts in, according to her journal entry, while studying with her
sister. In anotherjournal entry she writes, "Sometimes I get a little disheartened because
of class but lately I feel more motivated. Especially now that I have more time." This
remark shows that the participant is more motivated at this point in the semester, not
because the instructor has made any changes in his teaching style but because she
reportedly has more time. This student's levels of motivation depend more on factors
within her control, in this case, her time, rather than the quality of instruction in her
French class. She also writes:
I made an easy A in French last semester and I don't feel like it will be that way
this semester. Luckily, though, when I do study the course material is really
helpful. The book is easy to understand that [sic] the oral section online is very
useful. I also like that the syllabus shows us everything that will be going on ... I
hate the fact that it is just random things what [sic] we spend our class time on.
This subject could be really interesting but it isn't right now.
It seems as if B17 is unable to recognize the reasons for her apathy in class. During
the interview, she stated that her lack of motivation was due to the fact that her instructor
spoke in French for long periods of time and she tended to switch off, that he went "on
and on about things that are irrelevant" and also because he "grades rather harsh for a
beginning class," yet when she has a little more time for her studies, she feels motivated,
finds the textbook easy to understand when she takes the time to read through it, finds the
course material helpful, and the online oral section useful. Not only does this student feel
motivated when she does her part, puts some effort into her studies, and takes
responsibility for her learning by studying for her language course, but at home, she
usually goes over the entire book, making study guides on grammar and vocabulary. The
first impression we get about B 17's lack of motivation and autonomy from her qualitative
data, is not substantiated by her behavior during the semester. Even though B17 thinks
she is unmotivated and apathetic in class, she does take responsibility for her learning,
determines what will help her learn, and sets out to accomplish her goals. We observe,
yet again, that the qualitative data do corroborate the quantitative data and show that B17
is highly motivated as well as highly autonomous, even though she reported being
dissatisfied with her class and thinks the instructor is ineffective.
We notice that B 10 was very positive in her assessment of the class and had a final
class grade of 97%, whereas B17 was not as positive, and had a final class grade of 87%.
It is possible that the attitude of a language learner plays a role in the class performance,
or even that class performance plays a role in motivation. Motivation and autonomy tap
different emotions in different language learners, and from the quantitative and
qualitative data provided, we observe that varying degrees of interactions between these
two affective factors may produce varying degrees of success in different students, or
varying degrees of success at different times within the same student. Success, in this
case, is evidenced by class grades. Research (e.g. Wang and Palincsar, 1989) shows that
when learners take responsibility for their own learning and make independent choices to
facilitate their goals, their motivation increases and they are better able to achieve their
goals. It is obvious that both B10 and B17 have taken responsibility for their learning,
and are also motivated to work hard and succeed in their French class.
In an attempt to observe differences between students who are both autonomous
and motivated, and those who are only motivated, six students, A18, B01, B02, B06,
B18, and B21, who were highly motivated during both administrations of the AMTB, but
were not also highly autonomous, were examined. Some of the comments they made in
their journals and interviews were that: they did not make any specific plans as to how to
improve their language skills, they did not try to improve their skills outside class, they
did not think about how to best study for the class, and they never paid attention to the
objectives of the class. These comments suggest that these participants were not aware of
what they could do to improve their language skills and how they could control their
learning. They did not come up with any goals and objectives of their own, nor did they
evaluate what they had learnt, which is characteristic of autonomous behavior. However,
the participants were definitely motivated to learn French, as is seen from their interest in
purchasing French materials such as cassette tapes and short story books, writing short
paragraphs in French, singing French songs, reading French newspapers and making
flashcards to learn vocabulary. They were enthusiastic to learn the language and made
the effort to do better in class. In other words, these participants were motivated, but not
autonomous. Hence we observe from both the quantitative and qualitative data that
although motivation and autonomy are closely related these two constructs are indeed
distinct and manifest themselves in different ways in second language learners.
Research Question 7
What are some possible reasons for a change, if any, in the levels of anxiety,
motivation, and autonomy experienced by the participants from test to retest?
In the discussion of Research Question 3, it was noted that there were no significant
changes in the levels of anxiety and autonomy experienced by the participants, but that
there was a statistically significant decrease in motivation in the sample from Week 5 to
Week 13 of the semester. It was also stated that the motivation levels of Group 2 were
higher than those of Group 1 at the beginning of the semester. Both groups experienced a
decrease in their levels of motivation during Week 13 of the semester, but only the
decrease in Group 2 was statistically significant. In the discussion of this question,
however, the two groups are not considered separately because there was a decrease in
motivation levels in both groups. Also, most of the qualitative data has been collected
from Group 2. Eight of the eleven students who wrote journals were from Group 2, and
five of the six interviews conducted were with students from Group 2.
In order to understand what some possible reasons may be for this decline in
motivation levels, we examine some background information about the two classes that
was obtained through classroom observations.
The two instructors who taught the two intact classes in this study had different
teaching styles. Instructor 1 used a deductive method of teaching in class, stating the
language rule he wanted to teach that day and then giving the students many examples to
help them understand the concept. The teacher would speak in French most of the time,
but did not require his students to answer him in French. As a result, the students mostly
answered comprehension questions or questions concerning the presentation in English.
When the instructor asked students to read a few sentences from the textbook or was
teaching vocabulary and asked for the French equivalents, the students would make an
effort to speak in French, usually uttering words or phrases rather than complete
sentences. The teacher would call on the students at random to answer questions, making
sure that each one had a chance to answer at least one question. In a 50-minute class of
about 20 students, the students got an opportunity to answer a question in French twice,
on an average. The teacher occasionally asked the students to read a few lines from the
textbook out loud, which gave them some practice in reading the language. The teacher
mostly focused on teaching grammar and vocabulary in this class, and stated, during the
interview, that the students were more interested in learning about culture than grammar.
This French class was observed at least once a week. During these times, the students
were never engaged in any pair work or group work. This was probably one of the
reasons why the students would come to the classroom and wait for the class to begin
without engaging in any small talk or having any interactions amongst themselves and
were "just there," as the instructor put it during the interview. When the teacher came in
early, he would usually greet the students and then read something or keep himself
occupied until it was time to begin the class. No efforts were made to walk around the
class and interact with the students, which, in turn, might have developed a sense of
rapport between the students and the instructor and also among the students. When the
students asked the instructor any questions he would always spend time answering them.
Instructor 2, on the other hand, used an inductive method of teaching, writing a few
sentences or examples on the chalkboard, and asking the students to examine the
structure and come up with the language rule to be studied that day, based on the given
examples. The teacher would speak in French most of the time and also required his
students to answer him in French. The students would try their best to answer in
complete sentences. This instructor would not call on individual students as much as the
previous instructor for answers, but would explain a concept for a few minutes then ask
the students to practice the concept or grammar item in pairs or in groups of three,
depending on the seating arrangements in class. In this way, the students got a fair
amount of practice in using the language regardless of whether or not they always used
grammatically correct structures with each other. This teacher also occasionally asked
the students to read aloud some dialogues from the textbook, which gave them some
practice in reading the language. Instructor 2 believed that grammar should be taught in
the cultural context and remarked, during the interview, that his students asked many
questions on vocabulary and grammar, especially conjugations. This was probably
because the focus of teaching in this class was not always on grammar and vocabulary, as
it was in the previous class, and the teaching style was inductive rather than deductive.
The dissatisfaction of the students in each of these classes indicates that neither the
inductive nor the deductive teaching style met the needs of all the students in the class.
This class was also different from the first with regards to student interaction. The
students would come to the classroom and interact, in English, with the others sitting in
their general vicinity while waiting for the class to begin. This was probably because
most of the time the instructor would require the students to practice or complete an
exercise in pairs or small groups, which involved interactions on a regular basis. The
students may have formed a bond which made it easier for them to interact with one
another even when they were not required to.
The decrease in the overall levels of motivation within the participants could have
occurred for a variety of reasons examined below. These comments were extracted from
the journal entries and interviews that were conducted with the students and instructors,
and support the possible reasons given for a decrease in motivation in their language
classes. It must be noted that the following comments are made by students whose
motivation levels decreased from test to retest.
The participants who were highly motivated in this study said that they experienced
test anxiety. B02 was anxious just before tests because that is when she realized that she
did not know as much as she should. She said that the teacher did not require them to
complete the assignments, and as a result, she had fallen behind and was very worried
about the final exam. B 06, B17 and B18 felt that being nervous was a waste of time
because classes and tests are within the student's control, so either they knew the answers
or they did not. B17 added that when a person is apathetic, there is no anxiety. B18
reported being frustrated with her test grades because she knew more French than what
her test reflected.
None of the students said that the material was very difficult to grasp, but they felt
that they might have been more inclined to do the daily assignments if there had not been
so much to do every day that it became very tedious. Even if they did the assignments,
they were not motivated to bother about quality because the instructors did not read what
was written, but only checked to see if the work was done. A18, who was highly
motivated only during Test 1, said "I am not sure that I'm retaining much of what I am
learning because I have to concentrate so hard on learning each new lesson and chapter
that I forget what we've studied previously." Also, the students were not very motivated
to do the writing exercises in the workbook because it did not prepare them for the tests,
which included multiple choice answers or word banks from which to choose answers.
Fast Pace of Class
It was also felt that the pace of instruction was too fast and because new material
was being taught every day, the students did not have time to assimilate or practice what
was taught, so they were not learning as much as they could. A17, who was highly
motivated only during Test 1, felt that if the pace of class was slower, she would not be as
confused or discouraged while learning French. This aspect of class was probably one of
the factors that lead to a decrease in her motivation level.
Some students disliked the fact that they had to take the French class because of a
language requirement policy of the university. In connection with this policy, B18 said
that she hated people telling her what to do. A18 said that she was taking the class as a
Satisfactory/Unsatisfactory option so that she didn't have to work very hard, and came to
class only for the attendance. This aspect and the fact that there was so much to learn in
class probably were the reasons for A18's decrease in motivation.
Communication with Instructor
In one of the classes, the students felt that the teacher could not understand the
questions they asked him in English, so they gave up asking questions after a while.
They also felt that he could not explain some concepts well because of his lack of fluency
in English, and wished that he could give them examples of the concepts taught and their
English equivalents. Two other students said that when they got an answer wrong, the
instructor just said "it's wrong" without giving them any hint as to how to get the right
answer. Their disappointment with the method of instruction in class probably dampened
their enthusiasm to learn French.
Dissatisfaction with Class
The students were also dissatisfied with the way the class was taught. One of the
instructors focused a great deal on teaching grammar and vocabulary, and called on the
students to answer questions. The students in this class felt that they needed to learn how
to communicate in French in full sentences, rather than to learn a lot of grammar and
vocabulary. The second instructor frequently asked the students to do some activities in
pairs or small groups, and the students in this class thought there was not much point in
interacting with their classmates because they did not know the right answers, anyway.
These students said they would have preferred to be called upon to answer questions and
did not think that it would be an anxiety-provoking experience. This may be so because
this group usually interacted with each other in class.
These were some possible reasons for the decrease in motivation observed within
the participants, especially within Group 2. The comments made by the students give an
insight into their frustrations. It is quite clear that the students were more motivated at
the beginning of the semester, but that as the semester progressed, some of them lost
interest and their motivation levels decreased.
There is no doubt that the students were frustrated in their French class, but the
interviews with the instructors revealed that they also had their share of frustrations.
There were times when they felt that they could not do much to improve the situation in
the classroom because it was beyond their control. They were very frustrated when the
students complained about the tests and said that the grading was harsh because the
students did not put much effort into their studies, were fine with a grade of 'B,' and
came to the tests unprepared. It was also frustrating when the students would walk into
class very late and even read the college paper during composition days instead of
spending that time in their writing. Teaching the students grammar was a challenge
because they were unfamiliar with the grammar terms even in English so they were not
really interested in grammar. The students also complained that they were not prepared
in class for the final oral exams, but the instructors believed that doing an oral exercise in
class was preparation for that exam. This mismatch of student-instructor perception of
the final oral exam might have been prevented if the students were reminded throughout
the semester that any oral communication they engage in during class is practice for the
final oral exam.
Some of the concerns the instructors had about the class were regarding the amount
of grading that needed to be done. Since it was not possible to grade all the work that
was done, they mainly checked for completion, hoping that the students had learnt
something from actually doing the exercises. Another area of concern the instructors had
was about the SAT II placement tests that were conducted only during the first two days
of the semester. When a student decided to join the French class on the third or fourth
day, they had to decide which class s/he should attend. This was a challenging task
because if a mistake was made in assigning the student to the appropriate class, they were
'stuck' with that student for the rest of the semester.
These remarks made by students and instructors alike give an insight into the
frustrations experienced on both sides and could serve as possible reasons why the
motivation levels within the participants decreased over the course of the semester. This
was only one side of the story, however, because there were some students who were
very happy with their instructors. At the same time, the instructors had some very
positive comments to make about some of the students. Nevertheless, because the results
show that there was an overall statistically significant decrease in motivation levels in
this group of students, some possible reasons for this change was explored.
In this study, out of a total of 32 participants, during Test 1 eight participants (25%)
experienced high levels of anxiety, 19 participants (59%) were highly motivated, and five
participants (16%) were autonomous. During Test 2, seven participants (22%)
experienced high levels of anxiety, 17 participants (53%) reported being highly
motivated, and four participants (13%) were autonomous. No correlations were found
between anxiety and motivation or between anxiety and autonomy, but there were
statistically significant moderate correlations between motivation and autonomy. Results
shows that the participants' motivation levels decreased from Time 1 to Time 2. As far
as correlations between the affective factors and class performance was concerned,
anxiety had a statistically significant moderate negative correlation with class