Examination of Horwitz, Horwitz,

and Cope’s Construct of Foreign

Language Anxiety: The Case of
Students of Japanese
WKIE AIDA
Department of Oriental and Afican Languages and Literatures
University of Texas at Austin
2601 University Avenue
Austin, T X 78712
Email: aidaQccwjcc.utexas.edu

THE PRESENT STUDY CONCERNS HOW Lambert, it requires approximately 1320 hours
language anxiety is related to Japanese lan- of instruction in an intensive program in lan-
guage learning. It uses Horwitz, Horwitz, and guages like Japanese, Arabic, Chinese, and Ko-
Cope’s theoretical model of foreign language rean to bring students to the same level ofprofi-
anxiety as a research framework. It has been ciency reached after only about 480 hours of
reported that foreign language anxiety is a instruction in languages like French or Spanish.
rather pervasive phenomenon (14; 31; 32; 46; 47; Therefore, the experiences that students have in
52). Although language anxiety could be viewed the classroom with such difficult languages may
as positive energy (or facilitating anxiety as be different from the experiences of students in
called by Alpert and Haber) that motivates languages that are more similar to English.
learners, many language teachers and re- Do students of Japanese feel anxious in their
searchers have been concerned about the possi- classrooms? If so, what are the sources of their
bility that anxiety may function as an affective anxiety? Are there gender differences in lan-
filter (28), preventing a learner from achieving guage anxiety? Does anxiety interfere with their
a high level of proficiency in a foreign language learning ofJapanese? The present study was de-
(4; 7; 17; 25; 27; 39; 42; 56; 62). However, most of signed to answer these questions.
the research studies have involved Western lan- Due to the importance of the economic and
guages such as French, German, Spanish, and political relationship between the US and Ja-
English, and there has been little investigation pan, the number of students interested in learn-
of non-Western languages like Japanese. In ing Japanese has been growing at a rapid pace.
order to develop a fuller understanding of the According to the results of the fall 1990 survey
nature of language anxiety and its implications conducted by the Modern Language Associa-
for language education, future research should tion, 45,117 college students were studyingJapa-
include non-Western languages. This study nese in United States institutions of higher edu-
takes a step in that direction. cation in 1990, representing a spectacular
As a Japanese educator, the author became increase of 94.9% from 1986 when 23,454 stu-
very interested in exploring the role of anxiety dents were registered in Japanese language
in Japanese language learning among college courses (6). Japanese became the fifth most
students. Learning Japanese is a very difficult commonly taught language in 1990, rising from
task for Americans. According to Jorden and seventh position in 1986. Therefore, it is impor-
tant for language educators to identify the vari-
ables that may increase or decrease retention
and success in Japanese language learning. Lan-
The Modern Language Journal, 78, ii (1994) guage anxiety is one of these important
0026-7902/94/155-168 $1.50/0
01994 The Modern Lanffuage Journal variables.
156 The Modern LanguageJournal 78 (1994)
EARLY RESEARCH ON that anxiety is not associated with achievement,
FOREIGN LANGUAGE ANXIETY
because we do not know whether the anxiety
Early research on the role of anxiety in for- measures used in Young’s study could accu-
eign language learning failed to demonstrate rately capture students’ anxiety levels in oral
any clear-cut relationship between anxiety and production.
a learner’s achievement in a foreign language.
For example, Chastain examined the relation-
ships between anxiety and course grades in HORWITZ, HORWITZ, AND COPE’S
three language programs: French (audiolingual CONSTRUCT OF LANGUAGE ANXIETY
or regular), German, and Spanish. While there Horwitz (24) and Horwitz et al. have attrib-
was a significant negative correlation found be- uted the inconclusive results of previous re-
tween course grades and test anxiety in the search to the lack of a reliable and-valid meas-
French audiolingual class, students in the regu- ure of anxiety specific to language learning.
lar French, German, and Spanish classes who They conceptualize foreign language anxiety as
experienced a higher level of anxiety were more “a distinct complex of self-perceptions, beliefs,
likely to receive better grades than students feelings, and behaviors related to classroom lan-
with a lower level of anxiety. Backman looked at guage learning arising from the uniqueness of
the relationship between anxiety and language the language learning processes” (25: p. 31).
progress among Venezuelan students learning The Foreign Language Classroom Anxiety Scale
English in the US. Students’ progress measured (FLCAS, hereafter) was developed by Horwitz
by a placement test, a listening comprehension (24) in order to capture this specific anxiety
test, and teachers’ ratings did not show a sig- reaction of a learner to a foreign language
nificant correlation with any of the anxiety learning setting. Horwitz et al. integrated three
measures. related anxieties to their conceptualization of
In Kleinmann’s 19’7’7 study of Spanish-speak- foreign language anxiety, i.e., communication
ing and Arabic-speaking ESL students, facilitat- apprehension (35), test anxiety (19; 50), and
ing anxiety was found to be correlated with stu- fear of negative evaluation (58).
dents’ oral production of linguistically difficult According to McCroskey (34), communica-
(thus challenging) English structures (e.g., in- tion apprehension is defined as a person’s level
finitive complements and passive sentences). of fear or anxiety associated with either real or
However, there was no evidence that debilitat- anticipated communication with another per-
ing anxiety negatively influenced oral perfor- son or persons. McCroskey (33) points out that
mance. The facilitating and debilitating effects typical behavior patterns of communicatively
of anxiety were also observed by Bailey through apprehensive people are communication avoid-
her review of students’ diaries. ance and communication withdrawal. Com-
Young (62) conducted a study to test whether pared to nonapprehensive people, communi-
oral proficiency was negatively influenced by catively apprehensive people are more reluctant
anxiety in three languages, i.e., French, Ger- to get involved in conversations with others and
man, and Spanish. She found some negative to seek social interactions. The extensive body
correlations between students’ OPI (Oral Profi- of research in this area, summarized by Daly
ciency Interview) scores and some of the anxi- and Stafford and by Richmond, supports Mc-
ety measures. However, when language ability Croskey’s claim. In 1985 McCroskey, Fayer, and
measured by a dictation test and a self-appraisal Richmond studied the relationships between
measure of foreign language oral proficiency communication apprehension and self-perceived
was controlled statistically (i.e., the variability competence in Spanish and English among
due to language ability was removed from the Spanish-speaking Puerto Rican college stu-
relationship between anxiety and oral perfor- dents who had received instruction in English.
mance), the correlations between anxiety meas- They found that students with low self-ratings of
ures and OPI scores were nonsignificant. Such competency in English were more likely to re-
results are very predictable since language abil- port higher levels of English communication
ity is likely to correlate with language achieve- apprehension. On the other hand, there was no
ment. When language ability is held constant as such correlation found between self-perceived
was done in Young’s (62) study, there is little left competence in the native language, i.e., Span-
in the OPI scores to covary with anxiety. How- ish and Spanish communication apprehension.
ever, these nonsignificant correlations obtained Similarly, Foss and Reitzel and Lucas report that
through the above procedure cannot warrant communication anxiety exists among students
Yukie Aida 157
in the ESL classroom; it seems to function as a evaluation is applied to foreign language
block for students’ mastery of English. It is very learners, we can easily imagine that students
likely that people experience anxiety and reluc- with fear of negative evaluation sit passively in
tance in communicating with other people or in the classroom, withdrawing from classroom ac-
expressing themselves in a foreign language in tivities that could otherwise enhance their im-
which they do not have full competence. provement of the language skills. In extreme
The second element of foreign language anx- cases, students may think of cutting class to
iety, test anxiety, is defined by Sarason (51) as avoid anxiety situations, causing them to be left
“the tendency to view with alarm the conse- behind. Horwitz et al. believe that these three
quences of inadequate performance in an eval- anxieties, i.e., communication apprehension,
uative situation” (p. 214). Students worry about test anxiety, and fear of negative evaluation, are
failing to perform well. Culler and Holahan and important parts of foreign language anxiety
other researchers (22; 60) speculate that test and have an adverse effect on students’ lan-
anxiety may be caused by deficits in students’ guage learning.
learning or study skills. Some students experi- Horwitz (23) reported that the FLCAS had a
ence anxiety during a test situation because correlation coefficient of .28 (p = .063, n = 44)
they do not know how to process or organize the with communication apprehension (measured
course material and information. Since daily by McCroskey’sPersonal Report of Communica-
evaluations of skills in foreign language class- tion Apprehension, 35), .53 (p < .01, n = 60)
rooms are quite common, and making mistakes with test anxiety (measured by Sarason’s Test
is a normal phenomenon, students may suffer Anxiety Scale, 51), and .36 (p < .01, n = 56) with
stress and anxiety frequently, which may pose a fear of negative evaluation (measured by Wat-
problem for their performance and future im- son and Friend’s Fear of Negative Evaluation
provement. Other researchers posit that test Scale). The FLCAS also correlated with final
anxiety occurs when students who have per- grades: r = -.49, p < .01 (n = 35) for two begin-
formed poorly in the past develop negative, ir- ning Spanish classes and r = -.54, p < .01 (n = 32)
relevant thoughts during evaluation situations for two beginning French classes. Higher
(40; 49; 59). Test-nervous students may not be FLCAS scores were associated with lower final
able to focus on what is going on in the class- grades. Price also reported in her dissertation
room because they tend to divide their atten- that the FLCAS scores of 106 students of
tion between self-awareness of their fears and second-semester French classes were positively
worries and class activities themselves. They correlated with test anxiety (r = .58, p < .001)
may say to themselves, “I’ll never be able to pro- and public speaking anxiety (r = .43, p < .001).
nounce it correctly,” “The teacher is ready to The FLCAS scores also correlated negatively
correct me,” or “Other students will laugh at with final grades (r = -.22, p < .05), final exam
me if I speak.” They become distracted and scores (r = -.29, p < .Ol), and oral exam scores
anxious during class, which interferes with their (r = -.27, p < .05). However, when students’
performance. Modern Language Aptitude Test scores were
Lastly, fear of negative evaluation is defined controlled, only the correlation between the
as “apprehension about others’ evaluations, dis- oral exam scores and the FLCAS scores re-
tress over their negative evaluations, and the mained significant.
expectation that others would evaluate oneself The main purpose of this study was to test
negatively” (58: p. 449). Research shows that Horwitz et al.’s construct of foreign language
people who are highly concerned about the im- anxiety by validating an adapted FLCAS for stu-
pressions others are forming of them tend to dents ofJapanese. It was an exploratory study to
behave in ways that minimize the possibility of discover the underlying structure of the FLCAS
unfavorable evaluations. They are more likely to and to examine whether or not the structure
avoid or prematurely leave social situations in reflects the three kinds of anxiety presented
which they believe others might perceive them earlier. It also assessed the instrument’s re-
unfavorably (29; 57; 58; 63). When they affiliate liability and the relationship of students’ anxi-
with others, they often fail to initiate conversa- ety levels to their performance in Japanese. It
tions or participate only minimally in the con- was hoped that the results of this empirical
versation, as by just smiling and politely nod- study using a non-Western language would
ding, o r listening to others talk and only shed new light on the concept of foreign lan-
interacting with occasional “uh-huh’s” (8; 30; guage anxiety and would expand its scope and
43; 45). When this notion of fear of negative implications.
158 The Modern LanguageJournal 78 (1994)
METHOD agree”, to (c) “neither agree nor disagree”, to
(e) “strongly agree”. A student’s endorsement
Subjects. In the fall of 1992, students who were in (a) “strongly disagree” was equated with a
enrolled in second-year Japanese I at the Uni- numerical value of one; (b) “disagree” was two;
versity of Texas at Austin were asked to partici- (c) “neither agree nor disagree,”three; (d) “agree”
pate in this study. Ninety-six students (fifty-six four; and, (e) “strongly agree” was five.
males and forty females) completed the ques- For each subject, an anxiety score was derived
tionnaires designed for this study. There were by summing his or her ratings of the thirty-
more than ninety-six students enrolled in the three items. When statements of the FLCAS
course, but some students failed to complete the were negatively worded, responses were re-
questionnaires or to pass the course. Three stu- versed and recoded, so that in all instances, a
dents did not pass the course because they high score represented high anxiety in the Japa-
failed to attend class regularly and to complete nese classroom. The theoretical range of this
several important exams and/or assignments scale was from thirty-three to 165.
(e.g., lesson quizzes, essay writing, oral presen- The background questionnaire included
tation). Therefore, only data obtained from questions on the student’s age, sex, ethnicity,
these ninety-six students were used for analysis. academic major and status, native language,
The mean age of this sample was 21.5 years. reasons why he or she was taking a Japanese
There were sixty-four native speakers of English course, whether or not he or she had been to
and thirty-two non-native speakers of English Japan and for how long, whether o r n o t he or
(i.e., five Spanish speakers, six Chinese, four- she was pleased with the final course grade
teen Korean, five other Asian language speakers, given for the second-semester Japanese class,
and two other non-Asian language speakers). and whether or not he or she had other family
When the native speakers of English and non- members who speak Japanese.
native speakers of English were compared on Instructors provided subjects’ final course
the level of anxiety (see the Proceduressection for grades (in percentages) for the second-semester
how to obtain a subject’s anxiety score), a one- Japanese classes. The final course grade was se-
way ANOVA showed that there was no signifi- lected primarily because it had been used as a
cant difference between the two groups: global measure of language proficiency by
F(1, 94) = .O?, p = .79, X = 96.2 for native many researchers (e.g., 7; 9; 18; 25; 55).
speakers of English and X = 95.5 for non-native
speakers of English). In addition, a Bartlett-Box
RESULTS
F test for homogeneity of the variance indicated
that the data of the present study satisfied the Reliability of the FLCAS. The present study,
assumption of equal variances: F = 1.21, p = .27. using ninety-six students of Japanese, yielded in-
Therefore, the two groups were treated as one ternal consistency of .94 (X= 96.7 and s.d. = 224,
sample. using Cronbach’s alpha coefficient. As shown in
Procedures. On the very first day of the fall Table I, the reliability, mean, standard devia-
semester 1992, subjects were asked to complete tion, and range obtained in this study were very
both a FLCAS and background questionnaire similar to those of Horwitz (23), who used stu-
(see Appendix). In this study, the term “for- dents enrolled in an introductory Spanish class.
eign language” used in the original FLCAS The mean of this study, 96.7, was slightly higher
was replaced with “Japanese language.” In re- than that of Horwitz’s (23) study, X = 94.5. It is
sponding to the statements on the FLCAS, understandable that students may feel more
subjects were asked to consider their experi- anxious in learning a non-Western, foreign lan-
ences in the previous year’s first-year Japanese guage like Japanese (26) than in learning com-
course. Therefore, students’ FLCAS scores re- monly taught Western languages such as
flect their anxiety in the first-year Japanese Spanish.
classroom. The instructions read as follows: There was no significant gender difference
“In this section, we would like you to respond found in language anxiety: t(94) = .41, p = .69.
to each of the following statements based upon The mean scores for males (n = 56) and females
your experience in your last yeark Japanese course (n = 40) were 97.4 and 95.6, respectively. The
(JPN507).” results of this study suggest that the FLCAS is a
Instruments. The FLCAS contains thirty-three reliable tool regardless of whether the language
items, each of which is answered on a five-point is a European Western language.
Likert scale, ranging from (a) “strongly dis- On the first day of the next semester (spring
Yukie Aida 159
TABLE I nalities, and percent of the variance are shown
Reliabilities of The FLCAS in Two Studies in Table 111. The solution accounted for 54.5%
~~~~

of the total variance. Eighteen items were

Horwitz
Present study et al., 1991 loaded on the first factor, accounting for 37.9%
of the variance. Examples of the items included
Sample size 96 108 in this factor are item 3, “I tremble when I lcnow
Students status first year first year that I’m going to be called on in my Japanese
Language Japanese Spanish
class,” and item 13, “It embarrasses me to volun-
Cronbach’s alpha .94 .93
Range 47-146 45-147 teer answers in my Japanese class.” The factor
Mean 96.7 94.5 one was assigned a label of Speech Anxiety and
Standard 22.1 21.4 Fear of Negative Evaluation. The items in-
deviation cluded in this factor indicate a student’s ap-
Test-retest r = 30, p < .01 r = 33, p <.01 prehension in speaking in a Japanese class and
reliability (n = 54; over (n = 108; over fear of embarrassment in making errors in
one semester) eight weeks) front of other students (see Table 11). Two
items, 8 and 18, were negatively loaded on this
1993), students who had passed second-year factor. In other words, item 8, “I am usually
Japanese I and were enrolled in second-year at ease during tests in my Japanese class,” and
Japanese I1 were asked to complete the FLCAS item 18, “I feel confident when I speak in my
again. Fifty-four students (thirty-one males and Japanese class,” are negatively associated with
twenty-three females) responded. Their two factor one. Unlike the speculation of previous
FLCAS scores were correlated to obtain test- researchers (e.g., 34; 5 8 ) , speech anxiety and
retest reliability over one semester. The correla- fear of negative evaluation may not be totally
tion between the FLCAS scores in the fall and independent concepts, but rather are probably
those in the spring was .SO, p < .01, n = 54, different labels describing one phenomenon
indicating that the FLCAS measures a person’s in a language learning situation. In their
level of anxiety with high accuracy at different factor analysis of various anxiety measures,
times. This high correlation suggests that the MacIntyre and Gardner (37) reported that
FLCAS may tap a person’s persistent trait anxi- McCroskey’s Personal Report of Communica-
ety (as called by Spielberger) in the foreign lan- tion Apprehension measure (34) and Watson
guage classroom and not a temporary condi- et al.’s Fear of Negative Evaluation measure
tion of state anxiety that is triggered by a loaded on the same factor. Their findings
specific moment or situation. are in accordance with those of the present
Factor Analysis. The second analysis was per- study.
formed to detect an underlying structure of the The second factor included four items (i.e., 10,
FLCAS’s thirty-three items, i.e., students’ rat- 25, 26, and 22) and accounted for 6.3% of the
ings of the original (unreversed and unre- variance. Item 22 was negatively loaded on this
coded) thirty-three statements. Principal com- factor. The author named this factor “Fear of
ponents analysis with varimax rotation was Failing the Class” and considered that it showed
performed on the thirty-three items. Orthogo- a student’s worry and nervousness about being
nal rotation was used because of the conceptual left behind in the class or failing the class
simplicity and ease of description. The initial altogether.
run produced seven factors with eigenvalue Items 32, 11, and 14 comprised the third fac-
greater than one. In a rotated matrix, however, tor, accounting for 5.6% of the variance. It was
there were only four factors with SSLs (the sum labeled “Comfortableness in Speaking with Jap-
of squared loadings, which is equal to the eigen- anese People” by the author. In the interview
value in the unrotated matrix) greater than with Young (61), Krashen says that foreign lan-
one. Therefore, the subsequent analysis spec- guage learners need to think of themselves as
ified the number of factors as four. With a fac- the kind of people who speak the foreign lan-
tor loading of .50 (twenty-five percent of the guage very well. This idea is similar to Gardner’s
variance) as a cutoff for inclusion of a variable concept of integrativeness. It is likely that indi-
in interpretation of a factor, six items (items 2, viduals who do not see the language as truly
6, 15, 19, 28, and 30, see Table 11) did not load foreign and feel comfortable with the native
on any factor. None of the items loaded on more speakers of the language have a lower filter of
than one factor with a loading of .50 or greater. anxiety.
Loadings of variables on factors, commu- Lastly, two items, 5, “It wouldn’t bother me at
160 The Modern Language Journal 78 (1994)
TABLE I1
FLCAS Items with Percentage of Students Selecting Each Alternative in Four Factors

SAa A N D SD
Factor One (Speech Anxiety and Fear of Negative Evaluation)
3d I tremble when I know that I’m going to be called on in my Japanese class.
7b 24 20 35 14
13d It embarrasses me to volunteer answers in my Japanese class.
5 20 19 35 21
27 I get nervous and confused when I am speaking in my Japanese class.
4 27 24 38 7
20d I can feel my heart pounding when I’m going to be called on in my Japanese class.
8 28 21 28 15
24c I feel very self-conscious about speaking Japanese in front of other students.
10 35 18 28 8
31d I am afraid that the other students will laugh at me when I speak Japanese.
3 12 13 45 28
7d I keep thinking that the other students are better at Japanese than I am.
16 28 26 22 8
12 In Japanese class, I can get so nervous I forget things I know.
12 32 17 35 4
23d I always feel that the other students speak the Japanese language better than I do.
9 26 21 34 9
18c.f I feel confident when I speak in my Japanese class.
7 27 28 32 5
33c I get nervous when the Japanese teacher asks questions which I haven’t prepared in advance.
12 44 21 23 1
16 Even if I am well prepared for Japanese class, I feel anxious about it.
9 35 21 24 10
1 I never feel quite sure of myself when I am speaking in my Japanese class.
14 28 13 34 12
21 The more I study for a Japanese test, the more confused I get.
1 6 10 43 40
2 9 ~ I get nervous when I don’t understand every word the Japanese teacher says.
6 27 23 37 7
4c It frightens me when I don’t understand what the teacher is saying in the Japanese class.
8 48 15 21 8
8f I am usually at ease during tests in my Japanese class.
12 28 13 39 9
9c I start to panic when I have to speak without preparation in Japanese class.
14 32 23 22 9
Factor Two (Fear of Failing the Class)
10 I worry about the consequences of failing my Japanese class.
30 27 8 18 17
25 Japanese class moves so quickly I worry about getting left behind.
18 40 9 24 9
26 I feel more tense and nervous in my Japanese class than in my other classes.
23 29 12 23 14
22f I don’t feel pressure to prepare very well for my language class.
27 45 12 13 3
Factor Three (Comfortabkness in Speaking with Native Japanese)
32 I would probably feel comfortable around native speakers of Japanese.
5 26 42 19 8
11 I don’t understand why some people get so upset over Japanese classes.
6 35 33 21 4
14 I would not be nervous speaking the Japanese language with native speakers.
10 47 19 17 7
Factor Four (Negative Attitudes Toward theJapanese Class)
5f It wouldn’t bother me at all to take more Japanese language classes.
1 5 14 35 45
Yukie Aida 161

SA= A N D SD
17 I often feel like not going to my Japanese class.
2 15 11 47 25
I t e m Not Included i n the Factor Solution
2e I don’t worry about making mistakes in my Japanese class.
16 47 12 19 7
6e During Japanese class, I find myself thinking about things that have nothing to do with the
course.
3 12 21 43 22
15 I get upset when I don’t understand what the teacher is correcting.
4 39 29 26 2
19e I am afraid that my Japanese teacher is ready to correct every mistake I make.
3 12 26 48 12
28 When I’m on my way to Japanese class, I feel very sure and relaxed.
3 19 41 29 8
30 I feel overwhelmed by the number of rules you have to learn to speak Japanese.
5 35 22 29 8
aSA = strongly agree; A = agree; N = neither agree or disagree; D = disagree; SD = strongly disagree.
bpercentages in this table are rounded to the nearest whole number, thus may not add up to 100.
CItems that are classified by Horwitz et al. (25) as ones indicative of speech anxiety.
dIterns that are classified by Horwitz et al. (25) as ones indicative of fear of negative evaluation.
eItems that are classified by Horwitz et al. (25) as ones indicative of test anxiety.
fItems that were negatively loaded on the factors.

TABLE 111
Factor Loadings, Communalities (h*), Percents of Variance for Four-Factor Principal Component
Analysis with Varimax Rotation on FLCAS Items.

Label Speech Fear of Comfort- Negative Label Speech Fear of Comfort- Negative
Anxiety Failing ableness Attitudes Anxiety Failing ableness Attitudes
with JPN with JPN
Item Factor 1 Factor 2 Factor 3 Factor 4 h2 Item Factor 1 Factor 2 Factor 3 Factor 4 h2

Item 3 .77 .69 Item 8 -.56 .48

Item 13 .76 .61 Item 9 .54 .49
Item 27 .75 .73 Item 10 .72 .54
Item 20 .73 .67 Item 25 .53 .60
Item 24 .73 .66 Item 26 .51 .56
Item 31 .71 .53 Item 22 -.5 1 .46
Item 7 .71 .60 Item 32 .74 .62
Item 12 .69 .58 Item 11 .60 .41
Item 23 .69 .57 Item 14 .59 .45
Item 18 -.67 .70 Item 5 -.77 .65
Item 33 .60 .42 Item 17 .73 .59
Item 16 .60 .59
% of
Item 1 .58 .60
variance 37.9 6.3 5.6 4.7
Item 21 .58 .53
Item 29 .57 .54 % of total variance accounted for by the
Item 4 .56 .62 solution 54.5

all to take more Japanese language classes,” In their review of the literature in regard to
and 17, “I often feel like not going to my Japa- anxiety and language learning, MacIntyre and
nese class,” constituted the fourth factor. Item 5 Gardner (38) postulated that language anxiety
was negatively loaded on this factor. This factor develops as a result of negative experiences the
explained 4.7% of the variance and was called student may have had in an earlier stage of
“Negative Attitudes Toward the Japanese Class.” learning a new language. The present analysis
162 The Modern LanguageJournal 78 (1994)
supports their hypothesis. Students’ negative at- beyond the required classes and that the attri-
titudes toward the language class can contrib- tion rate might be high at a transition point
ute to their overall levels of foreign language from a lower division class to an upper division
anxiety. classl.
The factor solution of the present study pro- The Relationship between Anxiety and Perfor-
vided partial support for Horwitz et al.’s con- mance. In the subsequent analysis, the relation-
struct of foreign language anxiety. It has shown ship between foreign language anxiety and stu-
evidence that speech anxiety and fear of nega- dents’ performance was investigated. First, the
tive evaluation are indeed important compo- correlation coefficient between anxiety and
nents of foreign language anxiety. Yet the pres- course grade was calculated with a Pearson
ent study did not support Horwitz et al.’s claim product-moment correlation. It produced a
that test anxiety is the third component of for- moderate negative correlation (r = -.38, p < .01)
eign language anxiety. Items 2, 6, and 19 which indicating that the higher the students’ levels of
were considered by Horwitz et al. to be indica- anxiety, the more likely they are to receive low
tive of test anxiety, failed to load on any of the grades. For the second analysis, each student
factors. In addition, eighty-three percent of the was classified into either a high anxiety group
students rejected statement 21, “The more I or a low anxiety group by a median split pro-
study for a Japanese test, the more confused I cedure, based upon his or her total score on the
get.” The subjects of the present study seem to FLCAS. The median score of anxiety for this
be less intimidated by the Japanese tests. These sample was ninety-five. A two by two ANOVA
findings are congruent with the results ob- was conducted using anxiety (high vs. low) and
tained by MacIntyre and Gardner (39), who gender (males vs. females) as the independent
found that test anxiety did not contribute to the variables and final course grade as the depen-
communicative anxiety of the language class- dent variable. There was a significant main effect
room. They concluded that test anxiety was a of anxiety: F(1, 92) = 7.35, p < .01 (see Table IV).
general anxiety problem; it was not specific to The high anxiety group received significantly
foreign language learning. Based on these find- lower grades (X = 85.6) than the low anxiety
ings, it appears clear that test anxiety is not con- group (X= 89.8). While students having a high
ceptually related to other components of foreign anxiety level were more likely to receive a grade
language anxiety as Horwitz et al. proposed, of B or lower, those with a low level of anxiety
and that items reflective of test anxiety could be were more likely to get an A.
eliminated from the FLCAS. Speech anxiety It was also found that there was a significant
and fear of negative evaluation are considered effect of gender on course grade: F(1, 92)
as relatively enduring personality traits (41), = 4.74, p < .05. Female students scored higher
whereas test anxiety is regarded as a state (X= 89.7) than did males (X= 86.1)2. There was
marked by temporary reactions (e.g., worry and no significant anxiety-gender interaction effect
nervousness) to an academic or evaluation situ- on course grade: F(1,92) = 3.20, p > .05. In both
ation (51). This distinction might also partially male and female groups, highly anxious stu-
explain the results of this factor analysis. The dents were more likely to receive lower grades
present study suggests that other factors such as than students having a low level of anxiety.
a student’s fear of failing the class, comfortable-
ness in speaking with native speakers of the lan-
guage, and negative attitudes toward the lan- TABLE IV
Anxiety by Sex ANOVA Results on Achievement
guage class influence the level of anxiety in the
( N = 96)
foreign language classroom.
The results show that a fair amount of anxiety Sum of Mean Sig.
exists in the Japanese classroom. A third or Source Souares df Souares F ofF
more of the students in the sample showed anxi- Main Effects
ety agreement with items reflective of foreign Anxiety 376.6 1 376.6 7.35 .008
language anxiety. There were six items (4, 5,10, Sex 243.1 1 243.1 4.74 ,032
25, 26, and 33) that were endorsed by over half
Interaction
of the students. Eighty percent of the students
Anxiety
disagreed or strongly disagreed with statement 5, by sex 164.1 1 164.1 3.20 ,077
“It wouldn’t bother me at all to take more Japa-
nese language classes.” This suggests that stu- Residual 4715.2 92 51.3
Total 5551.0 95 58.4
dents may be less likely to take a Japanese class
Yukie Aida 163
Association Between Anxiety and Demographic students who are feeling good about their
Data. A series of one-way ANOVAs were con- grades are likely to experience lower anxiety
ducted to investigate the relationship between than those who are not happy with their grades.
language anxiety and students’ demographic
data.
DISCUSSION
Elective us. Required Status. Students were clas-
sified into one of three groups: 1) Required The adapted Foreign Language Classroom
Group, including forty-one students who were Anxiety Scale was found to be a highly reliable
taking the Japanese class to satisfy the univer- instrument to measure the anxiety level o f stu-
sity’s language requirement, 2) Elective Group, dents learning Japanese in a college setting typ-
including forty-four students who were taking ified by the University of Texas at Austin. Fac-
the Japanese class for personal interest or enjoy- tors that had an impact on students’ anxiety in
ment, and 3) Major Group, including eleven learning Japanese were speech anxiety and fear
students who are majoring in Japanese or in of negative evaluation, fear of failing the Japa-
Asian Studies with specialization in Japanese. nese class, degree of comfort when speaking
An ANOVA result shows that there were no dif- with native speakers of Japanese, and negative
ferences in anxiety among the three groups: attitudes toward the Japanese class. In the pres-
F(2, 93) = 2.64, p > .05. However, when the Ma- ent sample of students of Japanese, test anxiety
jor Group was removed from the analysis and was not a factor contributing to students’ for-
the Required Group was compared with the eign language anxiety. The factors that were
Elective Group in the anxiety level, an ANOVA found important in this study for explaining the
yielded a significant difference: F(1, 83) = 5.5, construct of foreign language anxiety appear to
p < .05. The Required Group had a significantly support views of language anxiety proposed by
higher level of anxiety (X= 99.6) than the Elec- scholars such as MacIntyre and Gardner (38)
tive Group (X= 93.1). and Krashen and Terrell (cited in 61).
Experience inJapan. Comparison in the anxiety Consistent with research findings using West-
level was made between students who had been ern languages like French, German, and Span-
to Japan (n = 36) and those who had never been ish (e.g., 25; 32; 4‘7; 55), language anxiety was
to Japan (n = 60). The result of a one-way found to be negatively related to students’ per-
ANOVA was significant: F(1, 94) = 4.0, p < .05. formance in Japanese. A recent article by
Those with experience in Japan showed a signif- Gardner and MacIntyre reports that “the best
icantly lower level of anxiety in the classroom single correlate of achievement is Language
(X= 92.5) than those who had not been to Japan Anxiety” (p. 183). The author intends to exam-
(X = 98.1). Exposure to culture and people in ine in a future study whether the Gardner and
Japan may be a factor for this group difference. MacIntyre statement stands true for the sample
Family Who SpeaksJapanese. There were twenty- of students studying Japanese. The present
four students who had a family member with a study used final course grades as the dependent
command of Japanese. The anxiety levels of variable to examine the relationship between
these students were compared with the anxiety anxiety and language achievement. Since the
levels of students whose family members did not FLCAS appears to measure anxiety primarily
speak Japanese (n = 72). There was no anxiety related to speaking situations, use of a specific
difference found between the two groups: measure of oral skills may yield more profound
F(l, 94) = .1, p = .77. The presence of a family relationships between language anxiety and
member who speaks Japanese does not seem to achievement.
be related to the individual’s level of anxiety in Although the present study was successful in
the Japanese classroom. producing partial support for the findings of
Satisfaction with Grade in Japanese. Sixty-nine previous research studies using Western lan-
students indicated that they were pleased with guages, certain limitations of this study need to
the grade they received for the second semester be considered. First of all, the subjects were
of first-year Japanese. Twenty-seven students only those who had completed two semesters of
said that they were not pleased. A comparison Japanese. A study using students with a longer
of the anxiety levels of these two groups re- history of Japanese language learning may pro-
vealed a significant difference: F(1, 94) = 12.7, duce different results. In their 1991 article, Mac-
p < .01. Satisfied students exhibited a much Intyre and Gardner (38) cited the results of sev-
lower level of anxiety (X = 93.1) than non- eral studies, indicating that as “experience and
satisfied students (X = 103.4). It appears that proficiency increase, anxiety declines in a fairly
164 The Modern LanguageJournal 78 (1994)
consistent manner” (p. 111).Therefore, anxiety dents’ needs, language teachers can make it
may play a different role in foreign language possible for anxious students to maximize their
learning for advanced students. language learning by building a nonthreaten-
Secondly, subjects of this study were asked to ing and positive learning environment, as well
recall their experiences in the first-year Japa- as by helping them acquire effective study and
nese classes and to indicate their feelings about learning strategies.
those classes. There was a three month gap be- However, foreign language anxiety may not
tween the time when they completed the first- be alleviated simply through certain teaching
year course (spring 1992) and the time of their methodologies. Comeau points out in her thesis
anxiety assessment (fall 1992). For some stu- that the Natural Approach (54) which is de-
dents, the strong anxiety reactions they had ex- signed to lessen anxiety in the classroom has
perienced in the first-year class may have been not been proven successful in achieving this
lessened by fall 1992. Therefore, the accuracy of goal. In the study done by Koch and Terrell in
their refall of their anxiety experience cannot 1991 (cited by Comeau), sixty percent of their
be completely guaranteed. subjects with previous classroom language
Thirdly, readers should interpret the results study indicated that they felt more anxious or
of the factor analysis, keeping the following lim- equally anxious under the Natural Approach
itation in mind. The size of the variances for than under other methods. In her own study,
factors, two, three, and four was very small Comeau compared the anxiety levels of two
(6.3%, 5.696, 4.776, respectively), compared to groups of Spanish students: one hundred stu-
that of factor one, speech anxiety and fear of dents attending a school that uses the Natural
evaluation (37.9%).This suggests the possibility Approach and 116 students attending a school
that there was actually no more than one mean- that uses an eclectic/proficiency-based ap-
ingful factor in the present data. This study did proach. There was no significant difference in
not perform a data transformation to reduce the level of anxiety between students in the two
potential skewness in distributions of the items. schools, indicating no special advantage of the
If there were some items with skewed distribu- Natural Approach over other methods in reduc-
tions, the results could have been different. ing the anxiety levels of students. She suggested
Lastly, readers should note that due to the that anxiety interacts with learner variables
correlational nature of this study, the results of such as students’ expectation of grades and
the ANOVA do not prove that a cause-effect their own perception of language ability rather
relationship exists between anxiety and achieve- than with methodology itself.
ment in Japanese. It is possible that some uni- Therefore, future research should look into
dentified variables caused high anxiety and low potential interactions between anxiety and
achievement among students of Japanese. For other student characteristics such as learners’
example, a student’s help-seeking behavior may beliefs about their own language ability, self-
influence both anxiety and achievement. Stu- esteem, help-seeking behaviors, and knowledge
dents who are not comfortable in seeking help and use of language learning strategies. Anx-
from their instructors or teaching assistants ious students may be anxious in the classroom
may experience a high level of anxiety in the because they may not know how to ask ques-
classroom; and their failure to seek help may, in tions to clarify their assignments or how to
turn, result in lower levels of achievement. organize and process information to enhance
The findings of the present study and those their understanding of the material. Some stu-
of other language researchers suggest the im- dents may need assistance from the instructor,
portant role of teachers in lessening classroom but do not ask for help because they might view
tension and in creating a friendly, supportive help-seeking as a manifestation of weakness,
atmosphere that can help reduce students’ fear immaturity, or even incompetence. They might
of embarrassment of making errors in front of feel lost in the language classroom and anxious
peers. Students will appreciate and learn more about the teacher discovering their problem.
from teachers who are able to identify students It is also possible that anxious students may
experiencing foreign language anxiety and be able to handle anxiety-provoking situations
take proper measures to help them overcome if they possess high self-esteem. Greenberg and
that anxiety. In 1990, Appleby reported that stu- his colleagues (20; 21) proposed a terror man-
dents are most irritated by teachers who are un- agement theory, which posits that “people are
empathetic with their needs and who are poor motivated to maintain a positive self-image be-
communicators. Being responsive to the stu- cause self-esteem protects them from anxiety”
Yukie Aida 165
(21: p. 913). According to this theory, people ~

who are high in self-esteem are likely to be less NOTES

anxious; and threats to self-esteem cause anxi-
ety. Horwitz et al. noted that foreign language
learning could pose a threat to learners’ self- The author does not imply that the potentially
esteem because it deprives the learners of their high attrition rate is due solely to language anxiety. It
normal means of communication (since mak- could be influenced by other factors. Seniors would
ing errors are common in the language class- not likely delay their graduation by taking additional
room) and thus of the ability to behave fully as Japanese classes. Many juniors and seniors have to
normal people. Then, people with a sure sense take their major courses and may not have room for
an extra Japanese course. Some students may be in-
of self-worth could manage more effectively the terested in learning Japanese art, culture, or history
threats caused by the language learning en- but not necessarily the language.
vironment than those with low self-esteem. In a A possible explanation for this gender difference
study using a small group (n = 57) of students of in achievement may be the use of different language
Japanese, Aida, Allemand, and Kawashima learning strategies by men and women. In their study
found that students with high anxiety and high involving 1200 college students, Oxford and Nyikos
self-competence received slightly higher final found that females reported more frequent use than
course grades (X = 83.0) and oral skills scores males of three of the five learning strategies studied:
(X= 88.7) than did students with high anxiety formal rule-related practice stsategies, general study
strategies, and conversational input elicitation strate-,
and low self-competence (X = 79.6 for course
gies. On the other hand, males reported no more
grade and X = 86.0 for oral scores), although frequent use than females on any of the five strategy
the differences were not statistically significant. categories. Similar gender differences in the use of
Among students with high anxiety, those with learning strategies were found in Ehrman and Ox-
high self-esteem might be handling their anxi- ford’s study using seventy-eight sophisticated lan-
ety better than those with low self-esteem, re- guage learners as their subjects (e.g., Foreign Service
sulting in their higher scores on both course Officers, military officers, professional language
grades and oral skills grades. Future research trainers, and language instructors). Therefore, it is
employing a larger number of subjects may be possible that females in the present study might have
able to produce a clearer pattern of the rela- used more language learning strategies than males as
did the females in the studies by Oxford and her col-
tionship between self-esteem and anxiety.
leagues. Greater use of learning strategies may have
This study focused on issues pertaining to
positively influenced achievement levels for the fe-
anxiety in Japanese language learning. Since male students in Japanese courses.
the research area of foreign language anxiety is
still young, future investigators have much to
explore. The studies examining the relation-
ship between anxiety and the learner charac- BIBLIOGRAPHY
teristics mentioned above will help us increase
our understanding of language learning from
the learner’s perspective and provide a wider 1. Aida, Yukie, Carolyn Allemand & Hana Kawa-
shima. “The Role ofAnxiety and Social Compe-
range of insights.
tence in Japanese Language Learning.” Paper,
7th International Conference on Pragmatics
and Language Learning, Champaign, IL, April
ACKNOWLEDGMENTS 1993.
2. Alpert, Richard & Ralph N. Haber. “Anxiety in Ac-
ademic Achievement Situations.” Journal of Ab-
I wish to express my sincere appreciation to Dr. normal and Social Psychology 61 (1960): 207-15.
Elaine Horwitz, who read an earlier version of this 3. Appleby, Drew C. “Faculty and Student Perceptions
manuscript and provided me with valuable sugges- of Irritating Behaviors in the College Class-
tions, and to Carolyn Allemand, Hana Kawashima, room.” TheJournal of StaB Program, and Organira-
and Lin Yan Chan, who assisted me in data coding tion Development 8 (1990): 41-46.
and library research. Preparation of this paper was 4.Backman, Nancy. “Two Measures of Affective Fac-
partially supported by a grant from the Northeast tors as They Relate to Progress in Adult Second
Asia Council of the Association for Asian Studies. Language Learning.” WorkingPapen i n Bilingual-
Requests for reprints should be sent to Yukie Aida, ism 10 (1976): 100-22.
Department of Oriental and African Languages and 5. Bailey, Kathleen M. “Competitiveness and Anxiety
Literatures, the University of Texas at Austin, 2601 in Adult L2 Learning: Looking at and through
University Avenue, Austin, TX 78712. the Dairy Studies.” Clawroom Oriented Research in
166 The Modern Language Journal 78 (1994)
Second Language Acquisition. Ed. H. W. Seliger & dence That Self-Esteem Serves an Anxiety Buff-
M. H. Long. Rowley, MA: Newbury House, 1983: ering Function.” Journal of Personality and Social
67-103. Psychology 63 (1992): 913-22.
6. Brod, Richard & Bettina J. Huber. “Foreign Lan- 21. - , Tom Pyszczynski & Sheldon Solomon.
guage Enrollments in United States Institu- “The Causes and Consequences of a Need for
tions of Higher Education, Fall 1990.” ADFL Self-Esteem: A Terror Management Theory.”
Bulletin 23, 3 (1992): 6-10. Public Self and Private Sey Ed. R. F. Baumeister.
7. Chastain, Kenneth. “Affective and Ability Factors New York: Spring-Verlag, 1986: 189-207.
in Second Language Acquisition.” Language 22. Hodapp, V. & A. Henneberger. “Test Anxiety,
Learning 25 (1975): 153-61. Study Habits and Academic Performance.” Ad-
8. Cheek, Jonathan M. & Arnold H. Buss. “Shyness vances in Test Anxiety Research Vol. 2. Ed. H. M.
and Sociability.”Journal of Personality and Social Vander Ploeg, R. Schwarzer & C. D. Spiel-
P~chology41 (1981): 330-39. berger. Hillsdale, NJ: Erlbaum, 1983: 119-28.
9. Comeau, Jennifer L. “Foreign Language Anxiety 23. Horwitz, Elaine K. “Preliminary Evidence for the
and Teaching Methodology.” Master thesis., Reliability and Validity of a Foreign Language
Univ. of Texas at Austin, 1992. Anxiety Scale.” TESOL Quarterly 20 (1986):
10. Cronbach, Lee J. “Coefficient Alpha and the Inter- 559-64.
nal Structure of Tests.” Psychometrika 6 (1951): 24. - . “Foreign Language Classroom Anxiety
297-334. Scale.” Unpublished manuscript, 1983.
11. Culler, Ralph E. & Charles J. Holahan. “Test Anxi- 25. - ,Michael B. Horwitz & Jo Ann Cope. “For-
ety and Academic Performance: The Effects of eign Language Classroom Anxiety.” Language
Study-Related Behaviors.” Journal of Educational Anxiety: From Theory and Research to Classroom Impli-
Psychology 72 (1980): 16-20. cations. Ed. Elaine K. Horwitz & Dolly J. Young.
12. Daly, John A. & Laura Stafford. “Correlates and Englewood Cliffs, NJ: Prentice Hall, 1991: 27-39.
Consequences of Social-Communicative Anxi- 26. Jorden, Eleanor H. & Richard D. Lambert. Japa-
ety.” Avoiding Communication: Shyness, Reticence, nese Language Instruction in the United States: Re-
and Communication Apprehension. Ed. John A. Daly sources, Practice, and Investment Strategy. Washing-
& James C. McCroskey. Beverly Hills, CA: ton: The National Foreign Language Center,
SAGE, 1984: 125-43. 1991.
13. Ehrman, Madeline & Rebecca Oxford. “Effects of 27. Kleinmann, Howard H. “Avoidance Behavior in
Sex Differences, Career Choice, and Psycho- Adult Second Language Acquisition.” Language
logical Type o n Adult Language Learning Learning 27 (1977): 93-107.
Strategies.” Modern Language Journal 73 (1989): 28. Krashen, Stephen D. Principles and Practice in Second
1-13. Language Learning. New York: Pergamon, 1982.
14. Foss, Karen A. & Armeda C. Reitzel. “A Relational 29. Leary, Mark R. Understanding Social Anxiety: Social,
Model for Managing Second Language Anxi- Personality, and Clinical Perspectives. Beverly Hills,
ety.” TESOL Quarterly 22 (1988): 437-54. CA: Sage, 1983.
15. Gardner, Robert C. Social Psychological and Second 30. - ,Paul D. Knight & Kelly A. Johnson. “Social
Language Learning: The Role of Attitudes and Motiva- Anxiety and Dyadic Conversation: A Verbal Re-
tion. London: Edward Arnold, 1985. sponse Analysis.”Journal of Social and Clinical Psy-
16. -& Peter D. MacIntyre. “On the Measure- chology 5 (1987): 34-50.
ment of Affective Variables in Second Lan- 31. Lucas, Jenifer. “Communication Apprehension in
guage Learning.” Language Learning 43 (1993): the ESL Classroom: Getting Our Students to
157-94. Talk.” Foreign LanguageAnnalsl7 (1984):593-98.
17. - , Padric C. Smythe, Richard Clement & 32. McCoy, Ingeborg R. “Means to Overcome the
Louis Gliksman. “Second Language Learning: Anxieties of Second Language Learners.” For-
A Social Psychological Perspective.” Canadian eign Language Annals 12 (1979): 185-89.
Modern Language Review 32 (1976): 198-213. 33. McCroskey, James C. “The Communication Ap-
18. Gliksman, Louis, Robert C. Gardner & Padric C. prehension Perspective.” Avoiding Communica-
Smythe. “The Role of the Integrative Motive on tion: Shyness, Reticence, and Communication Appre-
Students’ Participation in the French Class- hension. Ed. John A. Daly &James C. McCroskey.
room.” Canadian Modern Language Review 38 Beverly Hills, CA: SAGE, 1984: 13-38.
(1982): 625-47. 34. - . “Validityof the PRCA As an Index of Oral
19. Gordon, Edward M. & Seymour B. Sarason. “The Communication Apprehension.” Communication
Relationship Between Test Anxiety and Other Monograph 45 (1978):192-203.
Anxieties.” Journal of Personality 23 (1955): 35. - . “Measures of Communication-Bound
317-2 3. Anxiety.” Speech Monographs 37 (1970): 269-77.
20. Greenberg, Jeff, Sheldon Solomon, Tom Pyszczyn- 36. - ,Joan M. Fayer & Virginia P. Richmond.
ski, Abram Rosenblatt, John Burling, Deborah “Don’t Speak to Me in English: Communica-
Lyon, Linda Simon & Elizabeth Pinel. “Why Do tion Apprehension in Puerto Rico.” Communica-
People Need Self-Esteem? Converging Evi- tion Quarter4 33 (1985): 185-92.
Yukie Aida 16’7
37. MacIntyre, Peter D. & Robert C. Gardner. “Lan- McCroskey. Beverly Hills, CA: SAGE, 1984: 145-
guage Anxiety: Its Relationship to Other Anxi- 55.
eties and to Processing in Native and Second 49. Sarason, Irwin G. “Stress, Anxiety, and Cognitive
Languages.” Language Learning41 (1991): 513-34. Interference: Reactions to Tests.” Journal of Per-
38. -& Robert C. Gardner. “Methods and Re- sonality and Social Psychology 46 (1984): 929-38.
sults in the Study of Anxiety and Language 50. - . Test Anxiety: Theory, Research, and Applications.
Learning: A Review of the Literature.” Language Hillsdale, NJ: Erlbaum, 1980.
Learning 41 (1991): 85-117. 51. - . “The Test Anxiety Scale: Concept and Re-
39. -& Robert C. Gardner. “Anxiety and Sec- search.” Series in Clinical and Community Psychol-
ond Language Learning: Toward a Theoretical ogy. Ed. Charles D. Spielberger & Irwin G. Sar-
Clarification.” Language Learning 39 (1989): ason. Washington: Hemisphere Publishing
251-75. Corporation, 1978: 193-216.
40. McKeachie, Wilbert J., Donald Pollie &Joseph 52. Schumann, John H. “Affective Factor and the
Spiesman. “Relieving Anxiety in Classroom Ex- Problem of Age in Second Language Acquisi-
aminations.” Journal of Abnormal and Social Psy- tion.” Language Learning 25 (1975): 209-35.
chology 50 (1985): 93-98. 53. Spielberger, Charles D. Manual for the State-Trait
41. Measures of Personality and Social Psychological Attitudes. Anxiety Inventory (Form Y). Palo Alto, CA: Con-
Ed. John P. Robinson, Phillip R. Shaver & Law- sulting Psychologists Press, 1983.
rence s. Wrightsman. San Diego, CA: Aca- 54. Terrell, Tracy D. “A Natural Approach to Second-
demic Press, 1991. Language Acquisition and Learning.” Modern
42. Mettler, Sally. “Acculturation, Communication LanguageJournal 61 (1977): 325-36.
Apprehension, and Language Acquisition.” 55. Trylong, Vicki L. “Aptitude, Attitudes, and Anxi-
Community Review 8 (1987): 5-15. ety: A Study of Their Relationships to Achieve-
43. Natale, Michael, Elliot Entin &Joseph Jaffe. “Vo- ment in the Foreign Language Classroom.”
cal Interruptions in Dyadic Communication as Diss., Purdue Univ., 1987.
a Function of Speech and Social Anxiety.”Jour- 56. Tucker, Richard E., Else Hamayan &Fred H. Gen-
nal of Personality and Social Psychology 37 (1979): esee. “Affective, Cognitive, and Social Factors
865-78. in Second Language Acquisition.” Canadian
44. Oxford, Rebecca & Martha Nyikos. “VariablesAf- Modern Language Review 32 (1976): 214-26.
fecting Choice of Language Learning Strate- 57. Twentyman, Craig T. & Richard M. McFall. “Be-
gies by University Students.” Modern Language havioral Training of Social Skills in Shy Males.”
Journal 73 (1989): 291-300. Journal of Consulting and Clinical Psychology 43
45. Pilkonis, Paul A. “The Behavioral Consequences (1975): 384-95.
of Shyness.” Journal of Personality 45 (1977): 58. Watson, David & Ronald Friend. “Measurement
596-611. of Social-Evaluative Anxiety.” Journal of Consult-
46. Powell, Jo Ann C . “Foreign Language Classroom ing and Clinical Psychology 33 (1969): 448-57.
Anxiety: Institutional Response.” Language Anx- 59. Wine, Jeri D. “Test Anxiety and the Direction of
iety: From Theory and Research to Classroom Implica- Attention.” Psychological Bulletin 76 (1971): 92-104.
tions. Ed. Elaine K. Horwitz & Dolly J. Young. 60. Wittmaier, Bruce. “Test Anxiety and Study
Englewood Cliffs, NJ: Prentice Hall, 1991: Habits.” Journal ofEducationa1Research 46 (1972):
169-76. 929-38.
47. Price, Mary L. “The Subjective Experience of For- 61. Young, Dolly J. “Language Anxiety from the For-
eign Language Anxiety Interviews with High eign Language Specialist’s Perspective: Inter-
Anxious Students.” Language Anxiety: From Theory views with Krashen, Omaggio Hadley, Terrell,
and Research to Classroom Implications. Ed. Elaine and Rardin.” Foreign Language Annals 25 (1992):
K. Horwitz & Dolly J. Young. Englewood Cliffs, 157-72.
NJ: Prentice Hall, 1991: 101-8. 62. - . “The Relationship Between Anxiety and
48. Richmond, Virginia P. “Implications of Quiet- Foreign Language Oral Proficiency Ratings.”
ness: Some Facts and Speculations.” Avoiding Foreign Language Annals 19 (1986): 439-45.
Communication: Shyness, Reticence, and Communica- 63. Zimbardo, Philip G. Shyness: What It Is, What to Do
tion Apprehension. Ed. John A. Daly &James C. About It. Reading, MA: Addison-Wesley, 1977.
168 The Modern Language Journal 78 (1994)
Are you taking this course to satisfy the university’s
APPENDIX foreign language requirement?
1 YES 2 NO 3 My major (Asian Studies
or Japanese)
Other reasons:
BACKGROUND QUESTIONNAIRE (please print)
What is your native language?
/
Your Last Name First Name
Are any of your family members of Japanese
heritage?
/
Work phone number 1YES 2N0
Home phone number
If yes, who?
Do any of your family members speak Japanese?
-/-/-/- /
1YES 2N0
age sex single married children (ages)
If yes, who?
ETHNICITY Circle one. Have you been to Japan? 1 YES 2 NO
1 White (not Hispanic) 2 Black (not Hispanic) If yes, how long in total? (Include every occasion
3 Hispanic 4 American Indian or when you were in Japan.)
5 Asian or Pacific Alaskan Native How many Japanese people do you know
Islander (include 6 Other Specify personally?
Asian Americans) How many of them do you consider as your close
What year are you in? friends?
1 Freshman 2 Sophomore 3 Junior 4 Senior How well do you expect to do in this class?
5 Graduate 6 other Specify (Place a check on the line.)
Double major? 1 YES 2 NO Very Very
If yes, please list your majors
~ / ~
well - : - : - : - : - : - : - : - :-poorly
If no, give the name of your single major ~

Errata
THE MLJ APOLOGIZES FOR MISSPELLING PROFESSOR JAVORSKY’S NAME ON HIS ARTICLE
in the last issue. The correct spelling appears in the citation below.

Ganschow, Leonore, Richard L. Sparks, Reed Anderson, James Javorsky, Sue Skinner &Jon Patton,
“Differences in Language Performance among High-, Average-, and Low-Anxious College Foreign
Language Teachers.” MLJ78,l (1994): 41-55.

We would also like to correct the authors of the citation number 51 (page 54). The correct authors
are Sparks, Richard and Leonore Ganschow.

We thank Leonore Ganschow for bringing these errors to our attention