Examination of Horwitz, Horwitz, and Cope's construct of foreign language anxiety: The case of students of Japanese PDF

Preview

Summary

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 S...

Description

Accelerat ing t he world's research.

Examination of Horwitz, Horwitz, and Cope's construct of foreign language anxiety: The case of students of Japanese Amms Kh The Modern Language Journal

Cite this paper

Downloaded from Academia.edu 

Get the citation in MLA, APA, or Chicago styles

Related papers

Download a PDF Pack of t he best relat ed papers 

Effect s of Language Anxiet y on T hree Proficiency-Level Courses of Spanish as a Foreign Lan… Maria Juan-Garau

Foreign language anxiet y in a Spanish universit y set t ing: t he relat ion wit h gender, age and grade Felix Guillen Mult ilingualism and Anxiet y: Evidence from Quadrilingual EFL Learners Elias Bensalem

Examination of Horwitz, Horwitz, and Cope’s Construct of Foreign Language Anxiety: The Case of Students of Japanese

zyxwvutsr zyxwvuts

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 language anxiety is related to Japanese language learning. It uses Horwitz, Horwitz, and Cope’s theoretical model of foreign language anxiety as a research framework. It has been reported that foreign language anxiety is a rather pervasive phenomenon (14; 31; 32; 46; 47; 52). Although language anxiety could be viewed as positive energy (or facilitating anxiety as called by Alpert and Haber) that motivates learners, many language teachers and researchers have been concerned about the possibility that anxiety may function as an affective filter (28), preventing a learner from achieving a high level of proficiency in a foreign language (4; 7; 17; 25; 27; 39; 42; 56; 62). However, most of the research studies have involved Western languages such as French, German, Spanish, and English, and there has been little investigation of non-Western languages like Japanese. In order to develop a fuller understanding of the nature of language anxiety and its implications for language education, future research should include non-Western languages. This study takes a step in that direction. As a Japanese educator, the author became very interested in exploring the role of anxiety in Japanese language learning among college students. Learning Japanese is a very difficult task for Americans. According to Jorden and

Lambert, it requires approximately 1320 hours of instruction in an intensive program in languages like Japanese, Arabic, Chinese, and Korean to bring students to the same level ofproficiency reached after only about 480 hours of instruction in languages like French or Spanish. Therefore, the experiences that students have in the classroom with such difficult languages may be different from the experiences of students in languages that are more similar to English. Do students of Japanese feel anxious in their classrooms? If so, what are the sources of their anxiety? Are there gender differences in language anxiety? Does anxiety interfere with their learning ofJapanese? The present study was designed to answer these questions. Due to the importance of the economic and political relationship between the US and Japan, the number of students interested in learning Japanese has been growing at a rapid pace. According to the results of the fall 1990 survey conducted by the Modern Language Association, 45,117 college students were studyingJapanese in United States institutions of higher education in 1990, representing a spectacular increase of 94.9% from 1986 when 23,454 students were registered in Japanese language courses (6). Japanese became the fifth most commonly taught language in 1990, rising from seventh position in 1986. Therefore, it is important for language educators to identify the variables that may increase or decrease retention and success in Japanese language learning. Language anxiety is one of these important variables.

zyxwvuts

The Modern Language Journal, 78, ii (1994) 0026-7902/94/155-168 $1.50/0 01994 The Modern Lanffuage Journal

zy

156

The Modern LanguageJournal 78 (1994)

EARLY RESEARCH ON FOREIGN LANGUAGE ANXIETY Early research on the role of anxiety in foreign language learning failed to demonstrate any clear-cut relationship between anxiety and a learner’s achievement in a foreign language. For example, Chastain examined the relationships between anxiety and course grades in three language programs: French (audiolingual or regular), German, and Spanish. While there was a significant negative correlation found between course grades and test anxiety in the French audiolingual class, students in the regular French, German, and Spanish classes who experienced a higher level of anxiety were more likely to receive better grades than students with a lower level of anxiety. Backman looked at the relationship between anxiety and language progress among Venezuelan students learning English in the US. Students’ progress measured by a placement test, a listening comprehension test, and teachers’ ratings did not show a significant correlation with any of the anxiety measures. In Kleinmann’s 19’7’7 study of Spanish-speaking and Arabic-speaking ESL students, facilitating anxiety was found to be correlated with students’ oral production of linguistically difficult (thus challenging) English structures (e.g., infinitive complements and passive sentences). However, there was no evidence that debilitating anxiety negatively influenced oral performance. The facilitating and debilitating effects of anxiety were also observed by Bailey through her review of students’ diaries. Young (62) conducted a study to test whether oral proficiency was negatively influenced by anxiety in three languages, i.e., French, German, and Spanish. She found some negative correlations between students’ OPI (Oral Proficiency Interview) scores and some of the anxiety measures. However, when language ability measured by a dictation test and a self-appraisal measure of foreign language oral proficiency was controlled statistically (i.e., the variability due to language ability was removed from the relationship between anxiety and oral performance), the correlations between anxiety measures and OPI scores were nonsignificant. Such results are very predictable since language ability is likely to correlate with language achievement. When language ability is held constant as was done in Young’s (62) study, there is little left in the OPI scores to covary with anxiety. However, these nonsignificant correlations obtained through the above procedure cannot warrant

that anxiety is not associated with achievement, because we do not know whether the anxiety measures used in Young’s study could accurately capture students’ anxiety levels in oral production. HORWITZ, HORWITZ, AND COPE’S CONSTRUCT OF LANGUAGE ANXIETY

Horwitz (24) and Horwitz et al. have attributed the inconclusive results of previous research to the lack of a reliable and-valid measure of anxiety specific to language learning. They conceptualize foreign language anxiety as “a distinct complex of self-perceptions, beliefs, feelings, and behaviors related to classroom language learning arising from the uniqueness of the language learning processes” (25: p. 31). The Foreign Language Classroom Anxiety Scale (FLCAS, hereafter) was developed by Horwitz (24) in order to capture this specific anxiety reaction of a learner to a foreign language learning setting. Horwitz et al. integrated three related anxieties to their conceptualization of foreign language anxiety, i.e., communication apprehension (35), test anxiety (19; 50), and fear of negative evaluation (58). According to McCroskey (34), communication apprehension is defined as a person’s level of fear or anxiety associated with either real or anticipated communication with another person or persons. McCroskey (33) points out that typical behavior patterns of communicatively apprehensive people are communication avoidance and communication withdrawal. Compared to nonapprehensive people, communicatively apprehensive people are more reluctant to get involved in conversations with others and to seek social interactions. The extensive body of research in this area, summarized by Daly and Stafford and by Richmond, supports McCroskey’s claim. In 1985 McCroskey, Fayer, and Richmond studied the relationships between communication apprehension and self-perceived competence in Spanish and English among Spanish-speaking Puerto Rican college students who had received instruction in English. They found that students with low self-ratings of competency in English were more likely to report higher levels of English communication apprehension. On the other hand, there was no such correlation found between self-perceived competence in the native language, i.e., Spanish and Spanish communication apprehension. Similarly, Foss and Reitzel and Lucas report that communication anxiety exists among students

zyx

zyxwvu

Yukie Aida

zyxwvutsrq

in the ESL classroom; it seems to function as a block for students’ mastery of English. It is very likely that people experience anxiety and reluctance in communicating with other people or in expressing themselves in a foreign language in which they do not have full competence. The second element of foreign language anxiety, test anxiety, is defined by Sarason (51) as “the tendency to view with alarm the consequences of inadequate performance in an evaluative situation” (p. 214). Students worry about failing to perform well. Culler and Holahan and other researchers (22; 60) speculate that test anxiety may be caused by deficits in students’ learning or study skills. Some students experience anxiety during a test situation because they do not know how to process or organize the course material and information. Since daily evaluations of skills in foreign language classrooms are quite common, and making mistakes is a normal phenomenon, students may suffer stress and anxiety frequently, which may pose a problem for their performance and future improvement. Other researchers posit that test anxiety occurs when students who have performed poorly in the past develop negative, irrelevant thoughts during evaluation situations (40; 49; 59). Test-nervous students may not be able to focus on what is going on in the classroom because they tend to divide their attention between self-awareness of their fears and worries and class activities themselves. They may say to themselves, “I’ll never be able to pronounce it correctly,” “The teacher is ready to correct me,” or “Other students will laugh at me if I speak.” They become distracted and anxious during class, which interferes with their performance. Lastly, fear of negative evaluation is defined as “apprehension about others’ evaluations, distress over their negative evaluations, and the expectation that others would evaluate oneself negatively” (58: p. 449). Research shows that people who are highly concerned about the impressions others are forming of them tend to behave in ways that minimize the possibility of unfavorable evaluations. They are more likely to avoid or prematurely leave social situations in which they believe others might perceive them unfavorably (29; 57; 58; 63). When they affiliate with others, they often fail to initiate conversations or participate only minimally in the conversation, as by just smiling and politely nodding, o r listening to others talk and only interacting with occasional “uh-huh’s” (8; 30; 43; 45). When this notion of fear of negative

157

evaluation is applied to foreign language learners, we can easily imagine that students with fear of negative evaluation sit passively in the classroom, withdrawing from classroom activities that could otherwise enhance their improvement of the language skills. In extreme cases, students may think of cutting class to avoid anxiety situations, causing them to be left behind. Horwitz et al. believe that these three anxieties, i.e., communication apprehension, test anxiety, and fear of negative evaluation, are important parts of foreign language anxiety and have an adverse effect on students’ language learning. Horwitz (23) reported that the FLCAS had a correlation coefficient of .28 (p = .063, n = 44) with communication apprehension (measured by McCroskey’sPersonal Report of Communication Apprehension, 35), .53 (p < .01, n = 60) with test anxiety (measured by Sarason’s Test Anxiety Scale, 51), and .36 (p < .01, n = 56) with fear of negative evaluation (measured by Watson and Friend’s Fear of Negative Evaluation Scale). The FLCAS also correlated with final grades: r = -.49, p < .01 (n = 35) for two beginning Spanish classes and r = -.54, p < .01 (n = 32) for two beginning French classes. Higher FLCAS scores were associated with lower final grades. Price also reported in her dissertation that the FLCAS scores of 106 students of second-semester French classes were positively correlated with test anxiety (r = .58, p < .001) and public speaking anxiety (r = .43, p < .001). The FLCAS scores also correlated negatively with final grades (r = -.22, p < .05), final exam scores (r = -.29, p < .Ol), and oral exam scores (r = -.27, p < .05). However, when students’ Modern Language Aptitude Test scores were controlled, only the correlation between the oral exam scores and the FLCAS scores remained significant. The main purpose of this study was to test Horwitz et al.’s construct of foreign language anxiety by validating an adapted FLCAS for students ofJapanese. It was an exploratory study to discover the underlying structure of the FLCAS and to examine whether or not the structure reflects the three kinds of anxiety presented earlier. It also assessed the instrument’s reliability and the relationship of students’ anxiety levels to their performance in Japanese. It was hoped that the results of this empirical study using a non-Western language would shed new light on the concept of foreign language anxiety and would expand its scope and implications.

zy

zyxw

zy zyxwvutsr The Modern LanguageJournal 78 (1994)

158 METHOD

Subjects. In the fall of 1992, students who were enrolled in second-year Japanese I at the University of Texas at Austin were asked to participate in this study. Ninety-six students (fifty-six males and forty females) completed the questionnaires designed for this study. There were more than ninety-six students enrolled in the course, but some students failed to complete the questionnaires or to pass the course. Three students did not pass the course because they failed to attend class regularly and to complete several important exams and/or assignments (e.g., lesson quizzes, essay writing, oral presentation). Therefore, only data obtained from these ninety-six students were used for analysis. The mean age of this sample was 21.5 years. There were sixty-four native speakers of English and thirty-two non-native speakers of English (i.e., five Spanish speakers, six Chinese, fourteen Korean, five other Asian language speakers, and two other non-Asian language speakers). When the native speakers of English and nonnative speakers of English were compared on the level of anxiety (see the Proceduressection for how to obtain a subject’s anxiety score), a oneway ANOVA showed that there was no significant difference between the two groups: F(1, 94) = .O?, p = .79, X = 96.2 for native speakers of English and X = 95.5 for non-native speakers of English). In addition, a Bartlett-Box F test for homogeneity of the variance indicated that the data of the present study satisfied the assumption of equal variances: F = 1.21, p = .27. Therefore, the two groups were treated as one sample. Procedures. On the very first day of the fall semester 1992, subjects were asked to complete both a FLCAS and background questionnaire (see Appendix). In this study, the term “foreign language” used in the original FLCAS was replaced with “Japanese language.” In responding to the statements on the FLCAS, subjects were asked to consider their experiences in the previous year’s first-year Japanese course. Therefore, students’ FLCAS scores reflect their anxiety in the first-year Japanese classroom. The instructions read as follows: “In this section, we would like you to respond to each of the following statements based upon your experience in your last yeark Japanese course (JPN507).” Instruments. The FLCAS contains thirty-three items, each of which is answered on a five-point Likert scale, ranging from (a) “strongly dis-

agree”, to (c) “neither agree nor disagree”, to (e) “strongly agree”. A student’s endorsement in (a) “strongly disagree” was equated with a numerical value of one; (b) “disagree” was two; (c) “neither agree nor disagree,”three; (d) “agree” four; and, (e) “strongly agree” was five. For each subject, an anxiety score was derived by summing his or her ratings of the thirtythree items. When statements of the FLCAS were negatively worded, responses were reversed and recoded, so that in all instances, a high score represented high anxiety in the Japanese classroom. The theoretical range of this scale was from thirty-three to 165. The background questionnaire included questions on the student’s age, sex, ethnicity, academic major and status, native language, reasons why he or she was taking a Japanese course, whether or not he or she had been to Japan and for how long, whether o r n o t he or she was pleased with the final course grade given for the second-semester Japanese class, and whether or not he or she had other family members who speak Japanese. Instructors provided subjects’ final course grades (in percentages) for the second-semester Japanese classes. The final course grade was selected primarily because it had been used as a global measure of language proficiency by many researchers (e.g., 7; 9; 18; 25; 55).

zyxwvuts zyxwvu zyxwv RESULTS

Reliability of the FLCAS. The present study, using ninety-six students of Japanese, yielded internal consistency of .94 (X= 96.7 and s.d. = 224, using Cronbach’s alpha coefficient. As shown in Table I, the reliability, mean, standard deviation, and range obtained in this study were very similar to those of Horwitz (23), who used students enrolled in an introductory Spanish class. The mean of this study, 96.7, was slightly higher than that of Horwitz’s (23) study, X = 94.5. It is understandable that students may feel more anxious in learning a non-Western, foreign language like Japanese (26) than in learning commonly taught Western languages such as Spanish. There was no significant gender difference found in language anxiety: t(94) = .41, p = .69. The mean scores for males (n = 56) and females (n = 40) were 97.4 and 95.6, respectively. The results of this study suggest that the FLCAS is a reliable tool regardless of whether the language is a European Western language. On the first day of the next semester (spring

zyxwvu

zyxwvutsrqpo zyxwvutsrq

Yukie Aida

159

TABLE I Reliabilities of The FLCAS in Two Studies ~~~~

Sample size Students status Language Cronbach’s alpha Range Mean Standard deviation Test-retest reliability

Present study

Horwitz et al., 1991

96 first year Japanese .94 47-146 96.7 22.1

108 first year Spanish .93 45-147 94.5 21.4

nalities, and percent of the variance are shown in Table 111. The solution accounted for 54.5% of the total variance. Eighteen items were loaded on the first factor, accounting for 37.9% of the variance. Examples of the items included in this factor are item 3, “I tremble when I lcnow that I’m going to be called on in my Japanese class,” and item 13, “It embarrasses me to volunteer answers in my Japanese class.” The factor one was assigned a label of Speech Anxiety and Fear of Negative Evaluation. The items included in this factor indicate a student’s apprehension in speaking in a Japanese class and fear of embarrassment in making errors in front of other stud...