技術學院應外系學生英文寫作焦慮量表之編製

全文

(1)81. 國立政治大學「教育與心理研究」 2008 年 9 月，31 卷 3 期，頁 81-107. 技術學院應外系學生英文寫作焦慮量表之編製蔡漢民*. 摘. 要. 本研究旨在發展一份英文寫作焦慮量表以瞭解臺灣技術學院應外系學生英文寫作焦慮的來源。研究工具包括一份開放式的問卷及五份經過適度改寫與翻譯，分屬不同領域的焦慮量表。開放式問卷的目的是要得知學生在英文寫作時產生的焦慮經驗。這些經驗所形成的主題便成了編製本研究所需量表的基礎。完成後的量表與前述五種分屬不同領域的焦慮量表則針對八所技術學院應外系共338位學生作測試。之後針對此量表作各種效度與信度檢測，最後則進行驗證性因素分析以評估探索性因素結構的適度性。研究結果顯示本研究所發展的英文寫作焦慮量表具有適度的計量特質。各種測試顯示四個因素的結構為本量表最佳的模式，分別是：(1)寫作考試的焦慮；(2)用字及文法結構上犯錯的焦慮；(3)擔心負面的評價；(4)英文寫作信心低落。驗證性因素分析進一步證明此模式的適度性。. 關鍵詞：英文寫作焦慮、量表、技術學院、應外系. *. 蔡漢民：崇右技術學院應用外語系副教授電子郵件：[email protected]. 收件日期：2007.8.28；修改日期：2007.11.13；接受日期：2007.12.12.

(2) 82. Journal of Education & Psychology September, 2008, Vol. 31 No. 3, pp. 81-107. The Development of an English Writing Anxiety Scale for Institute of Technology English Majors Han-Min Tsai* Abstract. This study aimed to develop an English writing anxiety scale to map out Taiwanese institute of technology English majors’ anxiety in English writing. The research tools included an open-ended questionnaire and five adapted and translated scales related to anxiety in different fields. The questionnaire attempted to elicit students’ anxietyprovoking experiences in English writing. The themes emerging made up the bases for developing the initial constructs of the scale. The resultant scale and the five translated scales were then given to 338 English majors in eight institutes of technology. Different types of validity and reliability were examined and confirmatory factor analysis was finally conducted to evaluate the fitness of the exploratory factor structure. The results show that the scale has adequate psychometric properties. Four factors were identified, namely, Fear of Writing Tests, Anxiety about Making Mistakes, Fear of Negative Evaluation, and Low Confidence in English Writing. Confirmatory factor analysis further substantiated the fitness of such a factor model.. Keywords: English writing anxiety, scale, institutes of technology, English majors *. Han-Min Tsai: Instructor, Department of Applied Foreign Languages, Chungyu Institute of Technology E-mail: [email protected]. Manuscript recieved: 2007.8.28; Revised: 2007.11.13; Accepted: 2007.12.12.

(3) 技術學院應外系學生英文寫作焦慮量表之編製. Introduction. 83. learners’ discomfort. This can then affect learners in several facets, including the. The existence and significance of. composing process, writing performance,. writing anxiety started to draw research-. academic achievement, and even occupa-. ers’ attention and interest in the 1970s.. tion choices.. The term writing apprehension or writing anxiety was first used by Daly and Miller (1975) when they developed an instru-. Potential Sources of FL Writing Anxiety. ment to measure writing anxiety. It was. FL writing apprehension includes. defined as an individual’s predisposition. complex, multi-dimensional factors. To. to withdraw from or hold unfavorable re-. gain a holistic understanding of its causes,. actions towards writing situations because. linguistic. of the accumulation of unrewarding,. socio-psychological phenomena, learner/. nervous experiences (Daly, 1978). Daly. instructor beliefs, classroom practice, and. and Miller’s Writing Apprehension Test. even cultural context should be taken into. (WAT), originally intended to discern first. account.. abilities,. cognitive. factors,. language writing apprehension, has been. Linguistic abilities: Language capa-. adapted for second or Foreign Language. bility may not be a major contributor to. use and research. The results have shown. native language writing anxiety, but it is a. the anxiety-provoking nature of foreign. striking source of anxiety to second or. language (FL) writing and the need to. foreign language student writers. Leki’s. view FL writing apprehension as distin-. (1999) review of writing anxiety research. guishable from other forms of anxiety. shows that FL learners’ expression of. (Cheng, Horwitz, & Schallert, 1999).. thought is often destroyed by their limited. Writing is difficult, as Byrne (1982). language knowledge and capacity, which. and Hughey, Wormuth, Hartfiel and Ja-. inevitably induces their writing apprehen-. cobs (1983) point out, because it involves. sion. Such mismatch between mature. complex psychological, linguistic and. thought and immature foreign language. cognitive factors. All of these indicate. proficiency was also proposed by Hor-. that writing requires demanding con-. witz, Horwitz and Cope (1986) when they. scious mental work and can easily evoke. tried to conceptualize foreign language.

(4) 84 教育與心理研究 31 卷 3 期. classroom anxiety. Sparks, Ganschow and. that writing is mostly formed in isolation. Javorsky (2000) argue that it is learners’. and the writer lacks the immediate feed-. linguistic deficit that results in poor per-. back and opportunity to rephrase or clar-. formance, which in turn provokes their. ify her/his statements once the product is. anxiety.. in the reader’s hand (Hughey et al., 1983). to. Such psychological helplessness makes. Hughey et al. (1983), cognition in writing. FL students inclined to feel threatened. refers to not only written forms and cer-. when their written texts have to be evalu-. tain syntax but also idea formulation and. ated, because what is in the texts might. organization. All of these have to be. not completely reflect what they think. learned through formal instruction. How-. (Raisman, 1982).. Cognitive. factors:. According. ever, first language interference (Hughey. The interactive relationship between. et al., 1983) and learners’ unwillingness. competitiveness and self-esteem can be. to take writing risks (Leki, 1999) can hin-. another source of many FL students’ writ-. der the development of such knowledge.. ing anxiety where learners compare them-. This often leads to or reinforces learners’. selves to others or to the idealized self-. writing apprehension. FL students’ unfa-. image. This might even cause, as Johan-. vorable prior experiences can also be a. son (2001) notes, English majors to adopt. cognitive factor contributing to writing. avoidance strategies in writing- associated. anxiety (Aldrich, 1982), which might. activities such as an on-line writing com-. produce a vicious cycle whereby learners. munity.. tend to avoid writing practice and thus achieve more frustrating results.. In addition to these issues, the clash between FL writers’ first language and. Socio-psychological issues: Such is-. target language culture, even though re-. sues include personal and interpersonal. search on this aspect is still scanty, can. components such as self-esteem, competi-. also be a cause of writing anxiety. For ex-. tiveness, social interactions and cultural. ample, Kaplan (1966) argues that in con-. identity. Writing is anxiety-provoking, as. trast to the directness of Western culture,. MacIntyre, Noels and Clément (1997) as-. Chinese writers tend to be indirect in their. sert, because it requires one to communi-. patterns of idea organization. This makes. cate with others. However, the dilemma is. it difficult, according to Scollon (Connor,.

(5) 技術學院應外系學生英文寫作焦慮量表之編製. 85. 1991), for Chinese students to express. ment, as the writing teacher Julie identi-. their purpose directly in the beginning. fied in Tsui’s (1996) study, can be an ad-. paragraph. What they write, therefore,. ditional cause of anxiety in a FL writing. might seem unconnected in the eyes of. class. What materials the teacher chooses,. Western readers. Such negative influence. and how s/he implements activities, man-. of first language rhetorical patterns over. ages the class tempo and reacts to stu-. students’ writing habits in FL becomes. dents’ written products will affect not. stronger when learners keep gaining un-. only instructor-learner interactions but. rewarding feedback from their writing.. also learners’ willingness to take risks.. Learner/instructor beliefs about writ-. For example, teachers’ undue focus on. ing: Learners’ conceptions about language. writing errors might make the writing en-. learning can create anxiety (Young,. vironment extremely discouraging and. 1991), especially when learners’ unrealis-. even cause, as Semke (1984) notes, some. tic beliefs and reality clash. In FL writing,. FL students to respond with hostility.. for example, students’ preoccupation with. Writing under time pressure and as a. producing grammatically accurate writing. lonely activity, according to Tsui, can also. is an unrealistic belief, which will not. debilitate the writing environment, mak-. only inhibit and block students’ writing. ing students feel that writing is difficult. (Johanson, 2001; Tsui, 1996) but aggra-. and boring. Testing aspects, such as un-. vate their writing anxiety when they find. familiar, ambiguous tasks and formats,. their mistakes exposed to others (Johan-. and highly evaluative situations (Young,. son, 2001). FL writing instructor beliefs. 1991), are likely to make language learn-. can also contribute to students’ anxiety.. ing environments stressful.. For instance, the teacher’s assumption about her/his role in a FL writing class and how s/he will treat learners’ errors, if. Instruments for Measuring Writing Anxiety. not corresponding to learners’ needs or. The most commonly used instru-. expectations, might give rise to learners’. ments for measuring writing anxiety-. anxiety.. related phenomena are Daly and Miller’s. Writing environments: An unsafe,. (1975) Writing Apprehension Test (WAT),. unfriendly, and unsupportive environ-. Rose’s (1984) Writer’s Block Question-.

(6) 86 教育與心理研究 31 卷 3 期. naire (WBQ), and Riffe and Stacks’. apprehension is multidimensional, Riffe. (1988) Mass Communication Writing Ap-. and. prehension Measure (MCWAM).. MCWAM. The 43-item MCWAM has sat-. Stacks. (1988). developed. the. The 26-item WAT mainly deals with. isfactory reliability and comprises eight. students’ aversion to writing, and attitudes. factors: general affect, blank page paraly-. as well as feelings towards writing and. sis, mechanical skill competence, career. evaluation. It has demonstrated reason-. and essential skill, evaluation apprehen-. able reliability and validity over time. sion, task avoidance, facts versus ideas,. (Bennett & Rhodes, 1988) and has in-. and audience salience.. spired numerous empirical studies on. Although the WAT has been widely. writing anxiety and its relationships with. used, its limitations lie in the fact that it. writing performance, willingness to take. was originally developed for first lan-. writing courses, and career choices. Be-. guage learners and that its construct valid-. sides, the WAT has been adapted for use. ity as a pure measure of writing anxiety. in certain FL writing contexts.. has been questioned (Mckain, 1991, dis-. Although writer’s block and writing. cussed in Cheng, 2004). Richmond and. apprehension are not the same thing, they. Dickson-Markman (1985) even question. are interconnected (Rose, 1984). The 24-. the construct validity of the WAT and. item WBQ provides more detailed depic-. suggest that it is a measure of self-. tion of writers’ response by focusing on. confidence in writing ability rather than. five cognitive subscales: attitude towards. of anxiety about writing. The other two. writing and evaluation, complexity of ma-. measures, while providing more dimen-. terials, premature editing, procrastination. sions in writing problems, were not spe-. in completing tasks, and inability to write. cifically intended for tapping FL learners’. (Phinney, 1991). Rose’s scale has proven. writing anxiety, either. Rose’s (1984) 24-. valid for English and Spanish bilingual. item WBQ attempted to measure factors. student populations and can be used to. associated with writer’s block, which, ac-. produce five scores to indicate each sub-. cording to Rose, is distinguished from. scale (Betancourt & Phinney, 1988, dis-. writing. cussed in Phinney, 1991).. MCWAM aimed to discriminate writing. Based on the assumption that writing. apprehension.. The. 43-item. attitudes and apprehension between mass.

(7) 技術學院應外系學生英文寫作焦慮量表之編製. 87. communication and non-mass communi-. tioned. The primary purpose of this study,. cation students, as well as among mass. therefore, was to develop an English writ-. communication students (Riffe & Stacks,. ing anxiety scale to map out the causes of. 1988). In light of the limitations of the. anxiety experienced by Taiwanese Eng-. current writing anxiety measures in use,. lish majors at institutes of technology.. to locate accurately sources of FL writing. The scale would go through different. apprehension, it is needed to design a. stages of psychometric examination re-. scale based on the understanding of the. garding the construct and criterion-related. FL context concerned and the items. validity. The exploratory factor model of. should represent the anxiety about FL. the scale would further be assessed in. writing.. structure fitness through confirmatory. Purpose of the Present Study Research in EFL (English as a For-. factor analysis. The following research questions were explored in this study: 1.What are the results of the prestudy open-ended questionnaire?. eign Language) writing anxiety in Taiwan. 2.What are the psychometric proper-. is still at a fledging stage and mostly fo-. ties of the English writing anxiety scale. cuses on the English majors of traditional. developed?. four-year universities. Little has been. 3.Does. the. confirmatory. factor. done on the students of institutes of tech-. analysis substantiate the structure fitness. nology. The main instruments employed. of the English writing anxiety scale de-. are the revised and translated versions of. veloped?. Daly and Miller’s (1975) WAT and Rose’s (1984) WBQ because reasonable reliability and validity have been shown in the. Method Participants. literature (e.g. Bennett & Rhodes, 1988;. Participants in this study consisted of. Cheng, 2004; Phinney, 1991). However,. 338 English majors from eight institutes. since these two instruments were not spe-. of technology in different parts of Taiwan.. cifically devised for tapping FL writers’. There were 37 male students (10.9%) and. anxiety, their construct validity for meas-. 301 female students (89.1%). 184 stu-. uring FL writing anxiety has been ques-. dents were in the first year of two-year in-.

(8) 88 教育與心理研究 31 卷 3 期. stitute of technology, and 154 students in. of the scale and a potential pool of items. the third year of four-year institute of. for the various constructs. Based on these. technology. All the participants had either. data, a preliminary 33-item English scale. completed or were taking an English. was devised on a 5-point Likert scale. composition course. The mean age for the. format, ranging from 5 points (strongly. sample was 21.52 (SD= 1.02). In addition,. agree) to 1 point (strongly disagree).. 14 teachers, of whom 8 were teaching. Each item of the initial scale was then. English and 6 teaching Chinese, from six. translated into Chinese. To make the. colleges were invited to check the English. wording clear and easy to read, six Chi-. and Chinese versions of the scale.. nese teachers and six English teachers. Procedure. were invited to read the Chinese and English versions, respectively. The English-. Prior to devising the scale, a Chinese. Chinese version was finally checked by. open-ended questionnaire (see Appendix. two English writing teachers experienced. A) on English writing anxiety was admin-. in translation.. istered to 114 institute of technology Eng-. To test the wording of the 33-item. lish majors. Data collected were subject. scale, the Chinese version was adminis-. to analyses mainly through the procedure. tered to 95 college English majors. After. proposed by Holliday (2002): look at the. certain modifications, the scale was then. overall character of the corpus of data→. piloted on another group of 107 institute. search for themes recurrently emerging→. of technology English majors. Item analy-. determine the character of each theme→. ses were subsequently carried out in SPSS. label the themes→select details to sub-. and five items were removed. The final-. stantiate each theme. Data sorted out were. ized 28-item Chinese version (see Appen-. then translated into English. Statistics on. dix B for the English and Chinese ver-. frequency were used where appropriate.. sions) was then administered to a sample. Related previous research on language. of 384 English majors enrolled in eight. learning anxiety was also reviewed.. institutes of technology. Attached to the. The resultant open-ended question-. scale were background items asking for. naire and literature review made up the. information about the participants’ (a). bases for developing the initial constructs. gender, age, and year in college, and (b).

(9) 技術學院應外系學生英文寫作焦慮量表之編製. 89. self- rating proficiency in English writing. Horwitz and Garza’s (1999) Foreign Lan-. on a 5-point basis, ranging from 5 points. guage Reading Anxiety Scale (FLRAS),. (very good) to 1 point (very poor).. Watson and Friend’s (1969) Fear of Nega-. The data collected were subse-. tive Evaluation (FNE), and Cohen and. quently checked and 46 of the sample. Waugh’s (1989) Computer Anxiety Scale. were invalid. Afterward, reliability, ex-. (CAS).. ploratory factor analysis and different. The WAT has been widely employed. types of validity of the scale were exam-. to explore the constructs of writing ap-. ined. extraction. prehension. The scale has demonstrated. method of principal axis factoring and. satisfactory reliability and validity over. oblimin rotation was used for factor. time (Bennett & Rhodes, 1988; Cheng,. analysis. To validate the hypothesized fac-. 1999; Reaves, Flowers, & Jewell, 1993).. tor structure, confirmatory factor analysis. The FLCAS aims to investigate potential. was subsequently conducted by using the. sources of anxiety in foreign language. computer program known as LISREL. classrooms. It consists of 33 statements,. (Joreskog & Sorbom, 1984). At the end of. and good internal consistency, test-retest. the semester, with the consent of the par-. reliability and construct validity have. ticipants, 147 of the 338 participants’ final. been shown by Horwitz et al. (1986),. course grades in English composition. Horwitz (1986) and Aida (1994). It also. were obtained to assess the predictive va-. reveals high item-total correlations (Sell-. lidity of the scale.. ers, 2000). The FLRAS is a 20-item scale. through. SPSS.. The. Instruments. measuring anxiety responding to foreign language reading. Adequate internal con-. For the present study, five instru-. sistency, concurrent and discriminant va-. ments were used to examine the conver-. lidity, and criterion-related validity have. gent and discriminant validity of the de-. been substantiated by its authors (Saito et. veloped English Writing Anxiety Scale. al., 1999). The FNE is a widely used 30-. (EWAS), including Daly and Miller’s. item scale with a view to examining the. (1975) Writing Apprehension Test (WAT),. degree to which people experience appre-. Horwitz et al.’s (1986) Foreign Language. hension when negatively evaluated, but. Classroom Anxiety Scale (FLCAS), Saito,. two items were deleted in the present.

(10) 90 教育與心理研究 31 卷 3 期. study. Its psychometric properties have. wan, the foreign language in WAT,. been validated by its authors (Watson &. FLRAS, and FLCAS was specified as. Friend, 1969). The CAS is a 16-item scale. “English”. The evaluator in WAT was re-. that measures the amount of computer. defined as “the teacher”, and the people. anxiety. It has a high internal consistency. and friends as “my classmates”. For ex-. (Cohen & Waugh, 1989) and has been. ample, the original item in WAT “I like to. used to measure decreased anxiety and. have my friends read what I have written”. improved understanding of the computer. was modified as “I like to have my class-. in terms of efficacy and communication. mates read what I have written in Eng-. (Thomas, 2002).. lish”.. The original versions of these in-. To make each statement clear and. struments were adapted through the fol-. easy to read, the second person pronouns. lowing stages: item selection, item modi-. were replaced by the first person pro-. fication, translation, translation checking,. nouns. After this, the instruments adapted. and pilot-testing. Afterwards, the reliabil-. were translated into Chinese and then. ity of the translated versions was as-. checked.. sessed. At first, the items (e.g. “I enjoy. Chinese version was re-checked. The next. writing”) judged to be very similar to oth-. stage involved pilot-testing the Chinese. ers (e.g. “Writing is a lot of fun”), or be-. versions to 95 college students to examine. yond most participants’ experiences (e.g.. the wording of each statement. Finally,. “I would enjoy submitting my writing to. the Chinese versions were administered to. magazines for evaluation and publica-. 116 college students to assess the reliabil-. tion”) were deleted. The next stage aimed. ity of these versions. All of the instru-. to modify complicated or ambiguous. ments adopted were answered on a 5-. items; for example, “I don’t mind reading. point Likert scale format, ranging from 5. English to myself, but I feel very uncom-. points. fortable when I have to read English. (strongly disagree). The items positively. aloud” was transformed to “I feel very. worded were reverse scored.. uncomfortable when I have to read English aloud”. Also, in order to suit the foreign language learning situations in Tai-. Subsequently,. (strongly. the. agree). to. English-. 1. point. Criteria for Selecting Items and Factors.

(11) 技術學院應外系學生英文寫作焦慮量表之編製. The selection of items for the EWAS. organize. ideas,. develop. 91. sufficient. was based on factor analyses. First, a. supporting details, and achieve unity and. minimum variable loading of 0.4 or. coherence; (2) worry about linguistic. greater was used for retaining an item.. errors; (3) fear of being evaluated, which. Second, following the suggestion of. might arise from experiencing too many. Floyd and Widaman (1995), to increase. negative reactions from their teacher or. the factor analytic validity of the scale,. classmates; (4) low confidence in writing,. only items with clear factor loading pat-. which might stem from students’ unpleas-. terns were selected. Items which cross-. ant past experiences in writing a piece of. loaded nearly equally on two or more fac-. work; (5) fear of revising written work,. tors were dropped.. which is concerned with students’ inabil-. Different criteria have been ad-. ity to deal with the teacher’s written. vanced to determine the number of fac-. comments or discern their own problems;. tors. The ones most frequently adopted. and (6) testing situations, which are asso-. are Kaiser’s eigenvalue criterion and Cat-. ciated with fear of low grades and pres-. tell’s scree plot (Dunteman, 1989). For. sure of time either in class writing prac-. the current study, the eigenvalue >1 and. tice or in exam writing.. scree test criteria were applied as guidelines for selecting the numbers of factors. In addition, each factor was expected to contain at least three interpretable variables.. Results Findings from the Openended Questionnaire Table 1 shows that Taiwanese institute of technology English majors’ writing. Psychometric Properties of the English Writing Anxiety Scale (EWAS) After item analysis, the psychometric properties of the resultant 28-item scale was examined and LISREL was finally conducted for confirmatory factor analysis.. Exploratory factor analyses. anxiety is mainly due to the following. The 28-item scale was subjected to a. causes: (1) global difficulties, which are. principal axis factoring extraction and an. tied to students’ inability to explore ideas,. oblique rotation with Kaiser normaliza.

(12) 92 教育與心理研究 31 卷 3 期. Table 1. Students’ Anxiety Sources in the Order of Frequency (N = 114). Anxiety sources 1.Global difficulties. 2.Local errors 3.Fear of being evaluated 4.Low confidence in writing 5.Fear of low grades 6.Fear of revising written work 7.Pressure of time. Details Unable to obtain sufficient ideas for a topic, not knowing how to start or organize an essay, unable to write sufficient supporting details for an idea, fear of digression, unable to deal with difficult and unfamiliar topics Insufficient knowledge of and fear of making errors in word usage, grammar and sentence structure Fear of their written work being evaluated by the teacher or read by classmates Fear of falling short of the teacher’s expectations, worry about the comprehensibility of their writing, no confidence in writing Worry about low grades in writing and writing exams Unable to make sense of the teacher’s error indicators or written comments, not knowing how to revise a piece of written work Time pressure from class writing and exam writing. tion. The estimate of Kaiser-Meyer-Olkin. accounted for 62.8% of total cumulative. measure of sampling adequacy (KMO =. variance. The eigenvalues for Factors I,. 0.932) was more than the suggested 0.6. II, III, and Ⅳ were 6.90, 1.75, 1.44, and. value (Tabachnik & Fidell, 2001), indicat-. 1.21, respectively. The factor pattern ma-. ing the sample collected was adequate for. trix for the final set of items is presented. factor analyses. Five factors with eigen-. in Table 3.. values > 1 were identified. However, the. The first factor consisted of 6 items. fifth factor was cut out because it in-. (Items 4, 6, 10, 12, 14, and 17) that pri-. cluded only two items with loadings. marily addressed students’ worries in rela-. greater than 0.4. Besides, scree test. tion to testing situations, including the po-. showed a four-factor model best fit. After. tential effects of the test itself, time pres-. removing Factor 5 and those items with. sure and the perceived difficulty of deal-. loadings less than 0.4, the remaining 18-. ing with a writing topic. This factor ac-. item scale (see Table 2) was again sub-. counted for 38.33% of the total variance. jected to the principal axis factoring ex-. and was labeled Fear of Writing Test. traction and the oblique rotation and the. (FWT). Items 1, 2, and 13 were included. four-factor solution was replicated. The. in Factor Two, accounting for 9.71% of. finalized KMO was .90. The cumulative. the variance. Item 13 was reverse scored.. variance explained was 56.05% for the. The items in this factor reflected students’. first three factors, while the four factors. fear of making mistakes in language use.

(13) 技術學院應外系學生英文寫作焦慮量表之編製. Table 2. 93. The EWAS Items with Percentage of Students Selecting Each Alternative (N = 338). 1. When writing an English composition, I worry about whether the words and expressions used are correct. 2. I feel anxious about whether the grammar used is correct when I am writing an English composition. 3. I get upset when I find my English composition filled with red marks from the teacher. 4. I feel nervous when reading instructions for the topic in English writing. 5. I think other students are better at English writing than I am. 6. Before taking a writing test, I feel jittery about what is going to happen. 7. I find it stressful to write my ideas down in English. 8. I have a terrible time organizing my ideas in writing an English composition. 9. I feel anxious if my English writing needs to be graded. 10. An unfamiliar writing topic makes me anxious. 11. Whenever I think that my teacher is ready to correct the mistakes in my English composition, I get worried. 12. Taking English composition exams makes me feel nervous. 13. I do not worry about making mistakes in sentence structure when writing in English. ® 14. I get nervous when the time for a writing task is running out. 15. The thought of my writing being judged by my English teacher makes me tense. 16. I think I can write good English compositions. ® 17. I worry about the consequences of failing my English writing tests. 18. I am afraid that my classmates will find fault with my English composition if I discuss it with them.. SA 38. A 59. N 2. D 1. SD 0. 43. 53. 2. 2. 0. 13. 44. 18. 23. 2. 9 22 17 18 14 15 28 7. 31 44 53 39 40 36 49 25. 28 24 16 20 22 25 14 26. 30 10 13 22 23 21 7 36. 2 1 2 1 1 4 1 6. 18 0. 41 1. 22 5. 18 63. 2 32. 25 12. 49 31. 16 28. 9 24. 1 4. 3 21 10. 24 42 25. 20 19 28. 36 16 32. 18 2 5. Note: SA=strongly agree, A=agree, N=neither agree nor disagree, D=disagree, SD=strongly disagree, ®= reversed item. Percentage may not total 100 due to rounding.. while writing. This factor was named Anxiety about Making Mistakes (AMM).. Lastly, four items (Items 5, 7, 8, and 16) constituted the fourth factor, in which. The third factor contained five items. Item 16 was reverse scored. The factor. (Items 3, 9, 11, 15, and 18) and was able. mainly reflected students’ low self-. to explain 8.02% of the variance. This. perception in English writing and their. factor was assigned a label of Fear of. difficulties in organizing their ideas or. Negative Evaluation (FNE) because it. expressing their ideas in English. This. represented students’ anxiety about get-. factor accounted for 6.74 % of the vari-. ting negative feedback from their teacher. ance and was identified as Low Confi-. or having their compositions read by their. dence in English Writing (LCEW).. classmates..

(14) 94 教育與心理研究 31 卷 3 期. ability over 8-week intervals was r = 0.79. Item Validity and Internal Consistency. (p < .01, n = 320).. Convergent and Discriminant validity. In the present study, the corrected item-total correlations (see Table 3) ranged from 0.509 to 0.781, implying. Prior to measuring the convergent. adequate item validity. Internal consis-. and discriminant validity of the developed. tency coefficient as estimated by Cron-. EWAS, the reliability of the adapted Chi-. bach alpha was high, achieving overall. nese version of the WAT, FLRAS,. 0.903 for the 18-item total scale, 0.845 for. FLCAS, FNE and CAS was assessed. Ta-. Subscale 1, 0.793 for Subscale 2, 0.847. ble 4 indicates that Cronbach α was satis-. for Subscale 3, and 0.789 for Subscale 4.. factory, ranging from .86 to .92. Such re-. The internal consistency coefficient at the. sults enhanced my confidence of using. second administration of the scale was. these scales for examining the convergent. also high (α= 0.92). The test-retest reli-. and discriminant validity of the EWAS.. Table 3. Factor Pattern Matrix of the EWAS, Corrected Item-Total Correlation, and Percentage of Variance for the Four Factors (N = 338). Items Item 12 Item 17 Item 6 Item 14 Item 10 Item 4 Item 2 Item 13 Item 1 Item 11 Item 18 Item 9 Item 15 Item 3 Item 5 Item 16 Item 7 Item 8 % of variance. Factor 1(FWT) .729 (.715) .680 (.654) .628 (.634) .595 (.548) .590 (.580) .573 (.622). Factor 2(AMM). Factor 3(FNE). Factor 4(LCEW). .948 (.761) .679 (.571) .601 (.590) .980 (.781) .660 (.612) .606 (.705) .602 (.673) .491 (.509). 38.33. 9.71. 8.02. .746 (.578) .742 (.665) .524 (.609) .449 (.544) 6.74. Note: Numerical values in boldface refer to the loading of each variable on the corresponding factor and those in parentheses represent the correlation of a single item with the sum of all other items in the same factor..

(15) 技術學院應外系學生英文寫作焦慮量表之編製. Table 4. 95. Reliability Assessment of the Adapted and Translated Scales (N = 116) Revised Chinese Version WAT FLRAS FLCAS FNE CAS. Cronbach Alpha .91 .86 .91 .92 .91. Note: WAT=Writing Anxiety Test; FLRAS=Foreign Language Reading Anxiety Scale; FLCAS=Foreign Language Classroom Anxiety Scale; FNE=Fear of Negative Evaluation; CAS=Computer Anxiety Scale.. As shown in Table 5, the overall. = 0.08). The four subscales of the EWAS. EWAS correlated highly with the lan-. had moderate to low correlations with the. guage-. WAT,. FNE, with coefficients ranging from 0.24. FLCAS, and FLRAS, with correlations. to 0.44. No significant correlations were. ranging from 0.56 to 0.74. With the ex-. found between the four subscales and the. ception of Factor 2 (Anxiety About Mak-. CAS. Even though with lower coeffi-. ing Mistakes), all the subscales of the. cients, Factor 2 was significantly associ-. EWAS correlated highly with the WAT.. ated with all the language-related meas-. Correlations ranged from 0.60 to 0.72.. ures, viz. WAT, FLCAS, and FLRAS, with. Furthermore, the three subscales were. coefficients 0.23, 0.21 and 0.13, respec-. moderately to highly correlated with the. tively. It also had significant associations. FLCAS and the FLRAS. Correlations. with FNE. However, no associations were. ranged from 0.46 to 0.59.. found between Factor 2 and the CAS.. related. anxiety. scales. To further examine the discriminant validity of the EWAS and its subscales,. Criterion-Related Validity. differences in the magnitude of their as-. Criterion-related validity aims to re-. sociation with the non-foreign language. flect predictive validity and concurrent. specific anxiety instruments of FNE and. validity (Onwuegbuzie, Bailey & Daley,. CAS were examined. Table 5 shows that. 2000). The former measures how well an. the total EWAS only moderately corre-. instrument predicts future performance,. lated with FNE, with coefficients of .45.. while the latter examines the correlation. However, no significant correlations were. between two instruments hypothesized to. found between the EWAS and the CAS (r. measure similar constructs. The concur.

(16) 96 教育與心理研究 31 卷 3 期. Table 5 Scale EWAS-1 EWAS-2 EWAS-3 EWAS-4 EWAS-T. Means and Standard Deviation of the EWAS and Correlations Among the EWAS and Other Measures (N = 338) M 21.89 12.95 15.95 14.11 64.89. SD 4.44 1.50 4.21 3.24 10.73. WAT 0.60** 0.23** 0.60** 0.72** 0.74**. FLCAS 0.53** 0.21** 0.54** 0.59** 0.64**. FLRAS 0.51** 0.13* 0.46** 0.51** 0.56**. FNE 0.39** 0.24** 0.44** 0.28** 0.45**. CAS 0.05 0.09 0.10 0.09 0.08. Note: EWAS=English Writing Anxiety Scale (1=subscale 1, 2=subscale 2, 3=subscale 3, 4=subscale 4, T=Total scale). **p < .01. *p < .05.. rent validity of the present study was based on the correlations between the EWAS and the WAT (Daly & Miller,. Confirmatory Factor Analysis. 1975). As shown in Table 5, the correla-. Confirmatory Factor Analysis func-. tions found were significant at r = 0.74 ( p. tions to test the structure generated from. < .01), meaning over 50% (0.742 ×100=. the exploratory factor model (Long,. 54.8%) of the variance of the EWAS. 1983). To assess the structure fitness of. overlapped with the WAT, thereby sup-. the EWAS under study, a maximum like-. porting the existence of concurrent valid-. lihood confirmatory factor analysis, as. ity. With respect to predictive validity,. computed in LISREL, was undertaken to. self-ratings on the perceived writing per-. test the following models.. formance in the demographic data were. 1.A two-factor model.. conducted by Pearson correlation and the. 2.A three-factor model.. result negatively correlated (r = -0.51, p. 3.A four-factor orthogonal model in. < .01, n = 338) with the EWAS, implying. which the full four factors were assumed. that the higher the English writing anxi-. to be uncorrelated.. ety, the lower the self-perceived compe-. 4.A four-factor oblique model in. tence on English writing. Further evi-. which the four factors were presumed to. dence of predictive validity was estab-. be distinct but correlated. This model was. lished via significant correlations (r = -. of main interest of the current study.. 0.29, p < .01, n = 147) between the scores. The indices used for measuring. on the EWAS and students’ final course. model fit included the following: chi-. grades.. square (x2), the ratio of. x2. to degrees of.

(17) 技術學院應外系學生英文寫作焦慮量表之編製. 97. freedom, Goodness-of-Fit Index (GFI),. ceptable range between 1 and 3 (Kline,. and. Index. 1994), and the adequate cutoff of <.08. (AGFI). To examine the effects of the. (Onwuegbuzie et al., 2000), respectively.. sample size on model fit (Hoelter, 1983). In addition, the values of GFI and AGFI. and to compare the fit of different models. were lower than the acceptable cutoff of. in the same population (Steiger & Lind,. ≥ .09 (Browne & Cudeck, 1993, cited in. 1980), the Critical N Index (CNI) and the. Onwuegbuzie et al., 2000). The three-. Root Mean Square Error of Approxima-. factor model, with reasonable x2/df ratio. tion (RMSEA) were also tested.. and RMSEA value, demonstrated an im-. Adjusted. Goodness-of-Fit. Results of the fit indices of various. provement over the two-factor model.. models are presented in Table 7. First, the. However, the GFI and AGFI did not meet. two-factor model indicated poor fit to the. the cutoff value. Also, the critical N index. data. Its estimates of x /df ratio and. was far lower than the rule of thumb. RMSEA were much higher than the ac-. value of ≥ 200 (Hoelter, 1983).. 2. Table 7 Model Model (1) Model (2) Model (3) Model (4). Results of Fit Indices for Alternative Models (N = 338) x2 1093.68 443.02 264.64 277.30. df 323 167 113 129. x2/df 3.38 2.65 2.33 2.15. Fit Indices GFI 0.78 0.88 0.91 0.91. AGFI 0.74 0.85 0.88 0.89. RMSEA 0.094 0.071 0.064 0.060. Critical N 119.65 162.60 193.14 206.73. Finally, it is challenging to choose. other indicator of better fit. More impor-. between the four-factor orthogonal and. tantly, the result of the critical N index. oblique models because the estimates of. also favored the better fit of the oblique. x /df, GFI, and RMSEA were so close and. model, with critical N value > 200.. 2. within the cutoff points. However, the smaller x2/df ratio of the oblique model. Discussion. shows that it was a better solution than. The EWAS proves satisfactory in. the orthogonal model. The value of AGFI. different psychometric properties. The re-. (0.89) in the oblique model, considerably. sults attained in Table 3 demonstrate that. closer to the cutoff point of 0.90, was an-. there were good item validity and internal.

(18) 98 教育與心理研究 31 卷 3 期. consistency in the overall EWAS and its. self-rating proficiency than with their fi-. four subscales. The test-retest reliability. nal course grades may show that students. further shows the acceptable stability of. tend to overestimate their anxiety levels. the overall instrument. Also, as shown in. and underestimate their perceived compe-. Table 5, the total EWAS and three of its. tence when faced with foreign language. factors reveal high correlations with the. learning (MacIntyre et al., 1997). The low. WAT, an instrument specifically written. correlation of the EWAS and the final. for assessing writing anxiety. Their corre-. course grades may indicate that using the. lations with the other two non-writing. latter as an achievement measure needs. specific foreign language learning anxiety. further examination in validity and reli-. instruments, the FLCAS and FLRAS,. ability (Cheng et al., 1999).. were found moderate to high. However,. The results of factor analyses lend. the EWAS and its subscales showed only. some credibility to the existence of con-. moderate to low correlations with FNE. struct validity of the devised EWAS. Con-. that measures the constructs of social. firmatory factor analyses further substan-. evaluation anxiety. There were no signifi-. tiate the fitness of the four-factor model.. cant correlations between the EWAS and. The resultant four factors also correspond. the CAS that tests the general computer. to some theoretical bases in the literature.. use anxiety. Even though the coefficients. The factor AMM appears to conform to. were low, Factor 2 correlated significantly. Rose’s (1984) Premature Editing that was. with the three language-related instru-. identified as a crucial cause of writer’s. ments, but no significant correlations with. block. Fear of negative evaluation has. the CAS were found. These results pro-. been identified as a vital dimension in in-. vide some evidence for the convergent. vestigating into foreign language learning. and discriminant validity of the EWAS as. anxiety (Aida, 1994; Horwitz et al., 1986). a foreign language writing anxiety in-. and writers like Aldrich (1982), and. strument.. Cheng et al. (1999) also found the exis-. In addition, the results demonstrate. tence of such a component in an EFL. an acceptable criterion-related validity of. writing class. Self-rating of writing was. the overall EWAS. However, the higher. found to be a significant component of. correlation of the EWAS with students’. student writers’ writing apprehension and.

(19) 技術學院應外系學生英文寫作焦慮量表之編製. 99. blocking behavior in Daly and Miller’s. own and make their voices heard, and en-. (1975) WAT and Rose’s (1984) WBQ, re-. courages creativity (Braine, 2003). Be-. spectively. In a study aiming to differenti-. sides, it centers on content and formative. ate writing and speaking anxiety compo-. meaning discovery and delays language. nents, Cheng et al. (1999) also corrobo-. forms to the final revision stage. Such a. rate such a factor in their adapted version. focus, according to Zamel (1987), con-. of the WAT. There is disagreement about. tributes to making L2 writers less. whether test anxiety is a general or situa-. stressed. Another favorable factor of this. tion-specific. (Aida,. approach is that the teacher, traditionally. 1994). The results of the present study,. seen as a threatening authority figure and. however, propose that test situations, such. assessor, acts as a facilitator to help learn-. as time pressure, difficult and unfamiliar. ers develop different strategies during the. topics, and fear of failure, be a component. writing process (Silva, 1990). The limita-. in English writing anxiety in the context. tion of the process approach in de-. of Taiwan.. emphasizing input, however, can be. anxiety. problem. Pedigogical Implications. bridged by the product approach which emphasizes knowledge input on linguistic. The results of factor analysis to the. and rhetoric forms. Such knowledge is. EWAS reveal that the potential anxiety. needed in an EFL writing context because. sources of English writing in Taiwanese. lack of it can be a component of students’. institutes of technology are in writing. anxiety (Leki, 1999; Tsui, 1996).. tests, making mistakes in language forms,. The anxiety sources identified in this. negative evaluation, and low confidence.. study also suggest that strategies, such as. This indicates that an integrative process-. allowing students to decide on topics to. product approach would be an appropriate. fit their interests, adopting multiple-draft. option to tackle EFL students’ writing. process, releasing students from testing. anxiety. The process approach is seen. situations in writing, avoiding overreact-. helpful in building up learners’ autonomy. ing to errors, providing frequent encour-. and self-esteem (Stringer, Morton, &. agement and positive reinforcement, and. Bonikowski, 1999). In addition, it allows. performing portfolio assessment, can be. learners to negotiate meaning on their. helpful for alleviating students’ writing.

(20) 100 教育與心理研究 31 卷 3 期. anxiety. Besides, techniques like holding. and that with some modification, this. individual talk periodically to overcome. scale can also be used in other tertiary. students’ unrealistic expectations and in-. contexts in Taiwan. Despite the satisfac-. volving students in verbalizing and show-. tory results of psychometric properties. ing classmates their fears in writing might. from this study, this scale needs further. help the students see that they are not. research from other tertiary contexts, such. supposed to be successful in writing. as four-year university and five-year jun-. overnight and they are not alone in writ-. ior college English majors. Investigation. ing anxiety, thus enhancing their willing-. into non-English majors is also desirable.. ness to take risks in writing.. Such studies will not only result in in-. Conclusion. triguing differences among different tertiary contexts but help to improve the psy-. This study investigated into issues. chometric properties of this scale. This. pertaining to writing anxiety in Taiwanese. can then make the scale more valid in di-. institute of technology English majors.. agnosing learners’ anxiety sources.. The finalized EWAS in this study presents. Developing such a scale, even. that the sources of English writing anxiety. though time and energy consuming, is. lie in the following: Fear of Writing Test,. worthwhile because it can increase our. Anxiety about Making Mistakes, Fear of. understanding of learners’ problems and. Negative Evaluation, and Low Confidence. help us look for appropriate pedagogies to. in English Writing. The total scale and the. improve an EFL writing class. However,. subscales of the EWAS were found to. it must be noted that a scale is not the. have adequate item validity, internal con-. only means to map out learners’ writing. sistency, convergent and discriminant va-. problems.. lidity, and criterion-related validity. Fi-. such as an open-ended questionnaire, es-. nally, the fitness of the four-factor model. say writing, interviews, and follow-up. was tested and substantiated through con-. discussion, are also needed. With data. firmatory factor analysis.. coming from different sources, the re-. Qualitative-oriented. means,. It is hoped that this scale can be used. searcher will be able to compare, synthe-. to define the anxiety sources of Taiwanese. size, and cross-examine the results,. English majors at institutes of technology. thereby providing a more accurate profile.

(21) 技術學院應外系學生英文寫作焦慮量表之編製. of students’ sources of anxiety and presenting more effective solutions to help students cope with their anxiety. Compared to the amount of anxiety research devoted to oral performance, anxiety as it pertains to FL writing is still very limited. However, recent studies (Cheng et al., 1999; Leki, 1999) have confirmed that writing in the FL context, like speaking and listening comprehension, can also be anxiety-provoking. Anxiety will affect not only students’ participation and performance in classrooms but also their selections in courses, majors and even career. As FL writing teachers, therefore, we must recognize the existence of students’ writing anxiety, know how to identify the sources of anxiety, and then present effective strategies to help. students. cope. with. anxiety-. generating situations and to make a writing class less stressful.. Reference Aida, Y. (1994). Examination of Horwitz, Horwitz, and Cope’s construct of foreign language anxiety: The case of students of Japanese. The Modern Language Journal, 78(2), 155-168. Aldrich, P. G. (1982). Adult writers: Some factors that interfere with effective writing. Research in the Teaching of English,. 101. 16(3), 298-300. Bennett, K., & Rhodes, S. C. (1988). Writing apprehension and writing intensity in business and industry. Journal of Business Communication, 25(1), 25-39. Braine, G. (2003). From a teacher-centered to a student-centered approach: A study of peer feedback in Hong Kong writing classes. Journal of Asian Pacific Communication, 13(2), 269-288. Byrne, D. (1982). Teaching writing skills. New York: Longman. Cheng, Y. S., Horwitz, E. K. & Schallert, D. L. (1999). Language anxiety: Differentiating speaking and writing components. Language Learning, 49(3), 417-446. Cheng, Y. S. (1999). Validation of a second language writing apprehension scale. The Second Biennial Pan Asian Conference, October 1-3, 1999. Cheng, Y. S. (2004). A measure of second language writing anxiety: Scale development and preliminary validation. Journal of Second Language Writing, 13(4), 313-335. Cohen, B. A., & Waugh, G. W. (1989). Assessing computer anxiety. Psychological Reports, 65, 735-738. Connor, U. (1999). Contrastive rhetoric: Cross-cultural aspects of secondlanguage writing. Cambridge: Cambridge University Press. Daly, J. A. (1978). Writing apprehension and writing competence. Journal of Educational Research, 72(1), 10-14. Daly, J. A., & Miller, M. D. (1975). The empirical development of an instrument of writing apprehension. Research in the Teaching of English, 9(3), 242-249..

(22) 102 教育與心理研究 31 卷 3 期. Dunteman, G. H. (1989). Principal components analysis. Thousands Oaks, CA: Sage Publications, Quantitative Applications in the Social Sciences Series, No. 69. Floyd, F. J., & Widaman, K. F. (1995). Factor analysis in the development and refinement of clinical assessment instruments. Psychological Assessment, 7(3), 286299. Hoelter, J. W. (1983). The analysis of covariance structures: Goodness-of-fit indices. Sociological Methods and Research, 11(3), 325-344. Holliday, A. (2002). Doing and writing qualitative research. London: Sage. Horwitz, E. K. (1986). Preliminary evidence for the reliability and validity of a foreign language anxiety scale. TESOL Quarterly, 20(4), 559-562. Horwitz, E. K., Horwitz, M. B., & Cope, J. (1986). Foreign language classroom anxiety. The Modern Language Journal, 70(2), 125-133. Hughey, J. B., Wormuth, D. R., Hartfiel, V. F., & Jacobs, H. L. (1983). Teaching ESL composition: Principles and techniques. Rowley, MA: Newbury House. Johanson, R. (2001). The self-reported perspectives regarding academic writing among Taiwanese graduate students specializing in TEFL. Texas Papers in Foreign Language Education, 6, 19-36. Joreskog, K. G., & Sorbom, D. (1984). LISREL-VI user’s guide. Mooresville, IN: Scientific Software. Kaplan, R. B. (1966). Cultural thought patterns in intercultural education. Language Learning, 16(1), 1-20.. Kline, P. (1994). An easy guide to factor analysis. New York: Routledge. Leki, I. (1999). Techniques for reducing second language writing anxiety. In D. J. Young (Ed.), Affect in foreign language and second language learning: A practical guide to creating a low-anxiety classroom atmosphere (pp. 64-88). Boston: McGraw-Hill. Long, J. S. (1983). Confirmatory factor analysis. CA: Sage. MacIntyre, P. D., Noels, K. A., & Clément, R. (1997). Biases in self-ratings of second language proficiency: The role of language anxiety. Language Learning, 47(2), 265-287. Onwuegbuzie, A. J., Bailey, P., & Daley, C. E. (2000). The validation of three scales measuring anxiety at different stages of the foreign language learning process: The input anxiety scale, the processing anxiety scale, and the output anxiety scale. Language Learning, 50(1), 87117. Phinney, M. (1991). Word processing and writing apprehension in first and second language writers. Computers and Composition, 11(1), 65-82. Raisman, N. (1982). I just can’t do English: Writing anxiety in the classroom. Teaching English in the Two-Year College, 9(1), 19-23. Reaves, R. R., Flowers, J. L., & Jewell, L. R. (1993). Effects of writing-to-learn activities on the content knowledge, retention, and attitudes of secondary vocational agriculture students. Journal of Agricultural Education, 34, 34-40. Richmond, V. P., & Dickson-Markman, F..

(23) 技術學院應外系學生英文寫作焦慮量表之編製. (1985). Validity of the writing apprehension test: Two studies. Psychological Reports, 56, 255-259. Riffe, D., & Stacks, D. W. (1988). Dimensions of writing apprehension among mass communication students. Journalism Quarterly, 65(2), 384-391. Rose, M. (1984). Writer’s block: The cognitive dimension. Carbondale, IL: Southern Illinois University Press. Saito, Y., Horwitz, E., & Garza, T. (1999). Foreign language reading anxiety. The Modern Language Journal, 83(2), 202218. Sellers, V. D. (2000). Anxiety and reading comprehension in Spanish as a foreign language. Foreign Language Annals, 33(5), 512-521. Semke, H. D. (1984). The effects of the red pen. Foreign Language Annals, 17(3), 195-202. Silva, T. (1990). Second language composition instruction: Developments, issues, and directions in ESL. In B. Kroll (Ed.), Second Language Writing. Cambridge: Cambridge University Press. Sparks, R. L., Ganschow, L., & Javorsky, J. (2000). Déjà vu all over again: A response to Saito, Horwitz, and Garza. The Modern Language Journal, 84(2), 251255. Steiger, J. H. & Lind, J. C. (1980). Statistically-based tests for the number of common factors. Paper presented at the An-. 103. nual Meeting of the Psychometric Society, Iowa City, IO. Stringer, S. J., Morton, R. C., & Bonikowski, M. H. (1999). Learning disabled students: Using process writing to build autonomy and self esteem. Journal of Instructional Psychology, 26(3), 196-200. Tabachnik, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Needham Heights, MA: Allyn and Bacon. Thomas, J. (2002). What technology do tomorrow’s science teachers need to know? A response to Henriques. Contemporary Issues in Technology and Teacher Education, 2(1), 1-4. Tsui, A. B. M. (1996). Learning how to teach ESL writing. In D. Freeman & J. C. Richards (Eds.), Teacher learning in language teaching (pp. 97-119). Cambridge: Cambridge University Press. Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33(4), 448-457. Young, D. J. (1991). Creating a low-anxiety classroom environment: What does language anxiety research suggest? The Modern Language Journal, 75(4), 426437. Zamel, V. (1987). Recent research on writing pedagogy. TESOL Quarterly, 21(4), 697716..

(24) 104 教育與心理研究 31 卷 3 期. Appendix A. A Questionnaire on English Writing Anxiety 英文寫作焦慮問卷. 同學您好：這是一份關於英文寫作焦慮的問卷。煩請您將英文寫作當中常造成您焦慮的原因寫下。您的描述將對我的研究很有幫助。您的資料將受到保密。假如有任何部分被引用，亦將以匿名方式來處理。謝謝您撥空幫忙。. 1. 2. 3. 4. 5..

(25) 技術學院應外系學生英文寫作焦慮量表之編製. Appendix B. 105. The 28-Item English Writing Anxiety Scale (English Version). 1. When writing an English composition, I worry about whether the words and expressions used are correct. 2. I feel anxious about whether the grammar used is correct when I am writing an English composition. 3. I get upset when I find my English composition filled with red marks from the teacher. 4. I feel nervous when reading instructions for the topic in English writing. 5. I feel panicky when I start writing an English composition. 6. I worry about whether the teacher understands what I write in English. 7. I think other students are better at English writing than I am. 8. Before taking a writing test, I feel jittery about what is going to happen. 9. I find it stressful to write my ideas down in English. 10. I have a terrible time organizing my ideas in writing an English composition. 11. I am afraid that what I write in English cannot meet my teacher’s expectations. 12. The thought of revising my composition gives me a lot of pressure. 13. I usually worry about how to revise my written work in English. 14. I feel anxious if my English writing needs to be graded. 15. An unfamiliar writing topic makes me anxious. 16. Whenever I think that my teacher is ready to correct the mistakes in my English composition, I get worried. 17. Taking English composition exams makes me feel nervous. 18. I am often distressed after a writing test. 19. I do not worry about making mistakes in sentence structure when writing in English. 20. A writing class is enjoyable to me when writing errors are not emphasized so much. 21. I get nervous when the time for a writing task is running out. 22. The thought of my writing being judged by my English teacher makes me tense. 23. I am happy to write sufficient details to support my ideas. 24. I think I can write good English compositions.

(26) 106 教育與心理研究 31 卷 3 期. 25. I feel nervous in a writing class when the teacher’s expectations are too high. 26. I worry about the consequences of failing my English writing tests. 27. I am afraid that my classmates will find fault with my English composition if I discuss it with them. 28. Even if I am well prepared for an English writing test, I still feel tense about it..

(27) 技術學院應外系學生英文寫作焦慮量表之編製. Appendix B. The 28-Item English Writing Anxiety Scale (Chinese Version) 英文寫作焦慮量表. 1. 寫英文作文時，我會擔心用字措詞是否正確。 2. 寫英文作文時，我會擔心文法是否正確。 3. 看到自己寫的英文作文到處都是老師批改的符號時，我會懊惱不已。 4. 看英文作文題目的說明時，我會很緊張。 5. 我一旦開始寫英文作文，就會緊張慌亂。 6. 我會擔心老師是否瞭解我所寫的內容。 7. 我認為其他同學英文作文寫得比我好。 8. 英文作文考試前，我會擔心考試的內容。 9. 我發覺很難用英文表達自己的想法。 10. 我在寫英文作文時，很難有條理地整理自己的想法。 11. 我擔心自己寫的英文作文內容無法符合老師的期望。 12. 一想到自己寫的內容需要再修改，我就覺得壓力很大。 13. 我通常不知道該如何修改自己所寫的英文作文。 14. 假如我的英文作文要被評分，我就會擔憂。 15. 看到不熟悉的作文題目，我會很焦慮。 16. 想到老師隨時會訂正我英文作文中的錯誤時，我就會焦慮不安。 17. 考英文作文我會很緊張。 18. 考完英文作文後，我還是常常焦慮不安。 19. 寫英文作文時，我不會擔心在句子結構上犯錯。 20. 假如不要太強調寫作所犯的錯誤，我就會很喜歡上英文作文課。 21. 當寫作的時間快要結束，我就很緊張。 22. 我一想到英文作文要被老師評閱，就會很緊張。 23. 寫出充分的內容來支持自己的論點，對我來說是件輕鬆愉快的事。 24. 我認為我可以寫出好的英文作文。 25. 老師對英文寫作課期望過高時，我就會感到緊張。 26. 我會擔心英文作文考試不及格的後果。 27. 假如跟同學討論我的英文作文，我擔心他們會挑剔我所寫的內容。 28.即使考前有充分的準備，英文作文考試時，我仍會緊張。. 107.

(28)