Vous êtes sur la page 1sur 90

VALIDATION OF WORLD ENGLISH (WE) LISTENING COMPREHENSION SCALE

by

YASUSHI KYUTOKU

Presented to the Faculty of the Graduate School of The University of Texas at Arlington in Partial Fulfillment of the Requirements for the Degree of

MASTER OF SCIENCE IN PSYCHOLOGY

THE UNIVERSITY OF TEXAS AT ARLINGTON MAY 2007

ACKNOWLEDGEMENTS I want to make acknowledgments for my supervisor, Dr. Bernstein, for all the supervision, instruction, advices, helps, kindness, patience, and encouragements in order to complete my thesis research. Also, I want to make acknowledgements for my committee members, Dr. Kimball, Dr. Kopp, and Dr. Gorfein. Without your advices, suggestions, and helps, I could not complete the thesis research. Again, thanks for devoting your precious time for my thesis. Also, I want to thank Dr. Dougall for advices, suggestions, and kindness. Finally, I want to thank Charles Aden, Chawki Belhadi, Chikako Davidson, Gabriel Davidson, Sachimi Kyutoku, Tomomi Ogura, and Ryo Yamada, and faculties, staffs, and colleagues in our departments for inspirations and supports.

March 19, 2007

ii

ABSTRACT

VALIDATION OF WORLD ENGLISH (WE) LISTENING COMPREHENSION SCALE

Publication No. _____

Yasushi Kyutoku, M.S.

The University of Texas at Arlington, 2007

Supervising Professor: Bernstein, Ira H World English (WE) includes any type of dialectical English spoken around the world (Hinkel, 2005). In the present study, a WE listening test was developed. The original form of the test consisted of six scales measuring, respectively, demographics, exposure to non-native English speakers, self perception of WE fluency, WE listening test, evaluations of non-native speakers' fluency, and attitudes towards non-native English speakers. Internal consistency, dimensionality, correlations among scales, and the effects of demographics on the test scores were analyzed based on the data collected from 32 participants in Study 1. Results indicated moderate to moderately high internal iii

consistency for the scales except for the attitudes toward non-native English speakers scale. Study 2 constituted an actual computer administration with 106 participants. SE listening comprehension test was added to the study. All of the scales except for the SE test showed moderate to high consistency and unidimensionality.

iv

TABLE OF CONTENTS ACKNOWLEDGEMENTS ...................................................................................... ABSTRACT.............................................................................................................. LIST OF ILLUSTRATIONS .................................................................................... LIST OF TABLES .................................................................................................... Chapter 1. INTRODUCTION ........................................................................................ 1.1 Standard English and World English ...................................................... 1.1.1 World English (WE) ................................................................ 1.1.2 Reasons to Study WE ............................................................... 1.2 WE Listening Comprehension Test and Relevant Scales ....................... 1.2.1 WE Listening Comprehension Test ......................................... 1.2.2 Relevant Scales ........................................................................ 1.2.2.1 Exposure Scale ............................................................ 1.2.2.2 Attitude Scale .............................................................. 1.2.2.3 Self-report Scale .......................................................... 1.2.2.4 Evaluation Scale .......................................................... 1.3 Purposes of This Study ........................................................................... 2. METHODS (STUDY1) ................................................................................ 2.1 Participants ............................................................................................. v 1 1 1 2 4 4 5 5 6 6 7 7 10 10 ii iii ix x

2.2 Materials.................................................................................................. 2.3 Design and Procedure ............................................................................. 2.4 Data Analyses ......................................................................................... 3. RESULTS (STUDY 1) ................................................................................. 3.1 Data Screening ........................................................................................ 3.2 Validation of Scales ................................................................................ 3.2.1 Exposure Scale ................................................................................ 3.2.2 Attitude Scale .................................................................................. 3.2.3 Self-report Scale.............................................................................. 3.2.4 WE Listening Comprehension Test ................................................ 3.2.5 Evaluation Scale.............................................................................. 3.4 Correlation among Scales ....................................................................... 3.5 Effects of Demographics on WE Listening Comprehension Test .......... 4. DISCUSSION (STUDY1)............................................................................ 4.1 Validation of Scales ................................................................................ 4.2 Correlation among Scales ....................................................................... 4.3 The Effects of Demographics on WE Listening Comprehension Scale . 4.4 Limitations of Study 1............................................................................. 5. INTRODUCTION (STUDY 2) .................................................................... 6. METHODS (STUDY 2) .............................................................................. 6.1 Participants.............................................................................................. 6.2 Materials.................................................................................................. vi

10 12 13 14 14 16 16 17 18 19 20 21 23 24 24 25 25 26 28 29 29 29

6.3 Design and Procedure ............................................................................. 6.4 Data Analyses ......................................................................................... 7. RESULTS (STUDY 2) ................................................................................ 7.1 Data Screening ........................................................................................ 7.2 Effects of demographics on WE Listening Comprehension Scale ......... 7.3 Validation of Scales ................................................................................ 7.3.1 Exposure Scale ................................................................................ 7.3.2 Self-report Scale.............................................................................. 7.3.3 Evaluation Scale.............................................................................. 7.3.4 WE Listening Comprehension Test ................................................ 7.3.5 SE Listening Comprehension Test.................................................. 7.4 Correlation Analyses among Scales........................................................ 8. DISCUSSION (STUDY 2)............................................................................ 8.1 Validation of Scales ................................................................................ 8.2 Correlation Analyses Among Scales....................................................... 8.3 Effects of Demographics on WE Listening Comprehension Test .......... 9. GENERAL DISCUSSION ............................................................................ 9.1 Conclusions, Limitations, and Future Directions.................................... Appendix A. WE LISTENING COMPREHENSION SCALE FOR STUDY 1................ B. WE LISTENING COMPREHENSION SCALE FOR STUDY 2 ............... C. WE CONVERATION SCRIPTS................................................................. vii

30 30 31 31 33 34 34 36 37 38 40 41 42 42 42 43 44 44

46 56 68

REFERENCES.......................................................................................................... BIOGRAPHICAL INFORMATION........................................................................

75 79

viii

LIST OF ILLUSTRATIONS Figure Page

1.1 Sensitive period for the second language learning .......................................... 3 3.1 PA for exposure scale ...................................................................................... 17 3.2 PA for self-report scale.................................................................................... 18 3.3 PA for WE listening comprehension test ........................................................ 20 3.4 PA for evaluation scale ................................................................................... 21 4.1 Distribution of WE listening comprehension test scores ................................ 26 7.1 PA for exposure scale ...................................................................................... 35 7.2 PA for self-report scale.................................................................................... 36 7.3 PA for evaluation scale.................................................................................... 38 7.4 PA for WE listening comprehension test ........................................................ 39

ix

LIST OF TABLES

Table

Page

3.1 Exposure scale.................................................................................................. 14 3.2 Attitude scale.................................................................................................... 15 3.3 Self-report scale................................................................................................ 15 3.4 WE Listening Comprehension Scale................................................................ 15 3.5 Evaluation scale................................................................................................ 16 3.6 Factor analysis for the exposure scale.............................................................. 17 3.7 Factor analysis for self-report scale ................................................................. 19 3.8 Factor analysis for WE listening comprehension test ...................................... 20 3.9 Factor analysis for evaluation scale ................................................................. 21 3.10 Table of descriptive statistics ......................................................................... 22 3.11 Correlation among the scales ......................................................................... 22 3.12 Effects of demographic data on WE listening comprehension scale ........................................................................ 23 7.1 Exposure scale.................................................................................................. 31 7.2 Self-report scale................................................................................................ 32 7.3 Evaluation scale................................................................................................ 32 7.4 WE listening comprehension scale .................................................................. 32 7.5 WE listening comprehension scale when only native speakers were included ......................................................................... 33 7.6 SE listening comprehension test ...................................................................... 33 x

7.7 Effects of demographics on listening scale ...................................................... 34 7.8 Factor analysis for exposure scale.................................................................... 35 7.9 Factor analysis for self-report scale ................................................................. 37 7.10 Factor analysis for evaluation scale ............................................................... 38 7.11 Factor analysis for WE listening comprehension test .................................... 40 7.12 Table of descriptive statistics ......................................................................... 41 7.13 Correlation among the scales ......................................................................... 41

xi

CHAPTER 1 INTRODUCTION

1.1 Standard English and World English English is spoken as the native language by 3.6 hundred million people and spoken as the second language by 1-1.5 billion speakers (Crystal, 2000). As a consequence, English is regarded as the most widely used international language. For example, factors such as movies, international travel, the internet, and recent developments in computer networks have all further widened the use of English. According to Kachru (1985), there are more non-native English speakers than native ones. Even in English speaking countries, such as the United States, there are increasing numbers of non-native English speakers because of increasing numbers of immigrants (Flowerdew & Miller, 1997). Non-native speakers are defined as people who already use at least one other language and who live in a community in which English is not normally used (Hinkel, 2005). 1.1.1 World English (WE) English languages can be categorized into three main types according to the degree of nativity (Hinkel, 2005). The first type of English is Standard English (SE), which is used in English speaking countries such as the U.S., the UK, Canada, Australia, and New Zealand. This is commonly referred to as normative, proper, 1

correct, or the Kings (Queens) English. A second type of English is non-Standard English, which is used in countries such as former colonies of English speaking countries where English is the official or second language but not the dominant one. There is debate whether specific kinds of native English, such as Ebonics, are standard or nonstandard. A third type of English is non-native English. Non-native English is used in countries where English is not official, dominant, or the first language. These three types of English languages are collectively called World English (WE) as opposed to SE, which includes only the first type of English. For example, WE speakers may say He like you car instead of He likes your car. WE speakers also have relatively strong accents derived from their dominant language that differ from those used by SE speakers. They further use a different accent and are less fluent in English than they are in their dominant language. Another characteristic of both non-standard and non-native English is the intrusion of vocabulary from the more dominant language. For example, I, as a native Japanese speaker, tend to say hai instead of yes (hai means yes in Japanese). 1.1.2 Reasons to Study WE Several studies reported that effective oral communication depends on the mutual cooperation of speakers and listeners (Ervin-Tripp, 1970; Grice, 1975). Although effective communication depends on mutuality, little attention has been paid to the enhancement of native English speakers WE skills (Davies, Hamp-Lyons, & Kemp, 2003). In other words, the standard of English proficiency in international settings is based only on SE (Chalhoub-Deville & Wiggleworth, 2005). For example, 2

non-native English speakers are often required to take an examination to qualify for academic admissions or for employment in English speaking countries. Even in nonEnglish speaking countries, many people are required to take English proficiency examinations such as TOEFL when there are academic or business demands to use English (Davies, Hamp-Lyons, & Kemp, 2003). The efforts of non-native speakers to improve proficiency can be a potential cause of miscommunication in various international settings in domains such as conference, academia, politics, and business (Davies, Hamp-Lyons, & Kemp, 2003). Further, it is unrealistic for non-native English speakers to achieve the SE fluency of native speakers (Bouton, 1994; Snow, 1998). Johnson & Newport (1989) reported a learning curve that describes the relationship between age of arrival to English speaking country and the test of English grammar score suggested that the age of learning a second language is important factor (Figure 1.1).

Test of English Grammar Score

290 280 270 260 250 240 230 220 210 200 native 3 to 7 8 to 10 Age or arrival 11 to 15 17 to 39

Figure 1.1Sensitive period for the second language learning

That implies that unless a non-native English speaker is exposed to English in early age, it is hard for him/her to ever achieve SE level of fluency. Also, and perhaps most importantly, the listening comprehension of native English and non-native speakers is influenced by the type of English (Major, et al, 2005). Thus, SE is appropriate where English is normally used while WE is more appropriate and practical in international settings (Davies, Hamp-Lyons, & Kemp, 2003). 1.2 WE Listening Comprehension Test and Relevant Scales Despite the importance of WE, it has received very little attention (Davies, Hamp-Lyons, & Kemp, 2003). As a result, there is not a widely used WE assessment test. Accurate and valid measurement of WE should be developed in order to enhance international communication. The present research investigated listening comprehension and four other related domains of WE proficiency. 1.2.1 WE Listening Comprehension Test The WE listening comprehension scale was modeled upon a widely used SE listening comprehension test in the U.S., the Test of English as a Foreign Language (TOEFL). The TOEFL is developed and administered by Educational Testing Service (ETS), and according to Bollag (2005) TOEFL is taken by most foreign students who are being considered for admission at English-speaking higher-education institutions, including those in the United States. Last year (2004) nearly 750,000 students took the test. In terms of predictive validity (Rosenfeld, Oltman & Sheperd, 2004) and construct validity (Powers et. al, 1999; Oltman, Stricker, & Barrows, 1990), TOEFL appears to be a valid measure of English fluency. The TOEFL listening test is 4

composed of questions in which examinees are given native speakers conversations, monologues, and sentence speeches. Respondents are instructed to pick the best answers based on multiple-choice questions. Similarly, the WE listening comprehension test is composed of questions in which conversations between a native Japanese speaker speaking English and an American speaker, monologues of a Japanese English speaker, and sentence-speeches of a Japanese English speaker are used as the test material. 1.2.2 Relevant Scales In addition to the listening comprehension test, exposure to foreign cultures (the exposure scale), attitudes toward non-native English speakers (the attitude scale), selfreports of WE listening comprehension skill (the self-report scale), and evaluations of non-native speakers English in the listening tests (the evaluation scale) were developed to investigate the domains that are related to WE listening comprehension. 1.2.2.1 Exposure Scale First, the exposure scale was designed to investigate the degree that exposure to non-native speakers and non-American cultures related to WE listening comprehension skills. In other words, this scale intends to measure the degree that the amount of experience to WE influences on the WE listening comprehension test score. Since Johnson & Newport (1989) suggested that the amount of exposure to spoken language should not be highly correlated with linguistic skill, amount of exposure to WE should not be highly correlated with listening performance.

1.2.2.2 Attitude Scale Second, the attitude scale was designed to measure attitudes toward non-native speakers and cultures related to WE listening comprehension skills. Although some may believe that attitudes of native speakers toward non-native speakers influence native speakers listening comprehension, Lindemann (2002) failed to find such a relationship. The present scale was used to reexamine this hypothesis. 1.2.2.3 Self-report Scale Third, a self-report scale directly asked respondents to judge participants own WE fluency in a similar manner to other metacognitive scales (Metcalfe, 1986; Nunnally & Bernstein, 1994). This scale was designed to investigate accuracy of selfperceptions of WE fluency. The scale investigated the following basic conversational components of WE comprehension in 5-point Likert scales. This scale first prompted respondents for their self-perceptions of their own WE comprehension skills. It then assessed the degree and frequency of difficulties in comprehending elements of speech such as accents (where you stress in a word or a sentence), pronunciation (how you say a word), and grammar (rules of a language). These factors were included in the scale because Harris (1994) suggested that the syntactical structure (grammar) and phonological structures (accent and pronunciation), and the combination of said (overall fluency) are important factors in the comprehension of spoken English. Various studies have failed to find a correlation between self-efficacy and actual performance in various cognitive domains such as memory tasks (Richard, Diefendorff, & Martin, 2006,

Shameem, 1998). This current study investigated whether metacognition (self-report) is related to WE listening comprehension performance. 1.2.2.4 Evaluation Scale Finally, the scale for the evaluation of non-native speakers fluency, pronunciation, accent, grammar, and the clarity of the speech in the listening comprehension part was designed to measure whether evaluation of non-native speakers was associated with examinees comprehension skill. 1.3 Purposes of This Study Listening and related components of WE were also investigated using standard classical test theory measures in 2 studies. Study 1 was conducted as a pilot study for the study 2. These studies involved determining Cronbachs of the exposure scale, attitude scale, self-report scale, evaluation scale, and the WE listening comprehension test and item/total correlations for the constituent items (rit). Cronbachs alpha is a statistical index of internal consistency that is determined by the correlations of items and the numbers of items in a scale. Higher implies the higher internal consistency. It has been suggested that an acceptable scale should produce a Cronbachs alpha larger than .7, and each rit should be larger than .2. in its target population (Nunnally & Bernstein, 1994). Second, exploratory principal component analyses should suggest that the scales are unidimensional using eigenanalysis and parallel analysis to establish a baseline for the assessment of dimensionality (Bernstein et al., 2007). As for principal component factor analyses, a table for each result presented factor pattern (= factor structure in the following analyses) and communalities (h2). Factor pattern is a 7

coefficient that describes the relationship between a factor and an item. Higher coefficient value implies the higher relationships. Communality is the sum of the squared factor pattern for each item. The larger communality implies that the amount of variance in an item is explained by a particular factor. According to Bernstein et al (2007): In PA, one factors a matrix containing the same number of observations and variables as the real data but simulated from a population in which all correlations are zero. Multiple matrices of this form may be generated. The results are averaged to provide for a more stable estimate and standard errors. In order for s solution to conform to a unidimensional solution, the first principal component obtained from the real data must be larger than the first principal component obtained from the simulation, but all subsequent components from the actual data must be smaller than their simulated counterparts. Third, correlations among the exposure scale, the self-report scale, the evaluation scale, and the listening comprehension test were analyzed. Fourth, the effects of demographic variables such as gender, nativity/non-nativity of English, and age upon the scales were also obtained.

Both studies to be reported were approved by the University of Texas at Arlingtons Institutional Review Board. Study 1 was a pilot study designed to provide a working set of scales.

CHAPTER 2 METHODS (STUDY 1)

2.1 Participants Thirty two students enrolled in introductory psychology at the University of Texas at Arlington were recruited. The data from one participant was omitted because that person did not finish the experiment. The sample consisted of 6 males and 29 females, with a mean age of 20.4 years (SD = 4.21). There were 27 SE speakers and 5 non-native speakers. They voluntarily participated in preliminary research in order to fulfill a course requirement. 2.2 Materials The first part of all the scales contained demographic information (age, gender and country of birth). Second, the degree of exposure to foreign cultures (exposure), attitudes toward non-native English speakers and cultures (attitude), self reports of WE listening comprehension skill (self-report), Listening comprehension tests (WE test), and evaluations of non-native speakers English in the listening tests (evaluation) were designed to investigate the domains possibly related to WE listening comprehension. These items may be found in Appendix A. a. Scales that preceded the WE listening comprehension scale: 1) Demographic data 10

2) The exposure scale (Part A of Appendix A): measured the degree of exposure to non-native speakers. In order to eliminate problems associated with the exposure scale, unit response to items were standardized. 3) Attitudes toward non-native English speakers were assessed using five-point Likert scales (Part B of Appendix A). 4) Self-perceptions of WE comprehension skill and the degree and frequency of difficulties in comprehending elements of speech (Part C of Appendix A) measured in 5 point Likert scale such as accents (the position of stressed syllable in a word), pronunciation (the way of vocalizing a word), and grammar (rules of a language). b. A scale the presented after the completion of the WE listening comprehension scale The evaluation scale (Part H of Appendix A) measured participants evaluations of the non-native speakers fluency in a conversation (Part D of Appendix A). As above noted, the listening comprehension scale of WE (Part D to G of Appendix A) was modeled upon the TOEFL. In the TOEFL, examinees are presented with a conversation by native English speakers in SE and asked to pick the best multiple choice answers. Instead of native English speakers, a conversation between a male native English speaker using SE and a Female whose native language was Japanese and who spoke WE was used. Participants were asked comprehension questions using a 411

alternative multiple-choice format. The Japanese speakers proficiency level was certified as Able to understand and use English well enough for everyday needs and situationsoverseas by the Society for Testing English Proficiency Inc (STEP, 2004). A digital audio file was used to reproduce the recorded conversation. Audio presentation has been found to be a more effective and efficient assessment method of listening comprehension than video (Read, 2002; Conian, 2001). Also, auditory presentations remove extraneous variables such as visual information out of the procedure. Participants listened to digitally recorded conversations through headphones, and they responded to paper and pencil version of the self-report scale and listening comprehension scale. 2.3 Design and Procedure After informed consent was collected, participants received instructions for the experiment. Then, they proceeded to complete the demographic survey. After the demographic survey, they responded to parts A (exposure), B (attitude), and C (selfperceived competency) of the survey, in turn. They then proceeded to part D, where they heard the conversation once. After listening to the conversation, they then answered the listening comprehension test of the conversation. They then proceeded to Parts E, F, and G in the identical manner as Part D. Participants debriefed and released after completing part H. Thus, all participants went through identical procedures. The procedure took about 25 to 35 minutes.

12

2.4 Data Analyses Listening and related components of WE were also investigated using standard classical test theory measures in 2 studies. This involved determining the internal consistency (Cronbachs ) of the exposure scale, attitude scale, self-report scale, evaluation scale, and the WE listening comprehension test and item/total correlations for the constituent items (rit). It has been suggested that an acceptable scale should produce a Cronbachs alpha larger than .7, and each rit should be larger than .2. in its target population (Nunnally & Bernstein, 1994). Second, exploratory principal component analyses should suggest that the scales are unidimensional using eigenanalysis and parallel analysis to establish a baseline for the assessment of dimensionality (Bernstein et al., 2007). Third, correlations among the exposure scale, the self-report scale, the evaluation scale, and the listening comprehension test were analyzed. Fourth, the effects of demographic variables such as gender, nativity/nonnativity of English, and age upon the scales were also obtained.

13

CHAPTER 3 RESULTS (STUDY 1)

3.1 Data Screening The means and standard deviations for each item of each scale are presented in Table 3.1 for the exposure scale, Table 3.2 for the attitude scale, Table 3.3 for the selfreport scale, Table 3.4 for the WE listening comprehension test, and Table 3.5 for the evaluation scale. Each scale was subjected to component analysis. The results will be presented to emphasize the loadings on the first (general) factor.

Table 3.1 Exposure scale


Variable a1 a2 a3 a4 a5 a6 a7 a8 a10 N 24 24 24 24 24 24 24 24 24 Mean 48.9 9.6 3.8 3.6 5.2 111.9 3.4 3.2 3 SD 114.7 7.6 1.6 1.6 8 299.6 1.4 1.4 1.6 rit 0.55 0.70 0.73 0.70 0.48 0.32 0.62 0.73 0.83

Note: Item 9 was eliminated because it measured the degree of exposure to written language rather than spoken language.

14

Table 3.2 Attitude scale


After revision Variable b1 b2 b3 b4 b5 b6 b7 b8 N 31 31 31 31 31 31 31 31 Mean 3.4 4.6 4.3 2.8 4.3 3.8 3.4 2.6 SD 0.76 0.62 0.92 1.46 1.07 1.19 1.3 1.2 rit 0.38 0.37 0.36 0.32 0.38 0.31 0.2 0.28

Table 3.3 Self-report scale


Variable c1 c2 c3 c4 c5 c6 c7 c8 c9 N 32 32 32 32 32 32 32 32 32 Mean 3.7 3.4 3.4 3.3 3.3 3.1 3.2 3.4 3.2 SD 0.78 0.94 1.01 0.92 0.87 1.00 0.91 1.07 1.03 rit 0.48 0.65 0.61 0.41 0.58 0.75 0.55 0.65 0.67

Table 3.4 WE listening comprehension scale


After Variable t11 t12 t13 t15 t18 t24 t27 t29 t30 N 32 32 32 32 32 32 32 32 32 Deletion Mean 0.50 0.97 0.97 0.75 0.44 0.63 0.84 0.78 0.75 SD 0.51 0.18 0.18 0.44 0.50 0.50 0.37 0.42 0.44 rit 0.47 0.53 0.53 0.44 0.33 0.28 0.47 0.41 0.40

15

Table 3.5 Evaluation scale


Variable h1 h2 h3 h4 h5 N 30 30 30 30 30 Mean 4.0 3.6 3.1 3.6 3.6 SD 0.93 0.86 0.90 0.89 0.77 rit 0.67 0.70 0.66 0.65 0.60

3.2 Validation of Scales 3.2.1 Exposure Scale In order to eliminate problems associated with scale units, response to items were standardized. The internal consistency of the 10-item Exposure scale was high (Cronbachs = .88, n = 24, Table 3.1). Item 9 was eliminated form the analysis because it was more related to written than spoken English. Consequently it was dropped from the scale, leaving 9 items (Cronbachs = .88, n = 24, Table 3.1). An exploratory principal component factor analysis was performed. The scale was unidimensional based upon a parallel analysis criterion (see figure 3.1), and the resulting pattern appears in Table 3.6.

16

5 4.5 4 Eigen Value 3.5 3 2.5 2 1.5 1 0.5 0 1 2 3 4 5 6 7 8 9 Number of factors Scree Plot Randomized data

Figure 3.1 PA for the exposure scale

Table 3.6 Factor analysis for exposure scale


Factor Pattern Items 1 2 3 4 5 6 7 8 10 Factor 1 0.65 0.77 0.81 0.79 0.59 0.36 0.71 0.81 0.89 Communalities 2 h 0.54 0.60 0.74 0.72 0.83 0.84 0.54 0.69 0.81

3.2.2 Attitude Scale The internal consistency of the 10-item Attitude scale was low ( = .54, n = 31). Items 9 and 10 had unacceptable values of rit (-.13 and .01, respectively) so were 17

eliminated. However, the internal consistency remained poor ( = .61, Table 3.2) after this deletion, the scale as a whole was eliminated. Consequently, the results of factoring the scale will not be presented. 3.2.3 Self-report Scale The internal consistency for the 9-item self-perceived competency scale was high ( = .86, n = 32, Table 3.3). Item level analysis based on rit suggested all items were related to the scale (Table 3.3) so all original items were retained. An exploratory principal component factor analysis was performed. The result was that the scale was unidimensional based upon a parallel analysis criterion (see figure 3.2), and the resulting factor pattern appears in Table 3.7.

Parallel analysis
5 4 Eigen value 3 2 1 0 1 2 3 4 5 6 7 8 9 Number of factors Scree Plot Randomized data

Figure 3.2 PA for self-report scale

18

Table 3.7 Factor analysis for self-report scale


Factor Pattern Items 1 2 3 4 5 6 7 8 9 Factor 1 0.59 0.73 0.72 0.51 0.66 0.83 0.66 0.75 0.76 Communalities 2 h 0.66 0.70 0.80 0.91 0.83 0.74 0.60 0.80 0.85

3.2.4 WE Listening Comprehension Test In order to achieve the final version of scale, there were four iteration processes. This iteration process continued until the both the criterion for the Cronbachs =.7 and rit = .2. First, item 8 that had no variance were eliminated (eg. Most participants answered that question correctly). Then, items that had lower than rit =.2 were eliminated in each step. In the step 2, items 6, 7, 20, 22, 26, and 28 were eliminated. In the third step, items 1, 3, 4, 9, 10, 14, 16, 17, 19, and 25 were eliminated. In the fourth step, items 2, 5, and 23 were eliminated. The final version of the WE listening test retained 9 items with moderate internal consistency ( = 72, n = 32, Table 3.4). An exploratory principal component factor analysis was performed. The scale was unidimensional based on parallel analysis criterion (Figure 3.3) and the resulting factor pattern appears in Table 3.8.

19

Parallel analysis
3.5 3 Eigne value 2.5 2 1.5 1 0.5 0 1 2 3 4 5 6 7 8 9 Number of factors Scree plot Randomized data

Figure 3.3 PA for WE listening comprehension test

Table 3.8 Factor analysis for WE listening comprehension test


Factor Pattern Items 11 12 13 15 18 24 27 29 30 Factor 1 0.52 0.82 0.82 0.58 0.43 0.41 0.63 0.56 0.53 Communalities 2 h 0.63 0.96 0.96 0.38 0.68 0.56 0.48 0.44 0.59

3.2.5 Evaluation Scale The internal consistency of the 5-item evaluation scale was high ( = .85, n= 30, Table 5). Each item met the rit = .2 criterion, so all items were retained. An exploratory principal component factor analysis was performed. The scale was unidimensional 20

based on parallel analysis criterion (Figure 3.4), and the resulting factor pattern appears in Table 3.9.

Parallel analysis
3.5 3 Eigen value 2.5 2 1.5 1 0.5 0 1 2 3 4 5 Number of factors Scree plot Randomized data

Figure 3.4 PA for evaluation scale

Table 3.9 Factor analysis for evaluation scale


Factor Pattern Items 1 2 3 4 5 Factor 1 0.80 0.82 0.80 0.78 0.74 Communalities 2 h 0.63 0.67 0.64 0.61 0.55

3.3 Correlation among Scales After the validation of scales, data screening for correlation analyses among scales was performed. Descriptive statistics appears in table 3.10.

21

Table 3.10 Tables of descriptive statistics


Scale age standardized logexp self eval WE mean 20.34 0 29.94 17.59 6.63 SD 4.21 1 5.92 3.57 2.04 Skewness 3.17 1.67 0.18 0.18 -1.31 Kurtosis 12.24 3.28 1.2 -0.64 2.72 n 32 32 32 32 32

Table 3.11 contains the scale intercorrelations. As can be seen, the correlation between the self-perceived competency scale and the listening test was very low. However, there was a moderate positive correlation between the listening test and the evaluation scale.

Table 3.11 Correlation among the scales


Pearson r Variable WE Self Logexp Eval WE __ Self -0.06 __ (n=32) Logexp .10 0.15 __ Eval 0.45* 0.14 -0.07 __

Note: WE = WE listening comprehension scale, Self = Self report scale, Logxp = standardized and log transformed Exposure scale, and Eval = the Evaluation scale. *P<.05

22

3.4 Effects of Demographics on WE Listening Comprehension Test Demographics variables such as gender, native/non-nativity of English, and age were unrelated to both the listening test (Table 3.12). There were no trends of note the mean for males was 6.50 and the mean for females was 7.16, and effect size was small, Cohen's d = .28. It would require about 320 participants to achieve a significant difference with = .05, and power = .80. The mean for native English speakers was 6.59 and mean for non-native speakers was 6.80, and effect size was very small, Cohen's d = .01. It would require about 2874 participants to achieve significant difference with = .05, and power = .80.

Table 3.12 Effects of demographic data on WE listening comprehension scale


ANOVA source Age Gender C.O.B. Error df 1 1 1 28 F 1.68 0.65 1.12 (4.42) p 0.21 0.43 0.30

Note: Values in parentheses are mean square errors

23

CHAPTER 4 DISCUSSION (STUDY 1)

4.1 Validation of Scales The results of the pilot study were that all scales save for attitude met the usual standards for reliability. Thus, attitude scale did not meet the statistical criterion. Also, there was theoretical concern about the attitude scale. In other words, it was not clear whether attitude scale was measuring actual attitude or overt response of participants. For example, because of social desirability, participants might have reported socially desirable attitude. Because of these reasons, attitude scale was eliminated from the analyses. These other scales were then re-examined in a larger sample, but five changes were made from study 1. First, item 9 was eliminated from the exposure scale, and all items were changed to be measured in Likert scale. Second, the attitude scale was eliminated due to low internal consistency as described above. Third, items 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 14, 16, 17, 19, 20, 21, 22, 23, 25, 26, 28 in the WE listening test were modified (Appendix 2) if no one picked a particular multiple choice response or everyone had the same response in a particular question. Therefore, either the questions or multiple choice answers were modified. For example, multiple choice b) of items 2 in WE test was modified from Father to Friend. Also, items such as item 8 were eliminated because everyone could answer that question correctly. Father was 24

changed to Friend because no one picked that answer. The 9-item exposure scale (Part A of Appendix 2), the 9-item self-report scale (Part B of Appendix B), the 42-item WE listening comprehension test (Part C to H of Appendix B), the 5-item English listening comprehension scale (Part I of Appendix B), and the 5-item evaluation scale (Part J of Appendix B) were also modified. Fourth, grammatical mistakes and typos were corrected in the modified scales. 4.2 Correlation among Scales Exposure scale was transformed due to the high deviation from normality. These previous analyses also suggested two other points: (a) perceived selfcompetency was not significantly correlated with test performance, and (b) scores on the listening test were positively correlated with the evaluation of the fluency of the non-native speakers. 4.3 The Effect of Demographics on WE Listening Comprehension Test Demographics such as gender and age did not affect listening test scores. Notably, native English speakers did not outperform non-native speakers in the WE listening comprehension scale. This contrasts with findings obtained using SE listening comprehension scales such as the TOEFL (Stricker, 2004). However, because of the imbalance in numbers of native and non-native group and, especially, the small sample size, a null conclusion may be premature.

25

4.4 Limitations of Study 1 The present study has two obvious limitations. As noted, the sample size was very small. Second, listening test scores may have been range restriction restricted since all scores distributed fell between 20 to 30 points (Figure 4.1).

Distribution of Scores
7 6 Frequency 5 4 3 2 1 0 19 20 21 22 23 24 25 26 27 28 29 30 Test Scores

Figure 4.1 Distribution of WE listening test scores

This might reflect the sample or the test materials. A practical solution to this limitation was to modify scales using a larger sample. Since Cronbachs alpha is not an inferential measure, it should not be affected by sheer sample size. However, Tabachnik and Fidell (2001; 2007) suggest a minimal sample size of 90 for a multiple correlation analysis of the present form (however, since SE listening comprehension test was added 26

later to test discriminant validity, sample size larger than 105 was necessary for the study 2).

27

CHAPTER 5 INTRODUCTION (STUDY 2)

The purpose of the study 2 was to investigate the modified scales of study 1 in a larger sample. An SE listening comprehension test was also added to investigate discriminant validity. The analyses paralleled those of study 1.

28

CHAPTER 6 METHODS (STUDY 2)

6.1 Participants One hundred and seven students enrolled in introductory psychology at the University of Texas at Arlington were recruited. The data from one participant was omitted because that person did not finish the experiment. The sample consisted of 67 females and 39 males, with a mean age = 20.0 years and SD = 2.91. There were 90 SE speakers and 16 non-native speakers. They voluntarily participated in preliminary research in order to fulfill a course requirement. 6.2 Material Five changes were made from study 1, as previously noted. First, item 9 was eliminated from the exposure scale, and all items were measured using a Likert scale format. Second, the attitude scale was eliminated. Third, items 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 14, 16, 17, 19, 20, 21, 22, 23, 25, 26, 28 in the WE listening test were modified (Appendix B). The modified versions of the 9-item exposure scale (Part A of Appendix B), the 9-item self-report scale (Part B of Appendix B), the 42-item WE listening comprehension test (Part C to H of Appendix B), the 5-item Standard English listening comprehension scale (Part I of Appendix B), and the 5-item evaluation scale (Part J of 29

Appendix B) were also attached as appendix B. Fourth, grammatical mistakes and typos were collected in the modified scales. Fifth, administration and data collection of the scales was automated using the E-prime program. 6.3 Design and Procedure After informed consent was collected, participants were assigned to a computer. They were asked to start answering the demographic survey and modified exposure scale, the self-report scale, the WE listening comprehension test, and the evaluation scale. After completion, participants were debriefed and released. Thus, all participants went through the identical procedure. The task took approximately 40-45 minutes. 6.4 Data Analyses Data analyses for the study 2 was identical to the study 1 except that the attitude scale was excluded for the study 2 and SE listening comprehension scale was added to the study 2.

30

CHAPTER 7 RESULTS (STUDY 2)

7.1 Data Screening The means and standard deviations for each item of each scale are presented in Tables 7.1 to 7.6. Table 7.1 contains the exposure scale. Table 7.2 contains the selfperceived competency scale. Table 7.3 contains the Evaluation scale. Table 7.4 contains the WE listening comprehension test, and Table 7.5 contains WE listening test for only native speakers. Table 7.6 contains SE listening comprehension scale. Severe violation of assumptions was not detected while screening data.

Table 7.1 Exposure scale


Variable a1 a2 a3 a4 a5 a6 a7 a8 a10 N 104 104 104 104 104 104 104 104 104 Mean 3.3 3.0 3.9 3.6 1.7 1.6 3.2 3.0 3.0 SD 1.8 1.8 1.1 1.3 1.3 1.3 1.3 1.2 1.3 rit .69 .71 .71 .56 .52 .30 .72 .64 .64

31

Table 7.2 Self-report scale


Variable b1 b2 b3 b4 b5 b6 b7 b8 b9 N 105 105 105 105 105 105 105 105 105 Mean 3.4 3.1 3.2 3.1 3.2 3.0 3.1 3.1 3.2 SD .86 .87 .93 .94 .90 .91 .93 .92 .93 rit 0.67 0.79 0.72 0.65 0.69 0.79 0.83 0.69 0.70

Table 7.3 Evaluation scale


Variable j1 j2 j3 j4 j5 N 106 106 106 106 106 Mean 2.9 2.7 2.6 3.0 2.6 SD 0.90 0.82 0.74 0.79 0.90 rit 0.73 0.74 0.53 0.47 0.70

Table 7.4 WE listening comprehension scale


Variable t1 t2 t3 t4 t8 t9 t12 t13 t14 t15 t16 t21 t30 t31 t33 t34 t35 t37 t41 N 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 106 Mean .86 .88 .44 .51 .92 .74 .72 .72 .34 .91 .75 .68 .87 .52 .41 .55 .74 .75 .42 SD .35 .33 .50 .50 .30 .28 .44 .45 .48 .29 .43 .47 .34 .50 .49 .50 .44 .44 .50 rit .40 .36 .25 .29 .24 .30 .23 .23 .29 .25 .21 .21 .23 .27 .29 .30 .29 .30 .29

32

Table 7.5 WE listening comprehension scale when only native speakers were included
Variable t1 t2 t3 t4 t8 t9 t12 t13 t14 t15 t16 t21 t30 t31 t33 t34 t35 t37 t41 N 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 Mean .85 .88 .44 .50 .73 .61 .72 .34 .91 .75 .60 .97 .52 .41 .55 .74 .42 .42 .50 SD .35 .32 .50 .30 .28 .44 .45 .45 .48 .29 .43 .47 .34 .50 .49 .50 .44 .44 .50 rit .43 .39 .32 .31 .26 .29 .22 .30 .31 .28 .24 .25 .22 .27 .32 .27 .28 .32 .32

Table 7.6 SE listening comprehension test


Variable item 1 item 5 N 106 106 Mean .65 .58 SD 0.48 0.50 rit 0.19 0.19

7.2 The Effects of Demographics on WE Listening Comprehension Test Demographic variables such as gender, native/non-nativity of English, and age again did not related to scores on the WE listening test. As for the effect on WE test, there were no trends of note the mean for males was 13.72 and the mean for females was 13.52, and effect size was very small, Cohen's d = .059. It would require about 480 participants to achieve significant difference with = .05, and power = .80. There were 33

no trends of note the mean for native English speakers was 13.64 and mean for nonnative speakers was 13.31, and effect size was very small, Cohen's d = .097. It would require about 9000 participants to achieve significant difference with = .05, and power = .8. Since no significant effect of nativity on any scale based on one way ANOVAs, data from both native and non-native participants were pooled (Table 7.7).

Table 7.7 Effects of demographics on listening scale


ANOVA source Age Gender C.O.B. Error df 1 1 1 100 F .57 .00 .17 (11.56) p 0.45 0.99 0.68

Note: Values in parentheses are mean square errors

7.3 Validation of Scales 7.3.1 Exposure Scale The result suggested moderately high internal consistency of the 9-item Exposure scale (Cronbachs = .87, n =106. Item level analysis based on rit suggested all items were related to the scale (Table 7.1) so all original items were retained. An exploratory principal component factor analysis was then performed. The scale was unidimensional based on parallel analysis criterion (Figure 7.1), and the resulting pattern appears in Table 7.8.

34

Parallel analysis
5 4 Eigen value 3 2 1 0 1 2 3 4 5 6 7 8 9 Number of factors Scree plot Randomized

Figure 7.1 PA for exposure scale

Table 7.8 Factor analysis for exposure scale


Factor Pattern Items 1 2 3 4 5 6 7 8 10 Factor 1 .77 .80 .79 .67 .61 .36 .80 .75 .74 Communalities 2 h .60 .64 .69 .56 .76 .66 .66 .56 .54

35

7.3.2 Self-report Scale The result suggested an acceptably high internal consistency for the 9-item selfperceived competency scale ( = .92, n = 106). Item level analysis based on rit suggested all items were related to the scale (Table 7.2), so all items were retained. An exploratory principal component factor analysis was performed. The scale was unidimensional based on parallel analysis criterion (Figure 7.2), and the resulting pattern appears in Table 7.9.

Parallel analysis
6 5 Eigen value 4 3 2 1 0 1 2 3 4 5 6 7 8 9 Number of factors Scree plot Randomized

Figure 7.2 PA for self-report scale

36

Table 7.9 Factor analysis for self-report scale


Factor Pattern Items 1 2 3 4 5 6 7 8 9 Factor 1 .75 .84 .79 .72 .76 .85 .88 .76 .76 Communalities 2 h .56 .71 .62 .52 .57 .72 .77 .57 .58

7.3.3 Evaluation Scale The result suggested moderately high internal consistency of the 5-item evaluation scale ( = .83, n = 106). Item level analysis based on rit suggested all items were related to the scale (Table 7.3), so all items were retained. The scale was unidimensional based on parallel analysis criterion (Figure 7.3), and the resulting pattern appears in Table 7.10.

37

Parallel analysis
3.5 3 Eiigen value 2.5 2 1.5 1 0.5 0 1 2 3 4 5 Number of factors Scree Plot Randomized

Figure 7.3 PA for Evaluation scale

Table 7.10 Factor analysis for evaluation scale


Factor Pattern Items 1 2 3 4 5 Factor 1 .85 .86 .69 .62 .83 Communalities 2 h .72 .74 .48 .38 .69

7.3.4 WE Listening Comprehension Test In order to achieve the final version of scale, there were four iteration processes. This iteration process continued until the both the criterion for the Cronbachs =.7 and rit = .2. First, items (6, 18, 23, 26, 28, 29, 32, 38, and 40) that had very low variance were eliminated (eg. Everyone answered that question correctly). Then, items that had lower than rit =.2 were eliminated in each step. In the second step, items 7, 10, 11, 17, 38

19, 20, 22, 24, 25, 26, 27, and 36 were eliminated. In the third step, item 43 was eliminated. After the elimination, 20 out of 42 items were retained. The result suggested moderate internal consistency ( = 71, n = 106). Item level analysis based on rit suggested all retained items were related to the scale (Table 7.4). Internal consistency of the WE listening comprehension scale for native speakers showed a similar result, one that was not discussed here (Table 7.5). An exploratory principal component factor analysis was performed. The scale was unidimensional based on parallel analysis criterion (Figure 7.4), and the resulting pattern appears in Table 7.11.

Parallel analysis
3.5 3 Eigne value 2.5 2 1.5 1 0.5 0 1 2 3 4 5 6 7 8 9 Number of factors Scree plot Randomized data

Figure 7.4 PA for WE listening comprehension test

39

Table 7.11 Factor analysis for WE listening comprehension test


Items 1 2 3 4 5 8 9 12 13 14 15 16 21 30 31 33 34 35 37 41 Factor Pattern Factor 1 .54 .52 .37 .40 .36 .43 .36 .53 .32 .38 .34 .30 .30 .31 .37 .39 .42 .44 .43 .39 Communalities 2 h .66 .63 .62 .77 .52 .75 .54 .65 .71 .66 .70 .55 .66 .61 .52 .58 .67 .73 .58 .65

7.3.5 SE Listening Comprehension Test There were originally 5 items on this scale, but only 2 items were retained due to low internal consistency ( = .31, n =106, Table 7.6). Due to low internally consistency and small number of items, meaningful analyses was not able to be performed further. Low internal consistency seemed to be caused by both low correlations among items and very small number of items in the scale. Based on the Spearman-Brown prophecy formula, about 11 more items were needed to have a reliability of .70.

40

7.4 Correlation among Scales Descriptive statistics after the validation was shown in Table 7.12.

Table 7.12 Table of descriptive statistics


Scale age exp self eval WE SE mean 19.96 26.18 23.07 13.8 13.6 1.23 SD 2.01 8.86 2.41 3.22 3.38 0.75 Skewness 2.64 0.09 -0.69 0.26 -0.37 -0.42 Kurtosis 9.64 -1.09 0.026 1.24 -0.33 -1.11

Results from correlation analyses (Table 7.13) suggested that there were a medium positive correlation between the WE listening comprehension test and English listening comprehension test, between the WE listening comprehension test and selfreport, and the listening test and the evaluation scale. The significant correlation between WE test and self report was unexpected based on the study 1.

Table 7.13 Correlation among the scales


Pearson r variable WE SE self exp eval WE __ SE 0.29** __ Self 0.36*** 0.05 __ (N=106) exp 0.02 0.03 -0.06 __ __ eval 0.22* -0.14 0.13 0.09 ___

Note: WE = WE listening comprehension scale, SE = Standard English listening comprehension test, Self = the Self report scale, Exp = the Exposure scale, and Eval = the Evaluation scale. *P<.05, **P<.01, ***P<.001

41

CHAPTER 8 DISCUSSION (STUDY 2)

8.1 Validation of Scales Except for SE test, results from the present study indicated moderate to moderately high internal consistency for every scales. Based on the Spearman-Brown prophecy formula, about 11 more items were needed to have a reliability of .70 for SE test. Principal component analyses suggested the unideminensionality of these scales. 8.2 Correlation among Scales Correlation analyses suggested that the self-report scale was significantly correlated to performance. That implies that the self-report reflects the WE listening scale. This conflicts with the findings from Study 1. However, self-report is still far from accurate measure of WE listening skill because it accounted for only 13% of variance of the WE test (R2 (106)= .13). Second, those who had higher scores on the listening test evaluated the fluency of the non-native speaker in the test better than those who scored lower on the listening test as was the case in Study 1. Significant positive correlation between WE listening test performance and self-report scale seems to be incompatible with those of the study 1. That might be attributed to the relatively 42

unreliable WE scale in Study 1. The fact that WE scale in study 2 appeared to be better scale might be attributed to the modification of items and larger numbers of items in the scale. 8.3 Effects of Demographics on WE Listening Comprehension Scale Based on the ANOVA and power analyses, demographics such as gender, age, and nativity of English did not have a significant effect on listening test scores. Indeed, effect sizes of demographics were very small. Notably, native speakers did not outperform non-native speakers in the WE listening comprehension scale as compared with other SE listening comprehension scales such as the TOEFL (Stricker, 2004) as the Study 1. However, because of the unequal number of participants in the native (90 participants) and non-native (16 participants) groups, this finding was not strongly generalizable.

43

CHAPTER 9 GENERAL DISCUSSION

9.1 Conclusions, Limitations, and Future Directions Because of internal consistency, unidimensionality, and a weak correlation between the WE and the SE listening scale, WE listening comprehension scale appears to be valid test to measure WE listening proficiency. However, some modifications should be considered in further investigations. First, more items should be included in the WE listening scale in order to make it more useful and reliable. Second, the SE test should include more items to perform meaningful analyses. Third, a more basic level of WE such as a word level or phoneme level should be analyzed in the investigation of WE comprehension because it was difficult to analyze data in the conversation level. Fourth, the low correlation between the exposure scale and WE listening test may be caused by the construction of the exposure scale. In other words, the exposure scale measured exposure to non-English speaking culture in general. Since this current study included only a Japanese speaker in WE listening scale as a non-native English speaker, exposure to Japanese English speakers rather than exposure to general non-native speakers should be measured. Fifth, it will be interesting to use self-report scale after the WE listening test as well as prior to WE listening comprehension test. Finally, 44

participants may have guessed what the non-native speaker said by using native speaker's speech as contextual cue in the conversation sections. Thus, participants might have used native speaker as contextual cue to answer the questions. Therefore, listening skill to comprehend conversation between non-native speakers' English rather than conversations between non-native and native speaker should be measured in conversation section. Logically that will make the listening scale more internally consistent because listening skill in the comprehension of only a non-native speaker is measured in the monologue and the sentence sections as opposed to conversation part that included a native speaker.

45

APPENDIX A

WE LISTENING COMPREHENSION SCALE IN STUDY 1

46

Please answer the following survey and the WE listening comprehension scale as honestly and accurately as possible. Please answer all parts of the following survey. In this survey, non-native English speaker is defined anyone whose first or dominant languages is not English. Demographic Survey
1. Age 2. Gender ( ( ( ) ) )

3. Country of Birth

Part A: Answer the following questions based on your experience.


1. How many non-native English speakers do you know? ( ) 2. How long have you known them in years (list oldest ones)? ( ) years 3. How close you are to the one you consider yourself closest to? Not close Very At all Neutral Close 1 2 3 4 5 4. How often do you see that one person? Very Rarely Neutral 1 2 3

Very Often 5

5. How many times have you visited countries in which English is not the dominant language? ( ) 6. How many days have you spent in countries in which English is not the dominant language? ( )7. How familiar are you with non-American culture? Very Very Unfamiliar Average Familiar 1 2 3 4 5 47

8. How familiar are you with the conversation of non-native English speakers? Very Very Unfamiliar Average Familiar 1 2 3 4 5 9. How familiar are you with the writing of non-native English speakers? Very Very Unfamiliar Average Familiar 1 2 3 4 5

10. How fluent are you in a language other than English? Not at All Average 1 2 3 4

Very Fluent 5

Part B: Imagine an average non-native English speaker, and answer following questions.
1. Rate your attitude towards this persons English fluency. Not Very Fluent Neutral 1 2 3 4 2. How favorable is your view of non-English cultures? Very Neutral Negative 1 2 3 4 Very Fluent 5

Very Positive 5

3. Do you want to go to a country in which English is not the native language? Yes No 1 2 3 4 5 4. Do you want to live in a country in which English is not the native language? Yes No 1 2 3 4 5 5. Are you interested in a non-American culture? Yes 1 2 3

No 5

48

6. Do you pay much attention to foreign cultures? Yes 1 2 3

No 5

7. When you do not understand a non-native speakers English, do you think that it is usually something wrong with their fluency? Yes No 1 2 3 4 5 8. When you do not understand a non-native speakers English, do you think that it is usually something wrong with your listening skill? Yes No 1 2 3 4 5 9. Do you regard Midwestern American or as the standard form of international English? Strongly Strongly Disagree Neutral Agree 1 2 3 4 5 10. Do you regard British as the standard form of international English? Strongly Strongly Disagree Neutral Agree 1 2 3 4 5

Part C: Imagine an average non-native English speaker, and answer following questions.
1. How good are you at understanding the English of someone who is not a native English speaker? Very Very Poor Average Good 1 2 3 4 5 2. On average, how difficult is it for you to understand the overall English of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5

49

3. How often do you find it difficult for you to understand the overall English of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5 4. On average, how difficult is it for you to understand the English grammar of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 5. How often do you find it difficult for you to understand the English grammar of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5 6. On average, how difficult is it for you to understand the English pronunciation of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 7. How often do you find it difficult for you to understand the English pronunciation of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5 8. On average, how difficult is it for you to understand the English accent of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 9. How often do you find it difficult for you to understand the English accent of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5

50

Before you start Part D, please listen to conversation 1. Then please answer following questions as accurately as you can.
1. What was the customer looking for? a) T-shirt b) Statue of Liberty Models c) Some souvenir model 2. For whom did the customer want the product? a) Mother b) Father c) Grandmother

d) Pewter

d) Herself

3. Why did the customer not want the first product the clerk recommended? a) She did not like the design. b) She did not like the material. c) She did not have enough money. d) She did not think it appropriate. 4. Which product was the customer first interested in? a) T-shirt b) Statue of liberty model c) Silver ring d) Chocolate gift 5. Which word did the customer not understand? a) Statue b) Copper c) Plastic

d) Pewter

6. Which kind of material of the product did the customer buy? a) Wood b) Copper c) Plastic d) Pewter 7. What kind of wrapping did the customer want? a) Birthday b) Gift c) Christmas d) Souvenir 8. Which country did the customer come from? a) Japan b) Jordan c) Korea

d) China

9. How many times has the customer visited the U.S.? a) 1 b) 2 c) 3 d) 4 10. How long does the customer plan to stay in the U.S.? a) 3 days b) 5 days c) 7 days d) More than 7 days 11. When will the customer leave? a) Today b) Tomorrow c) 2 days from now

d) Next week

12. Other than New York, where did the customer visit? a) Los Angels b) Philadelphia c) Pittsburgh d) Washington D.C. 51

13. How long did the customer spend there? a) A day b) 3 days c) A week

d) 10 days

14. What was the customers purpose of visit to the U.S.? a) Business b) To see her mother c) To study d) For fun 15. Why did the customer not take the tour? a) She could not afford it. b) She did not like it. c) She did not know about it. d) She did not have a chance. 16. How did the customer like New York? a) She liked it. c) She did not mention like or dislike.

b) She did not like it. d) She liked D.C. better.

17. Which show did the customer watch? a) Phantom of the Opera b) Miss Saigon d) She did not watch any. 18. Why did the customer go to Hard Rock Caf? a) It is her favorite place. b) It was her first time to visit there. c) She likes burger there. d) She likes their goods.

c) Cats

Before you start Part E, please listen to Monologue 1. Then please answer following questions as accurately as you can.
1. What is her occupation? a) Instructor b) Counselor c) Professor d) Principal

2. Where did her husband come from? a) Africa b) Japan c) Canada d) The U.S.

3. Why was she reading a book about cats? a) She likes cats. b) She is studying about cats. c) It is her assignment. 52

d) She has a cat. 4. What is her dream? a) To live in foreign countries c) To marry her boy friend b) To be a teacher d) To write a book

Before you start Part F, please listen to Monologue 2. Then, please answer following questions as accurately as you can.
1. According to the conversation speakers friend Yoko lives in. a) America b) Russia c) Asia d) Europe

2. How does the speaker evaluate her friend Yoko? a) She thinks Yoko is a great person. b) She does not like Yoko. c) Yoko is nice, but she has to develop more social skills. d) Yoko cannot speak French. 3. What is the most difficult part of living in Switzerland? a) She does not like winter. b) She does not like the area where she lives. c) Living costs are too expensive where she lives. d) She does not like to work hard. 4.Which country did the speaker imply Yoko comes from? a) Japan b) The U.S.A. c) China d) Switzerland

53

Before you start Part G, please listen to Sentences.


1. Why did she not go to the concert? a) Tickets are too expensive b) She did not like the band. c) She did not have the time. d) She could not get a ticket. 2. Why did she not go to Johns house? a) It is too far for her. c) She does not like John 3. How is she regarding Pennsylvania? a) She is not there right now. b) She is there.c) She will live there. d) She wishes to live there. 4. What does the sentence imply about Mary a) One should feel sorry for her. b) One should apologize to her. c) One doesnt need to apologize to her. d) She should be ignored. . b) It is too cold. d) She does not have time to go.

Part H: Please answer the following questions.


1. Rate the speakers fluency in conversation 1. Very Poor 1 2 Average 3 4 Very Good 5

54

2. Rate the speakers pronunciation in conversation 1. Very Poor 1 2 Average 3 4 Very Good 5

3. Rate speakers accent in conversation 1. Very Poor 1 2 Average 3 4 Very Good 5

4. Rate the speakers grammar in conversation 1. Very Poor 1 2 Average 3 4 Very Good 5

5. Rate the speakers clarity in conversation 1. Very Poor 1 2 Average 3 4 Very Good 5

55

APPENDIX B

WE LISTENING COMPREHENSION SCALE IN STUDY 2

56

Appendix B: Modified WE Scales

Please answer the following survey and the WE listening comprehension scale as honestly and accurately as possible. Please answer all parts of the following survey. In this survey, non-native English speaker is defined anyone whose first or dominant languages is not English. Demographic Survey
1. Age 2. Gender ( ( ( ) ) )

3. Country of Birth

Part A: Answer the following questions based on your experience.


1. How many non-native English speakers do you know? 1) 1-3 2) 4-6 3) 7-9 4) 9-11 5) More than 11

2. How long have you known them in years (list oldest ones)? 1) 1-3 years 2) 4-6 years 3) 7-9 years 4) 9-11 years 5) More than 11 years 3. How close you are to the one you consider yourself closest to? Not close Very At all Neutral Close 1 2 3 4 5 4. How often do you see that one person? Very Rarely Neutral 1 2 3

Very Often 5

5. How many times have you visited countries in which English is not the dominant language? 1) 1-3 2) 4-6 3) 7-9 4) 9-11 5) More than 11

57

6. How many years have you spent in countries in which English is not the dominant language? 1) 1-3 years 2) 4-6 years 3) 7-9 years 4) 9-11 years 5) More than 11 years 7. How familiar are you with non-American culture? Very Unfamiliar Average 1 2 3 4

Very Familiar 5

8. How familiar are you with the conversation of non-native English speakers? Very Very Unfamiliar Average Familiar 1 2 3 4 5 9. How fluent are you in a language other than Standard English? Not at Very All Average Fluent 1 2 3 4 5

Part B: Imagine an average non-native English speaker, and answer following questions.
1. How good are you at understanding the English of someone who is not a native English speaker? Very Very Poor Average Good 1 2 3 4 5 2. On average, how difficult is it for you to understand the overall English of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 3. How often do you find it difficult for you to understand the overall English of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5

58

4. On average, how difficult is it for you to understand the English grammar of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 5. How often do you find it difficult for you to understand the English grammar of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5 6. On average, how difficult is it for you to understand the English pronunciation of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 7. How often do you find it difficult for you to understand the English pronunciation of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5 8. On average, how difficult is it for you to understand the English accent of someone who is not a native English speaker? Very Very Difficult Neutral Easy 1 2 3 4 5 9. How often do you find it difficult for you to understand the English accent of someone who is not a native English speaker? Very Not often Often Neutral At all 1 2 3 4 5

Before you start Part C (question 1-17), please listen to Part C. You may listen to the Part C only once. Do not look at or start answering the questions until you finish listening Part C.

59

1. What was the customer looking for? a) T-shirt b) Statue of Liberty Model souvenir

c) Some souvenir

d) Specific

2. For whom did the customer want the product? a) Mother b) Her friend c) Grandmother

d) Herself

3. Why did the customer not want the first product the clerk recommended? a) She did not like the design. b) She did not like the material. c) It was too expensive. d) She did not think its appropriate. 4. Which product was the customer first interested in? a) T-shirt b) The medium model c) Silver ring model 5. Which word did the customer not understand? a) Statue b) Copper c) Souvenir

d) The large

d) Pewter

6. Which kind of material of the product did the customer prefer? a) Pewter b) Copper c) Plastic d) None of the above 7. What kind of wrapping did the customer want? a) Birthday b) Gift c) Christmas 8. How many times has the customer visited the U.S.? a) 1 b) 2 c) 3 d) 4 9. How long does the customer plan to stay in the U.S.? a) 3 days b) 5 days c) 7 days d) More than 7 days 10. When will the customer leave? a) Today b) Tomorrow c) 2 days from now

d) Paper

d) Next week

11. Other than New York, where did the customer visit? a) No where b) Hard Rock Cafe c) Broadway d) Washington D.C. 12. How long did the customer spend there? a) A day b) 3 days c) A week 13. What was the purpose of visiting there? a) For shopping b) For museum tour 60

d) 10 days

c) For sightseeing

d) For the

14. What did she buy there? a) Souvenir b) Ticket

c) Food

d) A badge

15. What was the customers purpose of visit to the U.S.? a) Business b) To see her mother c) To shop d) For fun 16. Why did the customer not take the tour? a) She could not afford it. b) She wanted to go by herself. c) She did not know about it. d) She did not have a chance. 17. Where she did not go? a) Steak house c) White House

b) Museum d) Statue of Liberty

18. How did the customer have fun in New York? a) Food b) Musical c) Sightseeing d) Museum 19. Why did the customer go to Hard Rock Caf? a) It is her favorite place. b) It was her first time to visit there. c) She likes burger there. d) She likes their goods.

Before you start Part D (question 20-27), please listen to Part D. You may listen to the Part D only once. Do not look at or start answering the questions until you finish listening Part D.

20. What did she expect that his view of Japanese food? a) Delicious b) Too heavy c) Too light d) Too salty 21. What did she think about Japanese foods in the U.S.? a) Delicious b) Too heavy c) Too light d) Too salty 22. What did they disagree about the tastes of rice and tofu? a) He insisted rice and tofu have taste. b) She insisted that rice and tofu have taste. c) She does not like rice and tofu. d) She does not like rice and tofu. 61

23. How did she think about his sense of tasting foods? a) He does not have good sense of tasting any foods. b) He has good sense of tasting any foods. c) He does not have good sense of tasting Japanese food. d) He has a good sense of American foods. 24. How did she evaluate popular American-Japanese food? a) Delicious b) Too heavy c) Too light d) Too salty 25. Which part of his explanation of difference between Japanese and American recipes she did not understand? a) Metric system b) Ingredients c) Type of a food d) Cooking procedure 26. What was her example of wrong American food in Japan? a) American coffee b) American tea c) American burger juice 27. How did they settle down about argument? a) They accepted differences between cultures. c) They changed topic.

d) American

b) They did not ague. d) He accepted her view.

Before you start Part E (question 28-31), please listen to Part E. You may listen to the Part E only once. Do not look at or start answering the questions until you finish listening Part E.

28. What is her occupation? a) Instructor b) Counselor c) Professor d) Principal

29. Where did she get married? a) Africa b) Japan c) The U.S. d) None of the above

30. Why was she reading a book about cats? a) She likes cats. b) To write a book. 62

c) It is her assignment. d) She has a cat. 31. What is her dream? a) To write a book about cats c) To marry her boy friend b) To be a teacher d) None of the above

Before you start Part F (question 32-33), please listen to Part F. You may listen to the Part F only once. Do not look at or start answering the questions until you finish listening Part F.

32. Why did her husband wanted to see the chef? a) Because he is a great chef. c) Because he is a famous chef. 33. What kind of restaurant was it? a) Japanese b) American c) Chinese d) Korean b) Because he is a TV star. d) The speaker did not mention why.

Before you start Part G (question 34-37), please listen to Part G. You may listen to the Part F only once. Do not look at or start answering the questions until you finish listening Part F.
34. According to the conversation speakers friend Yoko lives in. a) America b) Russia c) Asia d) Europe

35. How does the speaker evaluate her friend Yoko? a) She thinks Yoko is a great person. b) She does not like Yoko. c) Yoko is nice, but she has to develop more social skills. d) Yoko cannot speak French. 63

36. What is the most difficult part of living in Switzerland? a) She does not like winter. b) She does not like the area where she lives. c) Living costs are too expensive where she lives. d) She does not like to work hard. 37. Which country did the speaker imply Yoko comes from? a) Japan b) The U.S.A. c) China d) Switzerland

Before you start Part H (question 38-42), please listen to Part H. You may listen to the Part H only once. Do not look at or start answering the questions until you finish listening Part H.

38. Why did she not go to the concert? a) Tickets are too expensive. b) She did not like the band. c) She did not have the time. d) She could not get a ticket. 39. Why did she not go to Johns house? a) It is too far for her. c) She does not like John 40. How is she regarding Pennsylvania? a) She is not there right now. b) She is there. c) She will live there. d) She wishes to live there. b) She can go there any time. d) She does not have time to go.

64

41. What does the sentence imply about Mary? a) One should feel sorry for her. b) One should apologize to her. c) One doesnt need to apologize to her. d) She should be ignored. 42. According to the speaker a) This fall will be colder. b) Usually, it is colder in fall. c) It is a nice day. d) She complaints about warm weather.

Before you start Part I (question 43-), please listen to Part I. You may listen to the Part H only once. Do not look at or start answering the questions until you finish listening Part H.
43. Why does the second speaker wanted to go to England? a) Because it was on magazine. b) To see gardening plant. c) To see a person. d) Because of special event. 44. Why does the first speaker want to go to Cambodia? a) Because it was on magazine. b) To see gardening plant. c) To see a person. d) Because of special event. 45. Which country was not mentioned by them? a) Japan b) England 65

c) Jordan d) Iran 46. Which places they agreed that they want to go? a) Parts of the U.S. b) Asian countries c) Egypt d) Both a and c 47. Which place at least one of them has been to? a) Cambodia b) Egypt c) Hawaii d) England

Part J: Please answer the following questions.


1. Rate the non-native speakers fluency in the Part C to H. Very Poor 1 2 Average 3 4 Very Good 5

2. Rate the non-native speakers pronunciation in the Part C to H. Very Poor 1 2 Average 3 4 Very Good 5

3. Rate the non-native speakers accent in the Part C to H. Very Poor 1 2 Average 3 66 4 Very Good 5

4. Rate the non-native speakers grammar in the Part C to H. Very Poor 1 2 Average 3 4 Very Good 5

5. Rate the non-native speakers clarity in the Part C to H. Very Poor 1 2 Average 3 4 Very Good 5

67

APPENDIX C

WE CONVERSATION SCRIPTS

68

Conversation 1 A: Hi, may I help you? B: Oh, yes please. A: Are you looking for something in particular? B: Yes. I want to buy a souvenir for my mom. A: How about an I love New York T-shirt ? B: Well, my mom doesnt wear T-shirts. Do you have anything else? A: I see. We have some Statue of Liberty models. B: It sounds good. Can I see them? A: Sure, just over here. We have a lot of different sizes. B: ......... A: Do you see any that you like? B: Mmm, this middle size one looks good for her. Ill take this one. A: O.K. That ones pewter, so its 15 dollars. B: Whats pewter? A: Like silver but soft. B: Do you have any different materials? A: Of course. We have gold, silver, wood, copper, plastic. What would you like? B: Which do you recommend? A: This copper one is the most realistic. B: How much is it? A: This ones 12 dollars. C: O.K. Ill take it. G: Would you like that gift-wrapped? C: Oh, can you? Please. A: No problem. Birthday paper or Christmas paper or anniversary or just normal? B: Well, Christmas one, please. A: Sure. Just a moment, please.....So, where are you from?

69

B: I'm from Japan. Have you been to Japan? A: No, I haven't. Is this your first time to America? B: Yes. I've been here for 10 days, and I'll leave the day after tomorrow. A: Are you here on vacation? B: Yes, I am. A: Have you spent the whole time in New York, or did you go anywhere alse? B: I went to Washington D.C. A: How long did you spend there? B: For 3 days. I went to see the White House. A: DId you take a tour? B: I wanted to. But I couldn' t make a reservation. A: Oh well, did you go to the Smithsonian? B: Well I don't know it. A: It's the big group of museums near the capital building. B: O.K. I'll go there when I go to Washington D.C. next time. A: So, you spent the other 7 days in New York? B: Yes. I enjpyed New York very much. A: Thanks. So, did you see a Broadway show? B: Yes, I saw "Cats". It was wonderful. A: Yeah,everyone likes that one. Are you stying in Manhattown? B: Yes. A: Did you go to the Hard Rock Cafe? B: Yes. I like Hard Rock Cafe. I collect their badges. A: How many do you have? B: I have 7. 5 are fom Japan,1 is from Washinton D.C. 1 is from New York. A: Are you from Tokyo? B: No, I'm from Osaka, third largest city in Japan. A: Oh, I'm sorry. O.K. Here you go. One gift-wrapped Statue of Liberty.

70

B: O.K. Here you go. A: Thank you, and 3 dollars is your change. Have a nice day. b: Thank you.

71

Conversation 2

A: What did you think the first time you ate Japanese food? B: Ahh, lets see, that was when I was in high school I thought Wow, this is good. Its kind of like Chinese food. A: Didnt you think that Japanese foods flavor was light? B: Not really, it was pretty salty and strong. A: Because you had it in America. When I went to America, I thought This isnt real Japanese food. B: Yeah but Ive had Japanese food in Japan thats had pretty strong flavors, havent you? A: Yeah, some are salty. But when I had food in America, everything tasted heavy. Actually your cooking tastes a little heavy. B: Thats just because youre used to bland foods like rice and tofu. A: You usually say that rice and tofu have no flavor but they do! I can tell! I can taste it! B: Its just your imagination. A: You can ask your students! Im sure theyll agree with me. And also, water has flavor. B: Only if the flavor is added, like that C1000 Lemon Water. A: You dont understand. B: Dont understand what? A: Even though you like cooking and going to good restaurants, why cant you tell their flavor? You can tell Cokes flavor at McDonalds and so on. Like the syrup is too little or too much. B: Thats different, thats like how you can tell the difference between my ribs with five spice powder and the ribs without it. A: OK, youll understand some day, I hope. By the way, what did you have the first time you had Japanese food? B: I had teppanyaki, at a restaurant in Fort Wayne (the city we lived in in Indiana).. A: Teppanyaki isnt so Japanese for me. I know its popular in America but Japanese food is tempura or sushi for most Japanese. Isnt tempura famous in America? B: Its kind of famous but less than sushi or teppanyaki. I had heard of sukiyaki and teriyaki in America, too. I had teriyaki before coming to Japan but not sukiyaki. Anyway dont you think saying Japanese food is sushi and tempura is stereotyping? Dont you want other Japanese food to be famous?

72

A: Yes, other Japanese food can be famous, but also Im proud of sushi and tempura. What Japanese food can be popular in America you think? B: I think fried noodles and okonomiyaki could be popular. I think yakiniku could be really popular, too, but it probably wont have a chance because of too many safety regulations. A: Yeah, okonomiyaki is very good, and battered octopus balls. OK, we can open a restaurant in America! B: With okonomiyaki and octopus balls? No way, too Osaka! How about Nagoya-style wings and noodles? You know the wings would be popular, remember when I brought them to that party? A: I liked your wings, but you know I dont like Nagoya so much. By the way, you read recipes in Japanese and in English. Is there any difference between them? B: Only in the names of the ingredients and in the measurements, since Japan uses the metric system. A: Whats the metric system? B: You know, meters, grams, liters A: What system does America use? B: We use the English system. Feet, pounds, ounces Anyway, you know a lot of so-called American food in Japan is wrong too, right? A: You mean American coffee? Light-flavored? B: Thats one example. Its like hot brown water, real American coffee is definitely not like that. A: Things from overseas can be different in different countries. B: So you shouldnt complain about Japanese food in America then, right? A: OK, Ill try to not complain.

73

Monologue 1 Hello. I am going to tell about myself. I am a teacher for a cram school. Teaching is interesting for me. And I am married. My husband is American. His name is Gabriel. He is very sweet. And I like reading books. Now I am reading a novel about cats. I like cats. Sometime I want to have cats. At last, I will tell you about my dream. I want to be a translator someday. I want to publish my books. Thank you. Monologue 2 Hello, I am going to tell you about my experience. Last Tuesday was my husbands birthday. So, we went to a good restaurant in Nagoya last Sunday. Its a famous chefs Chinese restaurant. His name is Chin Kenichi. He became famous from Japanese TV program, Iron Chef. We really enjoyed lunch. When we paid, my husband asked the staff if Mr. Chin was there. But, later the staff came to us and said its actually secret, but if you wan to see Chin, He will be here on December 19th and 20th. We were very happy to hear it. Probably my husband will go there that day. Thank you. Monologue 3 Hi, I will tell you about my friend Yoko. Ive known her for 5 years. Now, she lives in Switzerland because her husband is Swiss. It can be difficult to live in foreign countries. And, so it is for Yoko. First, she speaks French there. She had to study French very hard. Second, its very cold in winter and winter starts much faster then in Japan. Third, everything is expensive there. And she doesnt like the new or different things. So it is difficult to get Japanese things. There are some more reasons, but Yoko always does her best. She never gives up. So, I really respect her. Thank you. Sentence 1 I would have gone to the concert if I had had a ticket. Sentence 2 I dont like to go to Johns house because its 5 miles from here. Sentence 3 I dreamed about living in Pennsylvania. Sentence 4 You dont have to say sorry to Mary. Sentence 5 Its warm for fall.

74

REFERENCES Bernstein, I.H., Rush, A.J., Carmody, T.J., Woo, A., & Trivedi, M.H. (2007). Clinical vs. self--report versions of the quick inventory of depressive symptomology in a public sector sample. Journal of Psychiatric Research, 41, 239-246. Boldt, R. F., Larsen-Freeman, D., Reed, M.S., & Courtnery, R.G. (1992). Distributions of ACTFL Ratings by TOEFL Score Ranges. TOEFL Research Report, 92(59). Bollag, Burton. (2005). New Test of English as a Foreign language Puts an Emphasis on Speaking. Chronicle of Higher Education, 52(2), A49. Bouton, F. L. (1994). Conversational implicature in a second language: Learned slowly when not deliberately taught. Journal of Pragmatics, 22, 157-167. Chalhoub-Deville, M., & Wiggleworth, G. (2005). Rater judgment and English Language speaking proficiency. World Englishes, 24 (3), 383-391. Conian, D. (2001). The use of audio or video comprehension as an assessment instrument in the certification of English language teachers. System, 29 (1) , 1-14. Crystal, D. (2000). Emerging Englishes. English Teaching Professional, (14), 3-6. Davies, A., Hamp-Lyons, L., & Kemp, C. (2003). Whose norms? International proficiency tests in English. World Englishes, 22 (4). 571-584.

75

Educational Testing Service (ETS). (2005). Test of English as a Foreign Language. Ervin-Tripps, S. (1970). Discourse agreement: How children answer questions. In Hays (Ed.), Cognition and the development of language. New York: Wiley. Flowerdrew, J., & Miller, L. (1997). The Teaching Academic Listening Comprehension and the Question of Authenticity. English for Specific Purposes,16(1), 27-46. Grice, H. P. (1975). Logic and conversation. In P. Cole & J. Morgan (Eds.), Speech acts: syntax and semantics. (vol.3, pp41-58). New York: Academic Press. Harris, J. (1994). English Sound Structure. Cambridge, MT: Blackwell. Hinkel, E (ed.) (2005). Handbook of Research in Second Language Teaching and Learning. Mahawah, NJ: LawrenceErlbaum associates. Johnson, D. W., & Newport, E. L. (1989). Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60-99. Kachuru, Y. (1985). Discourse Analysis, non-native Englishes and second language acquisition research. World Englishes, 4 (2), 223-232. Kertoy, M. K., & Goetz, K. M. (1995). The Relationship between Listening Performance on the Sentence Verification Technique and Other Measures of Listening Comprehension. Contemporary Educational Psychology, 20, 320-339.

76

Lindemann, S. (2002). Listening with an attitude: A model of native-speaker comprehension of non-native speakers in the United States. Language in Society, 31(3), 419-441. Major, R. C., Fitzmaurice, S. M., Bunta, F., & Balsubramanian, C. (2005). Testing the Effects of Regional, Ethnic, and International Dialects of English on Listening Comprehension. Language Learning, 55(1), 37-69. Mecalfe, J. (1986). Feeling of Knowing in Memory and problem Solving. Journal of Experimental psychology. 12 (2), 288-294. Nunnly J. C., & Bernstein I. H. (1994). Psychometric Theory (3rd). New York: McGRAW-HILL. Oltman, K. P., Stricker, J. L., & Barrows. S. T. (1990). Analyzing Test Structure by Multidimensional Scaling. Journal of Applied Psychology, 75 (1), 21-27. Powers, D. E., Scheldi, M. A., Leung, S. W, & Butler, F. A. (1999). Validating the revised test of spoken English against a criterion of communicative success. TOEFL Research Report, 99 (5), 63. Read, J. (2002). The use of interactive input in EAP listening assessment. Journal of English for Academic Purposes, 1(2), 105-119. Richard, E. M., Diefendorff, J, M., Martin, J, H., (2006). Revisiting the WithinPerson Self-Efficacy and Performance Relation. Human Performance, 19(1), 67-87. Rosenfeld, M. Sheppard, P. K., & Sheppard, K. (2004). Investigating the Validity of TOEFL: A Feasibility Study Using Content and Criterion Relater Studies. TOEFL Research Report, 3 (18), 71. 77

Shameem, N., (1998). Validating self-reported language proficiency by testing performance in an immigrant community: the Wellington Indo-Fijians. Language Testing, 15(1), 86-108. Snow, S. M. (1998). Economic, Statistical, and linguistic factors affecting success on the test of English as a foreign language (TOEFL). Information Economics and Policy, 10 (2), 159-172. The Society for Testing English Proficiency (STEP). (2004). Stricker, L. J. (2004). The performance of native speakers of English and ESL speakers on the computer-based TOEFL and GRE General Test. Language Testing, 21 (2),146-173. Tabachinick, B. G., & Fidell, L, S. (2001). Multivariate Statistics (4th). Needham Heights: Allyn and Bacon. Tabachinick, B. G., & Fidell, L, S. (2007). Multivariate Statistics (5th). Needham Heights: Allyn and Bacon.

78

BIOGRAPHICAL INFORMATION

Yasushi Kyutoku earned a high schools diploma from Seifu High School (Osaka, Japan) in March, 1992, and earned B.A. in psychology from the University of Texas at San Antonio in December, 2001. After the completion of thesis research, he will continue to study experimental psychology at the University of Texas at Arlington. He is interested in continuing to conduct researches related to psychometrics such as validation and application of scales.

79

Vous aimerez peut-être aussi