The present study was in an attempt to explore among the three factors, listening anxiety, listening self-efficacy, and metacognitive awareness of listening, which factor contributed most to English listening performance for students at two different
education levels (i.e. junior high school level and senior high school level). The research design of the current study was a correlational design using the survey approach. Elaborations of participants, instruments, procedures, and data analysis of the study are presented in this chapter.
Participants
Students from a public junior high school and a public senior high school in western Taiwan were selected as the participants in the current study. For one thing, an English listening comprehension test has been incorporated into the nation-wide senior high school and college entrance exams that the majority of the two groups of learners will take upon graduation, highlighting the importance of probing into these two groups of students’ listening comprehension performance. For another, there is a lack of studies in Taiwan that simultaneously examine the relationship among English listening anxiety, English listening self-efficacy, metacognitive awareness, and
English listening performance across the two education levels.
Table 1 shows the participants’ background information. Based on Table 1, the participants of the study were 121 eighth graders from normal class grouping classes and 123 eleventh graders from science and engineering major classes. More
specifically, students at the eighth and eleventh grades were chosen because the results obtained from them were assumed to be more representative of the education level they belonged to than their freshmen counterparts (i.e., the seventh and tenth
37
graders). They were also assumed to be more willing to respond to a survey than their senior counterparts (i.e., the ninth and twelfth graders) who were under great pressure to devote all of their time to preparing for entrance exams.
Table 1
Participants’ Background Information
Items Junior High Group Senior High Group
Numbers 121 123
Grade level 8 11
Class background normal class grouping science & engineering major
All the participants were informed that their responses to the questionnaire and their performance on the listening test designed for this study would not affect their course grades. They could also decide to take part in the study or not voluntarily.
Instruments
The instruments used in the current study for data collection included two English listening comprehension tests (respectively for junior high and senior high school students), a background questionnaire, a listening anxiety questionnaire, a listening self-efficacy questionnaire, and a listening metacognitive awareness questionnaire. For the listening self-efficacy and listening metacognitive awareness questionnaires, the researcher used a modified Chinese translated version. The original items from the English versions were translated into Chinese by the
researcher and proofread by a professor of English to check the accuracy, the clarity, as well as the comprehensibility. The listening anxiety questionnaire was originally written in Chinese, so it was directly adopted. For consistency, the commonly used 5-point Likert scale was adopted for these three questionnaires in the current study. A
38
pilot study was held to ensure the comprehensibility and the reliability of the three questionnaires (i.e., listening anxiety questionnaire, listening self-efficacy
questionnaire, and listening metacognitive awareness questionnaire). The pilot study involved forty-two 8th graders and forty-six 11th graders respectively from a public junior high school and a public senior high school in western Taiwan. No questions or doubts were expressed from the students during the pilot test and the reliability was acceptable. More detailed elaboration of the five instruments is presented in the sections below.
Listening Comprehension Tests
The participants’ listening performance was assessed by a published listening test on the General English Proficiency Test (GEPT), a widely recognized English test in Taiwan. There are several levels of listening tests in GEPT. According to GEPT, the elementary level test is designated based upon the 2000 basic words for junior high school learners, while the wording in the intermediate level test is based on the word lists designed for senior high school students. Thus in the current study, an elementary level listening test (see Appendix A ) was distributed to the junior high school
participants and an intermediate level listening test (see Appendix B) was utilized for the senior high school participants. Each of the listening tests, whose items were all in multiple choice format, contains three common parts: Picture Description, Question or Statement Response, and Short Conversation. Besides, the elementary level test contained an additional section, Short Talks. That is, the elementary level test contained four parts while the intermediate level test contained three parts.
The elementary level test contained 30 items in total. The first part (5 items), Picture Description, asks listeners to choose from three short oral descriptions in accordance with the picture they see. The Question or Statement Response part (10
39
items) asks listeners to choose an appropriate response from three written options after listening to a sentence. Short Conversation, containing 10 question items, asks listeners to select the best response from three written options after listening to a short dialogue and a follow-up question. The last part, Short Talks (5 items), tests whether the listeners could precisely choose the picture after listening to a short talk. The 30 listening questions in total would last for approximately 20 minutes. And the total score of the elementary level test is 30 (one point for each question).
As for the intermediate level test, each of the three parts (i.e., Picture
Description, Question or Statement Response, and Short Conversation) contains 15 items. In Part One, Picture Description, there are a picture and four oral descriptions for each test item. The test takers are required to choose the best description in accordance with the picture. In Part Two, Question or Statement Response, the test takers would hear a question or a statement, and then they are asked to choose the most appropriate response from four written options. In Part Three, Short
Conversation, the test takers are required to answer a question after hearing a short conversation. From four written options given for each item, the test takers have to choose the best answer that contains the most appropriate response to the question they hear. The total 45 listening questions would last approximately for 30 minutes.
And the total score of the intermediate level is 45 (one point for each question).
Background Questionnaire
The background questionnaire (see Appendix C) is made up of seven questions about the participants’ demographic information. Basic demographic information such as gender and grade level was enquired in the questionnaire. In addition to basic information, the questionnaire also probes into the participant’s year of learning English as a foreign language, perceptions of most difficult skill among the four
40
English skills (i.e. listening, speaking, reading, and writing), use of supplementary listening material, and English listening practice hours on a weekly basis.
Listening Anxiety Questionnaire
Cheng’s (2014) Second Language Listening Anxiety Scale (SLLAS) served as the listening anxiety scale in the current study (see Appendix D). The SLLAS
addresses three dimensions of learners’ listening anxiety (i.e. cognitive, physiological, and behavioral) and contained nine questions presented in 5-point Likert scale. The participants recorded their responses to each statement from Strongly Disagree (1 point) to Strongly Agree (5 points). The highest score of the SLLAS in the present study is 45 (5 points x 9 items). Based on the pilot test, the Cronbach’s alpha of SLLAS for junior high school students was .83 and that for senior high school students was .85, both lending support to the reliability of SLLAS.
Listening Self-efficacy Questionnaire
The modified Chinese versions of listening self-efficacy questionnaire (LSE), were used in the current study to assess the participants’ listening self-efficacy. The original version of LSE was developed by Yan (2012) and was used with reference to another listening comprehension test named College English Test 4 (CET4). The test CET4 was replaced with the test GEPT in the adapted Chinese versions of LSE in current study. Apart from translation into Chinese, modifications were also made in the statement of each section according to the test items of GEPT. Items varied based on different levels of GEPT, thus two modified Chinese versions of LSE were made to meet the requirement of the two education levels.
For GEPT elementary level LSE version (see Appendix E), modifications were made in each statement as follows. In section A, the statement “In terms of short conversations in College English Test 4, how sure…” was replaced with “ In terms of
41
the first section “Picture Description”, how sure…”. In section B, the statement “In terms of long conversations in College English Test 4, how sure…” was replaced with
“In terms of the second section Question or Statement Response, how sure…”. The statement “In terms of short passages in College English Test 4, how sure…” in section three was replaced with “In terms of the third section Short Conversations how sure…”. As for the fourth section, the statement “In terms of Compound
Dictations in College English Test 4, how sure…” was replaced with “In terms of the fourth section “Short Talks”, how sure…”. For GEPT intermediate LSE version (see Appendix F), other than deletion of the last section, modifications were made exactly the same as GEPT elementary LSE version.
Both LSE versions were on a 5-point Likert scale and the participants were asked to indicate how confident (1 = Almost Not Sure to 5 = Completely Sure) they were in accomplishing each part of the listening questions on GEPT from four perspectives (understanding main ideas, understanding details, understanding the meaning of unknown words, and catching the keyword). Based on different levels of GEPT tests, the LSE with 16 items was designed for the junior high school participants, whereas the LSE with 12 items was developed for the test-takers at senior high school level.
The highest scores of the LSE in the present study were 80 (= 5 points x 16 items) for junior high school level and 60 (= 5 points x 12 items) for senior high school level.
According to the pilot test, the Cronbach’s alpha of the modified LSE for junior high school students was .82 and that for senior high school students was .87. Both statistics were more than satisfactory.
Listening Metacognitive Awareness Questionnaire
The modified Chinese version of Metacognitive Awareness Listening
Questionnaire (MALQ, see Appendix G), originally developed by Vandergrift et al.
42
(2006), was utilized in the present study to assess the participants’ metacognitive awareness. According to Vandergrift et al. (2006), the questionnaire was aimed to assess five factors (i.e. problem-solving, planning and evaluation, mental translation, person knowledge, and directed attention). Apart from the translation into Chinese, there was no other modification in the current modified version of MALQ except replacement of the word “French” with “Chinese” in the items to suit the EFL
environment in the present study. The questionnaire also adopted 5-point Likert scale and the participants were asked to record their responses to each statement from Strongly Disagree (1 point) to Strongly Agree (5 points). According to the pilot test, the Cronbach’s alpha of the modified Chinese version of MALQ for junior high school level was .83 and that for senior high school level was .88. As there were 21 items in the questionnaire, the highest score for the MALQ in the present study was 105 (5 points x 21 items). Besides, item sixteen is negatively-worded so that the score for this statement should be reversely calculated.
43
Procedures
Figure 1 presents a flow chart of the research process, followed by descriptions of the procedures.
Figure 1. Flowchart of the Procedure
In the formal study, all the participants were required first to take the listening test and then fill in the questionnaires, including the background questionnaire, the SLLAS, LSE, and the MALQ within one class session. For junior high school
students, twenty-five minutes were scheduled for the GEPT elementary level listening test and then twenty minutes for the questionnaires. As for the senior high school participants, thirty-five minutes were distributed to the GEPT intermediate level listening test and then fifteen minutes for the questionnaires.
Data Analysis
Table 2 presents the research questions in the current study and shows the corresponding instruments and analysis to each question. As can be seen in Table 2,
Fill in the questionnaires (20 min.)
Take the elementary level listening test (25 min.)
Take the intermediate level listening test (35 min.) Junior High Group
(Class period: 45 min.)
Senior High Group (Class period: 50 min.) Procedures
Fill in the questionnaires (15 min.)
44
Pearson product-moment correlation was used to answer research question 1 and research question 2 by examining whether there is a significant correlation between the listening comprehension ability and the three learner factors. Multiple regression analyses was computed to answer research question 3 via finding out the relative contributions among the three learner factors to the listening comprehension test score.
Table 2
Research Questions and the Corresponding Instruments and Analysis Research
Questions
Instruments Analysis
Q1
GEPT elementary level listening test Questionnaires
Pearson product-moment correlation coefficient
Q2
GEPT intermediate level listening test Questionnaires
Pearson product-moment correlation coefficient
Q3
GEPT elementary level listening test GEPT intermediate level listening test Questionnaires
multiple regression analyses with the Statistical Package for Social Science (SPSS) Notes. Q1 = What is the relationship of English (FL) listening performance to English (FL) listening anxiety, listening self-efficacy, and listening metacognitive awareness among junior high school learners?
Q2 = What is the relationship of English (FL) listening performance to English (FL) listening anxiety, listening self-efficacy, and listening metacognitive awareness among senior high school learners?
Q3 = Of the three learner factors, English (FL) listening anxiety, English (FL) self-efficacy, and English (FL) metacognitive awareness, which one is the best predictor of English (FL) listening performance for learners at junior high and senior high levels respectively?
45
CHAPTER FOUR