The present study adopted a quantitative method to explore firstly the correlation among listening performance, listening anxiety, and tolerance of ambiguity and secondly whether listening anxiety or tolerance of ambiguity could be a significant predictor for one’s listening comprehension performance for students at high school level. In this study, a correlational design was adopted. Elaborations of participants, instruments, procedures, and data analysis of the study are detailed in this chapter.
Participants
In the current study, a total of 234 eleventh grade students, coming from three social science major classes and four science and engineering major classes in a municipal high school in Taipei, Taiwan were recruited, for they were considered to have enough learning experiences to represent students at high school level and assumed to have more willingness to fill out the questionnaires compared with their senior counterparts. All the participants were given the chance to choose whether to voluntarily participate in the study, and they were further assured that their response in the questionnaires as well as results of their listening test for the current study would not influence their final grades for the English course.
Instruments
The instruments employed in the study include a listening comprehension test, a background questionnaire, a listening anxiety questionnaire, and a tolerance of ambiguity questionnaire.
Listening Comprehension Test
In the current study, a published listening test on the intermediate level of General English Proficiency Test (GEPT) was employed to assess the participants’ listening
29
performance. The difficulty of the GEPT is divided into four levels—elementary, intermediate, high-intermediate, and advanced. According to the test developer and publisher, Language Training and Testing Center (LTTC), the wording in the intermediate test is based on a word list designed for senior high school students, in which approximately 5,000 words are listed. The listening test (see Appendix A) contains 45 questions in multiple choice format, and the test is divided into three types of tasks: Picture Description (15 questions), Answering Questions (15 questions), and Conversations (15 questions). The entire test lasts for approximately 30 minutes. In Part One, Picture Description, test takers are required to listen to four descriptions and choose the best one that corresponds to the given picture. In Part Two, Answering Questions, test takers are asked to choose the best response after listening to a statement or a question. In Part Three, Conversations, test takers need to choose from four written options to answer a question based on the conversation they hear. In a formal
intermediate listening test, based on the number of questions that are correctly answered, the score is converted with the use of a measuring scale, whose maximum score is 120. Yet as the measuring scale is left unpublicized, in the current study, the maximum score of the test was decided to be 45 (one point for each question).
Background Questionnaire
The background questionnaire (see Appendix B) consists of five questions. In addition to basic personal information and gender, the questionnaire also includes questions to understand the participants’ years of English-learning experience, the most difficult skill they considered to be among the four English language skills, their use of supplementary listening materials, and the hours they spent weekly practicing English listening.
Listening Anxiety Questionnaire
30
The current study adopts Cheng’s (2017) Second Language Listening Anxiety Scale (SLLAS) for measuring participants’ listening anxiety (see Appendix C). The SLLAS is composed of nine questions, covering three aspects of learners’ listening anxiety (i.e., cognitive anxiety, somatic anxiety, and avoidance behavior). The
participants were asked to respond to each statement in the format of five-point Likert-scale, from Strongly Disagree (1 point) to Strongly Agree (5 points). Participants’
scores achieved in this scale can range between 9 and 45. A higher score indicates a higher level of listening anxiety. According to Cheng (2017), the construct validity of the SLLAS was verified with the use of confirmatory factor analysis, which yielded fit indexes that met the recommended criteria (χ2/df = 2.47; SRMR=.08; RMSEA = .09;
CFI = .96; GFI = .93). The reliability of the SLLAS was also satisfactory (Cronbach’s alpha = .84; CRs of the three dimensions of SLLAS = .80 to .86). The original Chinese version of the SLLAS was used to collect data.
Tolerance of Ambiguity Questionnaire
In the present study, the Multiple Stimulus Types Ambiguity Tolerance scale (MSTAT-I) by McLain (1993) was used to measure the participants’ tolerance of ambiguity (see Appendix D). MSTAT-I contains 22 items that comprehensively examine features of stimuli and an individual’s reaction to perceived ambiguity.
According to McLain (1993), while the validity of the scale was pursued by relating MSTAT-I scores to measures of willingness to take risks, cognitive complexity, dogmatism, and receptivity to change, the factor analysis of the scale supported a single-dimensional solution, namely, general tolerance of ambiguity. A five-point response format, from Strongly Disagree (1 point) to Strongly Agree (5 points), was adopted. This scale yields scores ranging between 22 and 110, with higher scores indicating a higher level of tolerance of ambiguity. As stated by McLain (1993), the
31
convergent validity of MSTAT-I was established with its significant positive
correlations with other TA scales, including Budner’s (1962) 16-item scale, Storey and Aldag’s (1983) 8-item scale, and MacDonald’s (1970) 20-item scale. The coefficient alpha (α = .86) also lent support to the internal consistency of MSTAT-I. To facilitate students’ understanding of the items, the MSTAT-I was translated into Chinese by the researcher. Two senior high school English teachers were invited to confirm the accuracy and adequacy of the translation.
Procedures
As shown in Figure 3, which presents the flow chart of the research process, the listening test and all the questionnaires, including the background questionnaire, the SLLAS, and the MSTAT-I were given within one class session. In the fifty-minute class session, thirty-five minutes were first distributed for the GEPT intermediate listening test, and the rest of the fifteen minutes were allocated for the questionnaires. Before filling out the questionnaires, the participants were ensured that their responses would be kept confidential and that there would be no particular answers to the questions and statements given in the questionnaires. All of the questionnaires were collected and typed into the computer by the researcher for statistical analyses.
32
Figure 3. The Procedure of Research Process
Data Analysis
To answer the first research question, Pearson product-moment correlation was used to examine the correlation between learners’ listening comprehension ability and the two learner factors, namely, listening anxiety and tolerance of ambiguity. For the second research question, multiple regression analysis was applied to explore the relative contributions among the two learner factors to the listening comprehension test score. The correspondence of research questions and the instruments as well as analysis is indicated in Table 4.
Data Collection
GEPT Listening Test (Intermediate Level)
<35 minutes>
Questionnaires
<15 minutes>
Background Questionnaire
Listening Anxiety Questionnaire
Tolerance of Ambiguity Questionnaire
Data Analysis
1. Descriptive Statistics
2. Pearson Product-moment Correlation 3. Stepwise Multiple Regression
33
Table 5
Research Questions and the Corresponding Instruments and Analysis Research
Questions Instruments Analysis
Q1 1. GEPT intermediate level listening test
2. Questionnaires
Pearson product-moment correlation
Q2 multiple regression analysis
Note. Q1=What is the relationship of English (FL) listening performance to English (FL) listening anxiety and tolerance of ambiguity among high school learners?
Q2=Which of the two variables, listening anxiety and tolerance of ambiguity, can serve as a significant and better predictor of one’s English (FL) listening
performance for learners at high school levels?
34