Synthesis

To summarize, this study incorporated three different aspects of co-teaching which in past literature has not been conducted. Although the presence of FTAs has been studied individually (Yen, 2016), motivation towards language learning as a single aspect has been researched (Ushioda, 1996), listening anxiety and listening performance has been examined on its own (Elkhafaifi, 2005), rarely does a study include all three aspects. Furthermore, most of these studies have not been examined longitudinally.

Due to the research gap identified above, a study that included all aspects of the motivation, listening anxiety, and listening performance of junior high students in a native-teacher-led co-teaching context would be important, which was the intent of the present study. More specifically, the study firstly examined the differences in the levels of English-learning motivation and listening anxiety after a year. Then, it focused on how learners’ English listening performance changes while they are exposed to FTAs.

CHAPTER 3 METHODOLOGY

This chapter comprises five parts, each of which elaborated different aspects concerning the research. The first section introduces the design of the study. The second section describes the background information of the participants. In the third section, the instruments used in this study are presented. The fourth section outlines the procedures of this study. The final section illustrates the measures used to analyze data.

3.1 Research Design

This study employed both quantitative and qualitative approaches to answer the research questions. According to Cresswell (2013), research combing quantitative and qualitative approaches can facilitate a better understanding of the phenomena under study than one using a single approach. Qualitative method was used as a

supplementary approach to support the quantitative analysis. In this study, a

questionnaire adapted from The Attitude Motivation Test Battery (AMTB), Foreign Language Classroom Anxiety Scale (FLCAS), and Foreign Language Listening Anxiety Scale (FLLAS) were served as the main quantitative instrument to examine participants’ motivation towards English learning and listening anxiety with the presence of FTAs in the co-teaching context. The participants were also required to take a listening test designed by the researcher to further investigate the listening performance. To gain more specific information toward learners’ English-learning motivation, listening anxiety and the influence of FTAs on learners, qualitative

methods like classroom observation, interviews, and informal conversations were also employed. These qualitative methods were hoped to acquire more in-depth

information from the participants’ questionnaire, such as learners’ learning situation

influenced by the FTA, classroom interactions with the FTA and peers, and learners’

motivation and listening anxiety towards different lessons or particular topics etc.

(Please see Figure 1.)

Figure 1: The Diagram of the Research Design

3.2 Participants

The study involved 105 seventh graders (ranging from 12~13 years old) selected from four classes of two junior high schools in the northern Taiwan which attended the English co-teaching program in 2015. The participants were novice learners with low English proficiency in a FTA-led teaching context. They had only one co-teaching class period (40-45 minutes) once a week. Learners were required to fill in a questionnaire (combining motivation and listening anxiety scales) with a total of 26 items at the beginning and end of the school year. To avoid any misunderstandings or language barriers, the questionnaire was translated into learners’ native language. The questionnaire aimed to capture learners’ perceptions on not only learning motivation

The Impact of Foreign Teaching Assistants on English Language Learners’

Motivation, Listening Anxiety, and Listening Performance

Quantitative approach Qualitative approach

Classroom

observations Interviews Informal conversations

Field notes

&

Recordings Questionnaire Listening test

but also listening anxiety with the presence of FTAs in the co-teaching classes.

Moreover, all the participants were required to take a listening test at the beginning and end of the school year, aiming to keep track of their learning progress in listening after being exposed to native speaking teachers in English in classes. However, only 83 participants did return valid questionnaires and take listening tests at the beginning and end of the school year. A total of 83 reliable data received from these participants was further used in quantitative analysis. They included 35 (42.2%) male students and 48 (57.8%) female students.

To further investigate the learners’ motivation and listening anxiety change, classroom observations and formal interviews were done to explore the specific impact FTAs have on learner motivation and listening anxiety. Informal conversations immediately before and after the class were conducted to learn more about learners’

perceptions. Lastly, a total number of twelve students were chosen for interview. They were four students of the active group, four students of the quiet group, and four students of the asking for teachers’ help group. More detailed information about the selection of the three groups of students was presented in the section below. These interviewees were coded as pseudo names. Anna, Alex, Ben, and Bonnie were students of the active group. Cathy, Colin, Dave and Doris were students of the quiet group. Elsa, Edward, Felix and Flora were students of the asking for help group.

3.3 Instruments

In response to the quantitative and qualitative design of the study, three distinct instruments were adopted. For quantitative data, questionnaire, on the motivation and listening anxiety, and listening test were utilized. Qualitative data were obtained through classroom observations, interviews, and informal conversations. Diverse instruments allowed the researcher to triangulate quantitative and qualitative data so that the impact of FTAs on English learning was more specifically investigated.

3.3.1 Quantitative data 3.3.1.1 Questionnaire

Combining the two sections adapted from The AMTB and FLCAS as well as FLLAS, the questionnaire was designed to probe into learners’ perceptions of learning motivation and listening anxiety with the presence of the FTAs in co-teaching context.

The formal questionnaire (see Appendix A) contained three parts: (a) background information, (b) motivation scale (from No.1 to No.20 items), and (c) listening

anxiety scale (from No.21 to No.26 items). The background information was designed to indicate their demographic data (e.g., class, number, name, gender) and relevant language learning history, such as years of learning English and previous learning experience with native-speaking English teachers.

The motivation scale used in this study consisted of an adapted The Attitude/Motivation Test Battery (AMTB) established by Gardner (1985). The AMTB, originally composed in English and made up of over 130 items, has been adapted for many learning contexts all over the world (Dornyei, 2010). One modification of the Chinese version of The AMTB was done by Yen (2016) for the purpose of investigating the impact of FTAs on Taiwanese learners’ motivational changes toward English learning. The motivation scale in her study consisted of 20-items and had an estimated internal reliability of .894 (Crobach’s Alpha), which shows a high level of internal consistency among the 20 items. As the motivation scale in Yen’s (2016) study fitted the context of this study, the researcher used it as a tool to examine the motivation of the learners as they were taught by FTAs.

In addition, to elicit the participants’ anxiety toward listening, this study adapted from two commonly used listening anxiety questionnaires and the statements were generally taken from each of the questionnaire in order to form the listening anxiety scale employed in this study. One was from Horwitz, Horwitz and Cope’ (1986)

Foreign Language Classroom Anxiety Scale (FLCAS), consisting of 33 items. The FLCAS is considered the most prestigious questionnaire measuring learners’

perceptions of language anxiety in foreign classroom setting and is modified by various researchers (Phillips, 1992; Saito et al, 1999; Cheng et al., 1999). Because the participants in this study usually listened to FTAs’ English-medium instruction, out of 33 items, only one item in the FLCAS was chosen which directly measuring whether the participants suffered from listening anxiety toward the spoken target language messages.

Another questionnaire was from Elkhafafi’s (2005) Foreign Language Listening Anxiety Scale (FLLAS), consisting of 20 items. The FLLAS is recognized as a reliable questionnaire to specifically measure the FL learners’ perceptions of listening anxiety and have been used by many researchers (Cheng, 2007; Gonen, 2009; Xu, 2013; Zhang. 2013). Out of 20 items, 4 items in the FLLAS were adopted because they matched the participants’ learning situation that being exposed to the spoken language provided by FTAs and usually receiving a lot of input.

To meet the needs of the participants in this study, five items selected from FLCAS and FLLAS were modified. Modifications were as follows:

(a) One item (i.e., statement 4) in the FLCAS was adapted in this study. For this study, the words “the teacher” and “foreign language” were replaced by

“FTA” and “English” respectively. That is, statement 4: “It frightens me when I don't understand what the teacher is saying in the foreign language”, was altered as “It frightens me when I don’t understand what the FTA is saying in English.”

(b) As for the four items (i.e., item 2, 3, 14, 15) in the FLLAS, the words

“Arabic” and “the speaker” were replaced by “English” and “the FTA”

respectively. For example, statement 2: “When I listen to Arabic, I often

understand the words but still can't quite understand what the speaker is saying”, was altered as “When I listen to English, I often understand the words but still can't quite understand what the FTA is saying.” In addition, since the aim of this study was to explore students’ own perception of listening anxiety, the word “you” was replaced by “I.” Statement 14, for instance, “Once you get used to it, listening to Arabic is not so difficult” was changed into “Once I get used to it, listening to English is not so difficult.”

Furthermore, the researcher of this study added one statement related to listening test anxiety in the scale, that is, “I am usually at ease during listening tests in my English class.” Because the participants needed to take the listening section in CAP in the future and taking listening tests is recognized as a source of listening anxiety for FL learners (Chang, 2008), it would be worthwhile to examine whether the learners’

anxiety level at listening test would be reduced after being exposed to FTAs in English for a long term.

Last but not least, the researcher of this study classified the total of 6 items into two categories: anxiety level while listening to English spoken language and self-perception of English listening, as presented in Table 1. To ensure that the participants Table 1

Statement Types in the Listening Anxiety Scale

Category Example Statement

number Anxiety level while

listening to English spoken language

The hardest part of learning English is learning to understand spoken English.

No.21, 23, 25

Self-perception of English listening

Once I get used to it, listening to English is not so difficult.

No.22, 24, 26

could understand listening anxiety scale with a total of 6 items, it was translated into Chinese. A pilot study was held to ensure the consistency and the reliability of 6 items on the listening anxiety scale used in the current study. The pilot study involved sixty-one students, who have attended an English class with the presence of an FTA. No statements were needed to be revised and the reliability was acceptable. More detailed elaboration of the pilot study is presented in the section below.

3.3.1.2 Listening test

Since this study was also interested in how FTAs contributed to learners’

performance in listening, a listening test to identify their ability was needed. In the academic assessment field, validity is defined as the extent to which an assessment accurately measures what intends to measure (Carmines & Zeller, 1979). It was also the criteria for the researcher to choose a valid listening test that could appropriately measure what learners have learned, that is, daily spoken English taught by FTAs. The listening test in this study was modified by standardized tests as references, such TOEFL Junior tests and Cambridge English: Young Learners (YLE) tests, with the aim of measuring the progression of the learners on listening performance so that the impact of FTAs on listening ability could be more specifically examined.

The thirty-minute listening test was served as the first and the second test in this study. It was a tape-based aural test composed of thirty questions. For example, one of the questions, students were required to draw a line to match the conversation heard to the correct picture. The scripts were recorded in advance by one male and one female graduate student who majored in TESOL. All the instructions in the test were given in both English and Chinese. Questions were written and spoken in English. The 30 questions were stated twice.

3.3.2 Qualitative approach 3.3.2.1 Classroom Observations

In order to understand how learners’ motivation and listening anxiety would change over time with the presence of FTAs in the co-teaching context, classroom observations were carried out at least once every month for every school and for one period of class (i.e., 45 minutes). Classroom observations could give the researcher more insights into quantitative results. For example, the participants reported they enjoyed FTAs’ classes in the questionnaire (Statement 12: I feel that classes with foreign teachers are more interesting and fun.). The phenomena found from classroom observations, such as a relaxing classroom atmosphere through using authentic

materials and the students’ intentional opposite responses or jokes to the FTA, were the possible reasons for the students’ positive perception of the FTAs’ class to the questionnaire.

In addition, in response to the research questions, the researcher paid more attention on interaction between the FTAs and learners, learners’ motivation towards English learning, learners’ reactions towards topic lessons, learners’ listening anxiety, and kinds of the support or assistance the FTAs provided to learners, etc. The

classroom observations were audio-recorded. The recordings were transcribed after class.

3.3.2.2 Interviews and Informal Conversations

In order to clarify questions found in classroom observations, and to obtain further information on learners’ English-learning motivation and listening anxiety, semi-structured interviews, in which the interviewer prepared a list of questions in advance with the flexibility of adding more questions when needed (Weiss, 1995), were also used in this study. The list of interview questions is in Appendix B. The interviewee’s selection was based on separating students into each of the three categories. The reason was that during the classroom observations students in each school could be broadly dived into three types of behaviors: active, quiet, and asking

for teachers’ help. The active students were often talkative and voluntarily responded to FTA’s questions. Those who were shy and often not involved in the class were considered to be the quiet type. The students who often asked FTAs to check his/her answers during the activity were grouped as the third type of students labeled asking for teachers’ help. Based on the behaviors, students were put into one of the three groups according to the researcher’s own field work. Out of each group, students were selected based on FTAs’ recommendation of the person to be interviewed and the balance of gender. Finally, six students, who were active, quiet and often asking for teachers’ help in the class from each school were interviewed formally and informally outside the classroom. The researcher recorded all the interviews and analyzed the interview data in a coding process. First, the recordings were transcribed. Next, they were grouped together based on different questions. Then, content analysis was done to find out certain patterns.

Besides, the researcher made use of informal conversations before and after class. Informal conversations before the class were used to know learners’ English learning process in daily lives and their feelings of FTAs’ instruction. As for the informative conversations after the class, they were utilized to understand learners’

perceptions toward an incident at specific moments. Learners were asked to reflect upon what they were thinking and why they reacted that way. Since speech was quite transient, informal conversation were recorded and transcribed afterwards.

3.3.2.1.1 Field Notes and Audio Recordings

Field notes were taken during classroom observations, interviews, and informal conversations for the purpose of having a more detailed and rich description of learners’ learning context, such as the FTA’s instructional practices and classroom atmosphere. Learners’ interaction with FTAs and peers as well as nonverbal reactions such as facial expression, and gestures were also recorded in the field notes. Field

notes were not only served as a reminder of topics that could be discussed with learners after class but also provided more insight into what have had an impact on learners’ motivation towards English learning and listening anxiety.

The researcher used a digital recorder. All classroom observations, interviews, and informal conversations were recorded once the participants’ permission was granted.

3.4 Data Collection Procedures

Quantitative data were collected through the first and the second questionnaires and listening tests, while qualitative data were gathered through classroom

observation, interviews, informal conversations, field notes, and audio recordings.

At the preparation stage, original Foreign Language Classroom Anxiety Scale (FLCAS) developed by Horwitz, Horwitz and Cope’ (1986) and Foreign Language Listening Anxiety Scale (FLLAS) designed by Elkhafaifi (2005) were generally adapted considering the focus of the study and the learners’ background. In order to ensure the reliability and validity of the modified listening anxiety scale, a pilot test was administered to sixty-one students, who have attended an English class with the presence of an FTA. The modified listening anxiety scale using Cronbach’s alpha coefficient reveals high internal consistency (α=.705). Please refer to Table 2.

Table 2

Internal Consistency (Cronbach’s alpha) of the Adapted FLCAS & FLLAS Questionnaire in Pilot Study

Questionnaire No. of items Cronbach’s α coefficient (No. of responses) Adapted FLCAS &

FLLAS

6 .705 (61)

The result showed that a high level of internal consistency among the 6 items and

confirmed that this adapted FLCAS and FLLAS scale was reliable. The listening test was also completed at this stage. The listening test and CD were compiled at the same time.

At the beginning of the school year, the first questionnaire (motivation and listening anxiety scale) for the learners was given to the two designated schools, where co-teaching program was implemented. FTAs helped to distribute the questionnaire in the four co-teaching classes and the questionnaire took learners approximately ten minutes to finish. In the same month, the participants would have to take the first listening test so as to investigate their listening performance with the presence of FTAs. Learners would not be allowed to read the content of the listening test until the initiation of the test. The listening test were conducted in the whole class and lasted approximately 30 minutes. During the test, the students were asked to follow the instructions given by the speakers from the CD.

During the school year, the researcher did periodic classroom observations, took field notes, and audio-recorded the class with the permission of teachers (both FTA and LET) and school administration. Also, informal conversations with learners would be done before and after class. The former was used to get information about learners’ daily English learning situation and their perceptions of FTAs’ instruction;

the latter was used to clarify anything interesting or confusing incidents in the class and to avoid researcher’s over-interpretations of students’ in-class responses or reactions

At the end of the school year, the participants would be required to complete the same questionnaire and listening test, which also served as the second questionnaire and the second listening test of the study. In addition, six students, who were active, quiet and often asking for teachers’ help in the class, from every school observed by

At the end of the school year, the participants would be required to complete the same questionnaire and listening test, which also served as the second questionnaire and the second listening test of the study. In addition, six students, who were active, quiet and often asking for teachers’ help in the class, from every school observed by