Instrument and Procedures

Data were collected through the following two ways:

(1) Questionnaire Survey

A questionnaire was developed to investigate the perceived professionalism of

elementary school NETs and NNETs in Yilan and Hsinchu (see Appendix II). The overall content of the questionnaire drew on Brown’s (2001) good language-teaching characteristics (see Appendix I), “TESOL/NCATE Program Standards” (TESOL, 2003, see Figure 1 in the literature review section), Tsui’s conceptual framework on teacher knowledge and expertise (see Figure 2 in the literature review section), Arva and Medgyes’ interview sheet (2000, p.

371, see Appendix III), and other relevant literature.

The questionnaire contains the following four groups of questions:

1. Basic information of the participants, including their personal (such as gender or age) and professional background (for example, their highest degrees and specialization).

2. Perceived criteria/standards for professional English teachers in regard to the English language

3. Perceived criteria/standards for professional English teachers in regard to the classroom instruction

4. Perceived criteria/standards for professional English teachers in regard to the local culture and context

These four groups of questions represent four interrelated domains which constitute the perceived professionalism of both NETs and NNETs. Among the four, “Language” and

“Instruction” are the major domains, and “Personal & Professional” as well as “Culture &

Context” are the minor domains. Each domain took up 4 to 12 questions in the questionnaire and altogether there were 40 questions.

Under the language domain, for example, the importance of a professional English teacher’s proficiency of the English language was checked, the interrelationship between the English language teachers’ and students’ L1 (Chinese) and L2 (English) were also asked, as well as how these languages were used and perceived at home and in the larger society in Taiwan. Under the instruction domain, various pedagogical concerns were investigated, including classroom interactions, how instructional goals were achieved through the use of language (including both L1 and L2). Under the culture and context domain, for example, the questions included the role of the local culture and educational system, such as the understanding of the micro-level culture (common practice and norms) of learning and

teaching in the classroom, the pedagogical, curricular and administrative culture of the school, as well as the macro-level culture (beliefs and values) of the larger society in Taiwan.

The questionnaire took an on-line form and it was in English for both NETs and NNETs to avoid possible translation discrepancies. The on-line questionnaire was first pilot tested by elementary school ELT teachers and university professors/researchers in the field of TESOL. The website was given to these teachers and the professors/researchers. The feedback was collected from the on-line version, oral interviews in person or by telephone, and through e-mail messages. Based on the rich feedback and extensive discussions among the research team members, the questionnaire had gone through multiple revisions before it was finalized. The website was then given to the Yilan and Hsinchu NETs and NNETs.

After 6 weeks of time, the website was closed and the data were analyzed.

(2) Interviews

Interviews will be conducted with willing NETs and NNETs based on the response to the questionnaire. A note at the end of the questionnaire extended an invitation to

respondents who were interested in being interviewed afterwards. They were asked to give their contact information to indicate their interest. A 20 to 40 minute semi-structured interview was conducted in a face-to-face setting depending on feasibility and the

respondent’s preference. The purpose of the interview is to get information which could not be derived or was not clear from the questionnaire. Issues of NETs and NNETs and

professionalism will be further pursued. Arva and Medgyes’ interview sheet (2000, p. 371, also see Appendix III) was serve as an initial reference and was revised extensively.

Permission was asked to audio-record the interview and the audio-recorded interviews were transcribed within a week and analyzed afterwards.

4. Analysis

(1) Data grouping: By NET/NNET

Due to the small sample size (n=53 in total, including 31 NETs and 22 NNETs), the questionnaire data were analyzed with descriptive statistics to understand the trends and patterns in perceived professionalism of NETs and NNETs. The data were analyzed as a group, and then by NET/NNET. Further analyses were conducted by education level and academic specialty (e.g., English major vs. non-major), by geographical area, by teaching credentials and experience, by gender, and so forth, to see how various factors interact with the native vs. non-native distinction. Based on the survey data, the interviews pursued the information not available from the questionnaire. Results of the questionnaire survey and the interviews were cross-verified and examined; points of convergence and divergence between the questionnaire responses and the interview exchanges were further analyzed.

(2) Coding

The first part of the questionnaire, i.e., Questions 1-12, was personal information of the

respondents, including their age range, education (e.g. highest degree and specialty),

language background and language use (e.g., percentage of language instruction), co-teaching experience, and so forth (see Appendix II). Questions 13 to 40 were the second part of the questionnaire concerned the professionalism, i.e., qualities of good ELT teaching.

Respondents were asked to rate the qualities using a 5-level scale, and each level of the scale was then coded with a score as shown in the following table.

Table 1. Coding scheme of Question 13-40 in the Questionnaire

Agree 2

Partly Agree 1

No Opinion 0

Partly Disagree -1

Disagree -2

Part three of the questionnaire was a ranking task with the question: “What contributes most to the success of an ELT professional?” The respondents were given seven items (see Appendix II) and asked to rank the seven items from 1 to 7 in order of importance (with #1 as the most important). The ranks given were counted and percentages were calculated.