The Characteristics Of Good Quality Research

By Author: Mark Gary
Total Articles: 61

The design of this study reflects what Creswell (2003) calls a sequential explanatory multi-method strategy. This is a design which 'is characterized by the collection and analysis of quantitative data followed by the collection and analysis of qualitative data' (p. 215). Specifically, the study reported here adopted a survey approach in which largely quantitative data were first collected through a questionnaire. A sub-sample of the teachers who completed the questionnaire then participated in Merrell Sandal a second phase of data collection through which their questionnaire responses were explored and illustrated in more depth qualitatively [see Bryman for a detailed analysis of rationales for multi-method research]. I discuss the different forms of data below.

In the form of a questionnaire, the cross-sectional survey allows large amounts of data to be collected efficiently, economically, and in a standardized manner [see, for example, Aldridge and Levine (2001)]. The questionnaire used in this study (Appendix 1) had six sections, focusing on respondents' conceptions of what counts as research, views about the characteristics of good quality research, perceptions of their institutional culture in relation to research, engagement in reading research, engagement in doing research, and background information. The range of themes covered in the instrument was informed by issues raised in the literature reviewed earlier. The instrument was piloted with a group of 21 English language teachers and its length, wording and organization revised in line with their feedback. Questionnaire data were collected from a non-probability sample of 505 teachers of English in 13 countries. My goal was to obtain a broad perspective on the issues under study and with Merrell Boot this goal in mind I approached contacts in a number of ELT contexts around the world who were in a position to invite practising teachers (that was my only criterion for teacher participation) of English to complete the questionnaire. These contacts played a vital facilitative role in this study by providing access to respondents as well as by advising on which mode of administering the questionnaire hard copy, web-based, or e-mail attachment would work best in their particular contexts [all three modes of administration were used in the study see Couper (2005) for a discussion of the use of technology in survey research]. The closed questionnaire data were analyzed statistically using SPSS 12.

Of course, the many advantages of questionnaires are countered by certain limitations, particularly when respondents are being asked to report their beliefs [see Borg (2006c) for a discussion of meth-odological issues involved in using self-report instruments to study teachers' beliefs]. Questionnaires, also, often generate superficial answers and do not allow in-depth exploration of particular issues. In response to these concerns, direct questionnaire items of the type 'what are your beliefs/views about research?' or 'what is research?' were avoided; additionally, the questionnaires were supplemented with qualitative data, which I now describe.