The initial data collection started in October 2003 and helped me to develop my ideas. I quickly realized that I needed to elicit not just the teachers’ self-efficacy beliefs, but also broader aspects of their motivation, for example, their “sense of relatedness” (Dörnyei, 2001), and I acted accordingly from November 2003. Then, as I carried out preliminary data analysis, I became conscious of the need to explore growth in teachers’ self-efficacy beliefs in relation to growth in practical knowledge that is “the knowledge that is directly related to action… that is readily accessible and applicable to coping with real-life situations” (Calderhead, 1988, p. 54). In Progress Report 5, while writing about a teacher then part of the study, I reported:
He was quite convinced, for seven years, that what he was doing in the classroom was absolutely right, so convinced that he planned and taught without really thinking very much about what he was doing. Now there are doubts and questions in his mind, though he is very confident that he can find the right way. Before he was very confident that he could follow the right path. The ideas about the teacher’s role (in interpreting the course materials within the classroom context) are very different, though the self-confidence is fairly constant.
My understanding of research methods appropriate to my study was also developing. In the same report, I articulated the need to triangulate an elicitation of teachers’ self-efficacy beliefs with observation of practice, and with elicitation of practical knowledge, as the following extract demonstrates:
If, for example, a teacher talks confidently about being able to adapt teaching material to include a task framework centered on a core activity (Cameron, 2001), then how am I to verify the accuracy of this claim? One method would be through observation, focusing on behavior, while a second would be through interview, exploring the teacher’s understanding of the underlying concepts (Progress Report 5).
In the first few months of engaging in the research, therefore, my understanding of what I needed to elicit developed (i.e., teachers’ self-efficacy beliefs plus other aspects of motivation, practical knowledge and related cognitions, plus contextual information) and how to do it (using observations as well as interviews). My progress reports reveal that I was also developing sensitivity to issues connected with being an insider researcher. As early as September 2003, I was locating my research on Holliday’s (2002) “progressive qualitative paradigm,” discussing reflexivity, self-criticality, and the researcher’s ideological position; I was also reading and thinking about qualitative concepts such as “trustworthiness,” “validation” (Kvale, 1996), and different kinds of validity, including “catalytic validity” (Anderson & Herr, 1999) (Progress Report 2). Nevertheless, there was some unevenness in my reading and developing knowledge. In Progress Report 3, I discussed case study research explicitly for the first time, and with reference to Yin (1994), which was the only book on case study that was available in the remote region of Oman in which I was working; this must have been commented on in feedback by Simon and/or Gary, since I wrote in Progress Report 4: “I have not as yet been able to read Stake’s work, as suggested, but have requested the book.” Stake (1995) subsequently became a very influential source.
Moving Towards a Qualitative Case Study Approach
I am unsure whether, prior to 2003, I had read much qualitative case study research related to language teacher development. However, by November that year, Borg’s (2001) grammar teacher cognition case studies were clearly having an impact on my thinking, since, in Progress Report 5, I was speculating on the prior development and self-efficacy beliefs of two of Borg’s cases, therefore engaging in “naturalistic generalization” (Lincoln & Guba, 1985) prompted by the “thick description” (Geertz, 1973) in Borg’s writing.
Subsequently, while I reviewed the literature, my appreciation grew of the need for qualitative case study research into teachers’ self-efficacy (TSE) beliefs that is “teachers’ beliefs in their capabilities of supporting learning in various task- and context-specific cognitive, metacognitive, affective and social ways” (Wyatt, 2008, p. 5). My understanding of TSE beliefs had been shaped by my reading of Bandura’s (1986) theory and by my earlier experiences of researching language learners’ self-efficacy beliefs (Wyatt, 2000). However, on fully engaging with the TSE beliefs literature in 2004, I found much confusion. Though based on “a simple idea with significant implications” (Tschannen-Moran & Woolfolk Hoy, 2001, p. 783), numerous studies appeared to be conceptually problematic, with the construct frequently misunderstood and misapplied, as had already been recognized (Henson, 2002; Wheatley, 2002), and has been discussed more fully since (Klassen et al., 2011; Wheatley, 2005; Wyatt, 2014). Klassen et al. (2011, p. 37) complain, for example, of definitional entropy, whereby “carefully defined psychological constructs lose precision over time”; besides sharing this concern, I was troubled by the very global (i.e., insufficiently task-specific) way in which TSE beliefs were generally assessed in many quantitative studies, and by erroneous assumptions that TSE beliefs were relatively inflexible and therefore not open to change (Wyatt, 2008, 2014).
Qualitative case studies researching TSE beliefs were clearly in short supply. Only 2 of the 218 studies published between 1998 and 2009 that were reviewed by Klassen et al. (2011) were in this category, this despite various researchers (e.g., Labone, 2004; Tschannen-Moran et al., 1998; Wheatley, 2005) explicitly calling for the use of interpretive qualitative case study research methodology to develop in-depth understandings.
Given the scarcity of qualitative case studies of TSE beliefs, and the benefits I had derived from reading Borg’s (2001) grammar teacher cognition cases, I read the two qualitative TSE beliefs case studies I had identified with interest. Indeed, in my literature review (Wyatt, 2008), I then presented vignettes extracted from both Milner and Woolfolk Hoy (2003) and Mulholland and Wallace (2001) to demonstrate, in a field dominated by quantitative research methodology, the potential for learning about TSE beliefs from qualitative case study research.
Strengths of one of these cases (Milner & Woolfolk Hoy, 2003) includes its use of “triangulation” (Stake, 1995), with several forms of this drawn upon. Evident was: “data source triangulation,” which involves considering the consistency of the phenomenon observed over time; “methodological triangulation,” where different methods are combined; and “member checks,” whereby data are discussed with research participants. However, a weakness of the study was that “theory triangulation,” whereby alternative theoretical explanations are considered, seemed absent; this appeared problematic since the construct of self-efficacy beliefs was unfortunately misconstrued in this study (in terms of “lofty goals,” rather than specific tasks).
I also detected flaws in Mulholland and Wallace’s (2001) account of a beginning teacher’s struggle to use group work in a “hands-on” way in elementary science lessons in Australia. Concentrating on exploring the teacher’s development through “a psychological lens” (Tschannen-Moran et al., 1998, p. 203) in a way consistent with TSE beliefs theory (Bandura, 1986), the researchers nevertheless did so somewhat narrowly; I felt that analytical perspectives from the fields of teacher cognition and teacher education were lacking. Wheatley (2005, p. 760) argues for “a merging of fields” in research into TSE beliefs, drawing together expertise from different specialist areas. Reading Mulholland and Wallace’s (2001) story, I found myself wondering about the role of practical knowledge in the teacher’s development; how had this grown in relation to her developing TSE beliefs? I was also curious about the role of mentoring that encouraged deep reflection, and indeed about the role of reflection itself, a term which appeared only early in