Credibility of data interpretation from the view-

In research on teachers’ thinking and knowledge, it is generally assumed that

“language data,” meaning what teachers express through communication with researchers (e.g., data collected by interviews), represent their thinking and knowledge about the research problems at stake (Freeman, 1994). Consider-ing the contexts or perspectives, i.e., the practical and theoretical perspec-tives,(see the introduction to the present research and the section on position-ality) in which the present research was done, this representative assumptions and views are behind the data interpretation of the research. The most impor-tant measure that I took to increase the validity of my interpretation so that it would represent the teachers’ thinking (the reasoning underlying their practi-cal knowledge) was to get their feedback on my interpretation.

As mentioned, there was a seven-month interval between the first and second stages of data analysis. In the first round and in the initial stage of data collection, I analyzed and interpreted the data from semi-structured and stimulated recall interviews of two teachers. In this stage, I categorized the data and provided a short report and interpretation of categories. Then, I sent the report and the categorization in concrete language to the two teachers mentioned in order to get their feedback. To do this we had a focus group, in which we discussed the report and the interpretation. I recorded the discus-sion and then incorporated the teachers’ views and feedback into the interpre-tation and categorization. The most contentious part of the interpreinterpre-tation was at the word level where the teachers suggested some ideas about “phrases,”

“terms,” and “concepts” that I had developed to represent the meanings and content of categories. However, they generally believed that the interpretation made sense, and thus, “they could see and find their practical reasoning”

within the interpretation. This was the explicit and systematic way of check-ing my interpretation from the participants’ perspectives.

In addition, during data collection with the other participating teachers I used the same technique, but in less explicitly systematic way. As mentioned in chapter 3, I tried to transcribe all the interviews the day they were done.

While transcribing, I made my initial interpretation about the data and their potential categories. This kind of interpretation was general and as not yet formed into categories, but it contained sound meanings about teachers’ per-ceptions and could provide a useful, practical help for forming categories and making final interpretations. In the next possible, scheduled observation of the teachers’ classrooms, I discussed this initial and universal interpretation that had been drawn from their previous interviews, and generally asked if

Validity of the account 163

the teachers could see their views in the interpretation. This discussion was also audio-taped and incorporated into the final data interpretation.

The two actions described above may reflect the method that Lincoln and Guba (1985) called “member checking and describing it as “the most crucial technique for establishing credibility” (p. 314). In qualitative research (e.g., phenomenography) wherein the research mainly relies on the participants’

conception and understanding of the research phenomenon, member checking is very important. Because the main source of data in the present research was the teachers’ understanding and explanations of the research problem, I used this technique.

7.3 Consistency of data collection and analysis methods with the knowledge claims of the research

Even though I believe that there are no clear-cut rules and methods for differ-ent knowledge claim assumptions and research paradigms and researchers may use some degree of flexibility in choosing data collecting and analyzing methods, it is still important to show how the methods chosen are consistent with the knowledge claim of the study. As discussed earlier, the present re-search aims at describing “something” about teachers’ practical knowledge and has its theoretical roots in constructivist paradigms, specifically social constructivist paradigm. “Constructivists believe in pluralistic, interpretive, open-ended, and contextualized (e.g., sensitive to place and situation) per-spectives toward reality” (Creswell & Miller, p.125). From this perspective and in regard to the present research, producing useful knowledge is not solely the property of the educational researcher or teacher educators; teach-ers too can actively construct their knowledge. Thus, in order to undteach-erstand the different aspects of practical knowledge, which is mainly the property of teachers, researchers should rely primarily on the research methods that aim at describing the phenomenon from the teachers’ point of view.

The main data collecting tool was the interview, one of the useful ways to describe a research phenomenon from the participants’ point of view. Moreo-ver, during the data analysis and in two different ways, I tried to incorporate the teachers’ perspectives into the findings by “member checking.” However, it might be point of discussion whether the participants’ perspectives should be integrated into the results of the study and to what degree and how. This, I believe depends, to a great extent, on the research phenomenon and the re-search task. In any case it would be a good idea to solicit feedback from the participants about interpretation if the study relies mainly on the participants’

perceptions of the phenomenon.

Representing the epistemic nature of teachers’ practical knowledge 165

8 Discussion and implication