How Do We Select Teachers for Ethnographic research in Schools? The Implications of Heterogeneity

Ethnography is not neat, which is both unavoidable and appropriate given its purpose of explaining human behaviour and meaning. One perennial question in ethnographic access and data collection concerns the problem of self-selection bias and its mitigation. The question is important because it underpins the quality of data across a whole project. Bias in data also snowballs along with access and analytical cycles. Whilst theory has developed in some applications of ethnography, studies in Japan's schools often continue to represent teachers as relatively homogeneous. As such, there is little discussion on which teachers are selected as key participants and, implicitly, which are not. This not only detracts from representativeness, but moreover misses the opportunity to see how meaning is constructed. Whilst researching the contested topic of curriculum revisions to moral education in and around Japanese schools, I started to categorise geographical areas, schools, teachers and school administrators. This was initially intended as a methodological tool seeking to reduce various forms of self-selection bias. However, these developed into an analytical tool that supported theory on how meaning is constructed/contested between teachers in schools. Though perhaps it is ultimately unsurprising that tools developed for working with people can develop into tools for understanding people, this session presents theory and some imperfect methods of reducing self-selection bias in teacher selection and their contribution to a theory of policy enactment in Japanese schools, with the aim of prompting discussion.