`Analytics' are No Substitute for Methodology: A Response to Speer and Hutchby

Abstract

In their article in this issue, Speer and Hutchby argue that qualitative researchers' fears about the reactive effects of audio- or video-recording on the quality of data are misplaced. They claim that these fears stem from a methodological orientation which assumes that in order to study the world we must obtain data that are unaffected by the process of research, that we should observe the social world as if through a one-way mirror. Yet, they point out, most research strategies do affect the data produced. So, the authors argue, qualitative researchers are faced with an intractable dilemma. The solution proposed is to abandon naturalism and to investigate those occasions when people address the fact that their talk is being recorded, using a conversation analytic approach. In response to this, I argue that a concern with reactivity does not imply commitment to any naïve naturalism, and that the approach recommended by the authors does not avoid this methodological problem, unless one is prepared to accept that conversation analysis is the only legitimate form of social research; a proposal for which the authors do not provide a convincing case.