From “analytic nuisance” to interactional resource: Re-viewing small stories within interviews in a mixed methods study

Discussions about the use of interviews in applied linguistics research have been ongoing in methods-oriented literature for some time. The focus of much of this discussion has concerned studies employing qualitative frameworks in which interviews are the primary method for generating data. However, there has been relatively little discussion of studies employing mixed methods – in which qualitative interviews are adopted as secondary or tertiary methods of data generation and are used in conjunction with conventionally quantitative measures. In this article I reconsider the assumptions informing my own analytic and representational practices as they relate to my use of interviews in a descriptive, mixed methods study of L2 lexical inferencing. Drawing on Talmy's (2010) notion of interview as social practice, the notion of entextualization (Bauman and Briggs, 1990), and what Bamberg (2006) has referred to as small stories, I trace the process by which three instances of multiple and conflicting responses to questions concerning one participant's ‘self-reported background information’ went unreported in that study. The reanalysis presents an opportunity to discuss compatibility between representational formats and data generation methods, important differences between theorization of interviews and participation in them, and the role of qualitative interviews in mixed methods research more generally.

► Reconsiders the use of open-ended interviews in a mixed methods study.
► Discusses the significance of unreported biographical data generated in those interviews.
► Discusses compatibility between representational formats and data generation methods.
► Discusses differences between theorization of interviews and participation in them.
► Discusses theorizing interviews during research design for more principled uses of methods and data.