The analytic integration of qualitative data sources

In recent times there has been a considerable growth in research projects using more than one method (see for example, Corden and Sainsbury 2006; Dicks et al, 2006; Mason, 2006). This has led to renewed debate about the issues involved in using multiple methods in a single study, including questions concerning the different ways in which methods and data could or should be brought together (see for example, Caracelli and Greene 1997; Moran-Ellis et al, 2006; Pawson 1995). However, within these debates there is a tendency to focus attention on designs which bring together qualitative and quantitative methods, leaving aside research designs which utilise multiple qualitative methods, perhaps on the assumption that ‘qualitative data’ is an homogeneous category. In this chapter we examine the issues involved in integrating different types of qualitative data generated through three qualitative methods: ‘conventional’ in-depth interviews, photo-elicitation interviews, and narrative interviews. Drawing on data from the PPIMs project (Practice and Process in Integrating Methodologies project) - which explored the methodological issues that arise in multi-method and multi-level approaches to investigating the management of vulnerability in everyday life - we specifically focus on the process of achieving integration across these sets of data at the point of analysis and document an approach we call ‘following a thread’ (Moran-Ellis et al, 2004).