Search

This article critically examines recent arguments by Andrew Abbott, Jensen Sass, and Tariq Modood proposing a normative sociology, one that not only adopts an evaluative stance towards the phenomena it investigates, but also makes explicit and seeks to justify the values on which its evaluations rely. I argue that, while these proposals are to be welcomed in some respects, they fail to address two key issues: On what reasonable basis can it be assumed that there are single correct answers to value questions?; and What distinctive intellectual authority can sociologists claim to be able to discover those answers? I also point out that these recent advocates for a normative sociology pay insufficient attention to the opposing position, a commitment to 'value-neutrality', as proposed most notably by Max Weber. I argue that, while this is frequently ignored or dismissed out of hand, it represents a much more coherent and cogent view, even if its advocates have rarely fully lived up to its demands. [ABSTRACT FROM AUTHOR]

This article identifies and explores some unresolved problems surrounding the role of theory and evidence in qualitative research. It begins by distinguishing among various meanings of the term 'theory', focusing in particular on the view that portrays it as consisting of specific explanatory principles. An ambivalence towards theory on the part of qualitative researchers is highlighted, showing how the influence of positivism persists despite explicit rejection of it. The problems involved in non-positivist reinterpretations are discussed, notably the attempt to integrate theory with description. The focus of the paper then turns to the concept of evidence. Here again a tendency for qualitative researchers to inherit an empiricist approach is identified, and the implications of abandoning it are outlined. Finally, it is argued that theory must not be seen as the only or even as the most important product of qualitative (or indeed of quantitative) research; several types of product are possible, each placing different requirements on the researcher. [ABSTRACT FROM AUTHOR]

The concepts of internal and external validity, developed by Norman Campbell, are widely used to structure methodological thinking about social research. This article points to ambiguities in the interpretation of those terms, both as regards the relationships they refer to as well as the sort of object that is held to be capable of internal and external validity. In addition, it is suggested that the distinction between these types of validity is fundamentally misleading because it reflects a failure to distinguish relations between events and relations between variables. It also rests on the false assumption that we can separate the discovery of causal relationships from the question of whether these apply to other cases than the ones studied. In the final section, an alternative conceptualisation of validity is sketched, one that avoids the problems identified. [ABSTRACT FROM AUTHOR]

This is a reply to Peter Swanborn's comments on my paper 'A note on Campbell's distinction between internal and external validity'. I argue that Swanborn misconstrues my argument, interpreting it as recommending the abandonment of the concept of external validity, whereas it actually proposed abolition of the distinction between internal and external validity. I also explain that my argument did not involve a denial of the importance of the various issues included under these headings by Campbell, merely a proposed reconceptualisation of them. [ABSTRACT FROM AUTHOR]