Troubling theory in case study research

Abstract

The article begins by examining the variety of meanings that can be given to the word ‘theory’, the different attitudes that may be taken towards theories of these various types and some of the problems associated with them. The second half of the article focuses on one of these types, explanatory theory, and the question of what is required if case study research is to be effective in producing sound theories of this sort. A range of important distinctions is outlined and questions are raised about the possibility of, and need for, such theories.

Keywords:

• case study
• causation
• higher education research
• explanation
• theory

Notes

For an earlier attempt to identify some of these, see Hammersley (1995). There are some characterizations that are surprisingly indeterminate, for instance, Miller (2007) writes: ‘What we call theory is a series of ongoing debates about meaning, texts, knowledge, and subjectivity that extend from the Platonic dialogues, through Aristotle to Cicero, Seneca, Augustine, Aquinas, Dante, and so on to the present’ (p. 6).

For a useful discussion of the importance of abstraction in science, see Sayer (1992, chap. 3) and passim.

See Elliott (2009). This was not always so, a key example being the work of Simmel (1950) on the effects of group size, and on secret societies.

For useful discussions of Kuhn's work, see Bird (2000), Sharrock and Read (2002).

Incidentally, the events involved may be relatively small-scale or large-scale, in other words micro or macro.

For an excellent discussion of many of the issues, see Schofield (1990).

A classic illustration of one version of the problem is Skocpol's (1979) study of revolutions, where the number of cases in history that fit her definition of this phenomenon is very small.

See, for example, Cooper and Glaesser's application of Ragin's qualitative comparative analysis to large data sets (Cooper, 2005; Cooper & Glaesser, 2010).

On this distinction see Byrne and Ragin (2009).