generalisability
reliability
validity
Desirable Qualities
rigour
theoretical framing
triangulation/comparison
As Box 3.2 indicates, in my judgement case study arguably has rather more key strengths than it has key weaknesses. To start with, case studies are in-depth, detailed and particular; they allow a close focus on the case, which the researcher thoroughly studies. Second, and relatedly, the research is holistic, aimed at understanding everything – or, at least, as much as possible – about the particularity of the case in question. These characteristics are in contrast to much other research, which tends to focus on a limited range of variables or factors and inevitably oversimplifies, and does not get into the depth of, what is going on.
Third, the case being studied may be typical or exemplary, and these qualities can provide, as we have already seen, one answer to the charge of lack of generalisability. For, if the case is typical or exemplary, and this can be verified and demonstrated, then the likelihood of the findings from it being generalisable increase significantly (though the researcher might still need to perform two or more such typical case studies to be convinced and convincing). Or, alternatively, if the case being studied is critical or extreme, the need to argue generalisability diminishes, for it is the very particularity of the case – its unusual nature – that is important.
Finally, a very important advantage of case study research from the point of view of the small-scale researcher is that it is bounded and therefore much more feasible. When there are limits – of time and other resources – on what the researcher can afford to spend on a particular piece of research, it is highly pragmatic to be able to tightly and precisely define what is going to be researched.
Of course, at the same time, the researcher will still wish and seek to complete a piece of research that is useful and meaningful, but that should also be feasible. Research cannot always change the world (it rarely does so, even if ‘the world’ is conceived as being just that small piece of it of particular interest to the researcher at a particular moment in time), but it can always aim to be of interest beyond the researcher and the case concerned.
Box 3.2 suggests, in its identification of three desirable qualities for case study (and other forms of) research, how a useful and meaningful case study might be achieved (this is discussed in more detail in Chapter 4). First, the case study needs to be approached and carried out rigorously (see Chapters 8 and 9 for detailed guidance on how to do this). Second, the case study should have a theoretical framework, enabling the development of a fuller understanding of how it works (this is also discussed in more detail in Chapter 8). And, third, it is highly desirable that the findings from the case study are triangulated in some way, for example by comparison with other similar case studies or other kinds of evidence.
Other researchers have come up with similar lists to those in Box 3.2. Thus, writing some 40 years ago, Adelman, Jenkins and Kemmis (1976) produced a list of what they termed the ‘possible advantages’ of case study:
1 Case study data… is ‘strong in reality’ but difficult to organize…
2 Case studies allow generalizations either about an instance or from an instance to a class…
3 Case studies recognize the complexity and ‘embeddedness’ of social truths…
4 Case studies, considered as products, may form an archive of descriptive material sufficiently rich to admit subsequent reinterpretation…
5 Case studies are ‘a step to action’… Their insights may be directly interpreted and put to use…
6 Case studies present research or evaluation data in a more publicly accessible form than other kinds of research report. (pp. 148–149)
More recently, Simons (2009) provided the following summary of the strengths of case study:
Case study using qualitative methods in particular enables the experience and complexity of programmes and policies to be studied in depth…
Case study can document multiple perspectives, explore contested viewpoints, demonstrate the influence of key actors and interactions between them in telling a story…
Case study is useful for exploring and understanding the process and dynamics of change…
Case study is flexible, that is, neither time-dependent nor constrained by method…
Case studies written in accessible language, including vignettes and cameos of people in the case, direct observations of events, incidents and settings, allow audiences… to vicariously experience what was observed…
Case study has the potential to engage participants in the research process. (p. 23)
These lists both confirm and add to the strengths identified in Box 3.2, taking a more outwardly looking perspective on the broader purposes of undertaking case study research.
Perceived Weaknesses of Case Study and Responses to These
Box 3.2 identifies three key weaknesses of case study: generalisability, reliability and validity (or rather the perceived lack of these qualities). These weaknesses, particularly the perceived lack of generalisability, are persistent criticisms, and case study practitioners would do well to show awareness of them and respond appropriately. One appropriate response, of course, would be to emphasise the strengths of case study. Another, adopted in this section, is to explore and better understand the criticisms, and discuss how they might be addressed. We shall look first at generalisability, then at reliability and validity, and finally at some other issues that are raised less often.
These issues, as we have already indicated, have been widely and hotly debated with respect to case study. It is important to stress, therefore, that they are issues which need to be considered for all research designs. Indeed, like most such criteria, they originate from quantitative and scientific research, raising the issue of whether qualitative and social scientific forms of research can, or should, be judged in the same way. While, as I have argued, case study research may be quantitative, qualitative or both, these issues nevertheless have particular purchase because quantitative research is typically larger-scale, while case study research is expressly small-scale, so is always liable to be criticised in these ways.
Generalisability
The relationship between the singular and the general has long been the subject of discussion in case study research; indeed, it would probably be fair to say that it has been the most discussed issue of all (e.g. Bassey 1981; Kennedy 1979). It is an underlying problem with the case study design that needs to be recognised. Of course, if the case being studied has been chosen for its particularity or extreme characteristics, this may not be an important issue: the case is of interest because it is unusual, perhaps unique. Otherwise, however, unless the researcher can demonstrate that the case being studied is typical, generalising from the findings is problematic.
One common response to the generalisability issue has been to argue for the accumulation of single case studies on the same topic to allow for the identification of similarities and differences. This is, indeed, how case studies have long been used in certain disciplines. For example, writing in the 1980s, Tripp (1985) argued that:
we must find ways of utilising the cumulated wisdom of the case studies we have available… We also need to build archives of the cases similar to those of the legal system, and we need