TSElosophers meeting 18 October 2019. Kari Lukka, Milla Wirén, Otto Rosendahl
Flyvbjerg, B. (2006) Five Misunderstandings About Case-Study Research. Qualitative Inquiry, 12:2, 219-245.
Examination of the potential misunderstandings that still surround case study research is an excellent theme. We very much agree with Flyvbjerg that science needs meaningful, good quality case based research. The five misunderstandings he identifies go largely to the point:
1. General, theoretical (context-independent) knowledge is more valuable than concrete, practical (context-dependent) knowledge.
2. One cannot generalize on the basis of an individual case; therefore, the case study cannot contribute to scientific development.
3. The case study is most useful for generating hypotheses; that is, in the first stage of a total research process, while other methods are more suitable for hypotheses testing and theory building.
4. The case study contains a bias toward verification, that is, a tendency to confirm the researcher’s preconceived notions.
5. It is often difficult to summarize and develop general propositions and theories on the basis of specific case studies.
However, misunderstandings underlying Flyvbjerg’s analysis about these misunderstandings makes his own account inconsistent and even misleading. His response to the first point is to emphasize the human learning process, which requires context-dependent practice, but in terms of the importance of the theoretical (context-independent) knowledge, he seems to use a seriously outdated, narrow view on theory (Lukka & Suomala, 2014). This hides the potential of combining context-based empirical analysis of case studies with focused and motivated theorising.
Regarding the second point, Flyvbjerg as such correctly applauds the richness of narratives in creating understanding about phenomena. However, in his response to the second point, he undermines the need and role of generalising in the context of research, including case study research. In essence, he seems to be somewhat blind to the possibility and need of drawing insights from a context-specific case to a higher level of abstraction, which then makes it possible to create generalized theorems. We discussed this point through Flyvberg’s example of how London can only be familiarized through strolling the streets to gain in-depth understanding of the city, and pointed out that while indeed the insights of the nature of diverse alleys does require visiting them, having a map of London is still valuable for gaining other type of understanding of the overall city.
The third misunderstanding is well-analysed, as he points out the diversity of case types that can aid in unveiling specific types of phenomena and scientific propositions. He suggests various theoretical sampling methods as a starting point for testing and building theories, although again he neglects the deep theoretical insight that is needed for designing these non-random sampling methods.
The answer to the fourth point was in Tselosopher’s view blatantly wrong as he states that the case scholars are less prone to verification bias, and even goes as far as to say, ‘ in contrary’. His response to the fourth point does not consider that all studies, if not properly conducted, can be biased and that applies to case study research, too. This is especially a risk in case study research, where the researcher is often in close and long-lasting real-life contact with people in the field. Flyvbjerg should have started from accepting these premises and then examined openly the ways to avoid the risk. In our view, these include being conscious about this risk and then employing the principle of ‘critical independence’ (McSweeney, 2004). And here again the underlying problem is Flyvbjerg’s serious omitting of the role of theory and theorising, which makes him not to realise the possibilities of fruitfully combining rich case-based materials and theorising.
Responding to the final point, Flyvbjerg tries to defend fully descriptive, narrative case study reports, continuing the discussions in the first two points. However, their value as representing scholarly research needs to be questioned. Related to this point, he also unnecessarily limits his attention to the balance between keeping the case study account open and rich vis-à-vis summarising the findings, favouring the former. The much more relevant matter to consider would have been the challenge of using the rich case-based materials in elaborating a theoretically well-motivated research question and producing a meaningful theoretical argument as the conclusion of such analysis.
In sum, Flyvbjerg (2006) does not appreciate the potential and value of theorising in case study research. This makes him to oscillate between overstating and understating his defence for case studies: he concludes that cases are a great tool in a world without context-independent knowledge, but delimits their scholarly value with arguments that case studies cannot be summarized but only narrated. The outcome is therefore a series of further misunderstandings about theorizing. It is a pity since his topic is of great relevance. His article has been cited over 14000 times (Google Scholar on 23 October 2019), which generates a risk that many case study researchers have followed his misguidance along with his guidance.
Lukka, K., Suomala, P., 2014. Relevant interventionist research: balancing three intellectual virtues. Accounting and Business Research, 44, 204–220. https://doi.org/10.1080/00014788.2013.872554
McSweeney, B. (2004) Critical independence. In Humphrey, C. & Lee, B. (eds.) The real life guide to accounting research. A behind the scenes view of using qualitative research. 207-226.