One gets the impression in the introductory chapter of John Gerring's provocative new methodological treatise that the position of case studies in political science is a bit like that of adult videos in the entertainment industry: widely perceived as unseemly and marginal, yet in fact comprising an embarrassingly large share of the profession's total output. Gerring is no Potter Stewart, however; he is not content simply to know a case study when he sees it. His first task is thus to offer a definitive definition of the case study, to lay the groundwork for situating this much-maligned and misunderstood method in contemporary political science.
How exactly have case studies been misunderstood? Gerring makes three main claims in this regard. First, case studies have been wrongly treated as qualitative by definition: “To study a single case intensively need not limit an investigator to qualitative techniques” (p. 10). Second, case studies are never observations in isolation: “To conduct a case study implies that one has also conducted cross-case analysis, or at least thought about a broader set of cases” (p. 13). Third and perhaps most controversially, Gerring argues that “the strongest defense of the case study is that it is quasi-experimental in nature. This is because the experimental ideal is often better approximated within a small number of cases that are closely related to one another, or by a single case observed over time, than by a large sample of heterogeneous units” (p. 12).
Whether case studies indeed exhibit these strengths obviously depends on how case studies are defined. The problem is that existing conceptualizations of case studies have yielded a “definitional morass” (p. 17), rather than conceptual clarity. Gerring laudably acknowledges that the success of his entire enterprise hinges on the difficult task of draining this swamp. On this score, he advances our understanding, but does not quite provide conceptual closure. At the risk of highlighting the volume's minor flaws to the detriment of detailing its considerable strengths, I feel it necessary to underscore three problematic dimensions to Gerring's definitional exercise. Only then can the morass truly be considered drained, and the ground properly cleared for his insightful analysis to follow.
The first problem accompanies Gerring's main definitional advance: his insight that “[a] case study presupposes a relatively bounded phenomenon” (p. 18). Previous definitions have, surprisingly, overlooked this basic point. Yet he unnecessarily constricts the concept's connotation, defining a case as a “spatially delimited phenomenon” (p. 19) or as a “spatially or temporally delimited phenomenon” (p. 211).
Why is this problematic? Because cases are not necessarily bounded in spatial or temporal ways. Gerring himself notes the importance of “specific institutions (political parties, interest groups, businesses)” (p. 19) in case-study research. The boundaries of such institutions are conceptual and analytical, not spatial or temporal. The fact that an intensive investigation of Doctors Without Borders could qualify as a case study pointedly confirms that spatial boundaries are far from essential to the definition. The concern here is not that Gerring misunderstands what a case study is (he clearly does not). It is that his summary definition and substantive description are not as perfectly aligned as one might hope, particularly for pedagogical purposes.
Gerring's guiding distinction between case studies and cross-case studies raises questions as well. “All empirical work may be classified as either case study (comprising one or a few cases) or cross-case study (comprising many cases),” he suggests (p. 20). Yet in most places in the manuscript, including the glossary, Gerring defines the case study as “[t]he intensive study of a single case,” not a few. Works that examine multiple cases in depth employ a “case study research design” (p. 211), even though they are not case studies.
Besides being semantically awkward, this is substantively confusing. Why are all studies of multiple cases not cross-case studies? Apparently because they do not all use statistics. Robert Lane's in-depth study of 15 voters in Political Ideology (1962) is said to lack a cross-case dimension, since his cases “cannot be handled within a dataset format” (p. 35). Gerring reaffirms this perspective in the glossary, defining a cross-case study as “a large-sample study” that is “analyzed statistically” (p. 213).
Study is a verb as well as a noun. Hence, it seems peculiar to define any work that does not study cases in depth as a cross-case study. It seems even odder to deny the appellation to scholarship that intensively studies a significant number of cases and draws inferences accordingly. Are Ruth Berins Collier's Paths Toward Democracy (1999: n = 27) and Dietrich Rueschemeyer et al.'s Capitalist Development and Democracy (1992: n = 38) really not cross-case studies, simply because their cases are not “handled within a dataset format”?
Gerring thankfully retreats from this stance in his elegantly constructed and impeccably considered Table 2.4 (p. 28; see also his reference to “small-N cross-case analysis” on p. 189, also p. 204). In this sweeping typology of research designs, he rightly recognizes that the comparative method leverages cross-case as well as within-case variation for causal inference. This suggests a need for deeper definitional work. His case/cross-case distinction ultimately fails to adequately differentiate methods that involve the intensive and comparative study of multiple cases from quantitative methods that do not.
A third instance of definitional slippage emerges with Gerring's claim that “case study research may be either quant or qual, or some combination of both” (p. 10). To be sure, his elaboration of how to bring quantitative methods into case studies is a major contribution. But can a case study be strictly quantitative? When Gerring distinguishes case from cross-case studies by comparing four studies of American public opinion, he argues that “[t]he style of analysis differs in one respect: only in the case studies does qualitative analysis comprise a significant portion of the research” (p. 34). It thus appears that qualitative methods remain a “definitional entailment” of case studies, not merely a “methodological affinity” (p. 10).
To be clear, these definitional correctives do not undermine Gerring's main arguments, all of which this reader finds compelling and well supported. By pressing us to think of cases as bounded units, and not just as single instances, Gerring clarifies in a simple yet powerful way why case analysis and quantitative analysis are not contradictions in terms. By explicitly judging case studies in terms of how well they stack up against the experimental ideal (Chapter 6, with Rose McDermott), he reminds us that control rather than n is the key to causal inference. And by producing an accessible, insight-laden account of both when (Chapters 3–4) and how (Chapter 5, with Jason Seawright) to do case-study research, Gerring provides an engagingly written guidebook for graduate and undergraduate methods courses alike.
Perhaps the most appealing element of the volume is Gerring's informed confidence in the generalizing potential of case studies, however. Area specialists harboring theoretical ambitions will be especially inspired by his contention that “the narrowest terrains sometimes claim the broadest extensions” (p. 74). This book does more than any in recent memory to bring case studies out of the shadows and into their proper, proudly central place in political science.
