Configurational Comparative Methods

Core Assumptions and Goals

The whole ambition behind CCMs is, on the one hand, to make qualitative case analysis more systematic and to offer case-oriented researchers some tools that enable them to systematically compare thick, complex cases. On the other hand, it strives to offer an alternative way to envisage causal arguments made in mainstream quantitative (read statistical) social-scientific work. The overarching goal of CCMs is to unravel causal complexity by applying set-theoretic methods to cross-case evidence. In more concrete terms, the different QCA techniques developed within CCMs enable one to identify core combinations of conditions (input variables), which explain the variation of a given outcome (output variable) of interest. Therefore, QCA techniques are geared toward the identification of so-called specific connections between conditions and outcomes. In contrast to most statistical techniques, they are not geared toward the establishment of general, tendential, or correlational connections between each independent variable, on the one hand, and the dependent variable, on the other (Ragin, 2008; Rihoux, 2008).

Causality, or linkages between conditions and the outcome, more generally, are assumed to be [Page 389]multiple and conjunctural. This implies the following assumptions: (a) Most often, it is a combination of conditions (rather than a single condition) that generates the outcome; (b) several different combinations of conditions may produce the same outcome; and (c) there is no fixed effect of a given condition on the outcome—depending on how it is combined with other conditions, different values of this condition can produce the outcome. The concrete goal of CCMs, from this perspective, is to identify those different causal paths leading to some outcome of interest, each path being relevant in its own way, regardless of the number of cases it covers. Thus, this concept of causality does not take on board most of the core assumptions underlying mainstream statistical analysis, such as permanent causality, uniformity of causal effects, unit homogeneity, additivity, linearity, and causal symmetry (Dirk Berg-Schlosser, Gisèle De Meur, Benoît Rihoux, & Charles Ragin, 2009).

The logical foundations of CCMs can be found in John Stuart Mill's “canons,” in particular the method of agreement, the method of difference, and the joint method of agreement and difference, which are all logical ways to systematically contrast and match cases, so as to establish common causal relationships. In practical terms, in CCMs, this is translated into tests for necessity and sufficiency.

Techniques

The overall rationale behind the QCA techniques is that, through some step-by-step logical operations (based on Boolean algebra or set-theoretical logic), one is able to reduce complex data tables to shorter combinations of conditions explaining the outcome of interest. Therefore, at the heart of CCMs lies the quest for parsimony—but not in such a way that within-case complexity would be sacrificed. So far, three specific techniques have been developed under the umbrella term of QCA: crisp-set QCA (csQCA), the initial technique that uses Boolean, that is, dichotomous, sets; multi-value QCA (mvQCA), which allows the use of multiple-category conditions; and fuzzy-set QCA (fsQCA), which uses finer grained fuzzy-set membership scores.

...