Qualitative Comparative Analysis

Qualitative Comparative Analysis


On the one hand, comparative social science is defined by the existence (or at least the presumption) of meaningful “cases.” Comparativists treat cases as whole entities purposefully selected (e.g., the French Revolution), not as homogeneous observations drawn haphazardly or randomly from a large pool of equally plausible selections (e.g., a random selection of cases from the population of all revolutions—assuming such a population could be constructed). This gives comparative work a special focus on cases as “meaningful” in their own right. On the other hand, however, one of the primary goals of comparative social science (and social science in general) is to derive general statements about theoretically important relationships. Making general statements requires using general concepts. At the level of cases, concepts are most often represented through observable variables. Concepts and variables permeate almost all social scientific discussion of cases, no matter how much or how little homage is paid to their singularity, specificity, or meaningfulness as cases. Thus, comparative social scientists (especially) need tools that link case-oriented and variable-oriented discourse—tools that help them construct a rich dialogue of ideas and evidence.

The analytic challenge of case-oriented research is not simply that the number of cases is limited, but that researchers gain useful in-depth knowledge of cases that is difficult to represent using conventional forms of analysis (e.g., representations that emphasize the “net effects” of “independent variables”). The researcher is left wondering how to represent knowledge of cases in a way that is meaningful and compact, but which also does not deny their complexity.

Set-theoretic methods such as Qualitative Comparative Analysis (QCA) offer a solution. QCA is fundamentally a case-oriented method that can be applied to small-to-moderate size Ns. It is most useful when researchers have knowledge of each case included in an investigation, there is a relatively small number of such cases (e.g., 10-50), and the investigator seeks to compare cases as configurations. With these methods it is possible to construct representations of cross-case patterns that allow for substantial causal heterogeneity and case diversity.

Fuzzy set analysis can work in tandem with QCA.  The use of fuzzy sets is gaining popularity in the social sciences today because of the close connections it enables between verbal theory, substantive knowledge (especially in the assessment of degree of set membership), and the analysis of empirical evidence. Fuzzy sets are especially useful in case-oriented research, where the investigator has substantial familiarity with the cases included in the investigation and seeks to understand cases configurationally, that is, as specific combinations of aspects or elements. Using fuzzy-set methods, case outcomes can be examined in ways that allow for causal complexity, where different configurations of causally relevant conditions combine to generate the outcome in question. Also, with fuzzy-set methods it is a possible to evaluate arguments that causal conditions are necessary or sufficient. Examinations of this type are outside the scope of conventional variable-oriented analysis.

Expected Background

Students should have previous exposure to social research methods, including basic training in quantitative methods, at the post-baccalaureate level. The course will include instruction in the use of the software package fsQCA (for both Windows and Mac).

  1. Overview of Lecture Topics
  2. Part 1: Background
  1. Part 2: Basics
  1. Part 3: Crisp Set Analysis
  1. Part 4: Fuzzy Set Analysis
  1. Recommended Reading:
  1. WWW sites:

http://www.compasss.org (bibliography, working papers, etc.)

http://www.fsqca.com (free QCA software)