Symposium : The Set-Theoretic Comparative Method — Critical Assessment and the Search for Alternatives

The set-theoretic comparative method (STCM) — including very centrally Qualitative Comparative Analysis (QCA) — has formulated valuable goals for comparative research. This symposium raises serious concerns about the specific analytic tools advocated for pursuing these important goals. George Lakoff challenges the key argument that natural language is inherently set-theoretic and suggests that fuzzy-set scoring is in fact not fuzzy. Giovanni Sartori argues that the set-theoretic framing privileges technique over real insight, and Sean Tanner critiques the tools for causal inference advocated by STCM. Chris Krogslund and Katherine Michel explore the difficult challenge of finding simulation tests appropriate to evaluating these methods; Zachary Elkins formulates alternative approaches to measuring partial membership in categories; and Bear Braumoeller evaluates four alternative models for the study of interactions, including QCA. Finally, Andrew Bennett presents a major breakthrough in tools for process tracing. He moves beyond the traditional classification of four process tests, as well as the characterization of the four tests according to whether they provide necessary and/or sufficient criteria for supporting a particular causal inference. Instead, he uses Bayesian tools to map the out the probative value of tests in terms of a continuous spectrum, treating the traditional tests simply as benchmarks on this spectrum