Beyond the Co-Variational Template: Alternative Directions in Case Study Methodology.

Methodological reflections about case study research have increased markedly within the last years. Nevertheless, we argue that recent accounts do not entirely capture the different logics of various case study approaches. According to our account, there are three distinct ideal-types of case study methods: co-variational, causal process tracing, and congruence analysis. Whereas the first one has received increased attention and rigorous thinking, among others through John Gerring's recent book, the latter two have been so far treated as a mix rather than as distinct alternatives, e.g. in Alexander George and Andrew Bennett's work. We argue that causal process tracing and congruence analysis should be treated as conceptually distinct approaches. This would make them more coherent internally. Our paper seeks to support these two alternatives to the co-variational template by scrutinizing and comparing the ways of drawing inference, and the different understandings and directions of generalization of all three approaches. In the first part of the paper, which focuses on inference, we concentrate our efforts on the differences between causal process tracing and congruence analysis. In the second part of the paper which focuses on generalization we show, by contrast, that for both, causal process tracing and congruence analysis, there exists no "extensity/intensity" trade-off. We argue that the "thickness" of case studies creates only a trade-off when we want to generalize towards a wider population of comparable cases - as the co-variational template does. If we strive for analytical "depth" (as the causal process tracing approach does) or for theoretical "breadth" (as congruence analysis) case studies are not caught up in a trade-off but are very conducive to these goals. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]