Reducing Uncertainty: Information Analysis for Comparative Case Studies 1 Katya Drozdova Seattle Pacific University and Kurt Taylor Gaubatz Old Dominion University The increasing integration of qualitative and quantitative analysis has largely focused on the benefits of in-depth case studies for enhancing our understanding of statistical results. This article goes in the other direction to show how some very straightforward quantitative methods drawn from information theory can strengthen comparative case studies. Using several prominent “structured, focused comparison” studies, we apply the information-theoretic approach to further advance these studies’ findings by providing systematic, comparable, and replicable measures of uncertainty and influ- ence for the factors they identified. The proposed analytic tools are simple enough to be used by a wide range of scholars to enhance comparative case study findings and ensure the maximum leverage for discerning between alternative expla- nations as well as cumulating knowledge from multiple studies. Our approach especially serves qualitative policy-relevant case comparisons in international studies, which have typically avoided more complex or less applicable quantitative tools. The epic methodological battles of the late twentieth century have largely subsided in light of the eminently reasonable notion that there are benefits to be gained from both the empirical confidence that comes from broad aggregate studies and the in-depth understanding generated by more focused case studies (Coppedge 1999). This reconciliation has brought a rising interest in the use of “multi-methods” to pair quantitative and quali- tative work in the analysis of particular problems (Lieber- man 2005). The multi-methods approach has primarily focused on the parallel application of large-N and small-n analytics to the same empirical issue. In this paper, we argue for an even tighter integration of quantitative and qualitative methods and demonstrate a quantitative but simple and accessible approach to enhance small-n case study research. Our proposed approach applies where traditional statistics fall short. It complements and offers unique advantages over existing quantitative tools for small-n studies. Most importantly, we aim to aid qualitative schol- ars who typically would not use quantitative tools, but would benefit significantly from these improvements. To this end, we draw on information theory to propose a rig- orous yet simple and broadly accessible approach to uncertainty reduction (Shannon 1948; Cover and Thomas 2006). We especially focus on policy-relevant comparative case studies involving assessments of the relative impacts of multiple factors theorized to affect an uncertain out- come—under the constraints of few cases, significant challenges in gathering comparable data, and potentially very consequential policy implications of decisions that may be informed by such studies. Our approach is thus in the tradition of, and aims to strengthen, the case study methods frequently used in international studies and explicitly designed for generating and accumulating policy-relevant knowledge across multiple cases. The “structured, focused comparison” is a leading such method articulated by Alexander L. George (1979) as a response to the heightened interest in more systematic applications of qualitative methods to derive policy- relevant empirical generalizations (Eckstein 1975). In structured, focused comparison, a set of theoretically motivated critical variables is identified and then their variation is analyzed across several detailed qualitative case studies to derive systematic conclusions. While the structured, focused comparison method is admirably sys- tematic and analytic, we argue that its empirical applica- tions have often failed to provide analytic clarity. Drawing on recent advances in the field of information theory, we propose a straightforward method to provide a systematic quantitative understanding of the strengths and limita- tions of structured, focused comparisons and to reduce the uncertainty often associated with their results. We draw on information theory, which tackles uncer- tainty, to offer analytic tools where traditional statistics typically fall short in the very small-n world of compara- tive case studies. The small number of observations limits the applicability and effectiveness of typical statistical tests such as correlation or regression analysis. The informa- tion-theoretic approach makes no assumptions about the underlying distribution of data and thus is not limited by the Normal distribution assumption of many traditional 1 Author’s notes: The authors would like to thank Dr. Michael Samoilov, of UC Berkeley’s California Institute for Quantitative Biosciences (QB3), as well as the ISQ reviewers for helpful comments and suggestions. K. Drozdova would like to gratefully acknowledge the Hoover Institution at Stanford Uni- versity for the opportunity to serve as a visiting scholar and benefit from research in the Hoover Institution Library and Archives while writing this arti- cle. The data, the worked out examples, an annotated Excel demonstration, and an R script file are available at kktg.net/kurt/entropy. Drozdova, Katya and Kurt Taylor Gaubatz. (2013) Reducing Uncertainty: Information Analysis for Comparative Case Studies. International Studies Quarterly, doi: 10.1111/isqu. 12101 © 2013 International Studies Association International Studies Quarterly (2013) 1–13