Comparison is, directly or indirectly, the defining characteristic of political science. Comparison enables us to discover what is unusual about any given individual, event, group, process, or context. Comparing sharpens our awareness of assumptions that underlie our theoretical thinking, makes it clearer how concepts should be defined and operationalized, and may change what questions we ask. How do we choose what to compare and how can we defend our decisions? As the introduction to this symposium points out, methodological texts have focused overwhelmingly on designing controlled comparisons aimed at testing causal theories across a small number of cases. But comparative studies often have other goals, such as developing theoretical arguments, particularly so if they form part of a multi-method project. There is a discrepancy between the types of research designs scholars typically teach and the type of research they actually conduct (George and Bennett 2005, 10).