Fuzzy Sets and Social Research

Abstract

four in-depth methodological discussions of the use of fuzzy sets in social research. They have in common that they confront and compare fuzzy set methods with mainstream techniques. These contributions should not be read as introductions to fuzzy set analysis (see Smithson 1987; Ragin 2000) but as attempts to validate this new methodology and demonstrate some of its strengths by comparing it with estab-lished approaches. In brief, fuzzy sets extend Boolean or “crisp ” sets by permitting membership scores in the interval between 0 and 1. With crisp sets, cases are perceived only as members or nonmembers of a set. The problem is that many core concepts in social research are best under-stood as graded sets. Examples include such dichotomies as coordi-nated versus uncoordinated economies, national versus international politics, the public versus the private sector, states versus markets, consensus versus majoritarian systems, democratic versus nondemo-cratic, federal versus nonfederal, employed versus unemployed, male versus female, high versus low, established versus nonestablished, rich versus poor, and so on (see Pennings 2003). At a theoretical level, most researchers are fully aware of the problematic aspects of using these concepts as simple dichotomies. But this awareness has not been translated into the application of methodologies that are fully equipped to study diversity and complexity in a set-theoretic manner. Fuzzy sets can help social scientists conceptualize social and political phenomena as sets with imprecise boundaries between