A measure of multidimensional polarization for categorical diversity data

Numerous indices measuring the degree of heterogeneity or homogeneity of small and large social entities exist. However, only recently has interest in multidimensional inequality and diversity grown. The novelty lies in looking at multiple dimensions of diversity simultaneously by asking how it is patterned across different attributes. The group faultline concept, which has gained some popularity in organizational research, is an instantiation of patterned multidimensional diversity in the context of small groups. Faultlines are thought to increase the potential for intragroup conflict. The paper provides a conceptual and operational elaboration of the group faultline concept and develops an index of patterned multidimensional diversity for categorical data based on the notion of multidimensional bipolarization. The application of this index to some illustrative group constellations leads to the conclusion that it is a valid, albeit not perfect, operationalization of faultline strength.