Residential segregation is a multidimensional construct with five distinct axes of spatial variation

Evidence

In 1976, Cortese and colleagues ignited a vigorous new debate about how segregation should be measured by questioning the universal applicability of the index of dissimilarity. In response to their critique, numerous alternatives were proposed, yielding a proliferation of indices and considerable confusion in the literature. Massey and Denton (1988) sought to bring some order to the field by demonstrating that residential segregation was actually multidimensional, characterized by five distinct axes of spatial variation - evenness, exposure, clustering, concentration, and centralization — which comprehensively characterized the spatial circumstances of any racial and ethnic group. They showed that the index of dissimilarity measured the evenness dimension quite well, thus confirming Duncan and Duncan’s earlier work; and after clarifying the conceptual and empirical properties of the remaining dimensions, they went on to recommend a specific index for each one. — David Theo Goldberg, John Solomos, A Companion to Racial and Ethnic Studies, 2418

Categorization

• DEFINE: Formally defines residential segregation’s five dimensions: evenness, exposure, clustering, concentration, and centralization.