The Spatial Context of Residential Segregation

Abstract

Residential segregation is defined in a variety of ways to address a common concern: to what extent do social groups reside in separate or distinct places. The spatial pattern of segregation varies widely across cities, and distinct spatial patterns can be generated by different mechanisms and have different consequences for residents and their communities. However, the methods commonly employed to measure segregation ignore how a city is spatially organized. They do not take into account the spatial context of segregation patterns, including 1) the spatial arrangement of residential locations and neighborhoods, and 2) the physical connectivity or spatial boundaries between them. I have developed a new method that overcomes the limitations of conventional approaches and enables a closer examination of how individuals' residential contexts are shaped by their spatial environment. I demonstrate the contribution of my method using a series of stylized cities that represent patterns of racial segregation observed in U.S. cities, such as Detroit. The results reveal distinct spatial patterns of segregation for cities that conventional approaches measure as having the same level of segregation. It offers the first quantitative method for systematically studying how spatial boundaries structure patterns of residential segregation within and across cities. Better measurement is critical for advancing our understanding of segregation patterns and processes, and explaining how and why segregation matters for individual and community outcomes. My new approach bridges qualitative insight on the spatial context of segregation with the quantitative measurement of segregation for city populations.