Black–White and Hispanic–White Segregation in U.S. Counties

Abstract

Residential segregation in metropolitan areas has been the subject of much research, but this article analyzes patterns of white–black and white–Hispanic segregation in counties across the United States. Our purpose was to understand county variations in this one dimension of inequality. Conceiving of segregation as relative inequality of access to neighborhood resources, we measured segregation in 2000 by the index of dissimilarity (D) calculated by blocks, mapped the index values, and correlated them with census variables. Three filters enabled us to eliminate counties with characteristics that could have corrupted the analyses, leaving us with more than 1,000 counties in each analysis. Both minority groups were less segregated from whites in the West and South and in metropolitan counties. Lower segregation was strongly associated with higher minority socioeconomic status and higher percentages of minorities living in housing built in the 1990s, and Hispanic–white segregation was lower where more Hispanics were U.S.-born or English proficient. The racial threat hypothesis was supported only weakly and inconsistently. Mapping made it possible to identify regional and local patterns of high and low segregation as well as the lower segregation of suburban counties in some large metropolitan areas.

La segregación residencial en áreas metropolitanas ha sido tema de mucha investigación, pero en este artículo nos concentramos en el análisis de los patrones de segregación blanco-negro y blanco-hispano que ocurren a nivel de condado a través de los Estados Unidos. El propósito de nuestro estudio era entender las variaciones por condado en esta particular dimensión de la desigualdad. Viendo la segregación como la desigualdad relativa de acceso a los recursos del vecindario, medimos la segregación en el 2000 por medio del índice de disimilitud (D) calculado por manzanas, cartografiamos los valores del índice y los correlacionamos con variables censales. Mediante la aplicación de tres filtros nos fue posible eliminar los condados dotados de características que habrían podido viciar los análisis, lo cual nos dejó con más de 1.000 condados para cada análisis. Ambos grupos minoritarios aparecen menos segregados por los blancos en el Oeste, el Sur y los condados metropolitanos. La baja segregación estuvo fuertemente asociada con un mejor estatus socioeconómico de la minoría y con porcentajes más altos de minorías que residen en viviendas construidas en los años 1990; y la segregación hispano-blanco era menor donde había más hispanos nacidos en EE.UU., o que eran proficientes en inglés. La hipótesis de la amenaza racial solo apareció débil e inconsistentemente respaldada. El mapeo permitió identificar patrones regionales y locales de alta y baja segregación, lo mismo que la más baja segregación en los condados suburbanos de algunas de las áreas metropolitanas grandes.

Key Words:

• blacks
• counties
• Hispanics
• segregation
• whites

关键词:

• 黑人
• 县
• 西班牙裔
• 隔离
• 白人

Palabras claves:

• negros
• condados
• hispanos
• segregación
• blancos

Notes

¹Using data on geographical units in Census 2000, we calculated the mean area of blocks in each county that otherwise qualified in our study and arbitrarily eliminated from our study those counties with mean block areas of one square mile or greater (Minnesota Population Center 2004). It seemed likely that such small areas did not include two or more ethnically or racially distinct neighborhoods. This was because the county means of block areas were probably inflated by a few fairly large blocks, but these few must have been overwhelmed in numbers of very small blocks to produce a mean block area of a square mile or less.

² For each region we calculated the mean black–white segregation in central counties of metropolitan areas compared to the suburban counties. There were only very small and, we believe, insignificant average differences between central and suburban counties; the greatest difference was in the South, where the average segregation in suburban counties was 1.9 points less than in central counties. Because of the very small central–suburban county differences found for black–white segregation, we did not calculate these differences for Hispanic–white segregation. It seems likely that those suburban counties with low segregation as evident on the map were counties where minorities had higher socio- economic status and other characteristics found in our correlation analysis.