Census and PolicyMap: Racial and Ethnic Segregation
| Categories | Demographics & Population, Structural Vulnerability & Equity |
|---|---|
| Detail | Theil’s H index |
| Download Available | yes |
| Geographies | County, Metro Area / MSA |
| Public Edition or Subscriber-only | Public Edition |
| Source | U.S. Census American Community Survey (ACS), PolicyMap |
| Years Available | 2005-2009, 2010-2014, 2015-2019, 2020-2024 |
Description:
PolicyMap calculated Theil’s H index of segregation using the U.S. Census Bureau’s American Community Survey. For this index, PolicyMap used a total of 8 non-overlapping racial and ethnic categories provided by the US Census Bureau. These categories included the ethnic category Hispanic and the following 7 Non-Hispanic racial categories: White, African American, American Indian or Alaska native, Asian, Native Hawaiian or Pacific Islander, some other race, and two or more races. Theil’s H is an index ranging from 0 to 1 that estimates the extent to which racial and ethnic groups are evenly distributed in a sub-area as compared to a larger area. Values approaching 0 suggest that sub-areas have a composition similar to the larger area (i.e., even distribution, less segregation) and values approaching 1 suggest that the racial and ethnic composition of sub-areas within a larger area deviates from the larger area (i.e., non-uniform distribution, more segregation). On PolicyMap, sub-areas are defined at the level of the Census block and are compared to the following larger areas: block groups, tracts, counties and Core-Based Statistical Areas (CBSAs), which are an approximation of metropolitan areas. The calculation of Theil’s H is based on the methodology presented in the report, “The Multigroup Entropy Index” by John Iceland (2004). This methodology involves calculating the entropy, a measure of diversity, for each sub-area and larger area and calculating the population-weighted deviation in entropy values across all sub-areas within each larger area.
Areas where there are high concentrations of one racial or ethnic group occasionally produce a negative value. Rather than create confusion these values were suppressed. Geographies were suppressed when there were fewer than three sub-areas within a population. Geographies for which no data or limited data were provided or for which the population was less than 10 are represented as having “Insufficient Data.”