segregation and diversity vary across American metros
The graph below compares racial diversity and residential segregation in U.S. metropolitan areas with population’s of at least 500,000. One immediately noticeable thing is that there isn’t much of a trend between diversity and segregation. Some high-diversity places also have high segregation, but others have relatively low segregation. Ditto for low-diversity metros.
About the data
I calculated these statistics using 2015-19 American Community Survey block group data retrieved from IPUMS NHGIS, University of Minnesota. I included population counts for residents who identified as (non-Hispanic) white, Black, Asian, Native American, Pacific Islander, or Hispanic/Latino of any race.
The x-axis shows a simple Simpson’s diversity index for these groups. It shows the probability that two randomly chosen residents of a given metropolitan area would be from different groups.
The y-axis is a composite of the 13 individual dissimilarity indices calculated between every combination of 6 racial or ethnic groups indicated above. After calculating each index, I found the mean of them all, weighted to the share of the metro’s population included in each index.
Click here to download the underlying statistics for each metro area included in the graph.