Top 10 Most and Least Diverse U.S. Census Tracts

Diversity in the Real World, Following Up on the Parable of the Polygons

In December, I introduced the Parable of the Polygons, an incredibly engaging and revealing bit of computer-generated game theory (read: fancy graphics with shapes that move based on laws of social behavior), designed to illustrate how segregation can happen in our cities.

This got me curious about how segregation happens in our cities today, and what it might look like to find a neighborhood that was truly diverse. In this vein, I’ve compiled data from all U.S. Census Tracts with populations larger than 500 and created an index* which ranks each by its level of diversity. Above, you can see a map of the top ten most diverse (green stars) and top ten least diverse (red teardrops) tracts in the U.S.

Here are the most diverse:

Rank Diversity (0-1) State Neighborhood Tract
1 0.9626 AK Mountainview, Anchorage 6
2 0.9552 AK Airport Heights North, Anchorage 9.01
3 0.9128 AK Russian Jack Springs Park North, Anchorage 8.01
4 0.901 WA Portland Avenue Park Area, Tacoma 9400.07
5 0.9081 NY South Ozone Park, Queens 100
6 0.9076 NY South Ozone Park, Queens 96
7 0.9023 NY South Ozone Park, Queens 840
8 0.8973 NY South Ozone Park, Queens 838
9 0.8959 NY South Ozone Park, Queens 846.01
10 0.8898 NY E. Jamaica Estates, Queens 478

Here are the least diverse:

Rank Diversity (0-1) State Neighborhood Tract
1 0.0126 VA N Buchanan County 102
2 0.0132 TX Maverick County, near Eagle Pass 9502.04
3 0.0225 PA Central Cambria County/Beaverdale 131
4 0.0229 OH NE Holmes County 9763.02
5 0.0235 WV Western McDowell County 9539
6 0.0239 KY SW Breathitt County 9205
7 0.0241 VA SW Buchanan County 107
8 0.0256 AZ SW Yuma County, near San Luis 114.06
9 0.0258 TN NW Claiborne County/Clairfield 9704
10 0.0258 NY East Midwood, Brooklyn 754

More analysis will be following, but in the meantime I’ll let you do your own analysis of these results.

*Index is based on an entropy index as described in academic literature. My method grouped U.S. Census data into 6 racial/ethnic groups: White, Black, Native American, Asian/Pacific, Other/Multiple, and Hispanic. The proportion of each of these groups per tract is then put into the following formula:

Where p equals the proportion of each group of the total census tract population

I realize that this is not a perfect representation of diversity, but perfection must be balanced with ease of use.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s