The Spatial Causes of Racism

Nate Silver is definitely one of my favorite bloggers out there. He is the main voice behind FiveThirtyEight.com, a fantastic source for statistical analyses of political trends and voting patterns.

The above video is from a 9 minute presentation he gave at a TED conference this past February. During the talk, he discussed the topic of race during the 2008 election. More specifically, he asks the provocative question, “Can we predict racism?”

From a strictly statistical standpoint, the answer, apparently, is “Yes.” While he doesn’t get into the specifics of his methods, it looks like he ran a few linear regressions to look at the association between individuals admitting that race was a deciding factor in their vote and voting for McCain. In states like Arkansas and Tennessee, there was a very strong relationship between voting against Obama and saying that race mattered in that decision, suggesting racist voting patterns.

Around the 6 minute mark, he gets into a fascinating discussion of why this may be so. You can check it out for yourself, but the gist of his argument is that racist voters tended to live in mono-racial neighborhoods (i.e. didn’t live near any minorities). This social isolation predicts (statistically) racist voting.

In the social sciences, we typical discuss social isolation in terms of the urban poor. The general idea is that the urban poor tend to live in neighborhoods of concentrated poverty, and this spatial concentration, in turn, isolates these communities from mainstream resources, role models, and social networks.

Silver’s analysis is so fascinating because he applies this same concept to white, suburban neighborhoods. He compares the grid-style format of streets that typically defines a city to the cul-de-sac and dead-end model of suburban sprawl. He argues that the grid format of city blocks facilitates interaction among residents, while sprawl spatially disconnects residents from one another. In a sense, land-use patterns provide a spatial determinant for racism. In other words, the way your neighborhood was built may influence how you view people of different races and ethnicities.

I’m not convinced that we can actually predict racism, but I do agree that analyzing land-use patterns and more equitable, sustainable urban environments could make serious inroads in our fight against racism. Check out the video; there’s a lot of interesting questions to think about.