"What's the best level to analyze the vote? Should we be looking at individual data? County returns? State returns? There's no obvious right answer here. Yes, individuals, not counties or states, are the ones that cast votes. But people are not islands. They often think as members of communities and evaluate political events in terms of their impact on their geographic area."
In other words, voting takes place on the individual level, but one's sense of community and geography influences that.
Voters are certainly more likely to latch onto fixed boundaries rather than those lines that are apt to change every ten years.
But it goes beyond the simplicity of congressional districts vs. counties. People aren't necessarily tied together because of lines drawn in the sand or boundaries between county and county or state and state. The map below, for instance, from the CommonCensus Map Project is based on respondents' senses of spheres of influence -- of what major city has the most cultural and economic impact on the area you are in.
And that opens up an entirely different set of questions or at least an alternate unit of analysis: the virtual boundaries of these spheres of influence. As was mentioned by Jim Gimpel over at the Monkey Cage, but without the visual, this ends up looking an awful lot like the various media markets across the country. And there's some truth to that.
My natural inclination is to look at one of two places first: North Carolina and Georgia. In the case of the Tarheel state, there is a fair amount of overlap between the virtual spheres of influence boundary and the media market boundary. That's just the fragmented nature of the state; between the mountain, piedmont and coastal regions on the one hand and the three major urban areas (within the piedmont) -- Raleigh-Durham, Greensboro-Winston Salem and Charlotte -- on the other. Those boundaries hold up across the spheres map above and the media markets map below.
Georgia, though, is a different story. The state is divided into several media markets, but Atlanta subsumes those in terms of influence across the Peach state.
Interestingly, Brian Arbor in the comments to Gimpel's post draws a comparison between the idea of political/cultural spheres of influence and the loyal followings of various sports teams.
"When I lived in San Jose and Sacramento, you would see the Sharks and the Kings everywhere. This did not seem to be because people in those cities were massive hockey or basketball fans, per se. They were fans of a professional team that had their city’s name on their chest. In San Jose and Sacremento, this helped create an identity beyond being just a bedroom community to San Francisco. Common Census has a Bay Area map among its regional maps, and shows some of this effect."Whether politics, major cities or sports, though, this CommonCensus idea is an interesting one that forces us to reconsider the definition of boundaries -- thinking of them less in terms of lines and more so in terms of what ties groups of people or voters together.
[Yeah, you're right. I just wanted an excuse to put another set of maps together.]
Too Good Not to Mention: Coach K on Obama
Pack Your Bags: FHQ is North Carolina Bound
2012 GOP Presidential Candidate Emergence