Posted by Jay Livingston
The quote is usually attributed to political consultant James Carville. But how much “in between” is there? That’s crucial if you’re counting votes, which is mostly what Carville is interested in. But it’s also important for demographic variables that might not have much to do with voting.
I was thinking about this problem today because I’d just assigned students to look at the distribution of a variable across states. The trouble is that when you see a high or low score on some variable for a state, there are two important things you need to know:
- how concentrated is the state’s population; how much of it is accounted for by one or two large cities?
- how different are the metro and non-metro populations on this variable?
Neil Freeman at Fake is the New Real gives us some help by slicing metro areas (blue) away from states (brown) and then resizing each according to population. Here’s Pennsylvania, carved à la Carville.
New Yorkers often distinguish between the New York City area (NYC plus the Long Island and Westchester suburbs) and everything else, called “Upstate.” Here’s how that one looks (
(Note: the scale in the two graphics is the same. So Pennsylvania without its cities is more populous that New York without its cities. Pittsburgh metro is much larger than Buffalo or Rochester.)
Other interesting states:
- Illinois – Chicago and Downstate
- Texas – even without its big cities, Texas ranks fourth (after NYC, LA, and Chicago). There’s still a lot of non-metro Texas. Don’t mess with it.
- Nevada – Las Vegas (ranked #64) dwarfs the rest of the state (#93).
- New Jersey–
FakeFreeman takes out the urban areas, giving them either to the NYC or Philadelphia metro area. After that, there’s just not much left – geographically, at least (in population, non-metro NJ is ranked 89th, which puts it ahead of a half-dozen intact states).