Steven Malanga points out what seems like a pretty fatal flaw in Tom Frank's What's the Matter With Kansas?—namely, that Kansas isn't doing all that poorly in economic terms. Some of his stats could be red herrings (employment rates don't necessarily tell you all that much about how a state's doing [what if the jobs all suck, say?])—but if you look at the 2004 Census data, Kansas' median income is above the national average, it has more people insured than average, and fewer people in poverty. So it seems wrong to say, as Frank does, that Kansans are ignoring their own economic misery and voting purely on cultural issues when there isn't, in fact, much economic misery.

On the other hand, a lot of red states are sucking wind economically. Alabama, Kentucky, Oklahoma, West Virginia. Maybe Frank had the right idea, then, but just applied it to the wrong state. (As an added bonus, those other states have been known to actually respond to economic populism -- way back in the New Deal era -- whereas Kansas toed the conservative line through all the big realignments of last century.)

Of course, the correct way to look at this isn't always at the state level. The crucial question for politics is whether a key bloc of swing voters is a) doing rather poorly on economic issues, and b) voting Republican on cultural issues. Kansas voted for Bush by quite a bit, but it ultimately wasn't that many people who actualy swung the election, and the stories of those people could easily be "hidden" in the state-level data. Improbable? Aye. Not impossible, though.