More GOP birthers in heavily black states?

[Update (6/30/10): Serious questions have been raised about the validity of Research 2000's polls. The results discussed below should thus be viewed as potentially suspect until the matter is resolved.]

Two new polls are out measuring the state-level prevalence of the misperception that President Obama is not a citizen of this country.

Tom Jensen of Public Policy Polling has released a preview of a poll showing that 47% of North Carolina Republicans think President Obama is not a citizen -- an even more disturbing finding than his previous poll, which found that 41% of Virginia Republicans believed in the myth. By contrast, a Deseret News/KSL-TV poll found that only 13% of Utah Republicans -- and 9% of Utahns generally -- said that they believe Obama is not a citizen (via David Weigel).

These results are consistent with the national figures from a Daily Kos/Research 2000 poll, which found that the myth was endorsed by 28% of Republicans (and 11% of Americans) overall and that it was more prevalent in the South.

What explains the state-level differences in birther misperceptions that we observe? The Washington Independent's David Weigel suggests the difference may be linked to a lack of racial polarization in Utah:

So why does rock-solid Republican Utah have fewer “birthers” than, the deep South, or even fewer than blue Virginia and North Carolina? A lack of racial polarization has something to do with it. Utah, like the rest of the great plains and western states, got bluer in 2008 despite overall McCain victories and despite having a very, very white population. In Utah, Obama got 327,670 votes in 2008, up from the 241,199 votes that Sen. John Kerry (D-Mass.) got in 2004. For the first time since 1964, the Democratic candidate for president actually carried Salt Lake County. This happened with 31 percent of Utah whites backing Obama. Not even close to a winning margin; but in Louisiana, for example, Obama only won 14 percent of the white vote.

The reason, of course, for the lack of racial polarization in Utah is that it is overwhelmingly white. By contrast, states with large black populations (particularly those in the South) are often much more polarized along racial lines. Following up on my analyses of state-level Obama support by black population, I therefore plotted state-level GOP birther misperceptions against the state-level black population (with the aggregate US total added for context). While it is obviously far too early to draw any firm conclusions, the result is highly suggestive:

Again, the plot is only for illustrative purposes -- it is far too soon to tell if the relationship will hold with data from more states. But the fit to the state data is almost perfectly linear thus far (R²=.99).

(Cross-posted to Pollster.com)

Update 8/11 1:53 PM: Full results from the North Carolina poll are here (PDF).

Comments

[Sarang]

Yeah, but what about W.Va. or Kentucky (pretty low black population) vs. say New York state?

[Jinchi]

I think you're ignoring the elephant in the room. The difference between Utah and Louisiana isn't that Utah is homogeneous. It's that Louisiana and the other Southern States (aka the former Confederacy) have a history of racial conflict.

We saw the exact same trends in the election. If you had taken a road trip from Alabama north to Maine you would have found yourself traveling in states in which whites were progressively more supportive of Obama regardless of the diversity or prosperity of the population.

http://4.bp.blogspot.com/_QsEqf8R-X6I/SSO97dOFkfI/AAAAAAAAADY/ivwtu58T4gk/s400/Alabama_to_Maine.jpg

It has nothing to do with the percentage of blacks in each state.

[David]

This possible relationship would be unimportant, unless it's asserted that there's some sort of cause-and-effect relationship. Otherwise it's just an irrelevant coincidence.

I think it will be impossible to prove causality, because other factors may also correlate well with percentage of black population and with percentage of birthers, such as average income and some measure of education.

IMHO a more important variable might be the nature of conservatism in the state. Conservatism has various strains. One type of Republican might be more apt to be a birther than some other type. Unfortunately, this theory would be hard to convert into figures that could be analyzed statistically.

BTW I am not impressed with the very high value of R square. That's likely a coincidence. That is, even if there were a cause and effect relationship, one wouldn't expect so perfect a linear fit.

[NoMoreBlatherDotCom]

I think the chart would be improved if the DT variable were added in. (For those not familiar with the DT variable, that measures someone's ability to believe two contradictory ideas at the same time. For just one example, believing that HI's gov can't verify he was born there because it would be illegal at the same time as believing that she verified he was born there. Or, believing that HI has authenticated the picture on BHO's site at the same time as believing that they haven't. It's an art!)

[Robert David Sullivan]

I'm going to use Occam's Razor here and point out that Utah is one of the best-educated states in the US. Until I see that there are more birthers in racially diverse Maryland than in almost-all-white Kentucky, my hunch is that educational attainment is the strongest demographic factor in play here.