Monday, November 17, 2008

White IQ and change in white vote by state from '04 to '08 election

Half Sigma's contention that the GOP is losing the intelligent faction is disconcerting, but the more I look at it, the more inclined I become to put stake in his assertion.

GNXP's Razib has created a convenient comparative table of the white vote by state (including DC) in '04 and '08, showing Kerry's and Obama's respective performances. Obama's gains correlate positively with estimated white IQ by state at .34 (p=.01). In Massachusetts, Obama only garnered 57% of the white vote compared to Kerry's 59%. Our most intelligent state happens to be the one that fielded Kerry, so it's an explainable anomaly. Removing it from the analysis increases the correlation to .39. For each added white IQ point, Obama's performance relative to Kerry's similarly increased one point.

Stepping away from the rate of change for a moment, consider the relationship between the total white vote percentage for the Democratic candidate by state and the state's average white IQ. In '04, the white Kerry vote correlated positively with white IQ at .43 (p=.001). In '08, it has increased to .48 (p=.0004). For every one point increase in white IQ, Obama's support among whites increases a little over four points.

That states with more intelligent whites are more likely to vote Democratic than states with less intelligent whites are is what whiterpeople are after when they insinuate Democrats are smarter and more enlightened than Republicans.

But it does not follow that more intelligent states are more likely to vote Democratic (although smarter states shifted more to the left than duller states did). In fact, the opposite is true*. Bush's total share of the '04 vote and the state's estimated average IQ correlate positively at .31 (p=.03). In '08, McCain's share and state IQ correlate at .26 (p=.07).

In summation, smarter white states give more white support to Democrats than duller white states do. That trend became stronger this election. Smarter states are still more likely to vote Republican than duller states are, but the gap narrowed this time around. Both of these observations lend credence to Half Sigma's assertion that the Republican party is losing its relative ability to appeal to intelligence.

Relatedly, see what Steve Sailer's perspicaciously dubs "The McCain Belt". You might consider ordering Albion's Seed if you've not previously read it.

* At first blush, this may seem contradictory. But keep in mind that nationally the GOP still comfortably wins the white vote while losing the black and Hispanic votes. The dullest white state (West Virginia, white IQ of 97.1) is still more intelligent than the smartest black state (Washington, black IQ of 94.5). Further, even though whites in high IQ states like Wyoming and Idaho are less likely to vote Republican than whites in low IQ states like Alabama and Mississippi are, Wyoming and Idaho are both more reliably Republican than either Alabama or Mississippi because of the large black populations in the latter two states.


Peter said...

Good analysis, but given that the white IQ range among states (excluding the DC outlier) is only about five points, does it really provide a valid basis for drawing any conclusions?

Audacious Epigone said...


Yes, I think so. I've looked at several correlations between state IQ and other attributes--obesity, incarceration rates, electoral participation rates, etc. Even by state, the range (ignoring the outlier that is DC) is only ~8.

Recall Murray and Herrnstein, in the BC, showing that a few points in average IQ makes a significant difference in the level of pathology in a society:

The drop from IQ 100 to 97 would increase poverty rates by 11 percent and the proportion of children living in poverty by 13 percent. All else being equal, similar rises would occur in rates of children born to single mothers, men in jail, high school drop-out, and men prevented from working due to health-related problems. In contrast, if average IQ were to increase 3-points to 103, poverty rates would fall 25 percent, children living in poverty would fall 20 percent, and high school drop-out rates would fall 28 percent.

Stopped Clock said...

I think that some of those predictions would come true and some wouldnt. If I remember correctly, his methodology was to take the population as it is today and see how many people were unemployed, single moms, etc., and then start chopping away at the right tail of the bell curve until he had a new sample with a mean of 97, and then check those variables again. Then likewise the same for 103. But while Im sure that violent crime would go down in a high-IQ society, I think that there would still be poverty, because menial jobs will still be there and they wont pay any better than they do today, unless the population as a whole votes to raise the minimum wage. But minimum wage isnt directly linked to IQ.

I'd use Asia to prove my point but there is really only one country in all of Asia that has a high IQ and has enjoyed a modern economy for long enough for the comparison to be considered fair: Japan. Taiwan, Korea, Singapore, etc are rich, but their isolation and small size has probably held back their growth to some extent.

scottynx said...

stopped clock: "I think that there would still be poverty, because menial jobs will still be there and they wont pay any better than they do today,"

Some menial jobs would still be there, but less so. Lots of menial jobs only exist because of the large supply of menial-quality labor there is to take advantage of them.

The US probably would lot less farming if it weren't for the huge influx of menial-quality labor.

Audacious Epigone said...

Menial jobs would likely be reduced, because the productivity of those laboring in menial positions would probably increase. Also, a more intelligent society is more likely to come up with the improvements and innovations necessary to obselesce some of the menial occupations in the first place. And those with higher IQs will still be more productive outside of their menial working existence (going home and writing aspiring screen play specs instead of going home and sitting in front of the TV with a beer).

chantix said...

We suggest you to buy chantix to quit the very dangerous 'nicotine' addiction by having the anti-smoking magic pill 'Chantix' which is Pfizer manufactured and approved by FDA on May 11,2006.For more sophisticated information about the side-effects of smoking and the treatment of it,Please logon to Chantix Online.