« Fall in Health Costs: Eye Laser Surgery, an Uninsured Item | Main | Peculiar Peoples: Christians, Conservatives, Scholars »

November 27, 2004

Sailer on the IQ's of Bush and Kerry Voters, Different States

Steve Sailer's "The 2004 IQ Wars: So Much For The Candidates--What About The Voters?" is good. Exit polls are the standard way to check on the income and education of voters for particular candidates. As Sailer says, we have good reason now to mistrust the quality of the polls, but here is what they say:

Income. Not surprisingly, given Bush’s tax-cut agenda, voters with incomes over $100,000 went for Bush over Kerry 58-41 .

Education. In 2000, the self-reported educational level of the average Bush and Gore voters was virtually identical. In the 2002 House races, Republican voters did quite a bit better with the well-educated, winning 58-40 among college graduates and even winning a majority among those who had undertaken some graduate study. (The latter's ranks are inflated by Democratic-voting public school teachers who have done post-grad work in the easy field of Education.) v But in 2004, as you may have noticed, Bush ran a pretty dumbscale campaign. The Democrats normally win by a landslide among high school dropouts. This time, however, Bush wrestled Kerry to a draw among that segment. Bush ended up with an average voter with only a month and a half less schooling overall than Kerry's typical supporter.

Sailer quotes Professor Lindgren on the interesting fact that General Social Survey data (the standard sociology dataset) shows that conservative Republicans and liberal Democrats are both above average in education and do better on short word tests compared to moderates.

I was struck by Sailer's observation that "people with some graduate work" really means "public schoolteachers". He's quite right. Their Master's and Doctoral degrees, despite meaning very little, are extremely numerous, swamping real advanced degrees. So it is not the liberal professors who are driving the liberalism of people who have done graduate work. In fact, the professors are probably swamped even by the MBA's and MD's.

Much of his article debunks the comparison of Bush states with Kerry states, making obvious points such as that in the South, Bush voters almost certainly have higher IQ's than Kerry voters. But he also raises a problem with the plausibility of the data we are told about.

Everyone familiar with IQ testing scoffed at the validity of the hoax data that claiming the average IQ in Connecticut was 113 and in Utah was 87. To see why, it's important to understand how IQ tests are scored.

The mean is typically set at 100 and the standard deviation is 15. This implies that Utah's average person would fall 26/15ths (or 1.73) standard deviations lower than the average person in Connecticut.

Using the Normdist function in Microsoft Excel, you can easily put this on a percentile basis. This hoax data therefore implies that a Utah resident of average intelligence (50th percentile) would be only at the 4th percentile in Connecticut. The average person in Connecticut (50th percentile) would suddenly be at the 96th percentile if he moved to Utah.

When phrased like that, the numbers appear obviously wrong.

Posted by erasmuse at November 27, 2004 04:24 PM

Trackback Pings

TrackBack URL for this entry:


Post a comment

Thanks for signing in, . Now you can comment. (sign out)

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Remember me?