IQ By State
IQ link of the day: an IQ Test which Seema assures me is free
Thanks to Steve Sailer for his comments on my recent post, Chad IQ. You can see his evidence that the chart is a hoax and his data on intelligence by state in the sidebar of his main page, iSteve.com. [Update: He’s provided a handy permalink for the whole issue here.]
That being said, I don’t believe that it has quite been proven that the IQ-by-state chart is inaccurate (even if it is a hoax). Certainly there’s no reason to believe the data since no citation is given - the attribution that lured me in has since been retracted and was apparently based on a misreading of the original link. Even if it were legitimate it would be insufficient proof of the idea that smart people vote for Democrats and dumb ones for Republicans (because of the Electoral College and other complicating social factors). If I had to guess what the political distribution of intelligence is, I’d agree with something Gene Expression mentioned - Democrats tend to have an inordinate share of both very smart and very not-smart voters (the outliers), and Republicans a bigger share in the middle of the brain range.
So when I wonder whether the chart is accurate, what I’m really wondering is whether it’s plausible that IQ in the US is distributed in the way the chart implies. It’s not enough to cite 8th grade public school tests when trying to approximate adult intelligence by state. Americans move around the country in certain patterns and their children (if they have them at all) regress to the mean. It’s entirely possible that the hoax values are more accurate than anything short of administering IQ tests to a random sample of adults, simply because they were extrapolated from incomes (though doubt has been cast on the accuracy of the incomes as well).
The people with the real answers are the IRS and the Census Bureau, not the NAEP.
[Update:] You knew I couldn’t walk away from that one, didn’t you? I’ve decided to provide some real state-by-state numbers. I ranked the states by the percentage of college graduates (that is, people over 25 with a bachelor’s degree or more). Income and votes for 2000 are included for those it amuses. College education isn’t the best measure of intelligence but it is a measure of the intelligence of settled adult populations, which makes it superior to children’s test scores for my purposes.
The data was assembled by me from the sources cited. Feel free to reproduce it and colorize it with the traditional reds and blues.
2000 Census and Election Results
Education: http://www.census.gov/prod/2003pubs/c2kbr-24.pdf (Table 1, 2000)
Income: http://www.nemw.org/pci.htm (2000)
Electoral: hearsay
% College Grads Income '00 Electoral
Massachusetts 33.2 37,756 Gore
Colorado 32.7 33,371 Bush
Maryland 31.4 34,257 Gore
Connecticut 31.4 41,495 Gore
New Jersey 29.8 38,372 Gore
Virginia 29.5 31,084 Bush
Vermont 29.4 27,680 Gore
New Hampshire 28.7 33,398 Bush
Washington 27.7 31,780 Gore
Minnesota 27.4 32,018 Gore
California 26.6 32,466 Gore
Hawaii 26.2 28,417 Gore
Utah 26.1 23,878 Bush
Illinois 26.1 32,187 Gore
Kansas 25.8 27,694 Bush
Rhode Island 25.6 29,216 Gore
Oregon 25.1 28,100 Gore
Delaware 25.0 30,871 Gore
New York 24.7 34,900 Gore
Alaska 24.7 29,863 Bush
Montana 24.4 22,932 Bush
Georgia 24.3 27,989 Bush
Nebraska 23.7 27,627 Bush
New Mexico 23.5 22,134 Gore
Arizona 23.5 25,661 Bush
Texas 23.2 28,313 Bush
Maine 22.9 25,972 Gore
North Carolina 22.5 27,071 Bush
Wisconsin 22.4 28,573 Gore
Pennsylvania 22.4 29,697 Gore
Florida 22.3 28,511 Bush
North Dakota 22.0 25,109 Bush
Wyoming 21.9 28,463 Bush
Michigan 21.8 29,553 Gore
Idaho 21.7 24,076 Bush
Missouri 21.6 27,243 Bush
South Dakota 21.5 25,722 Bush
Iowa 21.2 26,554 Gore
Ohio 21.1 28,208 Bush
South Carolina 20.4 24,426 Bush
Oklahoma 20.3 24,410 Bush
Tennessee 19.6 26,099 Bush
Indiana 19.4 27,134 Bush
Alabama 19.0 23,768 Bush
Louisiana 18.7 23,080 Bush
Nevada 18.2 30,438 Bush
Kentucky 17.1 24,414 Bush
Mississippi 16.9 21,007 Bush
Arkansas 16.7 21,926 Bush
West Virginia 14.8 21,901 Bush
May 9th, 2004 at 5:24 pm
The great “IQ hoax”
Steve Sailer has a full round-up over at his site. Readers of Lefty blogs should spread the word so that this false meme is counteracted…. Update: Speak Stiltedly and Wear a Yellow Short has a post that ranks % of…
May 10th, 2004 at 12:28 am
Thanks for the college data. The problem with looking at college graduate percentages is that it doesn’t represent the average for the state, just the size of the educated elite. For example, California ranks high in % of college grads but it’s also #1 in percentage of resident who have never even attended high school.
Liberal states like California and New York tend to be more unequal than conservative states like Utah and Kansas.
It can be approached more directly by looking at national exit poll data of voters:
In terms of education and party voting, there existed in 2000 a U shaped pattern, with Gore doing best with the least and most educated.†
Vote by Education
No H.S. Degree
5% % of total voters
59% Voted for Gore
39% Voted for Bush
High School Graduate
21% % of total voters
48% Voted for Gore
49% Voted for Bush
Some College
32% % of total voters
45% Voted for Gore
51% Voted for Bush
College Graduate
24% % of total voters
45% Voted for Gore
51% Voted for Bush
Post-Graduate Degree
18% % of total voters
52% Voted for Gore
44% Voted for Bush
This comes from the Voter News Service exit poll of 13,000 voters.†
A couple of caveats: First, people exaggerate their education. It’s unlikely that 18% of voters actually had graduate degrees. Second, a large fraction of Gore’s votes from Graduate Degree holders came from schoolteachers, and many don’t consider Ed School masters and doctorates to be in the same class in terms of mental demands as other advanced degrees.
UP
May 10th, 2004 at 11:17 pm
I considered high school graduation rates (available at the same URL) but decided against them because high school attendance is mandatory while college is voluntary, and it’s more common to pass students for social or political reasons in high school than in college. But the reasons you mentioned probably swamp mine, as far as ranking the states goes. The most accurate approach would be to rank by county or district, but I’m not about to assemble those figures.
I don’t think the 18% figure is far off - it’s well known that voting (at all) is correlated with education, only 50% or so of Americans vote, and (same citation as above) about 9% of the US population has post-graduate degrees. In other words, it’s still the elite wing of the Democratic party carrying the liberal states, even though the high school dropouts (20% of the US population but only 5% of voters) tend to vote Democrat as well.
May 22nd, 2004 at 9:08 pm
I agree that the hoax data may not be all false. My page http://SQ.4mg.com/schools.htm uses 4th grade test scores to rank state IQ. The result is the most believable to date, with the lowest scores for the District of Columbia, the deep South, and states with large percentages of Latino immigrants.
May 23rd, 2004 at 2:25 pm
I saw that, but it would be nice to do a rough conversion between the test scores and IQ. I found your inclusion of the hoax IQ’s in the table confusing.