Thursday, November 06, 2008

Educational index by state

In refuting a defense of Sarah Palin based on the presumption that educational attainment is relatively rare in Alaska, Half Sigma asserted (with evidence) that Alaska actually fares slightly better than the national average when it comes to the percentage of the population with a bachelor's degree or higher. This spurred the creation of an educational index by state built not only on the percentage of the population successfully involved in higher education, but also on the percentage of the population that did not make it through high school.

Agnostic suggested tweaking the index to give more weight to states with large college-educated populations, rhetorically asking:

What if you defined the index as (SMART - DUMB) * SMART? Where the smarties and dumbies are about as common, the original and new index are about 0. But what about two states where there are many more smarties than dumbies, like Washington vs. Utah -- wouldn't you rather live where the fraction of smarties is larger, all else equal? (Which it probably isn't -- more expensive to live in WA than UT.)

Honestly, I just have an aversion to the behaviors of the hard underclass. Minimizing that is most important for immediate quality of life. I spend some free time in mixed working class areas and actually enjoy it. But that's being short-sighted on my part, I don't live there, I'm not raising a family, and anyway, the question was rhetorical so I shouldn't flatter myself in thinking it was directed specifically at me!

Doing this alters the index very marginally. Agnostic's is nearly identical to the original. The two correlate at .97.

Still, he makes a good point. Since the subject came up from Half Sigma's ongoing interest in the voting tendencies of smarter and dumber states, and 17% of this election's voters were postgraduates, I changed the formula so postgrads would be 'worth' 1.5 times as much as those with bachelor's degrees in determining a state's score. It is computed by multiplying the postgraduate percentage of the population by 0.5, adding it to the percentage of the population with a bachelor's degree or higher, and then subtracting the percentage of the population not having completed high school (all for people above the age of 25) and multplying by 100 for ease of viewing.

Following is the new index, colored according to the '08 Presidential election, with the inclusion of DC*. Light blue indicates an Obama margin of victory of ten points or less, and light red indicates the same for McCain:

1. District of Columbia -- 28.7
2. Colorado -- 25.4
3. Massachusetts -- 25.3
4. Connecticut -- 22.3
4. Maryland -- 22.3
6. Vermont -- 22.0
7. New Hampshire -- 21.4
8. Minnesota -- 20.0
9. Washington -- 19.9
10. Alaska -- 17.9
11. Utah -- 17.8
12. New Jersey -- 17.7
13. Virginia -- 17.5
14. Kansas -- 16.3
15. Montana -- 15.4
16. Hawaii -- 15.2
17. Oregon -- 15.0
18. Nebraska -- 14.0
19. Wyoming -- 13.4
20. Illinois -- 12.8
20. New York -- 12.8
22. Delaware -- 12.2
22. Maine -- 12.2
24. Wisconsin -- 11.1
25. Iowa -- 10.5
26. Idaho -- 10.0
27. Michigan -- 9.4
28. North Dakota -- 9.3
29. South Dakota -- 9.2
30. Pennsylvania -- 8.8
31. Arizona -- 8.6
32. Rhode Island -- 8.6
33. California -- 8.5
34. New Mexico -- 7.9
34. Ohio -- 7.9
36. Missouri -- 7.4
37. Georgia -- 6.9
38. Florida -- 6.5
39. Indiana -- 5.4
40. North Carolina -- 4.5
41. Oklahoma -- 4.2
42. Texas -- 2.9
43. Nevada -- 1.9
44. South Carolina -- 0.6
45. Tennessee -- (1.0)
46. Alabama -- (2.0)
47. Louisiana -- (3.2)
48. Arkansas -- (4.8)
49. Kentucky -- (5.1)
50. West Virginia -- (6.7)
51. Mississippi -- (7.3)

Far from being an uneducated wilderness, Alaska is the most educated reliably red state.

Parenthetically, Steve Sailer has helpfully pointed out that younger populations, all else being equal, tend to be more educated than older populations are. This means red states look even worse by comparison to blue states, as younger states are more likely to vote Republican than older states are. The correlation between McCain's share of a state's vote and its median age is an inverse .29 (p=.04).

* As a city included in an index comparing states, DC is an outlier by default. Levels of inequality in DC are higher than they are in any states as a whole (inequality, incidentally, is a great thing for Democrats). Not surprisingly, it is an anomaly in terms of the relationship between estimated IQ and educational attainment of the population. Without DC included, the correlation is .59 (p=0). DC, however, does the best in the country on the educational index even though it has a lower average IQ than any state does, including those in the South.


Peter said...

Note the caution pointed out at Siggy's. It almost certainly is not true that 17% of the electorate has postgraduate degrees. It is more reasonable to say that 17% of the electorate have taken some course work beyond the bachelor's level. This would include, for instance, teachers who have to do some graduate work to maintain their certificiations, and liberal arts degree holders who take a graduate course or two while barista-ing.

Audacious Epigone said...


I'd be interested in a link to that if you wouldn't mind. 9.5% of the US population has a graduate degree, which would mean postgrads turning out in numbers 79% higher than would be expected if each level of educational attainment voted at equal rates. That's exactly in line with college graduates as a whole--25% of the population has a bachelor's or higher, yet the group constituted 45% of the electorate during this election cycle (80% higher than expected).

I bet the average IQ of someone with an undergrad in chemical engineering is higher than the average IQ of someone with a graduate degree in education at just about any given school, but we're looking at the aggregate level of education, not necessarily intelligence.

Steve Sailer said...

Keep in mind that people lie about their education levels on exit exams.

Steve Sailer said...

Alaska and Utah are young states, so that pushes them up higher. North Dakota is an old state (no jobs for young people) so I suspect it's residents are pretty well educated relative to their ages.

Of course, this methodological problem will diminish as time goes by because the high school graduation rate hasn't gone up in over 35 years.

Steve Sailer said...

What people who talk about this are really interested in are differences in white people, not the overall population. West Virginia's whites lag all the other whites in years of education tremendously. I've got the numbers somewhere from the 2000 Census.

Audacious Epigone said...


But is it in a way that skews one way or the other? If the liars just inflate educational attainment across the board, relative comparisons between parties/candidates still work.

I know you have a thousand things to do with your new book hitting the press (I'm bummed I missed the Ron Smith show yesterday--by the time I paid my daily visit it was already after 4pm central), but if you have a link to white educational stats by state from '00, I'd really appreciate you dropping it.

Steve Sailer said...

Here's the 2000 Census on the Average Years of Schooling for Non-Hispanic Whites:

Avg Years Schooling NH Whites 2000

United States 13.3

1 District of Columbia 16.2
2 Hawaii 14.2
3 Colorado 14.0
4 New Mexico 13.9
5 California 13.9
6 Alaska 13.8
7 Utah 13.7
8 Maryland 13.7
9 Washington 13.6
10 Arizona 13.6
11 Massachusetts 13.6
12 Connecticut 13.6
13 Texas 13.5
14 New York 13.5
15 Minnesota 13.5
16 New Jersey 13.4
17 Oregon 13.4
18 New Hampshire 13.4
19 Virginia 13.4
20 Kansas 13.4
21 Illinois 13.4
22 Vermont 13.4
23 Wyoming 13.3
24 Montana 13.3
25 Nebraska 13.3
26 Idaho 13.3
27 Delaware 13.3
28 Nevada 13.3
29 Florida 13.3
30 Georgia 13.1
31 Wisconsin 13.1
32 Michigan 13.1
33 Iowa 13.0
34 Maine 13.0
35 South Dakota 13.0
36 North Carolina 13.0
37 Rhode Island 13.0
38 North Dakota 13.0
39 Ohio 13.0
40 South Carolina 13.0
41 Oklahoma 13.0
42 Pennsylvania 12.9
43 Missouri 12.9
44 Indiana 12.9
45 Mississippi 12.7
46 Louisiana 12.7
47 Alabama 12.7
48 Tennessee 12.5
49 Arkansas 12.5
50 Kentucky 12.2
51 West Virginia 12.2

Steve Sailer said...

So, to summarize, DC is off in the stratosphere at the top, states with nice scenery and good weather tend to be near the top, hillbilly states are at the bottom.

Leaving out DC, it's really only a range of a couple of years on either side of 13.3.

In contrast, Average Years Married between 18 and 44 ranges from 12.2 in California to 17.0 in Utah. So, there is considerably more difference in Years Married than in Years Schooled among the states.

Audacious Epigone said...


Tremendous, thanks so much! HS' commenter who argued educational attainment is given low priority in AK really appears to be off--it's the highest among red states.