Remember that Hoax Site Correlating IQ and Vote? Well...

[McC]

Apologies if this belongs in Off Topic rather than here.

In any case, after the 2004 election, a website circulated purporting to compare average IQ of a given state with the way that state voted (red or blue) in the election. The obvious trend was that red states are dumber. Alas, for it was just a hoax.

Except...not.

I decided to run similar numbers myself. There aren't any statistics for actual average IQ per state (that I know of), but the census does collect information on the number of people over 25 with bachelor (or higher) degrees. Population density is also easily accessible. Using the degree percentage as an indicator of how well-educated a populace is and using the population density as an indicator of how urbanized or ruralized (i.e. how likely a state is to experience diverse populations, which strongly correlates with progressive ideaology), some pretty damning numbers result.

Average Percent of State Population Holding Degree

Code: Select all

              2000    2004
Red states    24.4    24.3
Blue states   29.3    29.9

Average Population Density of State (per sq. mi.)

Code: Select all

               2000    2004
Red states     92.3    90.3
Blue states   338.0   368.7

Average Degree Density of State (per sq. mi.)

Code: Select all

               2000    2004
Red states     22.6    22.1
Blue states   106.6   116.8

There are a few states that buck the trend, but the overwhelming leaning is pretty decisive.

[Kanastrous]

Isn't looking at what the Republican base believes (creationism, etc) enough to underwrite the conclusion that in large part they are not so bright...?

[Darth Wong]

It would also be interesting to look at rates of teen pregnancy, cigarette smoking, high school dropout rates, etc. Just for curiosity's sake, if nothing else. But it is indeed the case that if you're a creationist, you're more likely to be a Republican, and creationist = stupid. Or to put it another way, even if we assume that nothing about being a Republican inherently gravitates toward stupidity, they've been trying so hard to pander to the stupid demographic over the last 20 years that they must have surely increased their percentage of stupid people by now.

[Phantasee]

I'm interested to see the outlier states.

[McC]

Degrees (Normalized to 0-100)


Population Density (Normalized to 0-100)


Degree Density (Normalized to 0-100)

[McC]

Ghetto edit: I should specify that those are using 2000 election figures, not 2004.

[Surlethe]

I don't know that degree density (per sq. mi.) is a useful metric. Degree density is equal to population density (people/sq mi) multiplied by per-capita degrees (degrees/people). Does it tell anything besides emphasizing that blue states have both higher per-capita degree rates and higher population densities?

[McC]

I don't know that degree density (per sq. mi.) is a useful metric. Degree density is equal to population density (people/sq mi) multiplied by per-capita degrees (degrees/people). Does it tell anything besides emphasizing that blue states have both higher per-capita degree rates and higher population densities?

Probably not. I've mostly included it for the sake of completeness.

[Anguirus]

This was actually in my high school government textbooks. Of course, it had to take a centrist perspective and carefully avoid correlating college degrees with actual intelligence. Instead, it was phrased as something like "going to college tends to make an individual more liberal."

[McC]

This was actually in my high school government textbooks. Of course, it had to take a centrist perspective and carefully avoid correlating college degrees with actual intelligence. Instead, it was phrased as something like "going to college tends to make an individual more liberal."

I wouldn't correlate it with intelligence, either. Of course, the phrasing your textbooks used (if you're being literal) is also poor.

I would argue that there's a difference between the entrenched cultural ignorance of most Republican voters and their intellectual capability. I would correlate education and liberal perspectives, though, and the latter as a natural consequence of the former. The more education you receive, the more liberal views you are likely to hold by simple virtue of the fact that most conservative views disintegrate under any kind of critical scrutiny.

[CaptainZoidberg]

http://dabacon.org/pontiff/?p=539

Assuming this site's data is correct, democrats seem to pool around the left and right tails of the bell curve, while the Republicans reside in the middle and above average ranges.

This correlates pretty well with my personal experience. Most of the professors I've had tend to be pretty liberal, while most of the engineers I've met tend to be pretty conservative.

[Anguirus]

The more education you receive, the more liberal views you are likely to hold...

...was in my textbook.

by simple virtue of the fact that most conservative views disintegrate under any kind of critical scrutiny.

Was not. That should give you a better idea of what I was trying to get across.

[GySgt. Hartman]

Average Percent of State Population Holding Degree

Code: Select all

              2000    2004
Red states    24.4    24.3
Blue states   29.3    29.9.

Sadly, I am really lacking in statistical training, but wasn't there something about significance, like what difference would be considered explainable by chance? Is 5% a lot? Can you tell us where you got your data from? Some nice numbers would be median, minimum, maximum, maybe the 1st and 3rd quartile for both blue and red, and maybe for the raw data.

I'm really not sure if "degree density" makes any kind of sense.

You posted a graph labeled "Degrees (Normalized to 0-100)". I assume that is the percentage of the population that holds at least a bachelor's degree, normalized so that the largest percentage is 100? Why would you normalize that? It only changes the numbers on the x-axis, but not the graph, right? so it actually removes information.

Anyway, I took the graph and the liberty to play around with it a little bit. I just drew a line at the number 50 and counted how many bars were above and below. For your pleasure:
Number of bars above 50:

Code: Select all

total: 22
blue:  14 (1 of them just barely)
red:    8 (2 of them just barely)

Number of bars below 50:

Code: Select all

total: 27
blue:  6
red:   21 (1 of them just barely)

Number of blue bars:

Code: Select all

total: 22
>50:   14 (1 of them just barely)
<50:    6

Number of red bars:

Code: Select all

total: 29
>50:    8 (2 of them just barely)
<50:   21 (1 of them just barely)

I ignored West Virginia, since it had no bar at all.

[Sikon]

It would be foolish to make any implicit assumption that educational level correlates primarily with intelligence alone as opposed to correlating in large part with other factors including socioeconomic status and background. The latter can correlate strongly with party affiliation for much of the public, along with typical cultural background, predominate ideological tribalism in a peer group, etc.

Still, let's look at education level versus self-described party affiliation in a poll.

*******************

http://dabacon.org/pontiff/?p=539

Assuming this site's data is correct, democrats seem to pool around the left and right tails of the bell curve, while the Republicans reside in the middle and above average ranges.

This correlates pretty well with my personal experience. Most of the professors I've had tend to be pretty liberal, while most of the engineers I've met tend to be pretty conservative.

Ah yes. That links to a good source of data, so let's graph some. Poll results (excluding independents):



Data from SDA Berkeley set with v3254 and v3114 for data references: here

Within that distribution, if it was broken down, one would find statistical trends like for college campuses versus suburbia, inner-city regions versus rural areas, etc.

[Darth Wong]

It would be nice if those degrees were separated by field of study, particularly with respect to whether they have anything to do with the natural sciences.

People who get degrees like "communication" are imbeciles, but they are imbeciles from families which are affluent enough to send them to school to get completely useless degrees.

People like to say there is no correlation between intelligence and education, but that's only true if you include all of the useless bird courses that universities offer in order to increase their revenue stream. The correlation is quite valid for other fields: there are no stupid theoretical mathematicians, for example. That doesn't mean these people are infallible, but let's be realistic: a guy with an IQ of 90 isn't going to be a theoretical mathematician. However he could quite easily get a degree in communications with a minor in journalism, which is why I don't think it's meaningful to treat "education level" as a simple function of how far you went, regardless of which path you picked.

[Admiral Valdemar]

It would be nice if those degrees were separated by field of study, particularly with respect to whether they have anything to do with the natural sciences.

People who get degrees like "communication" are imbeciles, but they are imbeciles from families which are affluent enough to send them to school to get completely useless degrees.

People like to say there is no correlation between intelligence and education, but that's only true if you include all of the useless bird courses that universities offer in order to increase their revenue stream. The correlation is quite valid for other fields: there are no stupid theoretical mathematicians, for example. That doesn't mean these people are infallible, but let's be realistic: a guy with an IQ of 90 isn't going to be a theoretical mathematician. However he could quite easily get a degree in communications with a minor in journalism, which is why I don't think it's meaningful to treat "education level" as a simple function of how far you went, regardless of which path you picked.

How extensive do you think the anomaly of a well educated person like Behe is, who is also a Creationist? He's obviously smart enough to be a very good bio-chemist. On the other hand, he accepts a tenet of otherwise conservative thinking that flies in the face against science.

As you say, any decent degree of intelligence will be shown by someone being educated in science, maths or engineering (which is basically the applying of the former two to construction) where you cannot bullshit your way through something by learning factoids. But it is clear there are instances where smart, learned people are flummoxed into accepting things against their better judgement, simply because of their upbringing and/or political affiliation.

[CaptainZoidberg]

It would be nice if those degrees were separated by field of study, particularly with respect to whether they have anything to do with the natural sciences.

People who get degrees like "communication" are imbeciles, but they are imbeciles from families which are affluent enough to send them to school to get completely useless degrees.

What I've observed thus far at WPI is that mechanical and civil engineers tend to be Mitt Romney style conservatives, physics/math majors tend to be Ron Paul style libertarians, and the pre-law / psychology / humanities majors tend to be liberals.

That Republicans are more likely to have bachelor's degrees might be caused by the fact that engineering and computer science do not require one to go to graduate school, but to get a job an English major would usually need an MS or PhD.

I would be curious to see if my observations hold across the board, or if I just have a sampling bias.

People like to say there is no correlation between intelligence and education, but that's only true if you include all of the useless bird courses that universities offer in order to increase their revenue stream.

Even with the BS degrees I would be surprised if there were no correlation. A psychology degree is way easier than an Electrical Engineering degree, but it still requires that one reads a textbook, writes papers, etc.

The correlation is quite valid for other fields: there are no stupid theoretical mathematicians, for example. That doesn't mean these people are infallible, but let's be realistic: a guy with an IQ of 90 isn't going to be a theoretical mathematician. However he could quite easily get a degree in communications with a minor in journalism, which is why I don't think it's meaningful to treat "education level" as a simple function of how far you went, regardless of which path you picked.

It might be a good idea to just look at the political affiliation of students from a given school, say MIT, where all of the majors are probably quite rigorous.

It would also be nice if the SAT test included a questionnaire on political affiliation.

How extensive do you think the anomaly of a well educated person like Behe is, who is also a Creationist? He's obviously smart enough to be a very good bio-chemist. On the other hand, he accepts a tenet of otherwise conservative thinking that flies in the face against science.

I've met quite a few people like that in my life - people with a great aptitude for math or science, but who refuse to apply their abilities to religion or ethics.

[Sikon]

There aren't a lot of studies of political affiliation versus field of study in college easy to find quickly. However, going to eric.ed.gov, setting for full text availability, keywords: political, thesaurus descriptors: majors (students) ... results appear of which one has some relevant data: this

Data from table 4 of that document, for the most recent survey, converted to a graph using a spreadsheet illustrated here:



This is from the above survey of undergraduates at the University of California Berkeley. It's not entirely ideal. For example, I would prefer a survey of all students nationwide and in a more recent year instead. Still, it is the most relevant data readily found, and it suggests some trends.

[CaptainZoidberg]

That graph doesn't surprise me one bit (actually, I expected professionals to be somewhat more liberal). Business majors have good reasons not to be liberals, considering that liberals blame virtually everything on big business.

Computer Scientists might be more liberal because of their experiences with open source, UNIX, etc.

[Darth Wong]

I'd be less interested in "self-identification" than voting records. For example, people who voted for Bush. Twice. After all, the thread title is about voting, not "self-identification as liberal".

Frankly, self-identification as "liberal" in America doesn't mean shit. People in America don't know what the term even means. It's considered an insult in many circles.