Mike the Mad Biologist

ScienceBlogling Razib has an excellent post, “Why it matters if liberals are much smarter“, about the recent work regarding IQ, political orientation, and religious beliefs. Razib’s point about the tails of the distribution is key: to the extent that the extemes of high (and low) IQ matter, a slight difference will mean that the skilled positions will be disproportionately held by one group. But what I haven’t been able to find out is how those differences arise.

What I mean by this is that IQ is actually composed several different things, and ranges from abstract geometry questions to language comprehension and general culture awareness. James Flynn has meticulously documented how over the last hundred years in many societies, including the U.S., IQ has risen to the point where approximately 30-40% of our grandparents would be considered mentally retarded. What’s happened is that more recent cohorts score much better on the abstract reasoning components of IQ tests, while the knowledge-based components haven’t really changed much at all (the irony is that this might largely be due to teaching children formally and informally how to do well on abstract reasoning tests).

It would be interesting to know where the differences between liberals and conservatives arise: is this primarily knowledge, abstract reasoning, or a general difference?

Anybody know? As best as I can tell, that hasn’t been reported.

Update, part I: Before you blow a gasket, Razib is a conservative.

Update, partII: What I’m trying to get at is why, on average, conservatives do worse on IQ tests. As Razib notes, this has profound implications for the relative proportions of high and low IQ individuals (2 standard deviations from the mean; the top and bottom 2.1% of scores, which, as a phenotype, does matter. Greater than 130 or less than 70 probably does mean something. Understanding why this is the case is important.

Comments

#1 John Danley
March 3, 2010

Lemme guess…

Reification: Not to be found in Conservapedia.
#2 Luther
March 3, 2010

Sounds like a conclusion looking for stats to cherrypick.

Take some nutcase notions and define them as “very conservative.” Do the opposite for “very liberal.”

I’ve never met a liberal engineer. Engineers are much smarter than liberal arts liberals. Therefore conservative are much smarter than liberals. QED
#3 DGKnipfer
March 3, 2010

If you’ve never met a liberal engineer you haven’t looked very hard. I know several, but I do see your point.
#4 Katharine
March 3, 2010

My tendency is to think that it has to do with the ability to combat things such as confirmation bias and to use metacognition. There’s also less of a concern about purity and ingroup-outgroup issues among liberals and atheists.

This reminds me of a study mentioned by a guy named Altemeyer who said that a bunch of liberals and a bunch of conservatives did a world simulation. The liberals ended up producing prosperity. The conservatives ended up blowing up the world completely, utterly, and horribly!
#5 WcT
March 3, 2010

@2 As a liberal engineer, I dispute your point, ergo liberals are indeed smarter than conservatives. QED
#6 Min
March 3, 2010

One possible explanation is the liberal cultural bias of those who devise the I. Q. tests. Verbal tests in general are sensitive to wording, and wording is sensitive to culture. And psychologists, the ones who determine the wording of the tests, tend to be liberal.
#7 razib
March 3, 2010

just to make sure people know: i am moderately conservative in my personal politics.
#8 Pierce R. Butler
March 3, 2010

Maybe the two groups are about the same, except that the test sample for the “conservative” side included S. Palin, G. Bush, or D. Quayle, and thereby mangled the average (regardless of the size of the cross section studied).
#9 Gray Gaffer
March 3, 2010

@6, you make the point for us. It takes brains to be a psychologist. If the majority of them are also liberal, well, QED.

However, my experience of IQ tests is that they test reasoning, not cultural bias. In fact, now I think on it, I’m not even sure I could identify a verbal cultural bias toward liberalism. Other than possibly: good reasoning tactics prefer independent rather than dependent axioms, which could be transferred to dealing with life without concern for what other people do in their own time and space, which is I believe a Liberal attitude. But that is a far stretch.

So who defined Liberal vs Conservative in the first place? And what do they have to do with fundamentalist Republicans vs Marxists? Or Right vs Left? Do the FR scholars make too much out of the Latin for Left – ‘Sinister’?
#10 Katharine
March 4, 2010

I’ve taken a handful of IQ tests; they never test on cultural or value sorts of things. It’s all speed of processing and numerical reasoning.
#11 geolith
March 6, 2010

Only saw an abstract, but the snip below suggests there is profit in exploring the covariation between brain areas involved in aspects of IQ and in cognition related to various political beliefs.

We found that three independent dimensions explained the variability of a set of statements expressing political beliefs and that each dimension was reflected in a distinctive pattern of neural activation: individualism (medial prefrontal cortex and temporoparietal junction), conservatism (dorsolateral prefrontal cortex), and radicalism (ventral striatum and posterior cingulate). The structures we identified are also known to be important in self-other processing, social decision-making in ambivalent situations, and reward prediction.

Soc Neurosci. 2009;4(5):367-83. Epub 2009 Jun 26.
Individualism, conservatism, and radicalism as criteria for processing political beliefs: a parametric fMRI study.
Zamboni G, Gozzi M, Krueger F, Duhamel JR, Sirigu A, Grafman J.

http://www.ncbi.nlm.nih.gov/pubmed/19562629
#12 snoo
March 29, 2010

@Katharine: Intelligence tests don’t test for “culture” but the test items do have cultural assumptions built into them. There is a considerable body of research about this and the most apparent difference is seen when non-Western, indigenous peoples are tested. Their interpretation of the question as well what they offer as suitable answers can be quite different. Also, as Min suggested above, language ability plays an integral role in how one performs in an intelligence test and culture is inherent in language.

Without having read the original Kanazawa study, I’d hazard a guess that the underlying link between IQ and liberalism/conservatism has to do with education and that liberals and conservatives (in the social, not economic sense) differ on abstract reasoning more than they do on knowledge. Modern education relies heavily on logic and on abstraction, i.e. learning theory and then being able to apply it in practice. It is likely that people who have more abstract reasoning ability are more successful in education which in turn also makes them “smarter” on knowledge and cultural awareness i.e. liberals. In contrast, conservative thinking is dominated by ingroup vs. outgroup thinking and as Katharine said above, less reliance on higher level reasoning.

But who knows whether IQ drives political orientation or whether some immanent politcal bias drives IQ scores?