By Scott Jameson
800 words
Long and short of this issue is that something has to explain why most of the really, really smart people are men. There are two hypotheses: men have a higher mean, and men have a higher standard deviation. They don’t really have to compete, and so some people believe that both are true. Some believe neither, of course.
Let’s start with three facts:
- Women tend to get slammed by men on Raven’s Progressive Matrices; the second graph in the post linked above details this. It’s a difference of 5 IQ points on average, quite a bit, certainly more than on other IQ tests.
- Women tend to lose even harder in visuospatial measures. John Loehlin pointed out in The Handbook of Intelligence that the gap here was a whopping 13.5 points.
- Raven’s is so g loaded because your score is primarily driven by spatial and verbal-analytic abilities.
The biggest subtest difference is spatial, and I think that likely explains the abnormally large differences in Raven’s scores. Other IQ tests, like the SAT, hardly use visual abilities. Women do about as well as men on the SAT. I’ve also seen the White-Asian gap smaller on the SAT than in other IQ tests, and that gap is also driven in large part by spatial scores. Conversely you might expect the SAT to go better for a hypothetical demographic that scores well in math and verbal abilities, but not especially well in spatial. By hypothetically I mean that these people make up like a fifth of the kids at the Ivy Leagues, even more than you’d expect from an average IQ of, I don’t know, 111ish.
Off topic: these differences are probably going to be slighter still now that they’re fastidiously removing every useful element of the test in an effort to make it less “biased” by race. I wonder if colleges will just throw up their shoulders and start looking for kids who do well on the ACT. Moving on.
There are other sex differences in subtest scores. Pulling from Loehlin again: “females tend to have an advantage on verbal tests involving the fluent production of words belonging to a category, such as synonyms.” Women are known to do better on verbal than on math.
Loehlin also points out that girls do better at math in early childhood, but that boys outstrip them by the time it, uh, matters, when they take standardized tests in adolescence.
I have a wild hypothesis that men and women respectively being more oriented towards mathematical and verbal thought corresponds to observed differences in interests. Women are known to read more often than men on average, whereas male dominated activities like sports and video games often have a distinctly mathematical bent. My spurious hypothesis is that doing these different things differentially develops their abilities, constituting an example of crystallized intelligence rather than fluid intelligence; alternatively, they were differentially selected for ability to perform well on tasks that their respective sex does more of, in which case the abilities are innate.
Even if they aren’t innate, it’d be an instance of secondary heritability because evidence tends to show male-female personality differences as innate; in this scenario they are innately prone to practicing different abilities to different extents.
Loehlin points to Hedges and Nowell’s 1995 meta-analysis, showing a higher male variation in IQ and elucidating a few more small subtest differences. I’ve lifted a meaty bit here:
On average, females exhibited a slight tendency to perform better on tests of reading comprehension, perceptual speed, and associative memory, and males exhibited a
slight tendency to perform better on tests of mathematics and social studies. All of the effect sizes were relatively small except for those associated with vocational aptitude scales (mechanical reasoning, electronics information, and auto and shop information) in which average males performed much better than average females. The effect sizes for science were slightly to moderately positive, and those for perceptual speed were slightly to moderately negative. Thus, with respect to the effect size convention, these data suggest that average sex differences are generally rather small.
In summary:
- There are sex differences in scores of various IQ subtests, including but not limited to female orientation towards verbal and male orientation towards mathematical ability.
- The largest of these differences is a substantial male advantage in spatial ability.
- On any IQ test that doesn’t weight subtests such that men and women perform equally by default, men tend to score a hair better.
- Men also have a higher standard deviation in IQ.
There are more male geniuses, particularly with respect to mathematical genius. There are also more mentally retarded males. I just explained why men tend to populate CERN, NASA, Silicon Valley, and lists of who’s died in the Running of the Bulls.
NotPoliticallyCorrect 
Thanks for sharing this great article. Many wouldn’t think about talking about the standard deviation, but it is what makes it possible to talk about intelligence beyond scientifically, ethically, and politically more problematic descriptives of average/mean and the populistic pointing to examples. Had to smile: More geniuses, but also more retarded men:-)
LikeLike
Thank you, I’m glad you liked it.
LikeLiked by 1 person
I think women have higher VIQ than men because they needed to talk to the children more and whatnot. I brought this up in a psychology class last year and the professor agreed.
I know Lynn has a lot of papers on this and there is a paper that apparently validates Lynn’s arguments on sex differences in IQ in Nigeria. I’ll link it later.
And I’m beginning to believe that g is a psychological construct, not physiological.
LikeLike
I should check that out.
LikeLike
If you watch Stefen Molyneux at all he covers all the aspects of IQ spread across races and sex pretty well.
Men tend to be all over the spectrum for their race, women tend to gravitate more towards the average for their race.
LikeLike
I don’t need to watch Molyneux to learn that, I knew it already. (The fact that he pushes Anonymous Conservative’s r/K bullshit is a turn off)
Second paragraph, knew it. Scott covered it in the post even…
LikeLike
Scott,
Here are some articles on sex differences in IQ in Nigeria.
Sex Differences in IQ in Nigeria
Testing Lynn’s Theory of Sex Differences in Intelligence ina Large Sample of Nigerian School-Aged Children andAdolescents (N>11,000) using Raven’s StandardProgressive Matrices Plus
LikeLike
Hmm.. A couple of points.
Once you correct for body fat men do not, in fact, have larger brains.
In fact females have a higher EQ by a fraction. The correlation to brain size is the same for men and women of the same corrected body mass. Basically, there is no bigger male brain to account for the higher male score claim.
Next point, there’s zilch on the Y chr that could make the brain smarter. It’s a very small chr.
Third: 70% of special ed students are male, almost certainly as a result of defects on the X chr. Omit them from a mass IQ test and the male average IQ looks higher than female, but its down to sampling error
You get more males in the higher scoring IQ groups for a couple of reasons. The first is they are less likely to pick up an X chr defect (only one copy). The second is; now and then in-utero testosterone steals more processing capacity from social problem solving than it should and swaps it over to vs/maths ability. This is why you get a lot of Asperger’s scientists. The price is being socially handicapped (like Sheldon Cooper, it’s a trope for a reason). Social intelligence isn’t measured on IQ tests.
These points do two thing. Truly screw up the claim for a higher male IQ, and pretty much prove that brain size is a critical factor in human intelligence.
This does put people claiming racial equality ‘because brain size is irrelevant to IQ’ in a really sticky spot. Because the brain size IQ correlation is same whatever your race, and we know damn well that average brain size varies by population.
Really the only way out of this logic trap for the egalitarians is to claim the smaller brain size is down to poor environment. All they would have to do is show that the heritability of brain size and IQ is lower in black adults. Such work would put this whole argument to bed overnight. And no-one is doing it.
LikeLike
1) EQ isn’t a thing.
2) A brain does not need to be larger in order to be smarter. That helps but it’s unnecessary.
3) The Y chromosome turns genes on and off throughout the genome. If it couldn’t produce mental differences then it couldn’t produce physical differences. Which it does.
4) Male IQ is slightly higher even if you leave the mentally retarded kids in the analysis.
5) You should look at atavisionary’s comment regarding the x chromosome. It’s not about picking up defects, it’s about genes coding for intelligence existing on that chromosome.
6) Inuit have larger brains and lower IQs than Whites do on average.
LikeLike
Source?
Not really… People with the brain size of erectus can have IQs in the modern range. Re microcephalics. So if this is true—and it is—then large brains are not needed for high IQs. See Skoyles (1999) for more information and a review of microcephaly and IQ.
I think brain size is irrelevant to IQ and I don’t claim racial equality. Brain size varies by race, in my opinion, due to differing climates and the amount of tools used/made. I covered it in this article.
If you want to get technical, it is down to environment, yet over thousands of years. Large brains would quickly overheat; smaller heads are conducive to smaller pelves which allows for better bipedality. They made fewer tools near the equator. Large brains evolved in higher latitudes since it conserved heat, the same for wide pelves. Not everything has to do with intelligence and IQ!
LikeLike
Scott,
Correct. Brain size does slightly help, can’t deny the correlation. But a smaller brain would mean a smaller pelvis which is conducive to endurance running and a higher survival rate of mother and child.
Don’t they live in the coldest areas? Winter temperatures in the Nunavut, for instance, range between –15 degrees Celsius to –47 degrees Celsius (for those of us using Freedom numbers, that’s 5 degrees Fahrenheit to —- 53 degrees Fahrenheit. People from colder climates have larger brains and eyes which helps them to process information due to the low level of light in Northern climates.
Here is a paper comparing different Eskimo groups’ skeletal measurements (Hawkes, 1916).
As I covered in my article on Neanderthal/Homo sapiens strength, Neanderthals had wide pelves—wider than Inuits, which is termed ‘hyperpolar’ or ‘hyperarctic’ (Weaver, 2003). Also see fig. 2 in Weaver (2003). Hip shape—obviously—corresponds to climatic zone, since it’s easier to dissipate or conserve heat based on skeletal morphology. So since they have larger brains then they must have larger pelves than Europeans/East Asians. (Which clearly doesn’t give an advantage to IQ, but it does mean they have larger eyes which would be great for surveying the barren lands of the Arctic. They are great navigators; having larger eyes is extremely helpful for that.)
LikeLike
@MAB
You would really have to substract more than fat from equation. For example muscle mass is highly independant from number of neurons. When a person body-builds then muscle mass increases, while neural connection to muscles stay more-or-less the same. Thus men should have significantly lower number of neurons per muscle mass than women. You should also do correction for thicker bones in men.
Many studies show that men have higher social intelligence than women, for example men form larger and more cohesive groups than women, are more cooperative than women, and work better as a team.
According to the vast majority of studies men have a bit higher verbal ability than women (1- 2 IQ points) and are also more variable in that ability, which leads to a situation where men are the majority of the best writers, poets, philosophers, comedians, song writers, as well as 90 percent of Literature Nobel Laureats.
LikeLike
Relevant.
Average IQ of students by college major and gender ratio
LikeLike