Tag Archives: gender

Brain size, relative intelligence, sex and fat.

After re-reading the ludicrous (and I’ll explain why it’s ludicrous) assertion that women have an average IQ five points lower than the male claims from Lynn, I thought I’d dig up some of the facts and figures surrounding this flawed work.

First of all I’m going to put a link to this paper on relative brain size and intelligence by Tom Schoenemann, a professor whose specialist field is the evolution of the human brain. It has several numbers relevant to this subject:

Male:          (55.5 kg body weight, 1361 g brain weight)
Female:    (51.5 kg body weight, 1228 g brain weight)

As you can see, the human male appears to have a brain size proportionally slightly larger than the female. However, one thing  routinely ignored in all these measurements is fat. The average human male  (Western) carries a body fat of about 16%, women carry about 22%. Which means you need to calculate the relative brain size compared to non-fat mass, as fat is a null factor and as far as anyone can tell requires no processing power to control. My IQ at nine stone would be exactly the same if I weighed ten stone, although my relative brain size would have decreased due to the extra fat I now carried. This does not apply to long-term obesity which affects the brain, but the mere gaining of a few pounds has no known effect on IQ.

So from the numbers above the  you would have 25.52 g of brain per kg of body mass (male) vs 23.84 g (female). The female comes out as 93.4% the relative size of the male from this.

Factoring in the difference in body fat… 

 male       55.5- 8.88 (16% fat)  =  46.62  for 1361g, or 29.19g/kg

female    51.5 – 11.33 (22% fat) = 40.17 for 1228g, or  30.57g/kg

And all of a sudden the ‘large relative difference’ (16.6% here) between male and female brain size does a vanishing act. Here women actually seem to have a slightly larger relative brain size, although this may well be from the body fat percentages I used here being slightly askew. I’m not claiming that the percentage for body fat is 100% accurate, and if anyone reading this can link me to a study with the exact figures I’d be grateful, but you get the ballpark idea here.

In the Lynn study he comments how women seem to be doing better than men in spite of having a lower IQ; which suggests to me that the tests he and his colleagues were using were hinky. One of the main uses for IQ tests is to predict academic ability, and really all that Lynn’s test did was establish that his did not measure academic ability well in women or men, which pretty much proved it was slightly biased in favour of the male, and therefore not an accurate measure of intelligence. Gender biasing an IQ test is easy to do if you put in a few extra maths questions and remove a few language questions (in favour of the male). Something similar happened in the early days of IQ testing when a series of IQ tests found women to have a notably higher IQ, until they ‘balanced’ the test out.

This really goes towards ‘what are IQ tests and do they measure general intellegence’ debate. So far (poll a few psychologists) the consensus is that IQ tests are a real indicator of your general intelligence level and are a good predictor of your life outcome. If they weren’t relevant to real life/academic success, the only thing an IQ test would indicate would be how good you are at IQ tests, and your score wouldn’t be even remotely related to how smart you are (see earlier point about the tests Lynn used).

Going back to the Schoenemann paper, he makes it very clear that so far relative brain size and IQ are very strongly correlated:

“It is quite simply a myth that brain size and  IQ are empirically unrelated in modern populations.”

So far all the studies I’ve seen show a correlation between general brain size and IQ of about .4, which is statistically significant. I’m wondering if a more focused MRI/IQ brain size study vs non-fat body mass would reveal a much higher correlation for humans than this.

But essentially, functionally identical relative brain size (when fat is factored in) for male and female makes Lynn’s claims for a 5 point difference extremely hard to support, even more so when he admits that the tests used did not accurately predict academic outcome for the women who took them. In fact, he himself has commented on how racial difference in IQ are supported by the difference in relative brain mass. So how, with no quantitive difference between sexes in relative brain mass, can his claims for a lower average female IQ be correct?

It can’t.

Shame on you for bad science, Dr Lynn.

Gender quiz.

 Who is Martha and who is Arthur?

So, I finally finished this mini-experiment.

I asked two sets of questions, one to ascertain if the colouring of the face affected the way people perceived their gender,  and one to see if naming them by placing the names ‘Arthur and Martha’ above them would have any affect (the colouring based test had no names).

When asked, most people picked Left as the male and Right as the woman, by a 4:1 ratio.

But when primed by the placing of the names, they picked Left as the woman by a 2:1 ratio. There were almost as many don’t knows as picked right as female.

The reason I did this?

I’ve seen some studies that suggest women have naturally lighter skin and hair colour, but slightly darker eyes. The general response, showing a stronger response to colouring than to the not very subtle naming of the faces, suggests people do associate lighter colouring with women. There was a fairly large ‘don’t know’ response when the face were named too, showing some confusion in my guinea pigs.

This observation that women have lighter skin might explain the preference for lighter skinned women in a lot of cultures. It’s not something that’s been introduced by conquering Europeans, it’s just a natural preference for a sexually dimorphic trait.

IQ and gender.

Hmm I’ve been reading up on IQ and gender on Wikipedia. Pretty funny really. If you know a lot about IQ tests and life outcome, it’s laughably simple to see how some of the mooks of this world will read SAT scores and university IQ tests as proof that women have a lower IQ, as the male average on all of these is a bit higher than female. The reason for the higher score is…drumr0ll… That most of the really low IQ people won’t be taking those test, and they are mostly men!It’s very simple, the lower the IQ, the more likely you are to drop out of school and not get around to sitting an exam that you know you’ll fail abysmally. If your IQ is less than 70 there is a 55% chance that you’ll drop out of high school, at 80 to 70 it’s 35%, at IQs above 120 it’s statistically insignificant (just me then). And the majority people with learning difficulties are male (about twice the number of women). It skews the numbers to make male IQ look higher. If you want to test a population to find the mean, you have to test ALL OF THEM. Anyone who understands statistics even a bit can see the bias.