>
Do you think that women are intellectually flawed? [...] If no, what is your best guess why they are underrepresented?My best guess is that relevant male math-type skills have a higher variance than female math-type skills. That means that even if the average ability is about the same, there are more men than women at the very top of the skill distribution and ALSO at the very bottom of the skill distribution. If you look for math morons, you'll find more men; if you look for math geniuses, you'll find more men. If you look for people with roughly average math ability, you'll find more women. If the variance is even a LITTLE larger, it can explain an arbitrarily large imbalance at the high end, and we have both (some) statistical evidence that it IS larger and plausible biological reasons to expect that it MIGHT be larger.
Women have two X chromosomes; men only have one. All the important stuff that makes us human is on the X chromosome. So for the sake of intuition, let us imagine that there were only THREE available genes for math ability, all equally prevalent in the population, and any X chromosome you inherit includes one (and only one!) of these three "math ability" genes. The genes are:
(1) "moron" (recessive)
(2) "genius" (recessive)
(3) "average" (dominant)
In that world, men would have a 1 in 3 chance of being a genius and women would have a 1 in 9 chance of being a genius.
Similarly, men would have a 1 in 3 chance of being a moron and women would have a 1 in 9 chance of being a moron.
In that world, having two X chromosomes makes women "more average" than men, which means that if some career requires the "math genius" gene, that career will be 75% male.
Does that all make sense?
Okay, now let's suppose that "genius" gene is a bit more rare. Let's say that only 10% of X chromosomes have the "potential math genius" gene. In that case, men would outnumber women 10-to-1 instead of 3-to-1 in professions requiring it. The rarer that gene is, so long as that gene is either recessive or to some degree averages out with its counterpart on the other X chromosome, it will be hugely rarer that women fully express the properties that gene codes for.
If this analysis is valid, it suggests that women are less likely to be at the very TOP of math-related fields (or will have to work a lot harder to get there) for essentially the same reason women are far less likely than men are to be colorblind.
(relevant study: http://www.ncbi.nlm.nih.gov/pubmed/7604277 )