[tomp]

It could be the butterfly effect. It starts with children. Better at math -> more interested in math -> better at math -> ... by they time young adults start seeking jobs, a small initial bias has turned into a large gap.

Personally, I don't think that's the whole story (according to my observations, it's also upbringing, girls socialize more than boys, who are more likely to keep to themselves, and differences in topics of interest, which I have no idea what they come from).

[rayiner]

First, it's probably not the case that interest and practice can have any impact on your underlying mathematical aptitude. Second, regardless, the 2:1 ratio for perfect scores on the SAT Math manifests at 16-18, so when people are already mostly developed.

[selmnoo]

> First, it's probably not the case that interest and practice can have any impact on your underlying mathematical aptitude.

What?! How did you arrive to that conclusion? This is a really bad myth. Math abilities for the most part have very little to do with genes, and almost all to do with hard word, motivation, and practice. See http://www.theatlantic.com/education/archive/2013/10/the-myt... for more.

[tomp]

No ,I was saying that they have an impact on the actual mathematical aptitude, and the (again actual) ratios of men/women in STEM industries.