LARRY SUMMERS REVISITED
Another really interesting article on women in science (and still more in the queue!) - this one is more upbeat.
A piece in this week's Economist subtitled, "Girls are becoming as good as boys at mathematics, and are still better at reading" describes research that shows that the gap between boys' and girls' math scores on standardized tests goes away in countries in which there is greater equality between the sexes. This suggests to me that things like stereotype threat play a big role in some of the observed test score differences. The good thing about stereotype threat is that it's relatively easy to overcome - basically you just have to create awareness of it as an issue.
Boys do continue do better than girls in geometry, but that's the only area in which I've seen any compelling looking evidence for sex-linked differences in ability. There's been some interesting research linking fluctuations in spatial reasoning to levels of estrogen and testosterone, but I have no idea how solid it is. In any case, I would imagine that the differences involved are relatively small and swamped by individual differences, since otherwise this kind of thing would not really be contentious.
Girls do consistently better than boys on tests of reading, and the study also found that the reading gap is even greater in countries with greater sexual equality. The implications are interesting:
I, for one, welcome our new female overlords.