Because people are quick to generalize from statistical averages. And because they're more likely to do that when it comes to women and minorities. Consider th

Because people are quick to generalize from statistical averages. And because they're more likely to do that when it comes to women and minorities. Consider the following: Color blindness is generally the result of an X-linked recessive disorder, meaning that it is much, much more likely to occur in men. Yet nobody thinks that one should generalize this to mean that all men (or even a majority of men) have a problem with that or that men as a rule need to be treated with suspicion for all kinds of jobs where that may be an issue. Now consider mathematical aptitude, which very likely may not be linked to biological sex at all, and if it is, the difference is extremely small (one way or the other). But every single study that shows that men may (may!) have a very small higher aptitude at math quickly gets generalized into meaning that "girls are bad at math" [1], resulting in women being treated as less qualified for STEM jobs/education, etc. When you've seen this pattern play out too often, you start getting sick of pop science journalism playing up one single study as a definitive result. It's already annoying in general because this is not how science works, but when it's then further used to reinforce cultural stereotypes, it's very frustrating. Don't get me wrong: There are differences that are linked to biological sex other than reproductive equipment (X-linked disorders are a simple example). But there's a really huge amount of shoddy science and/or uncorroborated studies that pop science journalists publish selectively; they are not the same as a well-corroborated scientific theory such as evolution. The Larry Summers kerfuffle is a case in point. The problem with Summers' talk was not that it's impossible that men are more likely to be at the extremes of IQ (both good and bad) than women. The problem was that Summers was treating an uncorroborated hypothesis (1) as established science and (2) as an excuse to have fewer women in leadership positions. To be clear, the hypothesis – and it's an old one [2] –, is not per se unreasonable. It relies on two assumptions: (1) That a large part of IQ-related genes are on the X chromosome and (2) that the lyonization process averages out their effect in women. Which would mean that men are both more likely to have extremely high and extremely low IQ. Lyonization, or X-inactivation, is what happens if you have a cell with two X chromosomes. Only one of them can be active, so during embryonic development in mammals, one of them for each cell is randomly and independently deactivated (except marsupials, where it's always the paternal X chromosome). This is, for example, why even women who are carriers for hemophilia (an X-linked disorder) are unlikely to suffer from it: while some of their cells have the defect, enough of their cells still produce the necessary clotting factors (though, because the process is random, that may not always be the case). If the same held for IQ, it would be possible for a similar averaging process to take place there, too. One problem with that hypothesis is that it has not really been conclusively shown so far (it is really, really hard to get rid of the cultural factors to determine very small changes in standard deviation). Another problem is that this is a very simplistic assumption of how IQ is developed. For example, if the model were that simple, then men with Klinefelter's syndrome (who have two X and one Y chromosome in each cell, and whose cells also undergo an X-inactivation process) should experience a similar averaging for their IQ: Instead, Klinefelter's syndrome is held to be responsible for a drop in average IQ of nearly one standard deviation (compared to both men and women). So, it can't be that simple. In fact, while the brain structure for men and women IS different (in ways that we don't fully understand yet), these differences do not seem to relate to the presence or absence of a lyonization process. So, while the hypothesis is not per se impossible, the president of a research university should really know better than treating it as established science and essentially saying, "it's just normal that we have more male professors". [1] Obligatory xkcd: https://xkcd.com/385/ https://xkcd.com/385/ [2] R. Lehrke, "A theory of X-linkage of major intellectual traits."