[Cushman]

I don't think you see my point here. You can't use statistics to judge individuals.

If that study were a hiring situation and you figured that, since "women are more empathic than men", you should only consider the female applicants, you would be tossing out nearly as many qualified applicants as you'd be keeping. It's just not a useful thing to say based on that data.

When you add in the fact that saying "women are more empathic than men" contributes to the broader social condition that empathy is encouraged in women and discouraged in men, and that mean women get a free pass because they're women and caring men can't get a job teaching children because they're men, it becomes a positively damaging thing to say.

And what do you get out of it?

[sarkozy]

I agree on the importance of caution when applying statistical conclusions in individual cases - especially hiring decisions. I don't deny that there are, as you said, "mean women and caring men". But I don't believe in denying the existence of gender differences simply to avoid supporting the status quo. It can be useful to be aware of these patterns because, with a broad brush, we can cautiously identify where people's skills are most likely to reside.

[Cushman]

Perhaps it's more an issue of naming.

If you have a man and a woman in a room together, it's going to be more or less a toss-up which of them happens to be more suited for a particular task, since what the body of research mostly shows is that, in general, men and women are pretty much the same.

Considering that, it strikes me as disingenuous to say something like "men are better at [x] than women" — which is vague by the most generous reading, outright false by the most literal — when what you really mean is "the mean for [x] among all men is 5% higher than the mean among all women."

I think the reason people don't say that instead is because it's obviously a very weak statement.