Teaching Python to kids

> If you want more extreme example - look at the number of females in armies. Not long ago it was vanishingly small, yet now it's growing.
It's funny that you mention armies, because that's one of the few remaining places where there is _actual_ discrimination. I was drafted and have lost nine months of my life to that.
Anyway, it's kinda obvious why men would enroll in the army and women wouldn't: men are less agreeable than women. This is why the prison population is overwhelmingly male. But then, it might also be because patriarchy :-)

> And it needs to be explained, it's meaningless in isolation without controls for other factors (like wealth).
You can discredit _any_ social science result by saying “but it doesn't correct for $FOO”, that's a trivial consequence of the fact that there's an infinite number of variables that the author of a study hasn't corrected for. My answer to that is that if you claim that correcting for $FOO actually makes a difference, you'd better provide data to support that, because otherwise the right thing to do is to ignore that objection.