Turns out, there's no statistically significant difference between male and female grades. http://minimaxir.com/img/online-class-charts/gender-grade.pn...
Basically your post confirms the theory of the person above. That grades aren't measuring mastery of knowledge but the skill to obey arbitrary rules.
> 113,571 female students (26.7%).
> The fact that it’s not even close to 50:50 is surprising.
It didn't surprise me, I went there & searched for it. How can it be surprising (you see these ppl all the time on campus!). And it shows that there are ~3 times more men at Harvard than woman.
So thank YOU so much for nearing to a problem: Girls do better early on (cf. orig. article) & men later in live (cf. your analysis)
Why did you screw that? Political correctness? You are dishonest (you'd have my downvote).
An important aspect: We need to make sure to not loose males [which would do fine later] during early years.