Further: almost nobody practicing computer science in the industry has a PhD. There are, believe it or not, places where having a doctorate is frowned on. For the degree that actually matters --- the bachelors --- math has as you say something approaching parity.
Tell me again about the innate limitations women have performing in STEM.
It drops from ~50% to just under 30%, where CS degrees hover between 18-22%. Frankly, that's not "substantially higher than CS".
> Tell me again about the innate limitations women have performing in STEM.
No one is making that claim except people like you who love tearing down absurd straw man arguments. Please read more carefully and stop making assumptions.
Finally, it's not a double-digit improvement, but of course we can quibble endlessly over how meaningful that exact number is, but it's irrelevant. The number is still piss poor compared to undergrad enrollment, indicating that women don't pursue higher math degrees for various reasons.
The link I initially provided cited evidence that they're interested in undergrad maths because of opportunities it provides outside of pure maths. It's not apparent that similar opportunities exist for CS undergrad degrees, which would be one explanation for the disparity in undergrad math and CS. So I'm still not seeing how anything you've said bears on this at all.
Certainly graduate maths have more women than does CS. So what? What exactly do you think this proves in light of the above?
Using CS as a tool to solve other problems is a different proposition than the study of CS itself. I don't see why they'd have to be connected in people's minds in the way you're suggesting.
For instance, undergrad math enrollment is almost at gender parity, but postgraduate math study has similar gender ratios to CS. This is because you need an undergraduate degree for teaching and other careers, and so the interest in pursuing math for maths sake appears to be much lower among women. This could be for a number of reasons, some of which might be fixable to improve ratios, but some may be innate.
http://slatestarcodex.com/2017/08/07/contra-grant-on-exagger...