He makes references to real science, but gets nowhere close to demonstrating that the effects he references should manifest themselves as an 80/20 gender ratio

He makes references to real science, but gets nowhere close to demonstrating that the effects he references should manifest themselves as an 80/20 gender ratio -- presumably worse, if his policy recommendations were adopted. Hell, I'm not even sure if the gender differences would make women worse at engineering, on average, anyway.