With math-related fields it IMO boils down to this: From what I have seen in academia during the last few years (yes, anecdotal, sample size of about two dozen

With math-related fields it IMO boils down to this: From what I have seen in academia during the last few years (yes, anecdotal, sample size of about two dozen female PhD students), I'd say that women are sorted out at almost all stages of the engineering/sciences career path. By the time you are a PhD student or Postdoc, the weeding out has produced a situation in which the women that remain are on average better than the male PhD students, because only the most talented/determined stuck it out. Then most of them leave academia because they might at some point in the future want children, which is a 90% chance career-ender.