As a mathematician, I find the conventional notion of a gender wage gap dishonest. The conventional wage gap is calculated by simply comparing the average wage of females with the average wage of males, without any regards for differences in the sources of income. Females and males strongly tend towards different employment areas (strongly influenced by the different academic interests mentioned above), which tend to align with lower paying jobs. When you compare males and females of the same job, then gender-based wage gaps tend to disappear.
Unless you look at the
actual data which shows there is still a wage gap present within fields.
It's true that a substantial chunk of the total wage gap is due to higher paying fields being predominantly male, but I don't think you can dismiss this as inherent preferences, unavoidable and acceptable.
A lot of that clustering is due to the social attitudes towards and within the jobs. Again, I posted that example of a Parcelforce job and an Amazon one. Very similar work, but one is traditionally male and regarded as physical so had no women compared to the Amazon pickers. Which, given the more established one is in a heavily unionised field and involved far more safeguards and progression into higher paying jobs, puts women at a disadvantage compared to men.
Then there is the whole reason I get so frustrated with these discussions of average strength, etc. It's because these things are completely irrelevant to any individual recruit (who should be weight against the requirements of the job) but
colour the attitudes of those who would recruit them. The attitude that women are, on average, weaker than men becomes an attitude that women are incapable or completely unsuited to a particular job or field. Which is no more true or false for any individual woman than it is for any individual man. But you can be sure that such attitudes have impacts on who enters those fields.
And this goes double for questions of academic ability or similar.