It is an possible theory, but in my view I tend to agree to an quite old theory that I saw referenced in an other Swedish study conducted by the department for

It is an possible theory, but in my view I tend to agree to an quite old theory that I saw referenced in an other Swedish study conducted by the department for higher education. They asked why so few men end up in the teacher profession, and the primary theory presented was based on a (if I remember right) 1950 study. A person on average tend to be more comfortable and confident in decisions if its mimics behavior of people that look/are similar to themselves. The teaching profession in Sweden actually start out with almost 50/50 during the first year at university. What the Swedish study found was that as students had setbacks, the students that decided to quit was mostly men. By graduation time a large portion of the male students has quit. Same thing occurs when people start seeking jobs, after the first year, second year, third year and fourth year. At the end the wast majority of men are gone from the profession, most citing "culture mismatch" and not feeling that they belong. I find this theory much more plausible explanation than hormones, and it also explains why removing gender roles might increase gender segregation. If a person is uncertain where they fit, making a similar decision to other people in the same situation will on average feel safer. It also explain why mentor programs are one of the most effective tactics to combat gender segregation.