The authors of a recent study published in the journal Science argued that "women are underrepresented in fields whose practitioners believe that raw, innate talent is the main requirement for success, because women are stereotyped as not possessing such talent" (Leslie et al., "Expectations of brilliance underlie distributions across academic disciplines," Science 347 (2015): 262-265).
This study has been widely reported on without much criticism, but Scott Alexander appears to have undermined the central claim of the study by showing that the proportion of women in STEM fields tracks very closely their relative performance on the GRE tests used to determine admission to STEM programs. In other words, according to Alexander, there is not evidence of discrimination against women trying to enter STEM fields in the first place.
Because it is outside of my area of expertise, I am unable to determine if Alexander's numbers or statistical analysis are accurate, but I think this is worth sharing anyway so that other people can examine his arguments.
Here is Alexander's figure showing the relation between average GRE quantitative score and percent women in a discipline; he says that the correlation between these two factors is r = -0.82 (p = 0.0003).