This is a gross generalisation, but: men are interested in things, and women are interested in people.
It may be a generalization, but it's supported by over 60 years of research.
While undoubtedly a statistical generalization and not necessarily appliccable on an individual basis, there exist several studies which support this conclusion.
Yes there are, going back to late 50s,, it's known as the "people vs. things gender difference." Now the studies are zeroing in sspecifically in STEM, and the latest research and studies are looking at specific areas within STEM (e.g., engineering, and medical services). The following article is an example of this:
The degree of women's underrepresentation varies by STEM fields. Women are now overrepresented in social sciences, yet only constitute a fraction of the engi...
www.frontiersin.org
From the Conclusion Section of the paper:
"We found drastically different levels of gender differences in basic interests within STEM fields. L
arge to very large gender differences in interests favoring men were observed in engineering-related fields (d = 0.83 for Engineering—professional level,
d = 0.89 for
Engineering Technicians, and
d = 1.21 for
Mechanics and Electronics). Small to moderate gender differences in interests favoring men were observed for mathematical careers (
d = 0.38 for
Mathematics, and
d = 0.23 for
Applied Mathematics). Gender differences in interests vary largely in the sciences, ranging from moderate, favoring men, in
Physical Sciences (
d = 0.56), to non-significant (
d = 0.19 for
Biological Science, d = 0.14 for
Science Technicians, and
d = −0.04 for
Medical Science), and to small to moderate, favoring women (
d = −0.33 for
Social Sciences, and
d = −0.40 for
Medical Services). These findings provide refined information about men and women's interests in sub-disciplines of STEM."
"The current study found the percentages of women within most STEM fields to mirror the gender differences in basic interests in those fields, lending support to the preference-based explanation for gender disparities in STEM careers."
What was important about this paper is that they point out the caveat that you can't simply stop at the people vs. things analysis. They found that women are still underrepresented in engineering, and point to factors other than people vs. things (emphasis added):
"For example, in mathematics and sciences, the actual gender composition is closely aligned with gender differences in interests; however,
there are discrepancies between the projected percentages of women based on interests and the actual gender composition in the engineering-related fields and
Medical Services."
"These results
indicate the existence of other factors that escalated the gender disparities in these STEM careers. A few potential factors suggested by the literature include preference for work-life balance (e.g.,
Ferriman et al., 2009), gender stereotyping and gender role schema in individuals' career decision-making (e.g.,
Konrad et al., 2000), and implicit bias in employers' selection process (e.g.,
Moss-Racusin et al., 2012). It is beyond the scope of this article to provide a detailed review of these alternative factors contributing to the gender disparities in the STEM fields (for a comprehensive review, see
Ceci et al., 2014)."
So people vs. things does actually explain a lot of the difference is career choices, but not all of it.