It’s been a busy summer! I am finally starting to think about catching up on reading.
The big news for women in mathematical sciences is that women now earn 30% up PhDs in the mathematical sciences, up from 5% in the 1950s. (I actually tweeted this one, but it deserves more than 140 characters). This is great news, of course. This tidbit made a splash after a paper was published in PNAS on 6/1/09 (by Janet S. Hyde and Janet E. Mertz). What is also interesting is that women received 11-15% of PhDs in the mathematical sciences between the 1890s and the 1940s! The authors note that men do seem to have more variability in their mathematics ability, resulting in more men who score at the top and the bottom than women, and that this transcends nationality (incidentally, this is what Larry Summers got in trouble for saying). However, Hyde and Mertz conclude that the evidence shows that this is likely the result of nurture not nature, and hence, it is changeable.
I was curious about the data from the US. NSF has a wealth of data on women in engineering and the sciences. Overall, women are growing at a faster rate than men in science and engineering fields. The figure below (from NSF), illustrates this nicely.
There is bad news. The data shows that the proportion of BS degrees in engineering awarded to women has gone down since 2000 (from 20.9% in 2000 to 19.5% in 2004). It looks more optimistic at the MS and PhD levels.
Figure C-1. Bachelor's degrees awarded in S&E and non-S&E fields, by sex: 1966–2006
Thanks to Larry from IEOR Tools for urging me to write about this!
Related posts:
Punk Rock Operations Research 
July 24th, 2009 at 8:24 am
Its my pleasure Laura. Your post is interesting and surprising. Great references and analysis.
July 24th, 2009 at 10:58 am
I grabbed the spreadsheet from the NSF site. It spanned 1997 through 2006 and skipped 1999 (so apparently the Y2K bug was more real than we thought, but happened earlier than expected). If I’m reading it correctly, the peak of 20.9% was in ’02 and the 19.5% was in ’06. More importantly, they’re only 1.6 standard deviations apart (measuring s.d. from just those nine years and ignoring any possible autocorrelation). Considering we’re cherry-picking the max (though not the mean), I don’t know that 1.6 std. devs. is anything to get excited about.
I’m actually a bit more concerned about the S&E men’s figure hitting a plateau — not because I’m worried women will outnumber men, but because we need the number of science and engineering majors to grow, and it may not be growing fast enough on the women’s side to fill the billets.
September 3rd, 2009 at 10:29 am
Laura —
Love the site. Have a question. Does the shift from 5% to 30% mean more women or fewer men or both?
— John Hall