More On Gender Imbalances In STEM: The Hidden (And Real) Good News

I made the mistake of looking at (and actually commenting on) Megan Geuss' silly Ars Technica opinion piece of the Ellen Pao verdict; there's nothing you haven't already seen in TechCrunch or Verge, i.e. It's Important We Have A Conversation About Sexual Bias In Tech (Now Stop Looking At What An Obvious Grifter Our Poster Girl Was). But that got me thinking about a post I happened upon a few days ago at Randal S. Olson's data visualization blog, showing the actual percentage of women graduating with bachelor degrees in STEM fields over time:

With the exception of engineering and computer science, women increasingly near parity with men in mathematics and the hard science fields, and have surpassed them in biology and the social sciences. Further, the so-called "pipeline problem" appears to be mostly a myth, according to a large-scale research study from Cornell, U. Texas, and Northwestern. Women are no more likely to "leak" (i.e. exit) from the "PhD pipeline" than men, something that will no doubt come as a disappointment to TechCrunch and anyone else invested in the idea that brogrammers and other modern bridge trolls are somehow chasing women out of these fields.

"There’s been a lot of focus on this idea of women in particular leaving academic science at far higher rates than men," says Miller, an advanced doctoral student in psychology at Northwestern and lead author of the study. "But in some cases … there’s been scant evidence of some of those gender gaps in persistence, and evidence that those gaps in persistence don’t exist at other time points."

That's not to say, however, that women and men are equally represented in pSTEM academia. Men still outnumber women about 3 to 1, Miller says. But the differences are not explained by gender bias in the pipeline – the percentage of women earning pSTEM degrees is now higher at the doctoral level than at the bachelor's degree level, the researchers found.

"We need to start reframing the conversation from instead of just trying to plug leaks, we need to get more students interested in the first place," Miller says.

Which, basically, is what I've been saying all along. Math instruction, particularly, is almost uniformly horrible at the K-12 level, even in fairly well-off districts. This is personal experience talking; if not for one particularly good instructor in college, I would not have ended up taking the major I did. But that is not the kind of thing feminist advocates wish to emphasize, because it means women might have some responsibility for learning math and how to write solid code (e.g.) instead of mewling pitiously that they are owed a high-status job despite inexperience.