I’ve occasionally touched upon the gender gap in mathematics, mostly in response to some recent study that has attempted to explain why mathematics (and the sciences in general) are so predominately male. An article that appeared in Slate last week makes me think it is time, once again, to discuss this topic.

After giving a brief overview of the observed gender gap in science and math careers, writer Shankar Vedantam then discusses the results of some recent experiments out of the University of Massachusetts at Amherst which revealed new features of this gender gap.

In both experiments, researchers (roughly speaking) found correlations between the unconscious attitudes that females in a variety of scientific majors had towards mathematics and the gender of proctors and professors in mathematics. Among the findings (more details can be had by viewing the article):

- Given a question posed to the classroom by the professor, the percentage of female respondents decreased from 11% at the beginning of the semester to 7% at the end when the professor was male, but jumped from 7% to 46% when the professor was female.
- The percentage of female students who asked for help from the professor went from 12% to 14% when the professor was female, but dropped from 12% to 0% when the professor was male.
- Female students wound up with less mathematical confidence when their professors were male, even if they performed better than their male peers when tested on their math performance.

These statistics are quite interesting, although it would be helpful if they were a bit more contextualized. For example, how did the percentage of male students who asked for help from the professor vary with the professor’s gender? Certainly a drop from 12% to 0% is telling, but if, for instance, a professor can’t retain even one female in his office hours by the end of the semester, that may say more about his teaching abilities in general than it does about any unconscious bias at work.

Even so, this correlation between female performance/self-identification and the presence of a female mentor is an intriguing one. While I don’t necessarily think that one can draw a causal inference from the data, it certainly would be nice if female students who are interested in mathematics had a larger pool of female role models from which to draw. Restricted to the realm of popular culture, the number of mathematically-inclined female role models is particularly slim. The only one I can think of offhand is Danica McKellar, who, though well-intentioned, still conforms to gender stereotypes that many parents may find problematic.

For my own part, I’m proud to say that I’m fairly good at retaining students in office hours from week to week. It may not be much, but maybe it’s a start.

Other things this study could show: (1) male teachers teach better, so students need less help, and (2) female teachers are better at eliciting in class responses. Without the data for male students, we can’t come to any real conclusions.

That this data was surely kept, but not proffered, is circumstantial evidence that it would buck the trend they are trying to show.

Also, while you get points for attempting to have a math problem instead of a captcha, mine (Sum of 5+7?) just upsets me.