When Lawrence Summers suggested that the reason there aren’t more women in the top academic positions in math and science is that they don’t have the aptitude for it, a firestorm was created that may have cost him his job as president of Harvard University. Sometimes lost in the hullabaloo surrounding the incident is the science surrounding that bit of speculation.
The entire August 2007 issue of Psychological Science in the Public Interest was devoted to the “science of sex differences in science and mathematics,” and Cognitive Daily will spend the remainder of this week discussing that article. The article was written by a formidable team of psychologists, led by former APA president Diane F. Halpern.
The authors begin by defining terms, demonstrating that even the terminology underlying this complex issue can be controversial and nuanced. Consider the terms “sex” and “gender.” Traditionally “sex” has referred to the biological differences between people, while “gender” covered social and environmental differences. But recent research has demonstrated that it is exceedingly difficult to separate environmental from biological differences. Brain development, for example, is influenced by hormones, which are in turn influenced by both social and biological factors. So is the structure of the brain a biological or environmental factor? Arguably, it’s both. The authors settle on the term “sex” for their article, but point out that both terms are favored by different groups.
Similarly, “abilities” and “achievement” are difficult to separate. We can’t rightly say that different people don’t have differing abilities, for two people could achieve the same (low) result on a math test — one after years of study and instruction, and one with no instruction at all. Surely the second person has more ability in this subject than the first. But ability can only get you so far. Every normal human is born with the ability to learn language. But if a child is raised without exposure to language, she won’t learn it.
Halpern et al. make a fascinating observation about IQ and intelligence: Modern IQ tests throw out questions that favor one sex over the other. Therefore the most commonly used IQ tests show no difference in results between the sexes. IQ tests, therefore, can’t tell us anything about sex differences.
Sex differences in abilities and achievement
So what can we know about sex differences? Take a look at this graph showing scoring differences on a literacy test given to fourth-graders in 33 different countries:
Girls scored significantly higher in every country in the study. A subsequent study of 15-year-olds in 25 countries showed a similar pattern: Reading literacy was significantly higher for girls in every country studied.
In the U.S., girls get higher grades than boys in every subject (including math and science), and females have a significantly higher college graduation rate than males.
So if boys are supposedly better, where are differences coming from? Well, boys do score better on the SAT test — both verbal and mathematical. In the verbal portion of the test, the male advantage is eliminated if the analogy portion of the test is eliminated; arguably this is more a test of mapping relationships than literacy. As for the math portion, one study suggests that if questions requiring mental rotation are removed, again the male advantage is removed.
Mental rotation tasks (see this CogDaily post), which require working with a three-dimensional representation of an object, have been found to have very large sex differences favoring males. The authors argue that the male math advantage in a number of different studies appears to be directly related to visuospatial skills, the most important being mental rotation. In tests on calculation or other mathematical problems that don’t require visuospatial skills, females perform just as well as — or better than — males.
What’s more, at least one study has found that it’s possible to teach these visuospatial skills. Such a course has been offered at Michigan Tech for many years, and students taking the course have not only shown measurable improvement on visuospatial tests, they have gotten better grades in subsequent engineering and graphics courses.
So while there is some reason to believe that females might not be as good at math and science as males on average, there’s also evidence that education can reduce or remove this gap. Further, the authors point out that success in math and science doesn’t depend solely on math skills. Strong reading and writing skills are essential in these fields as well, and here females have a clear advantage.
But there are other possible reasons that women are underrepresented in math and science. We’ll explore some of them tomorrow.
Halpern, D.F., Benbow, C.P., Geary, D.C., Gur, R.C., Hyde, J.S., & Gernsbacher, M.A. (2007). The science of sex differences in science and mathematics. Psychological Science in the Public Interest, 8(1), 1-51.