July 25, 2008

Data Don't Support Perceived Gender Gap in Math, Researchers Say

This story was updated on August 8, 2008.
The view that girls are worse than boys at mathematics is unfounded, a team of researchers at the University of Wisconsin at Madison and the University of California at Berkeley reported today in the journal Science. Their conclusion challenges the frequently cited argument that the poorer female performance in math explains the shortage of women in physics, engineering, and related careers.

In their study, the researchers examined extensive performance assessments of more than seven million students carried out by 10 states, as required by the No Child Left Behind law. The researchers found that, in standardized tests, the differences between the average math scores of boys and girls in grades 2 to 11 zero.

It is also commonly said that boys are overrepresented at both ends of math performance; they are more frequently the best and the worst achievers than girls. The researchers found there’s more variability in the boys’ math scores; but this variance was not large.

When the research team studied if there were gender discrepancies at the highest levels of mathematical performance, they got different results depending on if the kids were white or Asian American. In Minnesota, 11-grader white boys hit the 99th percentile more than twice more than white girls in the same grade. But for Asian American students, the pattern got reversed.

“It might be the different cultures’ emphasis on Math,” says Janet S. Hyde, leader of the study. Her group has two ongoing studies analyzing data on the math performance of American children and youth in other nations.

Finally, the researchers looked at SAT results. Men usually perform better than women on the test, but the researchers attributed that result to poorly done statistics. The sample group of SAT test-takers is not random: Only students who are applying to college take the SAT, and the two groups are different in size — more girls than boys take the test — so they can’t be compared fairly. —Maria José Viñas

Comments

1. This is something that anyone who ever worked with NCES transcript-based longitudinal studies knows. From high school through college, women get higher grades than men in math, and at every level through post-calculus. The problem lies in the background propaganda that women don’t do well in math and don’t belong there. So give this study more publicity and light, and we’ll see higher rates of participation in math and quantitative subjects among women.

   — Cliff Adelman    Jul 25, 03:35 PM    #

2. Also, the problem is that many institutions have math and science programs dominated by males and do not make participation of women in their programs either a goal or an amenable experience. This practice can bias participation which can bias the score studies unless they are normed; which hide male – female differences proportionately.

   — lef    Jul 25, 04:07 PM    #

3. Before we go jumping off the deep end in insisting that this data shows that it can only be bias that prevents more women from obtaining math, engineering, and science PhDs, I should point out that a number of people who have looked at this study have pointed out that boys are generally overrepresented at the highest end of the scores as well as the lowest end of the scores. In other words, boys paradoxically make the best and the worst students of math. When they are averaged out, the scores are very close to the girls’ average.

   However, keep in mind that the ranks of PhDs are generally taken from the top 5% of students of math. If boys are overrepresented in this group, is it any wonder that they make up a larger percentage of PhD recipients?

   — J. Ward    Jul 25, 04:51 PM    #

4. But before we go jumping off the deep end by assuming that the larger proportion of men receiving Ph.D.s in mathematics is due to the effect mentioned above, we should also familiarize ourselves with Abbe Herzig’s research on what happens to women in mathematics graduate school. A little Googling is sufficient to get the gist of it, but reading the full articles can be quite illuminating.

   With that sort of experience for women to anticipate, is it any wonder that men make up a larger percentage of Ph.D. recipients?

   — Bob M.    Jul 25, 07:58 PM    #

5. Of course, a lot of people contend that standardised tests measure nothing but the ability to take standardised tests. If they are correct, studies such as this can’t tell us anything about how much mathematics boys and girls know.

   — Gustave    Jul 26, 07:13 PM    #

Previous: Gunman Wounds 3 in Shooting at Community College in Phoenix
Next: U.S. Commerce Department Seeks Comments on 'Deemed Exports'