Get your copy of our weekly print products at any of these convenient locations.

UW study shows virtually no gender gaps in math scores

Todd Finkelmeyer  —  7/25/2008 12:42 pm

Crunch the numbers from a recent study, and the results might surprise you: Girls are just as good at math as the boys.

UW-Madison psychology professor Janet Hyde led a study that looked at SAT results and math scores from 7 million students who were tested in accordance with the No Child Left Behind act. And the numbers showed the average scores of boys and girls were virtually the same.

"Our country has a lot of stereotypes that boys are better than girls at math, and we have current evidence that both teachers and parents think that that's true," said Hyde. "But the data don't show that at all -- at least with these very current samples."

Hyde's research team, which included scientists from UW-Madison and the University of California, Berkeley, is to report its findings in Friday's issue of Science.

The team of researchers acquired math scores -- along with detailed statistics on the test takers, such as gender, grade level and ethnicity -- from state exams now mandated under No Child Left Behind. They then calculated the "effect size," which showed the degree of difference between girls' and boys' average math scores in standardized units. Hyde said an effect size of 0.20 or less is considered small, and the effect sizes they found ranged from 0.01 to 0.06 -- which is "basically zero, or no gender difference."

The research team also looked at gender differences at the highest levels of mathematical ability, and also tried to compare how well both girls and boys fared on questions requiring complex problem solving skills. In both instances, the study found little difference between the sexes.

The team, however, was surprised at the complete lack of complex problem solving questions on state assessments for No Child Left Behind. That forced the team to seek other sources of data for this part of the study.

Hyde said she is afraid that if teachers are gearing their lessons toward the No Child Left Behind exams, too many children aren't learning complex problem solving skills. Each state puts together its own set of No Child Left Behind exams.

"When you look at the items on these tests, which we did very extensively, they are low-level questions," said Hyde. "And so teachers are going to be teaching to that level, and not teaching more complex problem solving -- which is just what you need if you're going to succeed in science or engineering. So we really need to look into these state tests that are being used for No Child Left Behind."

Hyde's team also looked at SAT math scores. Although it has been widely publicized that boys do better than girls on the math portion of this test, Hyde and her co-authors downplayed those results.

"Yes, the bottom line is the boys do a little bit better on the math portion of the SAT," said Hyde. "But you have to remember that's voluntary test-takers -- it's people who select to take the SAT because they plan on going to college. So it's not a random sample of the population, and you really can't learn anything from those numbers.

"And the catch is that more girls than boys take the SAT -- by a pretty big gap. So if girls score lower, what does it tell you? It probably tells you you're dipping down farther into the talent pool -- it's not a fair comparison."

In January 2005, the issue of women in math and sciences became a hot-button topic following controversial remarks by then-Harvard President Lawrence Summers. At an academic conference, Summers suggested, among other possibilities, that differences in men's and women's "intrinsic aptitude" for math and science might explain the dearth of female mathematicians, engineers and physicists at the highest levels across the nation.

Although Summers later apologized and appointed two Harvard task forces to recommend ways to attract and retain women faculty, some wondered if, perhaps, girls' brains really aren't wired to be as good at math as a boys' brain.

Women in science and engineering often blame their low numbers on societal factors. Some argue girls and young women traditionally receive little encouragement to enter science and engineering fields, while others note that popular culture gives females few role models.

"I'm hoping that parents and teachers will see this (study) and say, 'Gosh, what I've been thinking about boys being better at math, maybe it's not true,'" said Hyde. "I mean the data just don't show that it's true -- so maybe people will revise their ideas."

In addition to Hyde, the study's other authors included UW-Madison's Sara Lindberg, Amy Ellis and Carolyn Williams and UC-Berkeley's Marcia Linn. The work was funded by the National Science Foundation.

Todd Finkelmeyer  —  7/25/2008 12:42 pm