From the The News & Observer / NewsObserver.com, , Friday, January 11, 2002. See
https://www.newsobserver.com/friday/front/News/Story/903028p-902294c.html . See, also,
https://www.eurekalert.org/pub_releases/2002-01/uonc-nsg011002.php for another article on this research. The complete citation for the journal article is: Leahey, Erin, and Guang Guo (2001). "Gender Differences in Mathematical Trajectories," Social Forces, Vol. 80, No. 2, December, pp. 713-732.

***********************************

Math gender gap doesn't add up

By Trish Wilson

A pair of UNC researchers say they have found the answer to the long standing question of why boys do better than girls in math: They don't.

After decades of research showing that boys consistently test higher than girls in math, investigators at the University of North Carolina at Chapel Hill are publishing a report today staking their position in what has become a testy debate. The problem, they say, is that previous studies have focused on narrow groups, such as super-smart seventh-graders and college-bound SAT takers.

"These studies were making broad generalizations, 'Here's the difference between boys and girls,' and we didn't think those were warranted," said Erin Leahey, primary author on the report she co-wrote with Guang Guo, an associate professor of sociology. "We felt uncomfortable with that. 'Hey, maybe if you used a really broad sample
you wouldn't find it.' And that's what we found."

Leahey and Guo examined for the first time the test results of 14,000 students in elementary through high school. They found that girls had higher average math scores than boys until about age 11 and higher reasoning scores at ages 11 to 13.

Boys did, however, tend to progress faster, and by the end of high school that learning edge resulted in a 1.5 percent advantage over girls, which is very small, Leahey said.

"We expected to find pretty big gender differences, and we expected to find them early," she said. Initially, she and Guo thought they had done something wrong. They recrunched their data. They started with young students and expanded the study to include older ones. They looked at overall math scores and then narrowed it down to geometry and reasoning.

No change.

Julian C. Stanley, a psychologist at Johns Hopkins University, was among the earliest to document the gender gap. In 1971, he launched the the Study of Mathematically Precocious Youth. In the early 1980s, he published a study in the journal Science, which said that when the SAT, the college admissions test, was given to gifted seventh-graders,
about four times as many boys as girls scored above 600 on the math portion. The top score is 800 on each portion.

When college-bound seniors take the tests, men typically score 35 points higher on the math test than women, and five points higher on the verbal test.

In Baltimore, Stanley, now retired, was less than impressed by the UNC results.

"We've known that for a long time," Stanley said. "I think they may be claiming more for their results than the literature shows. It's been known for years that girls do better in school. It's the reasoning component that they have more trouble with."

Among social scientists, he said, it is common knowledge that there is no math gap when looking at the general population of students. The difference only surfaces when focusing on the very brightest students, and only on reasoning tests, such as the SAT.

"Of all people who don't know about the field, it's sociologists," Stanley said. "Sociologists are typically very much against genetic hypotheses, which we're not invoking here, but they don't do very well with test data."

Leahey countered that her study did look at reasoning and still did not show a gap. As for Stanley's assessment of her discipline, she laughed. "I'm not into discipline-bashing. Many of my close friends are psychologists."

Perhaps the final answer can be found in Joan Troy's honors geometry class, which she teaches to freshmen at Cardinal Gibbons High School in Raleigh.

Troy, who has taught math in public and private schools, welcomed the UNC study but wasn't surprised by it.

"In my experience, I haven't seen a difference. I don't think at any time from when I began teaching in 1978 to today, there's been a time where I would absolutely say my best male students were better than my best female students."

In fact, many of her female students do better than their male counterparts, she said, because they tend to pay more attention to details.

That may be true in state tests as well. In North Carolina, math scores on end-of-grade tests for the 1999-2000 school year show that girls in grades three through eight scored slightly higher than boys. In third grade, 72.5 percent of girls scored at or above grade level, compared with 71.2 percent of boys.

In Troy's class -- 10 girls, 11 boys -- the honors students breathed no collective sigh of relief over the UNC findings. Few had heard that girls did worse than boys in math.

"I don't believe it," declared 14-year-old Greg Kabbes of Cary. "My mom is an engineer, and my dad is nothing with numbers."

On the other side of the room, Gary Burke looked up from his geometry problems. "I've always heard that it was the other way around. That girls are smarter than boys."

Staff writer Trish Wilson can be reached at 829-8927 or trishw@newsobserver.com , and
Leahey can be reached at (919) 843-6876 (office) or by e-mail at leahey@unc.edu .

*******************************************