Updated

For all the differences between the sexes, here's one that might stir up debate in the teacher's lounge: Boys learn more from men and girls learn more from women.

That's the upshot of a provocative study by Thomas Dee, an associate professor of economics at Swarthmore College and visiting scholar at Stanford University. His study was to appear Monday in Education Next, a quarterly journal published by the Hoover Institution.

Vetted and approved by peer reviewers, Dee's research faces a fight for acceptance. Some leading education advocates dispute his conclusions and the way in which he reached them.

But Dee says his research supports his point, that gender matters when it comes to learning. Specifically, as he describes it, having a teacher of the opposite sex hurts a student's academic progress.

"We should be thinking more carefully about why," he said.

Dee warns against drawing fast conclusions based on his work. He is not endorsing single-sex education, or any other policy.

Rather, he hopes his work will spur more research into gender's effect and what to do about it.

His study comes as the proportion of male teachers is at its lowest level in 40 years. Roughly 80 percent of teachers in U.S. public schools are women.

Dee's study is based on a nationally representative survey of nearly 25,000 eighth-graders that was conducted by the Education Department in 1988. Though dated, the survey is the most comprehensive look at students in middle school, when gender gaps emerge, Dee said.

He examined test scores as well as self-reported perceptions by teachers and students.

Dee found that having a female teacher instead of a male teacher raised the achievement of girls and lowered that of boys in science, social studies and English.

Looked at the other way, when a man led the class, boys did better and girls did worse.

The study found switching up teachers actually could narrow achievement gaps between boys and girls, but one gender would gain at the expense of the other.

Dee also contends that gender influences attitudes.

For example, with a female teacher, boys were more likely to be seen as disruptive. Girls were less likely to be considered inattentive or disorderly.

In a class taught by a man, girls were more likely to say the subject was not useful for their future. They were less likely to look forward to the class or to ask questions.

Dee said he isolated a teacher's gender as an influence by accounting for several other factors that could affect student performance. But his study is sure to be scrutinized.

"The data, as he presents them, are far from convincing," said Marcia Greenberger, co-president of the National Women's Law Center, which works to advance the progress of women.

Greenberger said she found Dee's conclusions to be questionable and inconsistent. More broadly, she said, boys and girls benefit by having male and female teachers as role models.

"I don't think there are many parents or students, looking back over their educational careers, who haven't been inspired by a teacher of the opposite sex," she said.

"And many have had very unhappy experiences with teachers of the same gender that they are. We have to be careful of too many generalizations," Greenberger said.

Student success cannot be narrowed to the gender of the teacher, said Reg Weaver, president of the National Education Association, the country's largest teachers' union.

Experienced teachers, good textbooks, smaller class sizes and modern equipment all influence how boys and girls do in class, Weaver said.

"Students benefit by having exposure to teachers who look like them, who can identify with their culture ... but this is just one variable among many," Weaver said.

Dee said his research raises valid questions.

Should teachers get more training about the learning styles of boys and girls? Should they be taught to combat biases in what they expect of boys and girls?

In the nature-nurture debate, he said, teacher gender belongs.

"Some people will react strongly to this," he said. "But I've taken pains to explain that we need to be cautious about drawing policy conclusions. As provocative as this all might seem, I really haven't gotten that much negative feedback."