In an analysis of some 10,000 students at nearly 100 schools, Riegle-Crumb found that the divide was anything but constant.

"What we find is that there are many schools where boys and girls take high school physics at the same rate," Riegle-Crumb said in an interview. "And that there are many other schools where more girls actually take physics than boys. And so when you look at the aggregate, you see a pattern where boys are taking physics more than girls, but there is a lot of variation around that."

There are some obvious things that could cause those variations. If parents of some kids are scientists, or highly educated, they might push their daughters to take tough courses in high school. Wealthy families might be able to afford tutoring, or have one parent stay home to help kids with homework. Better funded suburban schools might be at an advantage over inner-city schools.

But when Riegle-Crumb controlled for those and other possibilities, she found one reason remained: "What we found is that in communities that had a higher percentage of women in the labor force who are working in science, technology, engineering and math, that in those schools, girls were as likely as boys to take physics, or even more likely."

[RELATED: Girls and Math: Busting the Stereotype ]

Riegle-Crumb's finding about the importance of local role models meshes with a broad range of earlier work that shows the decision to pursue math and science is not about innate differences between boys and girls, but about social context and norms. Countries with greater gender equality, for example, reveal more equal math test scores among boys and girls.

Teenage girls growing up in communities where women are better represented in tech are more likely to see women commenting on tech issues in public forums and in school discussions — and more likely to run into a friend's astrophysicist mom at a birthday party.

By contrast, Riegle-Crumb said, girls growing up in communities where most working women are in jobs traditionally held by women such as child care or nursing might not see the possibilities that exist.

"If I am a young woman growing up in a community or culture like that, then that's what I see as, 'Well, this is what I am expected to do,' " Riegle-Crumb said. "And so it may not ever occur to me, that, 'Oh, you know, I don't actually have to do that. There's a vast array of things I could choose to do.' But if no one around me is doing those things, it's hard for me to even consider that possibility."

Copyright 2013 NPR. To see more, visit http://www.npr.org/.

Transcript :

LINDA WERTHEIMER, HOST:

NPR's social science correspondent, Shankar Vedantam, just got back to Washington from a trip to Silicon Valley where he noticed, as many others have, how few women are represented among the ranks of engineers at tech companies. He's here with us now to talk about a new study that takes a fresh look at the persistent gender disparity in science, technology, engineering, and mathematics, the so-called STEM fields. Shankar, welcome.

SHANKAR VEDANTAM, BYLINE: Hi, Linda.

WERTHEIMER: So does this research explain why we see so few women in tech?

VEDANTAM: Indirectly, Linda. A lot of people in technology want to see more female engineers but the problem, they say, is that not enough women are taking course that lead to engineering careers. In fact, if you go back even further, adolescent boys and girls often take different academic paths which leads to these disparities in career choices. We've known for decades, for example, that girls are less likely than boys to take high school physics courses.

I spoke with Catherine Riegle-Crumb. She's a sociologist at the University of Texas at Austin. She recently tracked 10,000 high school students across the country because she had a hunch that this gender gap in high school physics actually masks something really important. And what she found was actually very interesting. This gender gap in physics is not a constant. It varies quite widely across the country.

In some places boys do take more physics than girls but there are many schools where it's actually girls who are more often taking physics than boys.

WERTHEIMER: So does she go on, I hope, to explain why there would be such a variation?

VEDANTAM: So she looked at a bunch of factors that might explain it. You know, income and family education could play a role. If your parents are physicists, maybe you're more likely to take physics. Or maybe you have a parent who can afford to stay home fulltime with you and that's why you take tough high school courses. So Riegle-Crumb controlled for those things. She statistically eliminated the effects of wealth and family education background.

She even controlled for the location of the school - whether it was suburban or inner city - and she found that there was one factor that predicted why girls took physics.

CATHERINE RIEGLE-CRUMB: In communities that had a higher percentage of women in the labor force who are working in science, technology, engineering and math, in those schools girls were as likely as boys to take physics or even more likely.

WERTHEIMER: So let me be sure that I understand. She says that the number of women in the community who are doing math and engineering in their work, that when those jobs go up, somehow the girls from that community take more physics?

VEDANTAM: That's exactly what she's saying, Linda. And in some ways this matches what international studies have found. Countries that have greater gender equality, for example, seem to have more equal test scores in math among boys and girls. I think what Riegle-Crumb has done is empirically demonstrate the same thing happening within the United States.

So I asked her, you know, what causes this effect? You know, high school girls don't hang out at the local tech company to see whether there are women working there or not. And she said it's subtler than that. You know, having more women in the local tech workforce changes local norms. It changes how the playing field tilts. You know, who do you see on television? You meet the mom of a friend at a birthday party and it turns out she's an astrophysicist.

Compare that to a girl growing in a part of the country where women are in traditionally female occupations like childcare or nursing.

RIEGLE-CRUMB: So if I'm a young woman growing up in a community or a culture like that, then that's what I see as, well, this is what I'm expected to do. Right? And so it may not ever occur to me that, oh, you know, I don't actually have to do that. There's a vast array of things I could choose to do. But if no one around me is doing those things, it's hard to even consider that possibility.

WERTHEIMER: So does she think, Riegle-Crumb, does she think this work could possibly have policy implications? I mean is there some you can intervene in this process?

VEDANTAM: Well, she does. You know, a lot of tech companies want to increase the representation of women among the ranks of their engineers, but the thing is that companies think the source of the problem lies in education, that girls are not taking these courses. I think what Riegle-Crumb's research shows is that this is a chicken and egg problem. Fewer girls may be taking courses that lead to tech careers when they don't see female role models already in tech careers.

WERTHEIMER: So you supply role models.

VEDANTAM: Exactly. So you know, the answer to the question of why are there so few women in tech is because there are so few women in tech.

(LAUGHTER)

WERTHEIMER: Thank you very much, Shankar.

VEDANTAM: Thanks, Linda.

physicsphysics

Why Teachers Love Using Those Magical OK Go Videos in Class

How Classroom Culture Opens Up When Students Can ‘Patent’ Ideas

The Poetic Dance Between Simplicity and Uncertainty in Physics

What Would It Take For Humans to Fly Like Superheroes?

Why Aren't More Girls Attracted To Physics?