For computer science to grow, its most persistent stereotype has to fade.
That’s the takeaway from the work of a group of University of Washington researchers that’s showing just how easily the image of the geeky, socially awkward computer developer discourages women from considering careers in the field.
It’s an absurd deterrent, and it’s time it lost its power.
One of the group’s most recent experiments shows why.
In the experiment, published in Frontiers in Psychology, female college students had two-minute conversations with actors they thought were students majoring in computer science. Half the actors fit the computer geek stereotype, wearing “I code therefore I am” t-shirts and saying they liked solitary things like playing video games. Half the actors didn’t fit the stereotype, wearing solid color t-shirts and saying they liked to hang out with friends.
You can probably guess the rest. The students who talked with actors playing the stereotype were significantly less likely to say they were interested in majoring in computer science than students who talked with the other actors.
And get this — the gender of the actors didn’t matter. Even when the actors playing the stereotype were women, the female students were just as unlikely to show an interest in computer science as when the actors were men.
To the researchers, all this points to a truth worth overstating: We want to go where we feel we belong.
Sapna Cheryan, the study’s lead author, has been thinking about that since 2002, when she interviewed for internships as a graduate student at Stanford University. She was weighing two opportunities — one at Adobe, and one at another tech company. The tech company offered good money and interesting work, but she went with Adobe, she later realized, because the look of other company’s office — complete with Star Trek themed conference room names — made her feel like it wasn’t her scene.
“I remember thinking, ‘This is crazy,'” Cheryan said.
A few years later she ran a study that showed that the decor of computer science classrooms was enough to deter women from considering the field. Female students who walked into classrooms decorated with nature posters and plants were significantly more interested in computer science than female students who saw rooms decked out in science fiction posters and stacks of empty soda cans. Male students who entered the classrooms were equally interested in the field regardless of the decor. Cheryan later did the same study with virtual classrooms: Same effect.
“People can make judgments of a sense of belonging quite quickly and instantly,” said study co-author Andrew Meltzoff, co-director of the UW Institute for Learning & Brain Sciences.
Meltzoff started asking these questions after he spoke at a conference at Google in 2010. He loved the energy but was struck by his audience: a room full of men reshaping the world.
Changing the gender ratio in computer science isn’t about aesthetics to these researchers. It’s about equality of opportunity and quality of product. How will technology work for humanity if only some of humanity works in technology?
Which gets me thinking: When the money is so good and the opportunities so great, why do dumb, inaccurate stereotypes have so much power in the first place?
“Unfortunately, they’re kind of comforting,” said Cheryan.
It’s true. We’re social learners, and stereotypes are our shortcuts to understanding a complicated world, Meltzoff explained. If we had to make all our decisions based on first-hand knowledge and not a fluid set of social judgments, we’d be playing with sticks instead of smartphones.
We’re good at monitoring stereotypes we know to be damaging, so they don’t lead to things like racial bias. But we don’t think computer science stereotypes are damaging. We think they’re fun, and the stories we tell ourselves in our media (see my earlier column about the HBO show Silicon Valley) reflect that.
“If the Big Bang Theory were about actual scientists, it wouldn’t be funny,” Cheryan said.
The next thing to ask is if these studies implicate a rich geek culture as part of the problem, or suggest that women aren’t a part of it. I’d like a company for having Star Trek-themed conference rooms, personally. And these findings don’t ignore these nuances: 1 in 4 women in the study were drawn to geeky stereotypes, not repelled by them.
The solution isn’t to shun the image of the nerdy computer developer, the researchers said, but to make sure it isn’t the only image there is.
The computer science departments at Carnegie Mellon and Harvey Mudd raised the ratio of female majors from about 10 to 40 percent in five years by tackling stereotypes head on. They updated their recruitment process, used more diverse role models when reaching prospective students and revamped their classes so they didn’t seem to be just for “geeky know-it-alls.”
A separate study by the group, which includes co-author Allison Master, showed how female computer science teachers help girls overcome deterring stereotypes when they worry those stereotypes might keep them from succeeding.
Meltzoff sees a bright side to the fact that so much of what keeps women from entering computer science is in our culture: Culture can change.
“Educators should be one of the professions to break down stereotypes,” he said. “We shouldn’t be the stereotypers.”
Really, none of us should.