Most people think computer scientists are nerdy guys. Lisa Grossman asks if that image keeps women out of the field. Illustrated by Cat Wilson and Laura Vollset.
Illustration: Cat Wilson
Professor Sapna Cheryan led her student into a small classroom in Stanford University's computer science building. Star Wars posters adorned the walls, discarded computer parts and cans of Coke clustered on a table, and a life-size bust of Spock perched on the desk. “Sorry about the mess,” Cheryan said. “Just ignore that stuff, it's not part of our study. Here's your questionnaire. Let me know when you're done.”
The student took a dubious look at her surroundings and raised her pencil to answer the question: “How interested are you in computer science?”
Cheryan, now a psychologist at the University of Washington, has placed students in situations like this for nearly five years. She has found that women rate themselves as less interested in computer science than men in the “geek room” described above. But in a room decorated more neutrally with art posters, nature photos, and water bottles, their interest levels were about the same.
Many observers have lamented the lack of women in science, but the climate in computer science is worse than almost anywhere else. Women earned about half of the bachelor's degrees in science and engineering in 2006 (the last year for which the National Science Foundation has data), but only 20 percent of the degrees in computer science. That figure is dropping.
Cheryan's research offers an explanation: Women don't identify with the archetypal image of computer scientists. Cheryan's subjects describe this image as “nerdy, techie, stay up late coding, energy drinks, no social life.… They don't frequently take showers.” The geek room conjures this picture in our minds, Cheryan says, based only on the stuff we find lying around.
“You can walk into someone's house and immediately get a sense of whether you belong or not. You don't even have to know the people,” she says. “It communicates a sense of who put together that environment, who is currently in that environment, and who should be in that environment.” Her research also provides a potential solution: Change the perceptions by changing the stuff.
But for the women who already feel at home in computer science, this is a threatening thought. “You get used to people being quirky,” says Sherol Chen, a graduate student in computer science at UC Santa Cruz. “There's no punishment for being different here.” Changing the culture might wipe out that sense of belonging, they fear.
Does Star Trek guard the portal to computer science? Can changing the field's look open locked doors, without threatening the natural habitats of geeks?
Geekspace: The final frontier
The dismal numbers of women in science became headline news in 2005, when former Harvard University president Lawrence Summers suggested that men are hard-wired to be more analytical than women. The fallout was dramatic: Harvard faculty issued a “lack of confidence” statement, and Summers ultimately resigned.
But the inside picture was not so bleak. According to the NSF, the last 20 years have seen a massive influx of women into fields like chemistry, biology, and math. Women earned more than 60 percent of the undergraduate degrees in biology in 2006. The fraction of women getting math degrees has hovered around half since 1975. But the percentage of women completing computer science majors peaked at 37 in 1984, and has steadily declined since.
Social psychologists were already pointing at stereotypes as culprits in the late 1990s—but stereotypes of women, not of scientists. Psychologist Claude Steele, incoming Provost at
Columbia University and one of Cheryan's graduate advisers at
Stanford, studied the effects of the stereotype that women are bad at math. He and his colleagues gave questions from the advanced math Graduate Record Examination to students who valued their numerical skills. They told them that half the test generally showed gender differences (though they didn't mention which gender it favored), and the other half didn't.
Women and men did equally well on the supposedly gender-neutral half. But on the sexist section, women flopped. They scored significantly lower than on the portion they thought was gender-blind.
“All the negative stereotypes that go along with being female, like women are ditzy or less smart or not good at math, are activated when you walk into this test,” Cheryan says of Steele's work. “You're trying so hard to disprove these stereotypes, and there's a lot of anxiety and tension, because you realize that everyone's judging you for being a woman. And then you underperform.”
A room with a Q
Meanwhile, other psychologists wondered how much you can tell about someone from the stuff in their bedrooms.
Samuel Gosling of the University of Texas at Austin gave students permission to enter the dorm rooms or apartments of other students and asked them to rate the occupants on various personality traits. The participants generally agreed with each other's assessments of the strangers, and their descriptions were surprisingly accurate.
These two ideas came together for Cheryan the summer after her first year of graduate school, when she was job hunting in Silicon Valley. She interviewed at Motorola, a cell phone company, and Adobe, a graphic design company.
Motorola looked “totally stereotypical,” she says. “Their conference rooms were named after Star Trek things. I asked them how many women they had in this research office. There were none. I would be the first woman there.”
When the interviewer asked Cheryan if she would change anything about cell phones, she said she wished there was a way to make them stop ringing from a distance. “They were like, why would you need that? People just put their phone right here,” she says, indicating her front pocket. “I said, 'Where's my phone right now? It's over there, in my purse.' I basically said, 'You're not designing your phones for women.'”
Adobe, on the other hand, “was a 180 in terms of their environment.” Their building was bright and beautiful, and about one-third of their programmers were women. Cheryan worked there for five years.
“It was clear to me then that I totally drew these inferences based on environment,” she says.
A study in Star Trek
When I visit Cheryan's lab in Seattle, I expect her to sit me down at a desk like the one at Stanford: covered with science fiction, comic books, and video games. Instead I am led to a small, windowless room with Easter-egg yellow walls, containing only a desk and a computer.
Cheryan's “geek room” study has taken many forms. The current incarnation is in the popular virtual environment Second Life. The computer prompts me to imagine that I'm signing up for a computer science class, and I have a chance to check out the classrooms before picking one.
Video (20.4 mb): Lisa Grossman takes us into the virtual realm of Second Life, where psychologist Sepna Cheryan tests how students react to the "geeky" stereotypes of computer science classrooms and labs.
Requires QuickTime Player
I see the world through the eyes of my avatar, and I control its movements (I never discover its gender) with a keypad. I practice running around the halls for a bit before entering Classroom A, where I quickly discover that I can make my avatar jump on the desks. Fun!
Classroom A is the neutral room, and it feels like a dentist's waiting room. Colorful abstract paintings and photos of autumn in the woods hang on the walls. The teacher's desk has a potted plant and a lamp, and some of the students' desks sport blue plastic water bottles. Magazines like Time and Travel & Leisure sit in neat rows on a shelf. There's even a coffee maker. The only thing that marks the room as a place of programming is a message on the whiteboard: “Welcome to Intro to Computer Science.”
If Classroom A is a waiting room, though, Classroom B is a lounge. It's a mess. The Star Trek and World of Warcraft posters are crooked. Copies of Wired and science fiction novels lie in heaps on the floor. Loose computer parts and wires from video game controllers sit in sloppy piles on the students' desks.
But it does look more like a realistic classroom. The shelf that bore neat piles of magazines in the other room now offers textbooks on C++. And there are some endearing things, like an action figure of Yoda brandishing a green lightsaber. This room certainly has personality.
Cheryan brought her study to Second Life because of the amount of control it offered. In social psychology experiments, everything is a variable. The weather, the lighting, the experimenter's demeanor, or the subject's own mood can affect the outcome of the test. To make sure they're measuring something real, psychologists must run their experiments over and over again, on hundreds of subjects and in as many different configurations as possible.
For the “geek room” study, Cheryan and her team pre-tested every object they wanted to put in the room. They asked students to rate them on geekiness and masculinity. Even the neutral room took extensive pre-testing. Cheryan's lab manager checked out dozens of different water bottles before settling on one she thought would communicate nothing about the drinker. Second Life lets them change the posters and water bottles with a click of a button, giving the study much more flexibility.
Cheryan held her original study in a classroom in Stanford's computer science building to make sure the participants linked the objects with computer science. If they'd decorated rooms in the psychology building, it would have been too obvious that it was a setup, Cheryan says.
The students were led to believe the room always looked like that, and given a questionnaire to assess their attitudes toward computer science. Girls in the study consistently rated themselves less interested, less capable and less similar to the inhabitants of the “geek room” than the neutral room.
But what if it had nothing to do with the objects, Cheryan asked? What if they just thought all CS majors were boys? She did the study again, this time asking students to imagine they were joining an all-female team at a company. The only difference between the two teams was the objects in their office. Girls flocked to the non-geeky job.
Every time they changed the study, the results were the same: Most women avoided the geek space. When prompted, many said it gave them a masculine vibe. The more masculine they found the room, the less they liked it.
“The extent to which women don't like that room was pretty surprising,” Cheryan says. “No matter what we do to that room, even if we make it all female, women just don't feel like they belong there.”
Cheryan's work has been well received by other social psychologists. “It's new and exciting,” says Jennifer Crosby of Williams College, who studies how environmental cues affect racial minorities. “Environmental things send really strong messages about who belongs in a domain and who doesn't. To look at that systematically the way Sapna's doing is really important, well beyond the realm of computer science.”
Greg Walton of Stanford, who studies how feelings of belonging affect racial differences in academic success, says Cheryan's methods were “creative and clever. They were what we call externally valid—they did a nice job of mimicking the ways that people experience these cues in their real-world life.” Even virtual environments can have actual psychological effects, he says. “People interact with virtual worlds as if they're real,” he says. “It's a great tool for research.”
When I'm done exploring the Second Life classrooms, the computer asks me a series of questions: How much do you want to take each class? How similar are you to people in that room?
This question gets to the crux of the matter, as far as Cheryan's concerned.
“There's nothing blatant about it. I don't have any sexist posters in my environment, I don't have anything that's like, 'Women Not Allowed,'” she says. Even so, the subtle cues invite participants to wonder: Will I find people who look, act, or think like me here?
To boldly go…
But what about the women who do think like computer scientists? What of the girl geeks?
Cheryan has given talks where the audience doubted the existence of girl geeks. She's also given talks to girl geeks. There, she has received responses such as, “I'm a female engineer, and I like Star Trek! What are you trying to say?” She explains that her studies aren't supposed to give a picture of what computer scientists are actually like. The geek room is a caricature. “We couldn't have found a room in the CS building that really looked like that,” she says. But the perception it captures is real.
Cheryan's team chose objects for the geek room by asking Stanford undergraduates to describe things associated with a stereotypical computer scientist. These are liberal, politically correct kids, she says; if she had asked them to describe any other group, like black people or women, they would have refused to answer. “But describe computer science majors? No problem!” she says with a laugh. “They didn't acknowledge that there's anything wrong with this, or that there's a chance this might not be true. That was really interesting. It suggests this image is pervasive.”
To find out what girls who end up in computer science think of this image, I spend an afternoon in the CS grad student lounge at UC Santa Cruz with Sherol Chen and her classmates, Anne Sullivan and Gillian Smith. I walk in with environmental cues in mind, and I tick off a mental checklist. Video games? Check. Comic books? Check. Wired? Check. This room is totally geeky.
Photo: Lisa Grossman
Sherol Chen, Anne Sullivan, Gillian Smith (from left to right) and stuffed friends hang out in the computer science graduate student lounge at UC Santa Cruz.
But there's also something welcoming about it. The decorations in Cheryan's geek room tended toward dark and black; this room is bright and colorful. Plush Nintendo characters hold court on the sofa, including Kirby, a pink ball of fluff. Some of the objects from the neutral room make an appearance here, including potted plants and a coffee maker. And this issue of Wired features Martha Stewart making a cake shaped like a Wii.
When they hear about Cheryan's research, the girls are startled. “But we put up all this geeky stuff,” Anne says. “The guys don't decorate.” She and Gillian show off their geeky craft collection: a quilted pillow with the 1-Up mushroom from Mario, a mosaic of Yoshi, and characters from Final Fantasy made from plastic beads.
Could a room like this—geeky yet feminine—help bring in more women? Cheryan thinks it might. “The link between masculinity and computer science is really powerful,” she says. “If we can decouple geekiness from its masculinity, we should be able to attract more women into it.”
But she cautions against leaning too far to either side. Radically changing the image might make current computer scientists—men and women alike—feel unwelcome. “I think it would be dangerous to go in and completely overhaul the image without knowing how that's going to affect the people in the field,” she says. Rather than eradicating the geek stereotype, we should expand it, she adds. “The message is more about broadening the image than about turning it on its edge.”
Indeed, the idea of a more feminine-feeling CS department makes the girl geeks at Santa Cruz anxious. “It's my own form of sexism—I don't like girly-girls,” Gillian says. They like computer science because it is geeky.
“But I draw the line at Star Trek,” Anne says.
“Aww,” Gillian protests. “I like Star Trek.”
With a Little Help from My Friends
Illustration: Laura Vollset...
Researchers at UC Santa Cruz are exploring another way to keep women in computer science: a strategy called “pair programming.”
The approach requires two programmers to sit side-by-side at one machine. One types and controls the mouse while the other watches for logic errors and typos. Programmers who collaborate this way take longer but create cleaner code, according to computer scientist Laurie Williams of North Carolina State University. Williams also found that nearly all pair programmers enjoyed coding more as a social activity.
In the late 1990s, computer scientist Linda Werner of UCSC saw Williams speak about social programming at an engineering conference. Soon after, Werner read an article about the dearth of women in computer science. The author suggested several explanations, including the perception that computer science is a solitary activity.
A light bulb went off for Werner. “I said, Oh! Maybe pair programming could solve some of the problems we're seeing in an educational context,” she recalls.
To test the idea, Werner and her husband, computer scientist Charlie McDowell, joined UCSC psychologists Heather E. Bullock and Julian Fernald to earn a grant from the National Science Foundation. They split introductory computer science classes in 2000-2001 into two sections. One used pair programming on homework assignments, but not exams; the other students worked alone.
A year later, 59 percent of the paired female students were still computer science majors, compared with 22 percent of the women who programmed solo. The results were striking for men as well: the department retained 74 percent of male pair programmers, versus 47 percent of lone coders.
“I thought it would give a boost to women, but the results were gender-neutral,” McDowell says. “It helped everybody.”
Budget constraints prevented McDowell and Werner from following those students as their careers progressed. But McDowell has used pair programming in his introductory classes ever since. “It seems like a no-brainer,” he says.
The idea took off among other educators. McDowell has heard glowing reports from colleagues both near (San Jose State University) and far (University of South Australia). In 2005, the National Center for Women & Information Technology named pair programming a “Promising Practice.”
Meanwhile, McDowell is taking social computing to another level. In September 2008, he received a $600,000 NSF grant to provide scholarships for promising engineering students. The students—both women and men—will live together in the dorms, providing a built-in support network. “Students that feel connected and have a sense of community are more likely to stick around and succeed,” McDowell says.
Lisa Grossman B.A. (astronomy) Cornell University
Internships: New Scientist, Boston and Science News, Washington D.C.
When I was eight years old, my imagination was abducted by aliens, and I never quite managed to recover it. Consequently, I spent my adolescence devouring science fiction. When I read Carl Sagan's Contact, I recognized my future self in its heroine, Ellie Arroway: I was going to be an alien hunter. I could imagine doing nothing else.
So it came as a life-jarring shock when, while studying Martian soil composition, I realized that the once-stunning landscapes of the Red Planet had started to look like mere dirt. I'm turning to science writing to preserve my childlike joy in science, and to spread that joy as far as it will go. Rather than striving to be Ellie Arroway, I can aim to be Sagan himself and to inspire other little girls to be Ellies.
Cat Wilson B.A. (biology) Hendrix College, Conway AR
Internship: Scientific American, New York, NY
At my college, balancing science and art was difficult. All the science labs and studio classes shared time slots, so I would take whatever art was available. My eclectic schedule might contain courses like microbiology and metal sculpture, ecology and art of film, etc. It was not until years later, while working as an Oklahoma Aquarium educator, that I discovered the Science Illustration Program at UCSC. Finally, I found a career that engages both my creativity and curiosity.
Laura Vollset B.A. (English literature) and M.A. (literary studies), Newcastle University, U.K.
Internships: Monterey Bay Aquarium Research Institute, Moss Landing, CA, and Smithsonian Tropical Research Institute, Panama
I was born in the U.K., but I moved to New York State early in 2006. After two years in the U.S., I moved to Vancouver B.C., where I worked in a local gallery. Since 2006, I have pursued science illustration as something more than an extracurricular activity. I undertook illustration work on a pro-bono basis, and I also took classes in botanical watercolors, carbon-dusting, botanical sketching, and etching. I have been greatly encouraged by fellow illustrators in my decision to pursue science illustration as a career. I found the Science Illustration Program here at Santa Cruz both a challenge and an inspiration. Visit my web site.