Computer science (CS) is the most popular major at Stanford University, yet female students make up only 12 percent of the program. It is hardly news that women are underrepresented in all STEM fields, computer science included. But the relatively high number of young women in introductory classes shows that the problem is not lack of interest in the subject While the introductory CS class at Stanford is 40 percent female, the ratio drops significantly to 20 percent at the highest level course of the discipline. This phenomenon isn’t unique to Stanford; it’s universal. As other STEM fields, like biosciences, are becoming more even in gender representation, computer science is actually getting worse. In 1991, of the B.A. degrees offered in computer science, women received 29.6 percent. In 2010, this percentage dropped to only 18.2 percent.
The steep drop-off of women in CS courses demonstrates that problems arise once once interested women have already shown up for class. The culture within computer science classrooms — including sexist misconceptions about what computer scientists are supposed to look like and cyclical underrepresentation of female computer science faculty — contribute to an environment that makes an already challenging major even harder to access and thrive in for American women.
A 2010 report by the AAUW suggests that a powerful cause of the disparity between the representation of the sexes in STEM fields is rooted in the common cultural perceptions about who is supposed to pursue them. Two major stereotypes lead to the underrepresentation of women in STEM fields: the idea that women are inherently worse than men at math and the typical depiction of scientists as men. These popular misconceptions dissuade girls from entering STEM fields in higher education, and by contributing to unwelcome environments, stop those who do from ascending through the fields. The fewer women there are in these fields, the fewer women there are who want to enter them, and the harder it is for those who do to stay. This creates a vicious cycle of underrepresentation.
One factor that could help fix this situation would be hiring more female computer science professors. Of the 33 CS faculty members at Brown, only five are women; of the 30 faculty featured on Harvard’s computer science department page, five are women; of the University of Michigan’s 99 computer science and engineering faculty members, 16 are women. This outright deficiency in female faculty members creates adverse effects in regards to female representation within the student body. Kelly Mack, the vice president for undergraduate STEM education at the Association for American Colleges, argues that colleges can “never expect to diversify [their] student populations if [they] don’t diversify [their] faculty.”
The paucity of female professors department contributes to the self-perpetuating stereotypes AAUW documented. It amplifies the stereotypes about what a scientist looks like, which not only dissuades some women from entering the field but also creates a situation in which those who do have to play the defensive. As Adra Hren, a senior in Brown University’s computer science program, advanced in the CS track, the ratio of females to males in Hren’s classes decreased and she noticed that stereotypes seemed to come into play more: “When I’m in a class with two other women out of 50 students, if I ask a question or give a wrong answer, it’s not just me saying something stupid. I am seen as representative of a group. There’s more pressure,” Hren says.
A lack of representation can create an environment in which it is intimidating to learn. Victoria Chavez, a freshman studying computer science and applied mathematics at Brown, knows the feeling. “If I’m with a TA who I don’t feel comfortable with, sometimes I’ll say ‘oh yeah, that makes sense,’ even when it doesn’t. I’m more likely to keep asking questions when I am with someone who I can identify with.”
In 1991, of the B.A. degrees offered in computer science, women received 29.6 percent. In 2010, this percentage dropped to only 18.2 percent.
Such anecdotes are unfortunately countless. Bonnie McLindon, a computer science major at Stanford, shared a conversation she once overheard in class. “Dude, I think that girl sitting in front of us is a CS major,” said one of the boys, referring to McLindon. “Nah, look at that glittery shit in her hair,” responded the other. When women are made to feel as though certain characteristics associated with their gender cause others in their field of study to take them less seriously, it gives them a strong message about who is supposed to be there in the first place. More women teaching computer science classes would help show students that there is no one way that a computer scientist is supposed to look.
Nonetheless, increasing gender diversity amongst faculty isn’t terribly easy. A fairly obvious reason for this is that few female graduates yields few female professors. In addition, the women who do graduate with degrees in computer science face unfair disadvantages in entering the professional field. Studies by Fiona Murray and Leigh Graham of the Massachusetts Institute of Technology indicate that female scientists have fewer graduate and postdoctoral students working with them as well as smaller networks. They receive fewer invitations to serve on advisory boards and interact less with scientific industries. This so-called “false start” is compounded by an issue that affects women disproportionately: juggling family and work. While balancing cultural expectations and professional ambitions is difficult for women across professions, it is especially in the hard sciences. Since eighty-three percent of female scientists have partners who are also academic scientists — as opposed to only 54 percent of male scientists who do — they are often expected to shoulder a broader burden in the home life than their male counterparts.
Fortunately, students and faculty are working to address gender disparities in CS by addressing these gender-specific problems head on. Hren is a coordinator of WiCS—Women in Computer Science at Brown– a group that works hard to create a more positive, accessible perception of computer science at Brown. WiCS runs discussions on self-care, holds events with tech companies and resume groups and has an active mentorship program. It seems that mentorship is an important step to combatting the problem of gender disparity in STEM fields since so much of the intimidation that female STEM students feel results from a lack of female role models and connections. Similar groups are developing on other campuses too, like Stanford’s she++.
Beyond student groups, some colleges are tackling the problem from a more curriculum-based approach. In 2006, Harvey Mudd College began a program specifically designed to increase its female computer science majors. The strategy involved renaming courses, moving conversations with male students who dominated in class discussion to one-on-one meetings outside of class and offering programs to demonstrate how skills learned in computer science could be applied in the real world. Most importantly, female professors took female students who had completed introductory courses to the annual Grace Hopper Conferences, which focus on celebrating women in technology. After four years of the experiment, the number of female computer sciences majors at Harvey Mudd quadrupled from 10 percent to 40 percent. Other colleges, such as UC Berkeley and Northwestern University, have employed similar programs and yielded similar results.
It may be some time before gender disparities in CS disappear. Such a notoriously difficult major is a struggle for anyone, even those with a sea of similarly identifying potential role models and mentors in the department. Still, these limited success stories demonstrate that the most effective way to increase the amount of women in computer science is to change the culture within it and to increase the amount of women who remain in the field, climbing the ranks. While the task at hand is not easy, student-run groups like WiCS and she++ along with adjusted academic curricula on the part of administrations are a good start.