Women in Computer Science 2/2: Changing the culture
This is the story of how the School of Computer Science (SCS) at Carnegie Mellon University went from few female students before 1995, many of which later transferred out of the course, to 40% female students by the early 2000s, almost all of which finished with a degree. I report events as told to me be Lenore Blum, one of the participants of this year’s HLF, who was instrumental in initiating a number of the changes that led to the improvement. In Women in Computer Science 1/2: Beyond mere programming, I had told the story of how SCS widened the applicant pool, and broadened their admission criteria in a way that did not lower standards, but ensured that high-potential students now came from more diverse backgrounds than before. But those were only the first few steps.
At a time when women in computer science were few and far between, notable differences had emerged. As is usual at US universities, a sizeable fraction of Carnegie Mellon students, roughly 2/3 of all undergraduates, are housed in residence halls or on other university-owned or leased properties. First-year students are even required to live on campus. What followed was a game of numbers: Male computer science students were much more likely to have room mates studying the same subject. If you were a male student working late on an assigned problem and you got stuck, waking up your roommate, or someone in your dorm, is fairly easy. Would a female student call up a male classmate at midnight to ask for a meeting? Much more difficult, and prone to embarrassing misinterpretations. Similarly for the differences between fraternities and sororities. As long as women in computer science are rare, you are unlikely to find an older sorority member to tell you about which classes to take and who the best teachers are, let alone help you find a summer job related to your subject. In fraternities, the chances for all that were much better – not through ill will or explicit bias, but simply because of the numbers.
To counter these effects, and with support from then president of Carnegie Mellon Jared Cohon, School of Computer Science Dean Raj Reddy, and Computer Science Department Head James Morris, Blum founded the program Women@SCS for the women in the School of Computer Science. (Additional information about the program can be found in this paper.)
One key program is Big Sister / Little Sister, which matches first-year with more advanced female undergraduates, while the Grad/Undergrad Sisters program matches junior and senior undergraduates interested in grad school with graduate students. The goal in both cases is to create easy opportunities for exchanging information, and for advice and encouragement. Furthermore, social programs – such as class mixers, or socials for undergraduates/graduates create additional forums for discussion and support. Web pages featuring “Women of…” for the different institutes within the School of Computer Science create additional points of contact.
The students are actively involved in the organization through the Women@SCS advisory council, which not only turned out to be the driving force for the Women@SCS activities, but also provided the participants with leadership experience.
After the transition to the new culture – with the turning point being the class of 2002 – Blum and her then PhD student Carol Frieze, the current director of Women@SCS, asked male and female students the same questions as Margolis/Fisher had when the percentage of women was in the single digits. I talked briefly about this study in part one of these series; one of Margolis/Fisher’s finding had been a difference in attitude between male and female students: males more fascinated with computers themselves, females more with the applications.
Blum/Frieze’s result was markedly different from the earlier study: Now, women and men were much more similar than before – there were both men and women who enjoyed programming and computers for their own sake, and students of both genders who saw computers primarily as tools to achieve specific ends. The most common answer to the question asking for a description of computer science in general was now along the lines of “problem solving” and a “way of thinking” (the study can be found here).
Don’t let them miss opportunities!
Women@SCS also includes outreach activities, such as the “Outreach Roadshow”, an interactive presentation that can be used to introduce children, teachers and parents to the many areas of computer science. This turned out to be a draw for female students, who would spend a considerable amount of time taking the roadshow to elementary and middle schools. When Blum asked them why – after all, the curriculum is pretty hard, and does not leave much time for extra activities such as this – the most common answer was a bit sobering: The students answered that they themselves had come to computer science on a whim, through encouragement by teachers, just giving it a shot – and now they were loving it! For them, this immediately raised the question: What about the girls that weren’t so lucky with their teachers, or with the information accessible to them? Participating in the outreach programs was a chance of making sure that other girls would have these opportunities, too. Needless to say, participating in the outreach programs also entailed an educational experience for the students themselves – talking in front of an audience, filling the role of expert, speaking up for themselves. Women@SCS also offers workshops and training in speaking skills.
What benefits women benefits all
One thing that Blum has encountered repeatedly in her Women@SCS work is that women are not the only ones benefiting from the new structures and opportunities. In many cases, she said, Women@SCS made explicit what had been implicit benefits for the majority population – and that explicitness is an advantage for male students, as well. Whatever we discovered that was important, she says, was immediately taken up by the wider community, benefiting not only women but everyone. A good example is the advising session before each new semester, a students-only, no faculty admitted, event where students trade their experiences and advice about the upcoming courses and their teachers. Initially designed as a substitute for the informal advice more readily available to male than female students (see above), there was great interest about the event in the male student community, as well – and the event is now an integrated session.
On a larger scale, Women@SCS has now been joined by SCS4ALL, a student-run organization that works to extend the programs developed for Women@SCS to members of all underrepresented groups.
For Blum, the most encouraging lesson from her 15+ years experience with Women@SCS and the earlier initiatives is that changes to the microculture can have big effects. In the case of the School of Computer Science, there was no need to start in kindergarten and wait for a new generation of computer scientists to grow up over the years. Instead, changing the academic culture – without compromising the standards! – made a huge difference. A key finding that could well be applicable to other STEM fields seems to be that once there is a more balanced environment, both in terms of numbers and professional experiences available, the differences between men and women grow smaller – “Similarity is the Difference”, as Blum and Frieze call it in the title of two of their articles; bolstered by what they found in their repetition of the Margolis/Fisher study.
The jury is still out – or more precisely, systematic studies not yet in – on what happens when Blum’s students go into business, in particular to Silicon Valley, in which by all accounts the culture has not changed at all. Blum points me to an article in the New York Times, published a little over a week ago, exposing the work culture at Amazon – which is disturbing, to say the least, but whose 24/7 cult of availability is particularly hostile to all who have a family or, possibly, all who have a life beyond work. Informal information suggests that CMU women students are being successful, and Blum is confident that her students will have what it takes to carry Carnegie Mellon’s culture with them – and maybe plant some seeds in other places.
Even though that experiment is still ongoing, what is clear is that the Carnegie Mellon computer science program, and its ways of achieving a more diverse culture, is food for thought for anyone facing gender gaps in STEM institutions.
Many further resources can be found on the papers page at Women@SCS.