Q&A: What Women Want

During her childhood in Scotland and Canada, Maria Klawe was fortunate to receive her parents’ encouragement in a subject that was not supposed to interest girls: math. Over the course of her career—which included jobs as a computer science professor at the University of Toronto, a researcher at IBM, and a series of leadership roles at the University of British Columbia, Princeton University, and Harvey Mudd College—Klawe has made it her mission to change the culture of science and engineering by finding ways to encourage women and minorities to get involved.

Since becoming president of Harvey Mudd, you have had great success recruiting women to computer science.

It was actually set in motion the year before I arrived. We used to graduate two or three women CS majors per year, around 10%12% of the total. Then one of our faculty members, Christine Alvarado, looked around and said, “Let’s do something about this.” So a group of maybe four faculty decided to take our introductory course and change it from being a course about learning to program in Java to a course about computational approaches to problem solving.

They also switched from Java to Python.

Python is a more accessible and more forgiving language. Every student at Mudd has to take a computer science class in the first semester, so you have an opportunity to reach them all. The previous course had been very popular with the small percentage of students who already liked computer science and was absolutely despised by everyone else. And it went, literally in one year, from being the most despised course by almost everybody to the most loved course by everybody.

Outside the classroom, Mudd now takes incoming female students to the Grace Hopper Celebration.

This was also Christine’s idea—wouldn’t it be great if we could take incoming students to the Hopper conference and let them see what it’s like to be around all these confident, enthusiastic women in technical careers? So the first year, we took 12 students, and by the next year it was 20, and now we take, on average, about 35 first-year females, or roughly one-third of our incoming class. The total number runs a bit higher since we also take some upper class students as mentors. So far, we’ve been able to afford it; we have a donation from Alan Eustace at Google for $25,000 per year, and then the CS department and the president’s office each kick in a bit extra on top of that.

You created a new research program for female undergraduates as well.

The research program was funded by a grant and ran for four years. During that time, we offered research experience to a handful of female students in the summer between their first and second years. That’s when it’s hard to get a really great summer job, and we figured we could give students something interesting that might also change their idea of what you can do with computer science. So they have done robotics, epidemiology, educational games—a variety of things that have pretty clear applications to society.

You’ve talked before about the importance of teaching practical applications from the start, rather than waiting until students have mastered the building blocks.

We know from research that for women and minorities, the attraction of computer science is what you can do with it. It doesn’t mean they are not interested in complexity theory or other esoteric parts of the field, it just means that that tends to be the driving motivation. And in our experience, it’s not like women take one course or go to the Hopper conference and say, “I want to be a computer science major.” It’s more like, you take one course or go to the Hopper conference, and you take the next course. And then you take the course after that, and by then you’ve taken three courses and you’re going, “Oh, I’m actually good at this, and it gets me summer jobs. Maybe I should be a CS major.”

And now nearly half of Mudd’s computer science majors are female.

It depends on the particular year, because we’re talking about very small numbers. If we look at our graduating class this year, our CS majors are 37% female. Last year, it was 50%, and this coming year I haven’t looked, but it’s been running around 40% on average.

How applicable are the things you have done at Mudd to other institutions?

One of the things we learned is that changing the context of what students are doing—changing the way it’s presented—is something that absolutely any institution can do. Our students are doing just as much programming, and they are learning essentially the same concepts. It doesn’t cost more, and it’s really very easy. The other thing I think many places can learn from us is how important it is to streamline the classes.

By “streamline,” do you mean simplifying the progression from one course to the next?

No, I mean dividing students who have prior experience from those who haven’t. In an introductory CS course, there’s typically a broad range of prior experience. There are people who have been programming since they were 11, and there are people who have never done anything other than use a computer as a tool. And the students who are passionate about CS are just dying to talk to someone who really knows the subject, because most of the teachers at the high school and middle school levels don’t know that much about it. So they get into this first class and there’s a real computer science professor, and it’s like they can’t stop talking. It’s not done out of maliciousness—it’s done out of enthusiasm. But it’s very discouraging to other kids in the class.

So Mudd now has different introductory courses that vary according to students’ prior experience.

There’s zero experience, which is called CS5 Gold, there’s some experience, which is CS5 Black, and then there’s CS42, which is for people who have already had the equivalent of a CS course at the college level.

“We know from research that for women and minorities, the attraction of computer science is what you can do with it.”

This seems like a good moment to talk about “imposter syndrome,” which is a concept you have linked several times to women in computer science.

I talk a lot about the imposter syndrome because it’s unbelievable how common it is among women in technical fields where there aren’t a lot of other women. The idea is, you’re actually doing fine, but you have the impression that you don’t understand the material as well as others around you, or you get this scholarship or internship and think, “I only got it because I’m female, and they’re going to find out that I’m really not that good.” It’s something I discuss at the beginning of every academic year with our first-year students because I think it’s important to let them know that this is a normal way to feel, and it doesn’t mean that you are actually less experienced or prepared or able.

You have also spoken of the importance of being able to ask for help.

Asking for help is something I’m really pushing right now because probably the most important thing you could learn in college is how to learn. And one of the aspects of learning is knowing how to ask for help and how to work with other people. At Mudd, our students become very good problem solvers and very good learners and it’s partly because of the kinds of challenges we give them.

How did you discover your own talent for administration?

I’ve always wanted to run the world. At the same time, I was also very shy and geeky. But I was born liking all the things boys liked at a time when the world was very gendered, and I was lucky that I got a huge amount of encouragement from my parents. When I went to university, people would say, “Girls aren’t good at math,” and I was always the best student.

What was your first official leadership position?

It was at IBM research. My husband, Nick Pippenger, was working for IBM research and I was an assistant professor at the University of Toronto. U. of T. really wanted Nick and IBM didn’t want to lose him, so they made me an offer. But I soon realized my manager didn’t have any interest in me as a researcher. After about three years, I decided the only real solution to this was to become a manager myself. It took me a year to get approval to start a new group for discrete math. But within a year of my starting the group, my boss made me manager of five groups, including the original group I was in. And within about three weeks, my old manager thought I was the best manager he’d ever had, and we’ve been friends ever since.

That is a happy outcome.

It was when I became aware that I totally adored not just being in a leadership role but actually building something.

Before I left IBM, I realized one of the things I wanted to do was change the culture of science and engineering in a way that made it more supportive of people who are not the norm. And I’ve pretty much picked the things I’ve done since then with that goal in mind. I think of Mudd as a lab for innovation in undergraduate science and engineering education, where we can try things and then partner with other institutions to broaden them when we succeed. I’d say we are now in a situation where we have fixed the gender issue—our incoming class is 47% female, and our faculty is more than 40% female. We have got a long way to go with racial diversity, but we are working really hard on that.