The ACM Special Interest Group on CS Education (SIGCSE) cancelled their technical symposium in Portland the morning that the conference was scheduled to start. I was there. My wife (Barbara Ericson) and I arrived on the evening of Wednesday March 11, a half hour before Oregon’s Governor banned large meetings. The next morning I got the announcement that the conference was closed. I visited with others who had still arrived for the first day (all of us observing social distancing), met with collaborators during the day, then flew back home on Friday March 13.
While it was disappointing that the conference was cancelled, I’m happy to see that all the papers are posted in the ACM Digital Library (see SIGCSE 2020 proceedings link here), freely available to everyone until June 30. I’ve been spending time looking through the papers, wishing I’d have had the opportunity to hear the presentations and talk to the authors.
Let me tell you about the easiest paper to recommend — one of the Best Paper awardees for CS Education Research: Competitive Enrollment Policies in Computing Departments Negatively Predict First-Year Students' Sense of Belonging, Self-Efficacy, and Perception of Department by An Nguyen and Colleen M. Lewis of Harvey Mudd College (see link here). The punchline of the paper is in the title, and might be described as "If you send students the message that they’re unwanted, they’re going to feel unwanted."
An and Colleen look at a dataset from the Computing Research Association based on a survey of 1,245 first year students. They looked at four outcome measures:
- Perception of department as welcoming indicated by agreement with phrases like "My department cares about its students."
- Sense of Belonging indicated by agreement with phrases like "I feel like I belong in computing."
- Self-efficacy (which is a sense that you can achieve tasks in a domain) indicated by agreement with phrases like "I am confident that I can pass my classes."
- Growth mindset (which is a sense that you can always get better at something through effort) indicated by agreement with phrases like "Anyone has the ability to learn computing and be good at it."
They defined a department as having competitive enrollment if students had to apply to become a computing major, or if a student needs to meet grade thresholds (beyond simply passing) to become a computing major. They found that students in departments with competitive enrollment had a lower sense of being welcomed and a lower sense of self-efficacy. If students had prior experience, they still had a sense of belonging in computing, but that wasn’t true for students who didn’t have prior experience in belonging. Over all, female, Black, and Latinx first-year CS students had a lower sense of belonging and lower self-efficacy.
Departments are using competitive enrollment as a way of managing rapidly rising enrollment, or just to make sure that the students who get to the upper-level classes succeed. In any case, these moves are a barrier to students. The results from this paper suggest that the moves are having an impact on students.
Some of the discussion about this paper on Twitter points out that the effect isn’t all that strong — statistically significant, but not a large effect size. That makes sense to me. These are subtle and likely indirect effects. If a CS professor said to a student, "You’ll never make it in CS. You don’t belong," and then the student consequently showed a decrease in self-efficacy and sense of belonging, we would say that the mechanism would be pretty clear and the effect would be direct. In this case, the competitive enrollment barriers may not take effect until the 2nd or 3rd years of undergraduate, and this study is looking at all 1st-year students. The direct impact is maybe some form social pressure on students (e.g., talking to other students facing these barriers), or the student dreading a future date when they would have to be judged. Competitive enrollment tells students that not everyone is welcome, and 1st year students seem to be responding to that.
We can’t know the future of CS enrollment. Will CS still have huge enrollment next year, when the world is still dealing with Covid? How will competitive enrollment measures need to change in the future? This is an important study to realize that there are likely effects of barriers, even for students in their first year.
There is a lot more great stuff in SIGCSE 2020. I recommend taking a stroll around the proceedings.
Mark Guzdial is professor of electrical engineering and computer science in the College of Engineering, and professor of information in the School of Information, of the University of Michigan.