Computer science professor Andrew Campbell developed the StudentLife app, which tracks students’ mental health, academic performance and behavioral trends. In the spring of 2013, Campbell used the app to measure the ups and downs of 48 of his computer science students’ terms.
What were the most surprising results?
AC: Wow. Conversation, sleep, location, movement around campus outdoors/indoors: We have this deluge of data. We found correlations with different things — for example, mental health. The interesting breakthrough that we made was that the phone actually did this. We found that students in this group who slept less were more likely to be depressed.
So I’ve seen the website, and I notice there’s a number of intuitive results. Stress starts to plateau around week five. People stop exercising around week four.
AC: They stop going to the gym, that’s right. There’s some subtle things. Some people go to bed before 12, but most people go to bed between 12 and three. And we’ve got some vampires down here who go to bed at four, five, six. These are strong correlations, and they’re significant. The less sleep you got, the more likely you’re depressed. The fewer face-to-face conversations you’ve had during the day was the strongest indicator of depression.
I’d say that qualifies as a surprising insight.
AC: Kids who are depressed are more likely to retreat into themselves. But on the flip side of things, we also looked at loneliness, and we found that conversation frequency had no correlation with loneliness. When I talked to a psychologist, he said, “Yeah, that’s right.” As laypeople, we feel that somebody who is social wouldn’t be lonely. But it’s not the case. Here’s another result. We found that students who had longer conversations and more frequent conversations did better in their exams. Causality? No clue.
Let me shift tack a little bit and ask more generally about your interest in this as a hard scientist. I think that, fairly or unfairly, hard science professors get a worse rap than professors in the humanities for their level of interest in student life. Do you think that reputation is deserved?
AC: I don’t think it is actually. I think we’re all, to some degree, motivated by personal experience. My brother, many years ago — because I’m an old fart now — had his first depressive episode, and he dropped out. He went back to a different college and got a degree in computer science. Could the [app], perhaps, have allowed my brother to live a less disruptive life? Do hard scientists get a bad rap when humanities professors may be more connected with student life? Our department chair Tom Corman cares more about students than anyone.
Is Dartmouth doing enough to support its students?
AC: I don’t think so. That’s not to say that they’re trying to explicitly undermine people’s health. They clearly care about it. It’s part of the mission. But I would argue that we don’t really have a handle on what’s going on in our campus. There’s wonderful people over at Dick’s House, both on the clinical side and on the counseling side, who absolutely care about the student body. But because of stigma and because walking across campus is a huge struggle for some students, it doesn’t matter if the services are there. There’s a number of stakeholders. They lack the complete information to move things forward. And I honestly think that technology could fill that gap. I’m convinced it could.
This interview has been edited and condensed.
