boosting retention vs. invasion of privacy

In a recent article, the Chronicle of Higher Education covered a product from Degree Analytics that looks at a lot of “big data” – specifically WiFi location activity – to aid in student retention. The article is also about the privacy concerns when one starts digging into such data. Just because most if not all systems at least passively collect location data on WiFi networks doesn’t mean it’s appropriate to be using that information. Students haven’t specifically given permission to have their data accessed that way. I’m willing to bet a lot of money, too, that they aren’t aware that their movements could lead to such analysis that somehow “predicts” their success at the institution.

I’m not going to get into Degree Analytics specifically. I will admit, though, that we gave them a legitimate look here at Muhlenberg. And that I was pretty torn about the matter of boosting retention vs. a potential invasion of privacy.

I realize that this post will get a few up in arms, including some that I consider close colleagues and even friends. I would expect that to be the case, though, considering the topic. Certainly there were some here at Muhlenberg that were up in arms at the mere notion of using data this way. But we did take a look; this is a polarizing topic to say the least.

But here’s my take, and my conflict.

Retention is a big issue today. The connection to student academic success is obvious (though there are many, many other aspects to “success” than just academics). There’s the altruistic aspect to this – we want students to succeed because it’s the right thing to want and pursue. That’s far and away the bigger side of the issue, and I won’t belabor why that is so important. But there’s also a business side to this, all the more important considering today’s higher education climate – every student we retain from year X to X+1 is a multitude fewer we have to enroll as a first-year, provides revenue at a lower discount rate (presuming discount rate goes up with each class, as it generally does), and improves our graduation rate (which affects rankings).

So if we can retain even one more student for all the above reasons, altruistic and business…is that bad? Is it even…good? Good enough? How much is enough to justify using data however we want?

Let’s look at the other side of the coin, which is a doozy. First, students don’t realize that this kind of data on campus whereabouts based on WiFi connections is even collected. They certainly wouldn’t think we’d use it to literally track them, then draw conclusions about their “success” and intervene when we fear that “success” is in jeopardy. Second, just because we collect something doesn’t mean we should use it at all, much less in this way. By the way – at Muhlenberg at least we don’t “monitor” people through whatever data we do collect. Yes, our WiFi logs go back 90 days (but not farther, for any reason whatsoever), but we don’t comb through them pro-actively. We only use them if we get a DMCA complaint and we have to figure out who was connected to what IP address at a particular time. But that’s pretty darn specific. Degree Analytics is very, very broad.

So this is a pretty big invasion of privacy. A massive one, to many.

We didn’t do Degree Analytics here, but probably 30/70 cost vs. privacy. It wasn’t all privacy concerns. And I was among those torn about it. Because we do have to worry about retention. And maybe, just maybe, this is a way to improve and get all the benefits, to us and students, that better retention entails. But it’s a dangerous path that we’d be starting down, without question.