The state of Florida is conducting a master class in how not to have civil debate.
So it came as a blessed relief this week to see a relevant, civil, on-point scholarly debate play out in a relatively public venue. It’s about completion scholarships. It’s noteworthy both on its own terms, and as an example of exactly the kind of debates that academics should be having.
Some context: completion scholarships are typically small grants of money that colleges or universities will give to students who are nearly finished. The idea is to make sure that students don’t get derailed by finances at the last minute. We’re all better off if a student graduates than if that student walks away in their last semester due to a financial crisis. If a relatively small subvention can make a material difference for a significant number of students, it’s well worth doing.
Of course, the key word in that sentence is “if.” If completion scholarships are effective at scale, we should adopt them post-haste. If they aren’t, then we should redeploy those resources to something else that would make a positive difference. At heart, it’s an empirical question. Do they work?
Sara Goldrick-Rab posted a summary of a newly-released study that she did with a host of co-authors looking at completion scholarship programs at eleven public universities in several states. The study found that when the completion rates of students who received the scholarships were compared with the completion rates of similar students who didn’t – “similar” is key here – the differences were not significant.
In a piece by Goldie Blumenstyk at the Chronicle, Timothy Renick suggests that the success of the completion scholarship program at Georgia State suggests that the issue may be the details of implementation, rather than the concept itself. Goldrick-Rab rebuts that assessment; I’ll leave the details to the reader.
A few thoughts.
First, Goldrick-Rab and her colleagues are asking the right questions. Completion scholarships make sense intuitively, but sometimes the truth is counterintuitive. (I remember having a “huh” moment when I learned that the students with the lowest loan balances have the highest default rates. Truth doesn’t care if you anticipate it or not.)
Second, one study does not end the discussion. I’ll assume that the study is methodologically solid, but even if it is, its universe is limited. For example, it did not include any community colleges. Might the institutional context make a difference? This study, by itself, can’t say.
Third, it’s likely true that the details of implementation vary from place to place. Renick may be correct that Georgia State got the details right in ways that other places haven’t. Comparative work looking at places that make it work and places that don’t could be illuminating.
Finally, hooray for the debate. It’s about how best to support students completing their degrees. Are completion scholarships more effective than, say, extra advisors or local bus passes? I don’t know the answer to that, but I’d like to; it would be helpful in making difficult resource-allocation decisions. And the debate seems to be happening without any of the ad hominem or other bad-faith attacks that have become so common; it’s being conducted (as far as I know) by adults, as adults. That in itself is cause for hope.
I’ll admit that I hope completion scholarships wind up being vindicated. They’re so intuitively appealing that it’s hard not to root for them. But the real point isn’t to take sides; it’s to engage in informed investigation and debate to figure out the truth, and then to use that truth to better serve students. So, kudos to Goldrick-Rab, Blumenstyk, and Renick for showing us how it’s supposed to be done.