On Grinnell’s Completion Rate

In the aftermath of Grinnell’s decision to discontinue Posse, the President’s office produced a Draft Framework for Meeting the Diverse Needs of our Students (which isn’t quite the same as meeting the needs of our diverse students). The introduction of that document focuses on our completion rate. I think the document focuses on the retention rate because that rate is part of the strategic plan, and not because it’s particularly relevant to meeting diverse needs.

A recent document from the President’s office makes the following statement.

While our recent six year completion rates of 88 to 90 percent exceed the vast majority of colleges and universities in the United States, we also know that they do not compare favorably with the highly selective liberal arts colleges that we benchmark ourselves against.

Let’s take a look at that statement. First, what are our recent six-year completion rates? In 2011, the completion rate of the 2005 cohort was 88%. In 2012, the completion rate of the 2006 cohort was 90%. In 2013, the completion rate of the 2007 cohort was 86%. In 2014, the completion rate of the 2008 cohort was 89%. And in 2015, the completion rate of the 2009 cohort was 86%.

Those 86% completion rates are a bit depressing. And it’s certainly worth investigating why they are that low. There are clearly some obvious contributing factors: We had significant problems with our mental health services over the past four years, and that clearly affects student completion rate. I’d want to see the College address the issue of providing sufficient mental health resources on campus (as the SHACS evaluation team recommendations from four years ago suggested, and as the Draft Framework reinforces). I’d also like us to find out what else we are hearing from students who do not complete their Grinnell careers.

I’m not sure that I buy the claim that these do not compare favorably. Let’s look at the peer group. We do worse than Amherst (94% to 96% in the same range of years), Bowdoin (92% to 95%), Carleton (92% to 95%), Davidson (92% to 93%), Pomona (93% to 96%), Swarthmore (92% to 95%), Vassar (91% to 94%), and Williams (95% to 96%). We do better than Smith (85% to 87%) and Reed (74% to 82%). We do about the same as Kenyon (87% to 90%), Macalester (87% to 90%), Oberlin (85% to 89%), St. Olaf (85% to 89%), and Washington and Lee (88% to 91%).

So, would I say that we don’t compare favorably? Well, we are certainly lower than about 2/3 of the schools. What do we notice about these schools? All the schools that do better are on the coasts, except for Carleton. All the schools that do the same are in the midwest, except for Washington and Lee. As realtors suggest, location matters. Could it perhaps be that location has a direct or indirect effect? In the midwest, you may feel far from things, particularly if you’re not from the midwest. And maybe that’s not comfortable. (I’ve certainly heard a reasonable number of students from large cities say that they really miss having the availability of many of the things that go with big cities; some have said that it makes them want to leave.)

What about the populations at these schools? Grinnell generally stands out as a school which serves a larger financially needy population than our peers. We usually have more Pell grant recipients, more first-gen students (which I know does not necessarily tie to financial need), and such. These students are likely to have a tougher experience at college; they usually need to do more work for pay than other students, and they are likely to have fewer resources if something goes wrong. Both are issues that are likely to affect retention. Could that be enough to make the difference? I don’t know, but I think any reasonable analyst would look at issues like this before making big decisions about approaches to take to improve our retention rate, or even before assuming that we can significantly improve our retention rate.

I appreciate that we have started a program in which we try to interview every student who chooses to leave Grinnell so that we can gather better data on retention. Are people leaving because of academic challenges, because of our alcohol culture, because of frustrations about the ways in which we support students, because of health issues, for financial issues, or for a host of other reasons?

We should also accept that Grinnell is not the right institution for some students. Some may need more of an academic challenge (as hard as that is to contemplate) or access to graduate classes. Some may want access to different programs, such as architecture or engineering. Some may really benefit from being in a large city.

But we should make sure that the students who want to stay at Grinnell can stay at Grinnell. And we should probably try to make more students want to stay at Grinnell. Maybe we can improve our retention rate by putting in a better financial safety net for our students. Maybe we can improve our retention rate by addressing our alcohol culture. Maybe we can improve our retention rate by catching students in academic difficulty earlier, as the draft framework suggests.

Until we look at appropriate data, it’s not clear to me that we should be focusing significant resources on particular attempts to change retention (other than on mental health, which is a clear need). Let’s gather data. Then let’s make decisions about what goals are reasonable to pursue and what way to best pursue those goals. And let’s realize that making huge changes to our completion rate may not be be possible because of factors beyond our control, such as the effects of location.

Version 1.0.1 of 2016-10-13.