There can be lots of variation underneath a college’s overall graduation rate. Recently, a colleague here at The Chronicle stumbled across an unusually detailed breakdown of graduation rates that drove the point home.
The breakdown, which was compiled by the Indiana University system’s institutional-research office, might not seem all that revelatory at first. All colleges that receive federal financial aid report graduation rates for some subgroups—by race and sex, for instance—to the federal government. But the Indiana data go further, looking at categories like residency status, first-semester grades, and more.
That gives us a close glimpse of what the top-line data leave out. Let’s take a look:
The system’s overall graduation rate is 58.2 percent.
But what does that actually mean? All that figure really tells us is the percentage of students who were full-time, degree-seeking beginners in the fall 2007 cohort who went on to earn any degree from the system within 150 percent of the expected time (for example, a bachelor’s degree within six years or an associate degree within four).
For all its definitional precision, that number tells us only so much.
As you’d expect, things look rather different when we consider each campus separately. You’ll see, for example, that Bloomington, the flagship, has a much higher graduation rate than do the regional campuses, with IUPUI in between the two extremes.
There are two reasons for the disparity:
First, the regional campuses serve a different student population than Bloomington does. One way to illustrate the point is with SAT scores. As you can see, a substantially larger portion of new students scored at least a 1200 on the SAT at the flagship campus than at the others. (Note that SAT scores are not available for all students—here we’re looking at those with a 1200 or higher as a percentage of all students, whether or not they reported an SAT score.)
In addition, many students on the regional campuses have other responsibilities, like jobs or raising a family, on top of their academic work. “They’re trying to balance all of the other draws on their time,” says Todd J. Schmitz, the longtime executive director of university institutional research and reporting, which creates the graduation reports.
That raises some questions for the university system, Mr. Schmitz says. Like this one: “Can we offer them enough financial aid, without going bankrupt, to get them to back off from work commitments to focus on school?”
Second, the regional campuses serve a different purpose than the flagship does. When freshmen start at Bloomington, it’s reasonable to assume most of them aim to graduate from it. At the regional campuses, though, more students transfer to other colleges, and in some cases that may have been their plan all along. To help account for that, the system reports the “transfer-out rate” for each campus. (Transfer-out rates apply only to students who leave the university system entirely. Students who start at one system campus and graduate from another are counted as a graduate of their original campus.)
Those institutions with graduation rates that didn’t look so hot above? Factor in their transfer-out rates and you get a better picture of what they’re trying to accomplish.
It’s not just other colleges that might steal students away from the regional campuses, either. Their enrollment tends to dip when the economy improves, as students choose jobs over school. Just this fall, increased hiring at a Chrysler plant near the Kokomo campus has hurt its retention, Mr. Schmitz says.
Behind the Numbers
Graduation rates have become a popular measure of college quality, one expected to appear in President Obama’s proposed college-ratings system. But the rates are driven not only by what colleges do but by who’s on the campus in the first place.
One way to think about this is to look at the graduation rates for resident and nonresident students in the system. Over all, out-of-state students graduate at a higher rate, 74.2 percent, compared with 52.7 percent for Indiana residents. Typically, out-of-state students come from families with the resources to pay higher tuition. That sort of family background can often provide an edge in other ways, too.
The Indiana system breaks down its graduation rate in numerous ways: by race, gender, first-semester grade-point average, and more. In general, the rates follow the patterns that one would expect. For instance, the graduation rates for “other” races (including white) tend to be higher than those for African-American and Hispanic students, populations we know graduate at lower rates nationally.
None of this comes as a shock to leaders in the Indiana University system, Mr. Schmitz says, though they do use an even-more-detailed internal view of graduation rates to help tailor programs for at-risk students.
But there is the occasional surprise.
The reasons for departures from the expected pattern can be hard to pinpoint, Mr. Schmitz says. “If I had those answers,” he says, “I’d be a consultant bringing in the big bucks.”
As Mr. Schmitz’s comments suggest, graduation rates are descriptive, not predictive. But that doesn’t mean the data are without interpretive value. Colleges know that certain student characteristics are closely correlated with graduation. In the past, the conventional wisdom was that responsibility for student success rested mostly with the student, Mr. Schmitz says.
These days, though, colleges are held more responsible for it than ever before. For institutions like the Indiana University system, that means striking a balance between giving lots of people a chance at college—even if some of them seem like long shots—and offering better advising, more-detailed orientation programs, and other support to help as many as possible finish what they started. A good grounding in the data can help the university figure out who needs help and, over time, whether its efforts make a difference.
