Which Colleges Benefit Students the Most?

Students enter college with two primary goals in mind: earn a degree and get a job.

Not surprisingly, students at certain colleges are more likely to graduate and land a high-paying job in their field. A simple scan of the College Scorecard will provide data on both counts.

But are high graduation rates and good job outcomes more a function of the colleges themselves or the kinds of students who attend them? Which colleges actually add value to help students exceed expectations?

That’s what a team of researchers at Bain & Company set out to determine.

The Bain study begins by defining student outcomes as graduation rates and earnings, so we have to accept this premise before moving on. Certainly, there are more outcomes to consider, such as the actual learning gained, habits developed, connections made, and options opened, but we’ll go with this given the relative ease of measurement.

These two outcomes are also closely correlated, the study argues.

For graduation rates, Bain looked at six-year completion rates for the cohort entering in 2015. And it examined earnings 10 years out for cohorts entering in 2008 and 2009.

All told, the study’s sample includes 928 public and private institutions for which Bain could obtain data.

With all the numbers crunched, Bain then plotted its findings on a handy graph reflecting graduation rate as the X axis and earnings as the Y axis. Institutions toward the upper right excel in both.

Leading the pack for earnings is the Massachusetts Institute of Technology, with a median salary of $124,213. MIT also boasts a 96.2% graduation rate.

Situated similarly along the Y axis is the nearby Massachusetts College of Pharmacy and Health Sciences (MCPHS), with a median salary of $124,126. But its graduation rate is only 67.3%, so its plot point is farther west.

Mousing over other dots plotted toward the upper right reveals institutions such as the University of Pennsylvania, Princeton University, Stanford University, and Georgetown University.

It also highlights Babson College, with earnings of $111,604 and a graduation rate of 93.7%.

At the other end of the spectrum lie Chicago State University for earnings ($43,759) and Mississippi Valley State University for graduation rate (27.1%).

Most of the institutions are clumped along a trend line flowing from the lower left to the upper right, lending credence to Bain’s claim that graduation rates and earnings strongly correlate.

A few outliers, represented by dots floating in space, reflect anomalies such as the aforementioned MCPHS, along with Kettering University (high earnings, moderate graduation rate) and Vassar College (lower earnings, high graduation rate).

A main takeaway thus far in the study is that highly resourced colleges enroll students who graduate and secure high-paying jobs, while under-resourced colleges do just the opposite.

While this generalization might be true, it doesn’t tell the whole story of what’s going on at individual institutions.

Some might be better at providing economic mobility, and it can be misleading to assume that colleges and universities with higher earnings or graduation rates are doing more than other institutions to help their students succeed, the study notes.

To find those exceptional colleges, Bain adjusted for the nature of the student bodies, controlling for socio-demographics. Certain factors — such as family wealth, gender, community resources, race, and first-generation status — influence educational outcomes, the study argues.

Considering these factors, Bain created a model to predict graduation rates and future earnings. Colleges outperforming their predicted data provide a starting point for understanding which institutions excel at helping students succeed compared to others serving a similar student population, the study explains.

Following this adjustment, several colleges stood out.

Bain highlights Berea College, a small institution in rural Kentucky with a financially needy student body — nearly 90% of students receive Pell Grants. Yet it achieves a graduation rate 20 percentage points higher than its predicted rate.

Another is Florida International University, whose retention efforts have boosted its graduation rate to 5 percentage points beyond what Bain predicted.

The study provides another graph showing how institutions perform on graduation rates and earnings measures compared to Bain’s predictions. Colleges above the trend line exceed expectations, while those below it fall short. The graph doesn’t commingle the data, however, allowing for only one measure to be considered at a time.

With respect to graduation rates, a few colleges stand out, for better or worse. Chicago State’s actual graduation rate of 21% falls short of its predicted 28%. At the top end, Bain predicts Johns Hopkins University to have a rate of 99.3%, but its actual figure is 94.6%.

The University of Florida overperforms, boasting a graduation rate of 90.9% against a prediction of 79.8%. So does Virginia Tech (86.6% vs. 77.3%).

And some are right where they’re predicted to be, such as Stanford University at 95.5%, exactly what Bain’s analysis predicted.

Toggling over to earnings unveils a similar array of dots along the trend line, but with far more outliers. Median wages for MCPHS grads far exceed predictions, as do those for the University of the Sciences and Manhattan College.

Graduates of Wilmington College, though, should expect a median salary of $62,461 according to Bain’s predictions, but that figure is instead $46,654.

What’s more, elite colleges can overperform or underperform on the earnings measure. MIT’s median earnings exceed expectations by almost $9,000, while Brown University’s figure falls shy of the predictions by about $4,000.

Overall, the study points out, some institutions perform better on one outcome than the other. Having a higher-than-expected graduation rate doesn’t guarantee success in earnings, and vice versa.

What conclusions does Bain offer following its analysis?

For starters, it suggests it’s difficult for colleges to exceed Bain’s predictions. About 10% of institutions outperformed predicted graduation rates, and another 10% had median salaries at least $4,600 higher than predicted.

Of course, it’s not entirely clear how Bain arrived at these predictions. The study doesn’t offer much explanation of its methodology, saying only that it employed a nonlinear model that controlled for student body characteristics. Its sauce remains secret.

Among those colleges underperforming on graduation rates, a disproportionate number in the bottom decile are small or regional public institutions.

And for earnings, a disproportionate number of top performers are private.

The primary takeaway is that Bain’s findings prove that students from any background can succeed at high levels given the right support.

To that end, the study recommends a roadmap for action that includes making student outcomes a priority, addressing systemic barriers such as finances and mental health, providing career-connected learning opportunities, and using data analytics to monitor students’ progress.

Every college should be responsible for taking meaningful action toward improving graduation rates and employment prospects, the study concludes, helping students achieve the two primary outcomes of their college experience.