Knowledge is power when it comes to judging college performance

Students walk through the campus of Temple University in Philadelphia.,

Is college worth it? That’s a question that haunts not only many of the young adults considering their next educational step, but many policymakers, too. The value of a college education by number of years completed is, on average, high and rising. But those averages disguise a huge amount of variation, by institution type, degree major, and many other factors.

The lack of knowledge about college costs and benefits, especially among consumers, is a big problem. Without hard facts, rumors and myths spread like wildfire. Poor choices can result. So, this is an area where data really matter. The Obama administration made significant progress on this front, not least with the introduction of the College Scorecard. But significant data gaps remain.

Even Chetty’s data doesn’t paint a full picture

In a recent attempt to fill some of those gaps, Raj Chetty and his colleagues at the Stanford Equality of Opportunity Project have recently tracked the backgrounds and outcomes of students using federal income tax return data combined with financial aid records from the Department of Education, covering 1996-2014. This work has highlighted a group of universities and colleges, like SUNY-Stony Brook, that are particularly successful at attracting lower-incomes students are helping to propel them up the income ladder. But even this dataset has serious limitations. For one thing, the analysis only covered students of traditional age—between 18 and 22. Also the data on for-profit schools is patchy.

These are not trivial qualifications. In 2016, a majority of undergraduate students were considered ‘not-traditional. Meanwhile, for-profit colleges accounted for 706,769 undergraduates in 2016, almost all of whom were over the age of 24. However, because the current data collection system run by the National Center for Education Statistics (NCES) focuses on institutional-level measures of educational outcomes, like enrollment and completion, and generally pertains to full-time, first-time students. As a result, these data do not provide a good picture of the many non-traditional students who enroll part time, take time off, or transfer between schools.

Time to get rid of the data ban obscuring the picture?

One major reason for the paucity of data is that in 2008 an amendment attached to the Higher Education Act reauthorization barred the federal government from explicitly and systemically integrating and connecting the dots between employment and enrollment, major program, financial aid receipt, and graduation outcomes on an individual student basis. There are of course concerns here about protecting the privacy of students, but these are outweighed by the huge benefits that would come from more robust data collection.

The College Transparency Act of 2017, just introduced as a bipartisan bill in Congress, would end this ban, and allow for federal data systems that track employment and graduation outcomes on a student-unit record basis. Specifically, the legislation would permit NCES to coordinate with other federal agencies. As a result, policymakers researchers and consumers would have access to data with more granular detail on student enrollment, retention, completion, and subsequent labor market outcomes from specific institutions. Importantly, the bill explicitly prohibits the construction of federal college ranking or rating schemes. It also incorporates robust protection of student information, with the expunging of all information that could be used to identify individual, and strict rules against the commercial use of the data.

Data: the friend of fairness

Assessing colleges is a tricky business. All kinds of factors come into play, many of which are highly individual. But for most, cost and value for money are important. As Beth Akers, Kim Dancy, and Jason Deslisle show in “The Affordability Conundrum,” a proper evaluation of the costs and benefits of a college requires information not simply on upfront net costs to students, but how long it will take to complete, the ability to absorb incidental unanticipated costs, the major pursued, and what post-graduation labor market outcomes one may be able to expect.

Whatever the arguments for and against the specific legislation in Congress, it is clear that student-unit record data would provide valuable additional information to a debate fraught with myths and misunderstandings.

Perhaps some colleges will resist this additional transparency. To which the response should be the old Enlightenment plea, sapere aude: Dare to know.