Up for Vote #3: Vital Education Data Should Be Accessible and Confidential

This is the third blog post in the Up for Vote series focused on Congressional action to reauthorize the framework legislation for federal education policy. Our recommendations are based on AIR’s research. Here we focus on issues affecting the Higher Education Act (HEA). Next up: Title II, Part A.

The U.S. Department of Education has made landmark contributions to the development of a robust and secure education data infrastructure. For many years, its National Center for Education Statistics (NCES) has conducted high quality studies of the student experience from birth through adulthood. More recently, its State Longitudinal Data System Grant Program has helped states effectively invest hundreds of millions of dollars in technologies designed to help educators improve student outcomes and close achievement gaps.

That said, one facet of the Department’s operations has a conspicuous shortcoming. Analysts can’t get access to the vast data resources of the Federal Student Aid office (FSA), hindering their ability to build evidence about the impact of student aid, design policies aimed at widening access to and boosting success in higher education, and refine interventions that help students better manage educational debt.

The federal government provides about $150 billion dollars a year in grant and loan assistance to postsecondary students—and its student aid policies influence the higher education choices of states and colleges, the price that millions of students pay for college, and the debt that students carry after they leave school.

As Congress works on reauthorization of the Higher Education Act, the need for far better research and access to federal student aid data should be high on its agenda.

Consider the predicament of state government researchers, who presently have little access to FSA data. States and the federal government share responsibility for promoting access and quality in higher education. With better data on borrowers in hand (stripped of any individually identifiable data),  state researchers could understand how much debt students take on, see how changes to state tuition policies affect that debt burden, and target state student aid more precisely.

By linking FSA borrowing data to state-based earnings data, those same researchers could examine student debt-to-earnings ratios. That way, they could better assess the impact of state tuition policies, refocus operating subsidies to colleges and universities, and provide better information to students and their parents. With access to federal student aid application (FAFSA) data, state policymakers could better judge how initiatives to spur college enrollment and aid applications are working.

University-based researchers, too, have long sought access to FSA data – likewise, without success.

Some want to use federal student aid application data to evaluate whether small and inexpensive “nudges,” such as helping low-income tax filers fill out federal student aid applications, could boost college-going rates more efficiently than investing added billions in the Pell grant program.

Others want to use loan-repayment data to understand if, when, and why students make or defer payments, enter forbearance, become delinquent, or default on their student loans.

FSA has a difficult job. Every year, it processes tens of millions of aid applications and awards billions of dollars in grant and loan assistance through more than 6,000 partners in higher education. Rightly, FSA takes seriously its duty to ensure that the information that it collects is never used in ways that compromise the privacy of individuals.

But this duty does not preclude FSA from sharing its data—shorn of personal information—with those who could use it to improve postsecondary education policy and practice.  Together, well-established procedures and available technologies could make data accessible to researchers without risk of breaching data security or confidentiality. One model, pioneered by the Census Bureau, allows approved researchers to conduct student-level analyses in a secure “virtual workspace,” taking with them only the results of their work, and only after those results have gone through a thorough review to prevent the inadvertent disclosure of personal information.

Even more modest changes could be helpful, enabling analysts at higher education associations, advocacy organizations, and media outlets to do their work. FSA could, for example, leverage an existing tool developed by NCES–PowerStats–to make it possible for even novice users to quickly create custom tables and simple statistical analyses without ever touching underlying personal information.

If the Department believes additional statutory authorization is needed to implement these solutions, now is the time to act, as Congress reauthorizes the Higher Education Act.

With Congressional proposals for an austere aid future and the need for growing numbers of college graduates, we can’t afford to make student aid policy without the evidence about what works—data that FSA can provide.

Matthew Soldner is a senior researcher in AIR’s Higher Education practice. Prior to joining AIR, he worked at the Department of Education’s National Center for Education Statistics on federal student aid, student persistence and completion, and baccalaureate students’ post-graduation outcomes.