This year in the United States, over 21 million students will enroll in one of over 6,000 schools. How will they fare?
Will they graduate? Transfer? Wind up employed? Will their investment in college be worth it?
Given the current data infrastructure of higher education, we won’t know the answer to any of those fundamental questions. This severely hampers students’, parents’, and the federal government’s ability to make smart investments in college. The lack of data also prevents higher education from moving forward: without good data on outcomes, stakeholders are locked into an input-based regulatory structure that constrains innovation.
The system today
Higher education data today is collected by colleges and universities themselves, is largely self-reported, and until very recently, included little data on the outcomes of part-time students, or of students who had transferred from another institution. If a student attends multiple institutions, a reliable record of his experiences is weak at best.
This makes it nearly impossible to assess performance of schools. Do low graduation rates mean that colleges are failing students? Or are some institutions helping students traverse an expensive higher ed landscape in an affordable way? We have no idea.
Collecting data at the institutional level also means that we know very little about which programs at each school are effectively equipping students with the skills they need to succeed—and which are not. This makes it very hard for students to make good decisions about where to attend school. While there are school rankings for some majors, these ranking systems are input driven, and say little about the outcomes students can expect to achieve.
Current data practices make it especially difficult to assess innovative and online programs. The data we have on each school doesn’t differentiate between the outcomes of a school’s on-campus population and its online population. Because online students are more likely to attend part-time, the quality of the data on these programs are weaker overall. Even for fully online schools, reported outcomes may only reflect the experiences of the small number of students who enrolled as full-time, first-time freshman.
How can we get better information and make better decisions?
The system we have isn’t working. What would a new and improved approach to higher education data look like?
The best system would be one that counted all students, tracked their pathways through higher education (potentially across multiple institutions), gave us program level information, and information on outcomes after college.
This is called a student-level data system. Unfortunately, Congress prohibited the creation of such a system in 2008. Why?
Its fiercest critics come from two camps: privacy hawks and colleges themselves—private colleges in particular. Groups like the ACLU are mainly concerned about the potential for hacking and misuse of data. These are serious concerns. However, in this case, the data in question is largely information that the federal government already has access to through the IRS and other sources. A student-level data system wouldn’t give the federal government more information so much as it would allow the government to connect data that it already has.
On the other hand, a student-level data system has strong support from students. Groups like the Young Invincibles argue that more data is essential for students to make good decisions, and for colleges themselves to be able to innovate against student needs. They advocate for overturning the ban on a student-level data system and collecting data on all students.
Colleges and universities themselves have a mix of views on the issue. Some associations of colleges, like the Association of Public and Land-grant Universities and the American Association of Community Colleges, have come out in favor of a student-level data system. They argue a student-level data system is necessary to provide good data on student outcomes, and that the benefits far outweigh any downsides.
But the trade organization representing small private colleges, the National Association of Independent Colleges and Universities (NAICU), is vocally against such a system. They assert, “The systematic collection of data on individual students poses a serious and substantial risk to student privacy.” NAICU doesn’t make clear what that risk is and doesn’t respond to the arguments from student groups that the benefits of better outcome data are well worth the costs. Part of NAICU’s resistance to better data could be what that data might reveal: clarity around outcomes, particularly graduates’ salary information, could challenge the value proposition for private colleges, which tend to be far more expensive than public schools.
It’s up to the Senate
As Congress looks to reauthorize the HEA, data is likely to be a contentious issue. In the House, Committee on Education and the Workforce Chairwoman Virginia Foxx is an ardent privacy advocate, and the author of the 2008 ban on a student-level data system. The PROSPER Act, approved by the Committee in December, makes some changes to improve higher education data, but firmly forbids a student-level data system.
However, an earlier bill, the College Transparency Act, which proposed a student-level data system, received bipartisan sponsorship in both houses of Congress. Some suspect that the Senate may include the core proposals from the College Transparency Act in their HEA reauthorization bill, forcing the debate to a head.
A student-level data system would be a huge win for students, parents, and taxpayers—but colleges that don’t create value for students would likely wither once their outcome metrics were revealed. A more transparent higher education system would allow families to make better decisions about how they invested in college, and it would give schools an incentive to innovate against the outcomes that matter to students.
This post was originally published on The Christensen Institute’s blog here.