The Problems With Huge Data For Postsecondary
By Dr. Watson Scott Swail, President & Senior Research Scientist, Educational Policy Institute
There seems to be renewed interest in the collection of student unit record in higher education. Back in 2008 during the reauthorization of the Higher Education Act (HEA), a provision was added to the law that restricted the federal government from collecting student-based unit record data. This followed a multi-year dialogue about how to improve the Integrated Postsecondary Education Data System, better known as IPEDS, which collects massive amounts of data from every Title IV institution (i.e., institutions that are eligible to deliver federal financial aid). Data collected include student enrollment, retention, graduation, and even other areas such as institutional finances.
However, there was a huge revolt against the idea that the federal government would collect such private information about college students. The worry being that student unit record data, including student IDs, names, and addresses, would be poached by some non-government entity and used in a harmful way (like, Target would be in charge of the data!). For us policy wonks, we generally thought that this was kind of stupid on several accounts, the first being that there is enough encryption security firewall technology to band against this, and second, because the feds have individual data anyway through both the IRS and student financial aid systems. But let’s not quibble. The HEA language that followed was written by congressional devolutionists to ensure that the feds couldn’t have their hands on individual student records.Fast forward six years and much has changed, due in large part to President Obama’s call for a College Ratings System. Now there exists a renewed sense of what the federal government should be able to do, with one problem: they made it illegal to collect unit record data in 2008.
Last week, the US Department of Education held a one-day symposium on the technical issues related to the College Ratings System. When asked about how best to deal with some of these issues, many experts, including Virginia’s Tod Massa, stated that the federal government will require better data if they want to inform a federal rating system. In Massa’s words, “To the department, I say this: We need better data. Let me rephrase that: You need better data.”
Currently, IPEDS requires institutions to provide aggregate data on retention and completion (on time, 150, and 200 percent of time) for first-time, full-time students. It does not ask about part-time students and it doesn’t tell us anything about what happens with students who do not graduate or are possibly still enrolled. This is a considerable problem when many schools graduate less than 30 percent of their students. That leaves two thirds adrift in the data and adrift in public policy.
Several states have created data warehouses that track every student from kindergarten to and through college. Florida, for one, has one of the most comprehensive data warehouses in the nation and can track students who transfer to another institution. But even for states like Florida, the long-arm of the data system has its limits at the border. They are not able to track students who transfer to institutions in other states. The backdoor way of doing that now is through the non-profit National Student Clearinghouse (NSC), which tracks approximately 90 percent of higher education students. So, the state or institution can run a query through the NSC’s Student Tracker to find out if a particular student is enrolled at another out-of-state institution and if they graduated. The downside is that the variables available for query are limited. As well, most students, but not all, are trackable in the system. So the NSC works well, but is not a perfect solution.
The College Ratings System dialogue is a door opener for the US Department of Education to get the data they really want. To get at college ratings they need better data. To get better data, they have to change the federal law. Many members of congress, especially Republicans, are against this type of “federal intervention.”
However, on February 9, 2012, Senators Ron Wyden (D-OR) and Marc Rubio (R-FL) introduced the “Student Right to Know Before You Go” Act, a piece of legislation designed to build upon 1990s Student Right to Know Act, the original law requiring Title IV institutions to provide retention and graduation rates. The new legislation would require “more accurate and complete” data on these issues, including salary information of graduates. Three months later, to the day, Republican Duncan Hunter introduced a parallel bill in the House for similar legislation. When asked why now, Hunter said, “privacy is easier now. We’re at the point where you could have this information and aggregate it together with no privacy issues whatsoever.”
Both bills have sat in their respective committees. The impending presidential election in fall of 2012, followed by a record-low level of accomplishment by the 113th Congress (2013-14), kept this and many other bills from discussion or vote.
While there exist arguments against the creation of a federal unit record data system, there is no good argument that supplants reason. For the creation of prudent public policy, we need better data, which starts with a student unit-record data system that encompasses not only higher education, but starts at Pre-K and continues throughout the education continuum and into the workforce. The technology is available; we have a historically-proven ability to collect these data (e.g., IRS); and perhaps most importantly, it can happen without creating undue burdens on the institutions of higher education. In the end, the institutions are perhaps the greatest stakeholder in this matter, because they should want to and need to know what happens to the students that choose to leave their institutions.
Many questions remain, of course, including what happens with the National Student Clearinghouse? Who controls the data? And what variables and data are collected? But these are manageable issues that can be resolved with conferencing
It’s time to put politics and disingenuous diatribes against the development of a comprehensive, student-unit record data system aside. We need better policy, and better policy is only achievable through better data.
Educational Policy Institute
3172 Indian Plantation Drive
Virginia Beach Virginia 23456