Since its passage in 1974, the Family Educational Rights and Privacy Act (FERPA) has struck a careful and reasonable balance between the privacy of students and families and the need for timely and accurate information on the state of U.S. schools and school systems. But a provision in the FERPA overhaul “discussion draft” currently being circulated by Republican John Kline and Democrat Robert Scott threatens to upset this balance by giving parents the right to “opt out” of data-sharing agreements with “organizations conducting studies for, or on behalf of, educational agencies or institutions,” which are currently exempt from FERPA’s general prohibition on the sharing of personally identifiable information.
As written, this provision would do serious harm to efforts to evaluate and study existing education programs, because its widespread use would degrade the quality of the data on which many evaluations and studies are based. This would be a huge problem, especially if there were significant differences between students whose families chose to opt out and the broader student population (which there almost certainly would be). Such differences could (and likely would) bias the results of future studies that rely on education data, especially those seeking to understand the performance of students over time by linking data from different systems.
This isn’t just about the convenience of academics in universities, think tanks, and research firms. It’s actively menacing to the country’s ability to know things like:
How well are public schools preparing students for college and the workforce?
Why do some young people drop out of school, and what happens to them when they do?
What’s the “return on investment” for education spending, and does it vary by location?
Are existing career and technical education programs getting good results?
Do charter schools outperform traditional district schools?
How effective are school voucher programs?
Which education reforms are working?
Without reliable data on many, many students, answering such questions is nearly impossible. And make no mistake, we need answers! Despite the myriad studies and evaluations that have already been conducted, we still have a great deal to learn about the best way to educate our kids.
Besides being actively harmful, this opt-out provision is unnecessary, as demonstrated by the near- absence of major K–12 education data breaches in recent years. Thanks to the numerous precautions in place under current law and practice, the education data system already does many things right, In most cases, before agencies or schools can share education data with an outside organization, they must first remove all personal identifiers, including the student’s name, the names of family members, the student’s address, Social Security number, and student identification number, and such possible indirect identifiers as date of birth, place of birth, and mother’s maiden name. Furthermore, any outside organization that is entrusted with data that can be linked in any way to a student’s identity must destroy these data after completing its analysis. In short, current data-sharing arrangements do not violate student privacy. (Indeed, from the perspective of quality research and evaluation, they’re awfully restrictive!)
Given the technological changes of recent decades, FERPA may be due for an update. But any rewrite of the law must recognize the vital role of data analysis in a modern education system. Truly understanding what is or isn’t working in education means accounting for the performance of every child (or ensuring that whatever samples we have are representative). Anything less will cloud our understanding of the present and blur our vision for the future.
