Plain Talk with Eric Hanushek: How interpretations of FERPA are inhibiting research

Monday, November 26, 2007

Eric Hanushek

Eric Hanushek — member of the CALDER management team, is leading the study of Texas schools. He is also the Paul and Jean Hanna Senior Fellow at the Hoover Institution of Stanford University.

Read more about Eric Hanushek»

The Family Educational Rights and Privacy Act of 1974 (FERPA) protects the privacy of student education records. The federal law requires schools to get parental permission before releasing student records, with some exceptions. But confusion over how to apply the law has led to inconsistent policies and to concerns that misinterpretations will lock away valuable data and shut down research.

Why does the interpretation and practice of FERPA differ from state to state?

FERPA is a federal statute that was developed in an era long, long ago. Individual states have gone to their own lawyers to interpret the statutes, which leads to inconsistencies. Some states do everything they can to keep student records confidential, but also to allow for research. Other states have become increasingly stringent in their interpretation.

As for why we have variations, it's partly because some states are not very eager to do research on their schools and student performance. FERPA offers a handy excuse for not releasing data. There are some outrageous examples of states that won't release aggregate data even though there is no way the release violates the privacy of individual students.

None of the research that I do would ever come close to releasing identifiable information. Once you've done the statistical analysis, which is what's reported, it's impossible to recover any information about individuals. And there has never been an incident where privacy was violated by anyone using these data.

How have misinterpretations of FERPA affected your research?

They have made it extraordinarily difficult for us, particularly in Texas, because we can't get recent data on student programs. For a recent analysis of charter schools in Texas, we were unable to access data after 2002, which severely hindered our research. Texas and many other states are heatedly debating the role of charter schools and what their academic advantage is, and yet it's not possible to provide the analysis needed to decide the question.

Why are these conflicts coming up now?

It's only in the past few years that large state administrative databases have become readily available, partly because of state accountability rules and partly because of No Child Left Behind. States now have a wealth of data to use to understand why some schools work and why some schools don't. The federal government has given $115 million in grants to states to develop their longitudinal databases. These funds will be completely wasted if nobody can ever use the data.

No one wants to compromise confidentiality. It's a matter of sensibly interpreting current law within the context of a particular situation. The law was written 30 years ago, when nobody even dreamed about having regular data on student performance that could help improve school quality.

What is the solution?

The federal government needs to provide clearer guidance about FERPA and clarify what is permissible. And it's a matter of some urgency that this happens soon.

Right now, states and schools are under a lot of pressure to perform well because of No Child Left Behind. It seems only natural that schools and districts that aren't performing well would turn to the state and federal government for help. Unfortunately, the available research knowledge at the state and federal level on helping failing schools is remarkably thin. But we could use the state databases to provide much better answers as to why some schools seem to be doing well with, for example, disadvantaged populations, and why some aren't. We have little hope of actually reaching proficiency standards by 2014 if we can't provide better guidance to schools about how to do it.