States Using Education Micro-data in Research
This blog post originally appeared in Growthology on November 18, 2014 and is posted here for posterity and portfolio purposes.
For the last nine years, the National Center for Education Statistics (NCES) has hosted their Summer Conference around July in Washington, D.C.
Over a three day period, this event holds 100 or so unique sessions, sprawling ten conference rooms. Presenters and attendees include a cross section of educational groups that touch data: education technology vendors, education research institutions, state educational authorities, district educational authorities, philanthropic organizations, and not for profit organizations. Presentations vary in length between multiple hour seminars and single hour sessions.
Conferences of this ilk can and do provide helpful information, but for me, the meta-level themes can be just as interesting. The unsaid theme at this year’s conference was the high variance in the access and use of public education data across the country.
For example, representatives from California stated being very open to rigorous proposals for research collaboration with schools, districts, and the public when the data security and intention were a good fit, although they did not define their process to access student level data.
A review of the California Department of Education, Data Access Request website states they are “revising” their intake process, which may be why they did not fully define details at the conference. But however opaque, they did report receiving several research requests daily.
On the more secure side, to access Utah student level data you, as a researcher, would need to:
- Apply to the state,
- Go to a state-secured location and log into a secure computer,
- Access remotely a secure server in a different the secure location,
- Employ a two-factor authentication, and,
- Have any tables created reviewed by staff before release.
The potential solutions to allowing researchers to access data can often be viewed along a spectrum, running from: highly secure but cumbersome practices, to secure but more accessible.
The most secure data, hypothetically, has the most verifications and barriers to use. Utah appears to have chosen security over accessibility; not surprisingly, Utah is not processing a large number of research requests. State leaders at NCES indicated they may be revisiting their choice because of a desire to increase use of their data in research to understand education policies.
The variance in access discussed at NCES does not apply to the PhD researcher alone.
In later posts, I’ll be discussing some of the disparities in public accessibility of education data, standout states, and why I believe that transparency and accessibility of public de-identified data are so important.