Using Data to Guide Education Policymaking
Louisiana learns to use information about its students to create real-life benefits for them.
One of the most contentious topics in the public sector these days -- at local, state and federal levels alike -- focuses on K-12 education. Many agree that something must be done. You would have to be a hermit to avoid running across a depressing multi-hued chart that shows the U.S. lagging other nations on this front. When it comes to figuring out what actions should be taken -- and how we can best evaluate their worth, success and failure -- the academic fur begins to fly.
One potent trend has emerged: More educational data is being gathered now than ever before. This is partly driven by $250 million in American Recovery and Reinvestment Act funding and reforms pushed by the Obama administration. It's also been spurred by improvements in technology.
But data by itself gets you nowhere. Questions about how this information should be put to use are the ones begging for good answers. It's our tendency in this column to stay away from topics for which we can’t provide time-tested and proven models. Given the critical nature of this issue, however, it seemed more than worthwhile to look at the work that’s being done in Louisiana, a state that's clearly further along than many others in using educational data and offers at least a hint of benefits to come.
Louisiana's education system began using individual student identifiers back in the mid-1990s. The point was to track student progress as boys and girls moved through the education system. Having the data identifiers was a great start, but for some time it was really tough to make use of this material. "I literally spent thousands of man hours on data cleanup, data merging, data organizing," says George Noell, executive director of the Office of Strategic Research and Analysis for the Louisiana Department of Education. What Noell and other researchers were finding is that they could get tons of snapshots about individual years, but they couldn’t see the moving picture -- which would require easily tracking the information they wanted across years.
But the times and technology have changed. With better use of new tools, data now is collected and stored in a way that makes access much easier. Perhaps more important, computers now allow the information to be accessed in close to real time. Faculty can use data about this year's ninth graders to help students while they're still in ninth grade. In the past, information about the ninth-grade class could not be accessed in a timely fashion, which meant it could only be used to help inform decisions about future ninth graders.
Some readers are doubtless waiting for us to get to the question of standardized tests -- one of the most vigorously debated topics in education. But we’re going to skip right over that.
Louisiana’s data goes far beyond big standardized tests: It is focusing on attendance data, capturing information on a weekly basis so educators can look for shifts in attendance patterns at the district, school and individual levels. When unexcused school absences accelerate, administrators can deal with them right away.
This is important, says Noell, because "dropping out is not an event, it’s a journey." Statistically, it’s apparent that the journey leading up to that catastrophic life event is an increasing rate of absences. With attendance data, Louisiana can target high-risk students before they become a dropout statistic. (They can also blend information to focus on individuals who most fit the dropout portrait -- students who are behind their age group in school, for instance.)
Louisiana also gets immediate feedback on new programs it introduces. Consider the Ninth Grade Academy, a kind of school within a school that allows students to interact in smaller high school communities with more direct teacher attention. The theory is that students will be more successful in their high school careers if they get more help in the transition to high school. But as with most programs, success is often determined by how the program is implemented, and that varies from school to school. Louisiana’s data systems have enabled it to look for schools that seem to have better results on this initiative and those that aren’t doing as well. Educators can then investigate what the differences are.
The data also points more quickly to extant programs that aren’t successful. For example, the Options Program for ninth graders helped students work toward a GED diploma. A review of the longitudinal data demonstrates that it hasn’t worked well, and a working group is in place to figure out how to change or replace it.
Louisiana’s intense use of educational data has come with numerous lessons. One key lesson is that data has a price, so you need to know the reason you’re collecting information and who will be able to use it. Another is that you must understand what data systems can and can't do, and who they will benefit. "It’s not all going to be relevant to teachers and some that's relevant to teachers won't be relevant to policymakers," says Noell. "There are definitely good people who get this, but certainly our popular discourse at the federal level is not getting that different consumers need different levels of granularity in the data. The same data won't be useful to everybody."