Corrected: An earlier version of this story misspelled Mr. Doug Kosty’s name.
Spurred by the prospect of grant funding from the federal Race to the Top competition, as well as money made available from other parts of the American Recovery and Reinvestment Act, more states are taking strides to link K-12 data systems with postsecondary agencies to develop a more holistic picture of each student.
But despite the emphasis among policy leaders on K-20 data systems, the progress that states have made toward that goal varies greatly, and many challenges—both technical and political—remain.
Although the idea of linking precollegiate data with postsecondary databases has been around for years, the difference is “it is no longer something that seems to be a good idea to people just in education,” says Ben Passmore, the director of policy research for the Adelphi, Md.-based University System of Maryland, which represents 11 universities, two research institutions, and two regional higher education centers in that state.
The federal government has also put a heavy emphasis on beefing up statewide longitudinal data systems, he says. “This is something that’s going to happen, and it’s happened at a really breakneck pace over the past year.”
According to a 2009 survey by the Washington-based Data Quality Campaign, which encourages state policymakers to improve the use and availability of education data, 32 states have the ability to match student-level K-12 and higher education data, although most states, the survey found, do not have data systems that allow for two-way communication between the databases.
One essential step toward linking K-12 and postsecondary data is establishing a unique student-identification number that can help follow each individual student through the K-20 continuum, spanning kindergarten to graduate education, says Adam Levinson, the director of policy and strategic planning for the North Carolina Department of Public Instruction.
“The unique ID is the cornerstone of the longitudinal-data systems,” he says.
In fact, one of the challenges of linking K-12 with postsecondary data is figuring out how to track students accurately from one agency to the next. In many states, privacy laws prevent K-12 databases from storing students’ Social Security numbers, says Baron Rodriguez, the director of state data systems for the Data Quality Campaign. He points out, however, that most colleges and universities typically use those numbers to track students.
“In a way, the universities have an advantage, where the state [departments of education] may be hindered,” he says.
Establishing a unique statewide student-identifier program is the only way to address the challenge, says Rodriguez.
And once that student identifier is in place, it becomes possible to track students, as individuals, through the education system, says Bob Kieran, the assistant vice chancellor for research and planning for the Oregon University System.
Right now, in Oregon, student information is shared “through an ad hoc method,” he says. Researchers can conduct studies that determine how well students are doing in their first year of college as a group, for instance, but not on an individual basis.
“The data is not as specific as we need to draw conclusions about curriculum,” Kieran says. It’s hard to tell which classes make students more successful as college freshmen, for example, or which K-12 schools or teachers are most skilled at preparing students for college.
Linking K-12 and postsecondary data also provides the opportunity to track how effective certain teacher-training programs are, says Levinson, from North Carolina. With improved data systems, you can “look at the success of teachers who have been prepared by the institutes of higher education, who are now in classrooms in K-12,” he says.
But those kinds of conclusions are exactly what can make linking data systems so controversial, cautions Doug Kosty, the director of information services for the Oregon Department of Education. Using data alone to evaluate teacher-preparation programs or teacher performance can be seen as threatening because of the many factors that play into a student’s performance. “You need to focus on more than just test scores as ways to evaluate teachers and measure teacher effectiveness,” says Kosty.
Who Owns the Data?
And teachers aren’t the only ones who have to buy in to a data system partnership. One of the most challenging aspects of linking K-12 with postsecondary data is getting all parties involved to agree on how it should be done, says Passmore, from the University System of Maryland.
“When you start getting those people in a room and try to get them to agree on these things, it’s an extremely difficult task,” he says. “We have to come up with some kind of system that allows everyone to benefit, ... and that’s a tough tightrope to walk.”
Without a structure in place to make those decisions, combining data systems is impossible, says Levinson. “Without a really strong collaborative structure set up—and not just one where the spirit is collaborative, but one where you have clear governance for decisionmaking—you can’t do it really without that,” he says. “That’s the first big hurdle.”
Officials in Florida, which has been a leader in data collection and now has a pre-K-20 data warehouse in place, attribute some of the state’s success to the fact that, at one time, K-12 and postsecondary institutions were lumped into a K-20 agency. Though that structure has since been dismantled, its impact is still being felt.
“I think it’s key, from our perspective, that we had all the players around the table,” says Jeff Sellers, the deputy commissioner of accountability, research, and measurement for the Florida Department of Education. “All the major players were able to contribute and see the value that this kind of an offering brings.”
Nancy S. Shapiro, the associate vice chancellor for academic affairs for the University System of Maryland P-20 Partnership, says that before a partnership can work, K-12 and postsecondary institutions must hash out what are essentially political questions, such as: Who owns the data? Who can query the data? What kinds of questions can we ask of the data?
‘Complex Technical Nuances’
In addition to political questions, states face technical obstacles in linking data throughout their education systems.
“The limitations of technology, and the basic structures of the databases that have been created, have created challenges,” says Bruce Vandal, the director of the postsecondary education and workforce institute for the Denver-based Education Commission of the States. “Short of rebuilding your technology from the ground up, there are definitely hurdles in terms of getting data systems to speak to one another,” he says.
For example, in North Carolina, each entity—the university system, K-12, and the private colleges—have different data systems.
It’s possible to “enhance your existing database structure to be something more flexible,” says Levinson, and fortunately “there’s enough common and shared structure” for the data systems in that state to link together.
In addition, technical challenges can arise when figuring out how to exchange the data, says Levinson, but his state has decided to share data through a Web-based exchange that is compatible with each data system.
Part of the challenge, says Rodriguez, from the Data Quality Campaign, is that “there isn’t much overlap in the vendors” who provide data services to education agencies, causing “complex technical nuances” between each different system.
Still, because of the benefits of linking K-12 and higher education data, there has been a push to make systems interoperable, says Rodriguez.
Establishing a set of common data standards, regardless of the organization or agency, will lay the groundwork for interoperability, says Rodriguez, and that, in turn, will make it much easier for information to be shared.
A version of this article appeared in the January 06, 2010 edition of Education Week