Imagine the research possibilities if every student in the country carried a “virtual backpack” stuffed with statistics on his or her entire educational history.
The data, traveling with students as they moved from school to school, could be used to update parents on their children’s learning progress, register students in school, or import information when they moved to a new city or entered college.
Educators and researchers could plumb the backpacks for data to figure out how to target instruction more effectively or to determine which educational interventions work and which don’t.
Data backpacks are one facet of a futuristic vision for education data sketched out in , a new book published by the Thomas B. Fordham Institute, an advocacy think tank based here in the nation’s capital.
.
Released at a conference hosted by the institute, the book lays out the ways in which district, state, and national data systems fall short in providing the kind of data that educators and policymakers need. It also examines the barriers that stand in the way of improving those systems, and offers options on how educators could make those systems better and more useful.
“I really need data, and our research has shown that the quality of state data in many systems in which we’ve done work is pretty abysmal,” said Margaret Raymond, the director of the Center for Research on Education Outcomes at Stanford University and a proponent of the “virtual backpack” idea.
Progress in States
Still, states have come a long way in recent years, other conference panelists noted. State officials have been helped by federal grant programs aimed at bolstering their data-collection efforts and prodded by reporting requirements in the federal No Child Left Behind Act.
Just three years ago, according to the Austin, Texas-based Data Quality Campaign, no state had all the ingredients in place for an effective statewide system of longitudinal data collection.
Now, a that the group released at the Nov. 17 Fordham conference shows, six states have such systems in place, and 42 more are halfway there.
.
But state legislators also warn that the nation’s economic crisis could put a crimp on further progress.
“All this will be for naught,” said Aimee R. Guidera, the director of the campaign, “unless states can make the change from building information systems to really start thinking about what they want to do with all this information.”
One challenge is figuring out whom those data systems will serve, said Paul Manna, an assistant professor of government at the College of William and Mary, in Williamsburg, Va.
“Lots of people are interested in education data, but they’re not always interested in it for the same reasons,” he said.
Student-achievement data might be used, for example, to give teachers feedback on which students are struggling, to track schools’ compliance with federal and state education laws, or to decide which teachers deserve a pay raise. Different uses call for different kinds of data, Mr. Manna said.
Privacy Issues
Researchers looking to tap into troves of data piling up in states have also run into problems stemming from state officials’ differing interpretations of the Family Educational Rights and Privacy Act, a federal law aimed at protecting the privacy of students’ educational records. To ease the problem, the U.S. Department of Education in March proposed new rules clarifying the scope of the 1974 law. (“Groups Urge That FERPA Rules Give Researchers Access to Data,” May 21, 2008.)
But Chrys Dougherty, a senior research scientist at ACT Inc., the Iowa City, Iowa-based testing firm, and the National Center for Educational Accountability in Austin, Texas, said the regulations, if approved, would solve only “80 percent of the problem.”
.
“The bigger problem is convincing policymakers that the benefits outweigh the privacy risks,” he said.
At the federal level, Mark S. Schneider, who stepped down in October as the commissioner of the U.S. Department of Education’s top data-crunching agency, the National Center for Education Statistics, said inertia often cripples efforts to sharpen federal statistics-gathering efforts.
“Every question on a survey has a friend, and that friend has a friend who’s on the Hill,” Mr. Schneider, currently a vice president at the American Institutes for Research here, told the gathering.
He also noted that states and districts often have problems collecting basic information, such as student attendance or the number of computers in a school building.
One way the field could improve its data efforts, said Bryan C. Hassel, a chapter author and a co-director of Public Impact, a consulting firm based in Chapel Hill, N.C., might be to take a cue from the business sector and embed data-mining mechanisms in students’ and educators’ everyday activities.
On the Web, for instance, companies routinely gather useful marketing information by analyzing consumers’ “click trails,” tracking their browsing and buying habits when they make online purchases.
Likewise, Mr. Hassel said, Wal-Mart Stores Inc., the giant Bentonville, Ark.-based merchandising chain, studied prehurricane sales at its brick-and-mortar stores and discovered that, besides making expected purchases for items such as bottled water, shoppers stock up on Pop-Tarts before a hurricane strikes.
“The problem in education is not that we don’t have these sorts of daily experiences,” he said, “it’s that we don’t capture that information in any way.”
In the final chapter of A Byte at the Apple, Chester E. Finn Jr., the Fordham Institute’s president, outlines a long list of other possibilities for a more productive future for statistics-gathering efforts.
.
Pegged to the year 2025, his vision includes, in addition to Ms. Raymond’s virtual backpack, “swipe-able” student-identification cards that track attendance; periodic reports to parents that project whether their children are on track to graduate on time; computer-graded formative assessments; and cellphones or other hand-held electronic devices that students carry from class to class, allowing researchers to analyze how the activities pupils undertake during the school day contribute to their achievement.
Right Data Points?
For some conference-goers, what was missing from the discussion at the Fordham gathering was some attention to the quality of the data that newer, more sophisticated data systems could gather.
“I’m deeply concerned that having the wrong metric is worse than having no metric at all,” said Tony Wagner, a co-director of the Change Leadership Group at Harvard University. “Are we using the right data now?”
Several panelists conceded that the answer to that question was still no.
But, noted John E. Deasy, who recently left his job as the superintendent of schools in Prince George’s County, Md., to become the deputy director of the education division for the Seattle-based Bill & Melinda Gates Foundation, “there is value in saying whether students are in school or not, even if it’s not a perfect data point.”
