As part of an effort to improve the quality of educational research and make it less balkanized, the National Science Foundation and the Institute of Education Sciences have introduced a common set of evidence standards for federally funded work.
The criteria, rolled out last week at the American Educational Research Association’s annual meeting here, will guide all new research at the IES, the U.S. Department of Education’s main research agency, and all NSF research in science, technology, engineering, and mathematics education.
“We’re now in a time of amazing change; we’re all looking at the Common Core State Standards, new assessment techniques,” said AERA President-elect Barbara Schneider, a professor of education and sociology at Michigan State University in East Lansing. “It’s terribly exciting, but this all raises questions and concerns about what kind of research needs to go on, what kind of capacity needs to go on.”
The new guidelines distinguish three interconnected types of education studies:
• Core knowledge, including foundational research, early-stage, and exploratory research;
• Design and development research, to create interventions or strategies; and
• Impact evaluations, including how those interventions work under ideal situations, as well as real-life settings with different groups of students.
“We need stronger evidence provided at a faster pace,” said Janice M. Earle, the NSF’s education evaluations coordinator.
“We’re under more constrained federal resources,” she said, “and that provides a powerful impetus for collaborating and building purposefully on each other’s portfolios and investments.”
For each type of study, researchers must lay out their purpose and how the project contributes to existing evidence, as well as explain how it would further understanding of important policy or practice.
Moreover, researchers would have to be much more concrete about what they expect to produce and how other researchers would be able to build on the results.
The standards also align expectations for research on informal learning, such as in museums, clubs, and after school.
“We wanted to make sure there was a strong sense that all settings were fair game,” Ms. Earle said.
Clearer Goals
The guidelines have been in development for two years at the two agencies, the federal Office of Science and Technology, and the Office of Management and Budget.
But efforts to make education research more coherent across agencies goes back at least as far as the creation of the IES itself, in 2002, when then-Director Grover J. “Russ” Whitehurst sought to overhaul the Education Department’s research agency in the mold of the NSF and the National Institutes of Health, which were seen as having better-quality studies.
Those early efforts also sparked a long-running debate over whether quality necessitates experimental designs and randomized controlled trials alone.
The new joint evidence standards are in part an attempt to resolve that debate, by defining high standards of evidence and usable, relevant results for different types of research, from basic descriptive studies to nationwide evaluations of programs.
The new standards also lay out a more holistic view of how different types of studies can be used together to build a knowledge base.
All federal agencies have come under pressure from the OMB to beef up the quality of their research and program evaluations.
“This has been a pretty significant change for how we do business,” said Nadya Chinoy Dabby, the Education Department’s associate assistant deputy secretary for innovation and improvement.
“Most of our programs across IES and NSF have a ton of different evaluation requirements. Most of them aren’t particularly meaningful, they don’t connect to what you do next, don’t connect to what we do next, and so they sort of sit out there, and that’s appropriate because they’re also not particularly rigorous,” she said.
“But that doesn’t mean that we are systematically building a knowledge base, and it doesn’t mean that at the end of the day we are improving outcomes for our students,” she added.
“The part I think will be most useful is the articulation of expectations,” said Barbara Means, the director of the Center for Technology and Learning at SRI International, a Menlo Park, Calif., research group that has held grants from both agencies. “This is the first time I’ve really seen a clear articulation of what we should see at the end of a project.”
More Competitions?
The joint criteria were inspired in part, said IES Director John Q. Easton, by the tiered-evidence format of the $150 million Investing in Innovation, or i3, grants. The Education Department program, which provides different levels of funding based on the level of evidence on which a project is based, has become “a model across the federal government,” he said.
The joint standards also come as the Obama administration seeks in the president’s fiscal 2014 budget proposal to overhaul the federal government’s “fragmented approach” to STEM education by combining or restructuring 114 of the 220 education programs in that area in 11 of the 13 agencies that house them.
Part of that proposed overhaul would involve consolidating $180 million from the budgets of the IES, the NSF, and the Smithsonian Institution to create aligned programs, including $150 million in new competitive grants, which could raise the profile of the new evidence standards in the same way i3 did for the earlier tiered-evidence model.
It remains to be seen whether there will be any traction on those proposals in Congress. But Edith Gummer, the program director for the NSF’s learning-research division, said the evidence standards are final, and in the coming months, the agencies will publish and integrate them into new research competitions and priorities.
