69传媒

It can be tough to translate evidence into action in education research.

A principal or superintendent might sift through academic journals or vendors鈥� pamphlets for an effective reading program, but even a seal of approval from the federal What Works Clearinghouse is no guarantee that what helped students in one district will be successful with another.

To better inform that knowledge base, the Institute of Education Sciences is crafting a new research program, called 鈥渃ontinuous improvement research in education,鈥� to go beyond 鈥渨hat works鈥� and add more context to education findings.

鈥淜nowing what works plays a very important role in school improvement, but alone it鈥檚 not enough,鈥� said John Q. Easton, the director of the IES, the research arm of the U.S. Department of Education. 鈥淭here are questions about building the capacity to implement what works, building the capacity to measure, check, and adapt to changes.鈥�

The initiative, expected to be launched in 2014, would award four-year grants of up to $1.5 million each, though there鈥檚 no word yet on how many would be awarded. The IES wants researchers to focus on supportive school climates, high school transitions, or access to postsecondary education.

The initiative would build in part on a slew of new research models. They include"design-based implementation research"鈥攂eing developed at Vanderbilt University鈥檚 National Center on Scaling Up Effective 69传媒, in Nashville, Tenn., and SRI International鈥檚 Center for Technology in Learning, in Menlo Park, Calif.鈥攁nd the 鈥渞apid prototyping鈥� models tested at the Carnegie Foundation for the Advancement of Teaching, in Stanford, Calif., and the Institute for Healthcare Improvement, in Cambridge, Mass.

In both models, researchers and districts work together over years to identify and test solutions for specific problems.

鈥業terative鈥� Research

"[In] a typical research project, here鈥檚 my research question; I鈥檓 going to answer my research question. This is much more iterative, developed collaboratively, and you鈥檙e not sure how it鈥檚 going to go or what the time frame is,鈥� said Bridget T. Long, a Harvard University education economist and the president of the National Board for Education Sciences, which advises the IES and discussed the proposed initiative at a meeting here in Washington on Oct. 5.

Mr. Easton said the initiative would focus on cycles of improvement, in which researchers develop, test, and tweak interventions in the classroom or school context. (鈥淩esearchers Seek Faster Answers to Innovation Questions,鈥� Jan. 12, 2011.)

For example, NBES member Hirokazu Yoshikawa, the academic dean at the Harvard Graduate School of Education, said he used rapid prototyping, which includes 90-day, intervention-testing cycles, to adapt a teacher professional-development program in Chile.

The method was, Mr. Yoshikawa said, 鈥渟pectacularly successful in achieving buy-in and local ownership, but also, from a scientific standpoint, getting to a part of the science that I don鈥檛 think we鈥檇 ever gotten to, which is how do you track the day-to-day or week-to-week improvement process?鈥�

鈥淚t creates local tests of change,鈥� he said, 鈥渁nd it鈥檚 the practitioners who develop those methods themselves.鈥�

The projects would also focus on systemwide interventions, 鈥渆xamining how components of systems work together to generate desired outcomes,鈥� according to the draft request for proposals.

Prior federal research has pointed to local context as a sticking point for scaling up successful educational interventions. Recent federal longitudinal studies of school improvement found district policies and supports can mean the difference between schools that turn around and those that struggle.

Careful Collaboration

Anthony S. Bryk, an NBES member and the president of the Carnegie Foundation for the Advancement of Teaching, likened continuous-improvement research to the Toyota auto company鈥檚 famed 鈥渢otal quality management鈥� system, in which any worker can stop the factory line if he spots a problem.

鈥淲hat is the problem we鈥檙e trying to solve, what and why is the change we鈥檙e putting in place, and how will we know if that change is an improvement?鈥� Mr. Bryk said, describing the approach continuous-improvement research takes.

鈥淭hat鈥檚 a very different way of thinking about the work,鈥� he said. 鈥淚t鈥檚 seeing improvement as a learning journey, rather than the way most districts see it as, 鈥榃ell, we have this new idea, I have to roll it out fast, implement it at scale, and most of my attention is on crisis intervention.鈥� 鈥�

Barbara Means, the director of SRI International, cautioned that the federal research agency will have to structure the grants to force educators and researchers to be equal collaborators, in which 鈥渘either can totally steer the ship, but both must be accountable.鈥�

The IES has already been building up the supply of research partnerships that might be capable of taking up the work. The new network of federally financed regional educational laboratories has developed some 70 research alliances involving states, districts, and researchers.

鈥淲hat鈥檚 become really clear is this is really hard work,鈥� said Ruth C. Neild, the commissioner of the National Center on Education Evaluation, part of the IES. 鈥淓ngaging over a sustained time is harder for researchers, harder for districts and states, but we think the rewards will be greater.鈥�

If successful, said NBES member Robert Granger, the president of the New York City-based William T. Grant Foundation, the initiative could produce a new gold-standard model for education research, akin to the IES鈥� original focus on randomized controlled trials, but 鈥渨here the gold standard isn鈥檛 to run a trial; the gold standard is seeing consistent results across a number of conditions.鈥�