A national research-advisory board last week approved guidelines to govern the outside review processes that the Department of Education uses to screen and approve the studies and reports that it funds.
Taking a cue from federal science agencies, such as the National Institutes of Health and the National Science Agency, the guidelines call for standing panels of reviewers to screen many of the research proposals funded by the Education Department鈥檚 primary research arm, the Institute of Education Sciences.
They also formalize a special office within the institute that functions almost like the editor of a professional, peer-reviewed journal and coordinates reviews for most of the reports published with the Education Department鈥檚 imprimatur.
鈥淭his is a scientific review process that is truly independent of staff,鈥� said advisory- board member Jon Baron, the executive director of the Washington-based Council for Excellence in Government. 鈥淚t goes a long way toward establishing the institute鈥檚 credibility as an independent, nonideological source of authority on education research.鈥�
Critics have long said the Education Department鈥檚 peer-review procedures are partly to blame for the low quality of federally funded education research. A department-commissioned report in 1999, for instance, said that such procedures were sometimes haphazard and that reviewers often lacked expertise in research methodology.
When the institute was created in 2002, federal officials began to set in place more-structured procedures for handling outside reviews. But, under federal law, those procedures have to be formally approved by the National Board of Education Sciences, a 14-member board whose members are nominated by the president. At its meeting last week, the board essentially approved most of the mechanisms already in place, adding some small changes of its own.
鈥淲hat鈥檚 new here is the process of making these procedures transparent and available to the public,鈥� said Gerald E. Sroufe, the director of government relations for the American Educational Research Association, a Washington-based group that represents 22,000 researchers.
Two Review Tracks
The guidelines approved on Jan. 24 set up procedures for two kinds of research-review processes鈥攐ne for deciding which research proposals the department will fund, and one for signing off on reports on federally financed education studies.
Spanning about a year, the process for research proposals begins in March, when the department announces its research competitions. Panels meet to screen proposals twice a year, once in October or November and again in February or March.
Proposals are reviewed by either a standing peer-review panel or a single-session panel. Typically composed of 20 reviewers, the standing panels vet proposals for all the studies that fall under a particular topic area, such as reading or mathematics education. Serving anywhere from one to three years, most of the members of those panels have to be approved by the institute鈥檚 director, Grover J. 鈥淩uss鈥� Whitehurst.
鈥淏asically, we鈥檙e looking for highly qualified experts who are at least as good as the [principal investigators] who are applying,鈥� said Mr. Whitehurst.
However, one panel member questioned whether reviewers should include experts in the content area as well as researchers.
鈥淚f the report or the proposal is about math, then there should be mathematicians,鈥� said R. James Milgram, a mathematics professor at Stanford University. He said he and other professional mathematicians, in reviews of state and federal math assessments, have found mathematical errors in 20 percent to 25 percent of the questions.
But board members put off that issue until the panel meets again in May.
At members鈥� urging, the guidelines were amended to specify a process for 鈥渋nviting鈥� reviewers to disclose any methodological or ideological biases they have at the point when the review panels convene鈥攁 policy that other federal science agencies follow.
For reports produced by the institute鈥檚 four research centers, the guidelines call for two stages of review. The first stage occurs at the research centers themselves; the second takes place in the institute鈥檚 office of standards and review, under the newly approved procedures. Officials in that office either have the reports reviewed by staff or recruit outside experts鈥攖wo or more for each report鈥攖o critique them.
