In Brief

In recent years, the social and behavioral sciences have generated a huge body of research examining interventions' effectiveness, from preschool education to criminal justice. But such investigations, while widely welcomed, have presented a problem: how to organize and disseminate what researchers have learned about what works and what does not work, so that lay people, practitioners and policy-makers can make informed decisions.

A multidisciplinary group of about 150 scientists from more than a dozen countries took an important step in that direction in February, at the annual meeting of the Campbell Collaboration. The effort, more than a year in the making, aims to produce rigorous electronically available reviews of research on "best practices" in the social and behavioral sciences and education.

"The volume of research out there is increasing dramatically worldwide," remarks Robert Boruch, PhD, director of the University of Pennsylvania's Center for Research and Evaluation of Social Policy and chair of the collaboration steering group. If findings are to be responsibly applied to practice and public policy, he says, it's crucial to establish a respected means of keeping track of that research.

The Campbell Collaboration--named for the late social psychologist and methodologist Donald T. Campbell, PhD--will solicit systematic reviews of research on social and behavioral interventions in three broad domains: education, crime and justice, and social work and social welfare. The reviews will emphasize studies that randomly assign participants to treatment or control conditions and will be available in an electronic database. Reviewers will update their evaluations as needed, responding to critiques and incorporating emerging research.

A methods group will provide methodological and statistical advice, training and support to the substantive review groups and will conduct empirical research on the implications of different methodological and statistical approaches to reviewing studies.

A dissemination and technology group will provide information about why the Campbell Collaboration was developed. That group will also work with the methods group and the substantive review groups to develop and maintain a dynamic Web presence and will help integrate the needs of those who will use systematic reviews--for example, policy-makers and intervention programs.

Of the three substantive review groups so far established, the crime and justice review group has come out of the gate most quickly. It has already identified 15 review priorities, on topics ranging from the effectiveness of street lighting in preventing crime to the benefits of juvenile curfews. The group has received proposals for more than half of these reviews, says steering group member Joan McCord, PhD, a professor of criminal justice at Temple University.

University of Chicago psychologist Larry Hedges, PhD, who co-convened one of the methods subgroups, stresses that Campbell syntheses of research on interventions will be dynamic and interactive.

"It's not a matter of delivering experts' views of what the evidence means and shutting everybody else up," Hedges says. "Instead, this is about opening up a forum in which criticisms can be made and addressed. We're finally in a position to begin to think about social policy formation on the basis of research evidence, as opposed to convention or ideological perspectives."

--S. CARPENTER