Over the objections of dozens of researchers, the Department of Education plans to expand its push for “scientifically based” education research to all its programs.
Until now, the emphasis on such experiments has been limited mostly to reading programs, efforts funded under the No Child Left Behind Act, or studies begun under the department’s major research arm, the Institute of Education Sciences.
But the new research “priority,” which was published in the Federal Register on Jan. 25, expands the push for scientifically based research to a wider range of programs, including special education and rural education.
The push is part of a long-running effort by the department under President Bush to transform education into an evidence-based practice more akin to medicine. But the call for “scientifically based” research has also been controversial among researchers because some say it favors some scientific methods, such as randomized experiments, over others.
A draft of the department’s proposal, published in November, drew nearly 300 comments, most of them critical of the change. Commenters raised concerns about whether randomized controlled trials, which involve randomly assigning subjects to either treatment or control groups, are ethical or whether they are appropriate for answering most of the messy, complex questions that confront educators.
“It’s not like you can randomly assign kids or schools or teachers to classrooms. Most school boards are horrified at the idea,” said Catherine A. Emihovich, the dean of the education school at the University of Florida in Gainesville.
Some Flexibility
For its part, though, the Education Department says in its Federal Register notice that it has no plans to exclude other valuable forms of research. However, the notice adds, “the secretary considers random assignment and quasi-experimental designs to be the most rigorous methods to address the question of program effectiveness.”
It notes that department-funded studies might have to use different methods to study low-incidence special education populations, for instance, or for programs in rural areas where there may be too few subjects to conduct a full-fledged randomized experiment.
The final research priority, which is little changed from the draft version, is to take effect on Feb. 24. But offices within the Education Department will have some leeway to decide whether to use it, the notice says.
“The devil in all of this is in the details,” said James W. Kohlmoos, the president of the National Education Knowledge Industry Association, a Washington-based group that represents research organizations. “In deciding which programs this should apply to, will politics come into play? This could be applied to programs that are less favorable to the [Bush] administration’s overall positioning.”