Federal Class-Size Reports Do an About-Face
Just last month, the Department of Education released a report stating that the "consensus of research indicates that class-size reduction in the early grades leads to higher student achievement."
Ten years earlier, the same department--under different leadership--wrote that "the cost of reducing average class size by even a few students is very large and, of itself, the measure is not likely to enhance school outcomes."
The divergent messages raise the question of whether the public can rely on the federal agency to provide objective reports reflecting current debates in education research. Critics say the public can't.
The latest report "is a political document to backfill a policy that [the Clinton administration] proposed, but [that] doesn't have much support," contended Eric A. Hanushek, a professor of economics and public policy at the University of Rochester and a persistent critic of class-size-reduction policies.
Marshall S. Smith, the acting deputy secretary of education, countered that research in the past decade has documented successful experiments with class-size reduction and that the new report simply reflects those findings.
Others say the most important factor in any Education Department report is who's in charge of the agency. In 1988, the department was part of a Republican administration that actively sought to curtail federal education spending. Now, it's under the control of Democrats who are pushing President Clinton's plans to spend $12 billion subsidizing the salaries of 100,000 new teachers over the next seven years.
Some skeptics of the positive impact of class-size reduction are not disturbed by the Education Department's new study.
Releasing such reports "is a very appropriate thing for [the department] to do," said Douglas E. Mitchell, an education professor at the University of California, Riverside, who is not convinced that class-size reduction spurs student achievement. "It's in the nature of things that overdrawn statements will be made. I don't expect them to publish something that's so bland that it doesn't support anything."
Democrats complained in the late 1980s that the department, under President Reagan, tilted its research to reflect the conservative agenda of then-Secretary of Education William J. Bennett and his assistant secretary for research, Chester E. Finn Jr. Saying it wanted to take politics out of research, the Democrat-led Congress in 1994 created a nonpartisan policy board to oversee the department's office of educational research and improvement.
Since the latest class-size report includes no original research, it is not a product of the OERI. Instead, the synthesis of recent reports was published and distributed by the department's leadership, spurring questions about its credibility.
Indeed, the department put out the most recent report six days before Secretary of Education Richard W. Riley held a May 14 press conference with congressional Democrats heralding the formal introduction of legislation to enact the president's proposal. Mr. Riley's aides distributed the 17-page report called "Reducing Class Size: What Do We Know?" to members of the press.
Despite the report's unambiguous conclusion, Mr. Hanushek said, it cites several studies that differ from its thesis. But the contrary studies--including one of Mr. Hanushek's--are raised and then discredited by other reports that the department cites that favor class-size reduction. The rebuttals to pro-class-size-reduction research raised by Mr. Hanushek and others aren't mentioned in the department report.
Mr. Smith, the department's No. 2 official, said that the May 8 report is a fair sample of what researchers have discovered in the past 10 years. The most significant evidence supporting President Clinton's proposal, he said, was gathered in Tennessee's Project STAR. The longitudinal study has found students' test scores increased after being in small K-3 classes. The benefits tended to stay with them, later studies have found. ("Less Is More," July 12, 1995.)
Extensive reviews of Project STAR and separate studies have supported the belief that class-size reduction benefits students, Mr. Smith said.
"They've all seem to come out in the same direction...suggesting that there's something there," he said. "Had that research come out a different way, the report would have been put out" with those details in it.
Mr. Hanushek and other critics remain unconvinced.
"Project STAR is unable to say what it is teachers did differently in the small classes," said Tommy M. Tomlinson, who wrote the 1988 federal report and worked at the OERI until he retired in 1994. "It's sort of magic. They don't know what they did differently" and act as if the size of class was the only variable, he added.
If lowering student-teacher ratios does improve student achievement, the benefits may not be significant enough to justify the extensive costs, Mr. Tomlinson's report concluded.
Ironically, Mr. Tomlinson's report--"Class Size and Public Policy: Politics and Panaceas"--listed Mr. Smith as one of several people who reviewed portions of it. In a telephone interview last week, Mr. Tomlinson recalled one exchange of letters between the two, but said Mr. Smith never said he agreed with the report's conclusion.
For his part, Mr. Smith said he did not remember reviewing the 1988 report.
Washington insiders say decisionmakers are rarely swayed when a federal agency releases a report supporting one of its own initiatives. Independent research has a much greater impact on congressional debate, according to Christopher T. Cross, the president of the Washington-based Council for Basic Education and a former assistant secretary for research in the Bush administration.
So far, the department's new class-size-reduction report has had little impact. Shortly after its release, Mr. Clinton agreed to support a Senate tobacco bill that does not include the money for class-size reduction, as the president proposed.
Mr. Smith said the administration now is hoping to attach its proposal to a tax bill it will push Congress to pass this year.
Vol. 17, Issue 39, Pages 22,24