Student test scores rose faster under Florida’s high-stakes testing program than they did under the federal program that was partly modeled after it, a study concludes.
The study, which was presented by Harvard University researchers at a British economics conference last month, compares academic gains Florida elementary school students made in one year under the state’s A+ Accountability Plan with those that came a year later as requirements of the federal No Child Left Behind Act kicked in across the state.
Read the report, “The Efficacy of Choice Threats Within School Accountability Systems: Results From Legislatively Induced Experiments,” from the Program on Education Policy and Governance at the John F. Kennedy School of Government, Harvard University.
Although limited in scope, the study drew intense criticism last week from education researchers on this side of the Atlantic Ocean. They attacked the study as much for the way in which it was made public as for its findings. Academics complained that authors Paul E. Peterson and Martin R. West had acted “irresponsibly” by allowing their report to be passed along to reporters without getting it vetted first by independent scholars.
Critics are especially irked because the authors signed a full-page advertisement in TheNew York Times last summer that criticized the American Federation of Teachers, the nation’s second-largest teachers union, for using similar tactics with a study shedding a critical light on charter schools. (“AFT Charter School Study Sparks Heated National Debate,” Sept. 1, 2004.)
If outside scholars had examined this new study, some researchers said last week, they would have found it flawed, inconsequential, and a bit premature.
“Why rush in with a one-year result when we know so many one-year results are ephemeral?” said Gerald A. Bracey, an Alexandria, Va.-based education researcher and an associate professor of education at George Mason University in Fairfax, Va.
Different Approaches
Though the study was not published in a scholarly journal, the Harvard researchers said, they did send it to four or five colleagues and responded to their critiques.
“So we felt pretty comfortable in releasing it,” said Mr. West, a research associate with Harvard’s Program on Education Policy and Governance.
Mr. West and Mr. Peterson believe the bigger gains students experienced under Florida’s homegrown efforts may have been due to the plan’s more nuanced structure and the nature of the threats it presents to poorly performing schools. Under the Florida plan, students qualify for vouchers to attend private schools if the public schools they attend get F grades twice in any four-year period. (Schools also qualify for cash bonuses if grades improve.)
Students can also switch out of failing schools under the NCLB law, but only if they move to another public school in the district—an option that few choose.
“It’s not quite the same thing as a kid receiving a voucher,” said Mr. Peterson, a government professor and the director of the Program on Education Policy and Governance.
Also, he noted, small percentages of Florida schools—8 percent and 2 percent, respectively—qualified for embarrassing D and F grades in 2002. In comparison, he said, 75 percent of the state’s schools were cited for failing to make adequate yearly progress in 2003 under the federal program, putting poor-performing schools in more company.
In schools that were newly assigned F’s in the summer of 2002, the researchers found, students’ test scores improved by 4 percent of a standard deviation more the following school year than did test scores in comparable D schools. Scores in newly named D schools, likewise, improved by 4 percent of a standard deviation more than those in C schools did. The gains, which all came on state tests, did not translate to nationally normed tests that students also took.
‘Similar Beliefs’
The researchers found no similar gains for students in schools found lacking under the federal No Child Left Behind program.
Though the improvements under Florida’s program were modest, the researchers said, they could add up to substantial gains if they continue.
But many independent researchers argued last week that the findings were too small to be noteworthy.
One reviewer who disagreed with those critics, though, was Jay P. Greene, a Florida-based researcher for the Manhattan Institute, a conservative-leaning think tank based in New York City. He maintained that the Harvard study was important because it relied on data on individual students, rather than on schools, where changes in student mobility can skew findings.
“Both programs are based on conceptually similar beliefs,” Mr. Greene added, “yet this shows that the details of how it’s implemented matter.”
