Second Study Questions Research Linking Voucher Threat to Gains
Debate continued last week over a recent report that connected gains in student achievement in Florida to the state's voucher program, as another scholar came forward to question the conclusions of the Manhattan Institute study.
For More Information
|Mr. Greene's report, "An Evaluation of the Florida A-Plus Accountability and School Choice Program," is available from the Manhattan Institute for Policy Research.|
In an academic article published on March 19, a professor at the University of Colorado at Boulder contends that the substantial gains on state tests achieved by Florida schools facing the threat of vouchers might be due to schools' efforts to achieve minimum passing scores on the writing portion of the exam. The analysis came in response to the report released last month by the Manhattan Institute, in which researcher Jay P. Greene concluded that a "voucher effect" had motivated low-performing schools to achieve at higher levels.
"I saw Greene's report as partial," said Haggai Kupermintz, an assistant professor in the research and evaluation methodology program at the University of Colorado's school of education and the author of the new article. "I thought there was something more in the data than was revealed in his calculations."
Mr. Kupermintz's study follows an analysis by two Rutgers University professors that also questioned Mr. Greene's study. ("Choice Words," March 21, 2001.)
Writing Test Eyed
The Kupermintz study, which was published in a scholarly journal, Education Policy Analysis Archives, suggests that failing schools had followed a strategy to escape the threat of vouchers by working to achieve the minimum score necessary to pass the writing portion of the Florida Comprehensive Achievement Test.
Under the state's accountability program, schools are assigned letter grades based primarily on their performance on the reading, writing, and mathematics portions of the exam. Students in schools graded F for two out of four years qualify to receive publicly funded vouchers to attend private schools. Schools need only pass one section of the exam to escape the list of failing schools.
If schools made concerted efforts to teach students how to achieve the minimum passing score on the writing section—a 3 on a scale of 1 to 6—"then this would be reflected by the scores converging at the score of three, and this is what my data show," Mr. Kupermintz said.
The report also cites an article exploring Florida students' gains in writing published on June 21, 2000, in a Florida newspaper, The St. Petersburg Times, which asserted that "Florida educators have figured out how the state's writing test works and are gearing instruction toward it—with constant writing, and in many cases, a shamelessly formulaic approach."
"My findings are consistent with this explanation," Mr. Kupermintz said.
Mr. Greene was not available for comment last week, but in his original study he anticipated critics who might offer other reasons for the test-score gains he found.
"While one cannot anticipate or rule out all plausible alternative explanations for the findings in this study," Mr. Greene wrote, "one should follow the general advice to expect horses when one hears hoofbeats, not zebras."
Vol. 20, Issue 28, Page 22