Policy Summary

The following is a summary compiled by the Center for Education Reform of the following report:

AN EVALUATION OF THE FLORIDA A-PLUS ACCOUNTABILITY AND SCHOOL CHOICE PROGRAM
By Jay P. Greene 
For Florida State University; Program on Education Policy and Governance, Harvard University; Manhattan Institute
February 2001

CER Summary:

        Since the advent of opportunity scholarships for students in failing schools, advocates of these scholarships have suggested that their existence not only helped students who left to attend higher-performing schools, but also aided those students remaining in a failing public school by driving improvements in those schools.

        This study, conducted on behalf of Florida State University, the John F. Kennedy Program on Education Policy and Governance at Harvard University, and the Manhattan Institute, is the first to statistically examine the effect of vouchers on failing schools. The conclusion: The performance of students on academic tests improves when public schools are faced with the prospect that their students will receive vouchers.

        Florida is the only state that has linked school accountability to student choice as a way to improve the quality of the state's public schools. Since 1998, as part of Florida's A-Plus program, the state has administered the Florida Comprehensive Assessment Test (FCAT) in reading and writing to its 4th, 8th, and 10th graders and in math to its 5th, 8th, and 10th graders. Students enrolled in schools that receive a failing grade based on these tests results for two years in a four-year period are given the opportunity to transfer to another public school or to attend a private school with the help of tuition assistance from Florida's Opportunity Scholarship Program.

        As a result of the 1999 testing, two schools in Escambia County, Florida lost students to other public and private schools. In 2000, by contrast, no school received a second failing grade and, therefore, no new students qualified for Opportunity Scholarships. In fact, those schools that were in danger of receiving a second failing grade in 2000 saw their FCAT scores improve more dramatically than did schools that had yet to receive their first "F."

        In his report, "An Evaluation of the Florida A-Plus Accountability and School Choice Program" (February 2001), Jay Greene attempts to determine both the validity of the FCAT as a measure of student performance and the effectiveness of vouchers as a way to motivate schools to improve student performance.

        Greene compares the results of the 2000 FCAT, the statewide criterion-referenced exam with the high-stakes choice component, against the 2000 Stanford 9, the statewide norm-referenced exam with no stakes whatsoever. Greene found that in both reading and math, and at all grade-levels, there is a very high degree of statistical correlation between the results of the FCAT and the Stanford 9. Greene concludes that because there is a very high degree of correlation between the school level results of both tests, the FCAT is a valid measure of student performance.

        Confident that the FCAT is a reliable measure of student performance, Greene next undertakes to accomplish his second goal. Greene compares the degree to which schools improved based on how well their students performed on the 1999 FCAT.

        The scores of schools receiving a "D" or "F" in 1999 improved more dramatically than did the scores of school receiving an "A," "B," or "C." On a scale from 100 to 500, the "A," "B," and "C" schools improved by less than 5 points in reading and less than 12 points in math. By contrast, the "D" and "F" schools improved by more than 10 points in reading and more than 16 points in math. In writing, on a scale from 0 to 6, the "A," "B," and "C" schools improved by less than .5 points while the "D" and "F" schools improved by more than .5 points.

        When Greene examined the differences in improvement between the "D" and "F" schools, he found sharper contrasts than those separating the "A," "B," and "C" schools. In reading, the "D" schools improved their scores by 10.02 points, while the "F" schools improved by 17.59 points. In math, the "D" schools improved their scores by 16.06 points, while the "F" schools improved by 25.66 points. In writing, the "D" schools improved their scores by .52 points, while the "F" schools improved by .87 points.

        While it is obvious that the "D" and "F" schools improved more than did their higher performing counterparts, Greene also separates the improvement motivated by political embarrassment from the improvement motivated by the fear of losing students to choice by comparing the improvement in schools scoring low "Ds" in 1999 with schools scoring high "Fs" (those in danger of losing students to vouchers).

        The examination showed those with high Fs improved more than those with low Ds. Reading improved by 12.87 points for Low-Ds, vs. 15.52 points for High-Fs; Math improved 18.15 points for Low-Ds, vs. 24.24 points for High-Fs; Writing improved .59 points for Low-Ds, and only .75 points for High-Fs on a 1-6 scale score.

        Greene also dismisses other possible explanations for the improvements, noting that the correlation between the FCAT and the Stanford 9 remains strong, even when the F school results are isolated from the other scores (no evidence of intensified cheating at F schools). He also shows that dramatic improvements were consistent among failing schools, removing the possibility of a statistical "bounce," or improvement that could be accounted for by chance.

        Greene's conclusion: "The most obvious explanation for these findings is that an accountability system with vouchers as the sanction for repeated failure really motivates schools to improve."

        The entire study may be found at http://www.manhattan-institute.org.

John Kraman 
Research Associate

# # #

For more information, link to:  

        and related news articles at:

The Center for Education Reform [CER] is a national, independent, non-profit advocacy organization providing support and guidance to individuals, community and civic groups, policymakers and others who are working to bring fundamental reforms to their schools. For further information, please call (202) 822-9000.

CER Home Page CER News Alerts School Choice E-Mail CER