Voucher Threat Spurs Turnaround of Failing Schools

Published April 1, 2001

Vouchers will destroy public education, cry opponents. But if Florida’s early experience is any indication, the mere threat of losing their students to vouchers powerfully motivates stewards of substandard public schools to get their instructional acts together and focus on elevating student achievement.

The evidence that voucher opponents have been crying wolf comes from a study of the Sunshine State’s A-Plus accountability and school choice program. The analysis was conducted by Dr. Jay P. Greene of the Manhattan Institute and Harvard University, under contract with Florida State University.

Earlier studies by Greene and his associates in Harvard’s Program on Education Policy and Governance, headed by Dr. Paul Peterson, have documented gains for urban black students who took vouchers and transferred mainly to Catholic schools. But the new study establishes positive results as well for the children “left behind” in the public schools.

“This study of the Florida A-Plus educational program provides further evidence of the potency of even potential competition in improving performance,” commented Nobel Laureate Milton Friedman.

How A-Plus Works

The A-Plus Plan, initiated by Governor Jeb Bush and Lieutenant Governor Frank Brogan and approved by the Florida Legislature, assigns each public school a grade from A to F based on the performance of its students on the Florida Comprehensive Assessment Tests (FCAT) in reading, math, and writing. If a school receives two Fs within a four-year period, its students can receive vouchers to redeem at private schools, or they can transfer to better-performing public schools.

After the second round of the test in 1999, only two schools in the Pensacola area had received failing grades, and about 50 families opted for private school vouchers. However, an initial F had gone to another 76 schools, which therefore faced a clear and present danger of losing hundreds, possibly thousands, of students to the voucher option.

When all the scores were in from the year 2000 testing, all 76 schools had raised their students’ scores sufficiently to avoid a second F grade.

Vouchers Spur Improvement

Change in FCAT Scores
by School Grade
1999 School Grade # of Schools Reading Math Writing
A 202 1.90 11.02 .36
B 308 4.85 9.30 .39
C 1223 4.60 11.81 .45
D 583 10.02 16.06 .52
F 76 17.59 25.66 .87
The change for F schools compared to schools with higher grades is statistically significant at p < .01


Math and reading scales are from 100 to 500.


The writing scale is from 0 to 6.


That would be remarkable enough, but Greene’s study also revealed striking differences between schools facing an imminent threat of vouchers and those not under that gun.

For instance, on the reading test–which like the math test is scored on a scale of 100 to 500–the one-time F schools showed a gain of 17.59 points. Meanwhile, the C-graded schools reported a reading gain of only 4.60 points.

On the math test, the F schools gained a whopping 25.66 points, while the C schools were up 11.81 points.

The FCAT writing test is scored on a scale of 0 to 6. Again, the ability of the “failing” schools to break out of their low-performance rut was impressive. The F schools gained .87 in writing, as compared with .45 for the C schools.

“These results,” said Greene, “are particularly relevant because of the similarities between the Florida A-Plus choice and accountability system and the education initiatives proposed by President George W. Bush.”

President Bush proposes to convert federal Title I money to $1,500 parental choice stipends after children have been stuck for three years in schools that flunk achievement tests. Senate Democratic leaders have vowed to strip this limited choice provision out of the Elementary and Secondary Education Act reauthorization.

The evidence from Florida suggests Congress would be missing a significant opportunity to produce some tangible gains from the Title I program, which has seen precious few gains over its 35-year lifetime, despite the expenditure of more than $130 billion.