How Vouchers Passed Their Florida Test


Debates over school choice are often framed as battles between low-income parents struggling to find options for their children and under-funded public schools desperate to keep their per-student funding. But a just-released study by Jay P. Greene, a Manhattan Institute senior fellow as well as a Harvard research fellow, finds that Florida's recent two-year experience with limited school choice has actually improved public school performance.

This is good news not only for Florida's families and educators, but for the rest of America as well, because President George W. Bush's educational proposal is similar to Florida's reform.

In July of 1999, Florida Governor Jeb Bush signed a law known as The A+ Plan for Education. Under that plan, the state assigned each school a letter grade based on its students' performances on the Florida Comprehensive Achievement Test (FCAT). Schools earning F's would be given more money and forced to develop a plan for improvement. Students in schools that earn F's in any two years in a four year period, would be free to transfer to another public school or to take a $3,400 voucher to use at a private school.

That fall, two failing elementary schools saw 53 of its students walk out the door with vouchers in hand and another 85 transfer to better public schools. (The law retroactively set the testing period to 1997-1998, thus these two schools earned F's in that academic year and again in 1998-1999). In total, 78 Florida schools earned F's for the 1998-1999 academic year, a dismal performance that, if repeated, would have meant the freeing of 2000 students the following term. Yet the next year all 78 of these schools managed to improve student performance, earning a D or better. (Meanwhile, the number of 'A' schools increased from 203 to 551.)

On the face, it appears that school choice helped Florida's worst public schools improve. A study last year by education writer Carol Innerst for a consortium of non-profits, including the Urban League of Greater Miami and the Center for Education Reform, chronicled the response of educators working in failing schools. "People get lulled into complacency," Broward County public schools official Carmen Varela-Russo told Innerst. "The jolt of being labeled an 'F' school and the possibility of losing children to private schools and other districts was a strong message to the whole community. Labeling schools 'A,' 'B,' 'C,' 'D,' or 'F' caused some pain."

Yet the pain proved to be an effective motivator. Broward County schools hired reading and math coaches to assist the students in its seven failing schools, an educational enhancement that other districts adopted as well. One district extended the school year, from 180 to 210 days, allowing more time to work on fundamental skills.

Yet correlation does not imply cause, which is where Greene's study comes in. He wanted to test whether the academic improvements stemmed from vouchers, or from other factors.

Greene hypothesized that if the threat of losing students to vouchers prompts schools to improve, then schools earning F's in any given year ought to show more improvement than other schools. Schools earning passing grades risked only embarrassment–or nothing at all–if student performance remained stagnant or even declined. That's exactly what he found. Writes Greene, "Schools that received F grades in 1998-1999 experienced increases in test scores that were more than twice as large as those experienced by schools with higher state grades."

A cautious researcher, Greene considered other explanations for the schools' improvement, including statistical artifacts such as regression to the mean, and the deliberate manipulation of test scores. But he found little empirical support for these explanations. "It appears as if two forces were in effect to motivate schools to improve," writes Greene. "Schools had some motivation to improve simply to avoid the embarrassment of low FCAT scores. This motivation operated across all state-assigned grades. But schools with F scores had a second and very strong incentive to improve to avoid vouchers." Green labels this second incentive the "voucher gain."

Greene's findings come at a critical time, as Washington policymakers debate President Bush's education reform package. Like his brother's Florida plan, it relies on testing to identify problem schools and assists these schools with new money to improve performance. And like the Florida plan, the ultimate sanction is an exit option for students in the form of a voucher, should poor schools fail to improve over a three-year period.

However, the administration has indicated that it may be willing to back off its controversial voucher plan. Last week, Secretary of Education Rod Paige soothed an audience of worried school board members, assuring them that vouchers are a "fractional part of the program." (In fact, it's fraction approaching zero in the Senate, where an education reform package introduced Tuesday didn't even include Bush's voucher proposal.)

Paige and others pushing Bush's education plan might consider another rhetorical strategy, one that involves combining Greene's quantitative findings with Innerst's qualitative findings to make the case that the plan's voucher component is critical to the success of the education package. Without the limited vouchers, the plan turns into another elaborate scheme to reward failure; it merely sends ever more money to bad schools. And–unlike Florida's experience with school choice–neither students nor struggling public schools benefit from such an approach.