Gov. Bob McDonnell has the right idea with a new pay-for-performance system the state will be piloting in 25 schools across the commonwealth in the next school year. It would be foolhardy to expect miraculous results, but if the results are even mildly encouraging, Virginia public schools will have a new tool for recruiting and retaining the best teachers.

Participating schools must implement performance standards and a teacher-evaluation system that bases 40 percent of a teacher’s evaluation on student academic growth. The General Assembly set aside $3 million for the program, which will provide bonuses of up to $5,000 for teachers who win exemplary ratings.

“There is a growing consensus,” said McDonnell in a prepared statement, “that public schools must move beyond a compensation model that equally rewards mediocrity and excellence and is rooted in a past when our young people did not face fierce global competition. Bipartisan support – at both the state and federal levels – is allowing Virginia to implement performance pay in some of the Commonwealth’s most hard-to-staff rural and urban schools. I look forward to this program getting underway. It will benefit our students, parents and teachers.”

It’s a good idea in theory. Teacher quality is the single-most important in-school factor influencing student achievement. Research consistently shows that students learn more from the best teachers. It especially makes sense  for schools in the inner city and the countryside to reward their best teachers. As Roanoke Superintendent Rita D. Bishop said in the press release, “We have had some success with signing bonuses but still lose too many teachers to surrounding divisions. By participating in the performance-pay pilot, we are able to provide additional incentives for talented teachers to remain in four of our schools where they can make a real difference in the lives of students.”

If designed properly, the pilot programs should answer some important questions. Is the uncertain prospect of a $3,000 to $5,000 bonus enough to motivate teachers to do better? Will teachers buy in to the validity of the methodology for calculating gains of their students’ educational achievement? Should teachers even be rated on individual performance when teaching is increasingly team driven? New York City has just announced that it was dropping its three-year performance-bonus program, noting that it had no discernible effect on student achievement.

Virginia’s program is designed very differently than New York’s, which was geared to rewarding the collective performance of teachers and staff. Still, it would be remarkable if Virginia got all the details right on the first try. Before rolling it out statewide, the McDonnell administration should be prepared to tinker with the program for three or four years in order to get it right.

One last point: Ideally, the pilot project would follow the social-science protocol described in “Evidence-Based Social Policy” to ensure that the findings, whatever they are, leave no ambiguity. Otherwise, the politicians and educrats evaluating the results will revert to their default positions and little useful knowledge will be gained.