Jay Mathews: Charter Schools Provide Good Model on Teacher Pay

By Jay Mathews
Monday, June 1, 2009

It is hard for me to find a school leader with a track record for raising student achievement who does not admire almost everything Chancellor Michelle A. Rhee is doing with the D.C. schools. Yeah, I said almost. One important item on her agenda is not so popular -- merit pay for teachers.

Rhee is right when she says that the standard way of assessing and compensating teachers is a mess. It drives talented people from the field or leads them to avoid teaching in the first place. It deprives principals of vital tools for improving schools. Rhee learned this the hard way, creating the New Teacher Project to recruit and train teachers for urban schools. That nonprofit organization is releasing today a study of 12 districts that backs Rhee's point -- administrators shrink from assessing the relative quality of their teachers and tend to give nearly everybody a good evaluation.

To end this bad habit, Rhee has proposed paying teachers as much as $135,000 a year based on achievement gains, classroom practices, meeting school goals and choosing high-needs students, as long as they are willing to forgo tenure protection. The chancellor is part of a national movement, backed by some leading policy experts, to create for teachers the same kind of merit pay enjoyed by football players, stock analysts and shoe salesmen.

Even President Obama is cheering her on. "Too many supporters of my party have resisted the idea of rewarding excellence in teaching with extra pay, even though we know it can make a difference in the classroom," he said in March.

This makes sense to many people but not to educators who remember how they created good schools. Extra pay for better work sounds as logical as sharpened pencils and multiplication tables. But if done in the public and routinized way indicated by Rhee and the president, it could ruin the team spirit that has produced the most successful public schools, particularly in urban and rural areas.

Look inside award-winning schools and you find teachers, coaches, counselors and aides being paid the standard rate for their districts, with some extra money for those who work extra hours. No classroom stars make $135,000 a year. Staffers are selected for their enthusiasm, energy and willingness to cooperate with each other in addressing student needs. Those who teach subjects not tested by the state, such as history or music, still feel like they are part of the team. Every success in improving a child's reading proficiency, behavior in class or attendance record can be credited to many people. If overall success rates climb, everyone can share in the satisfaction, as important to them as money.

With an individual pay-for-performance system, suddenly it's every teacher for herself or himself. Too much weight falls on how small groups of students do in the testing room in May. "High-performing kids can test low, and low-performing kids can test high," said one successful public charter school leader, who asked not to be named in order to speak candidly about Rhee's proposal. "Although the test is usually accurate, I don't think it's fair to put a dollar figure on the backs of 9-year-olds performing on a test that is usually being proctored by the classroom teacher. Cheating will inevitably go up. Teacher performance will actually be the ability of a teacher to improve test prep but most likely not instructional improvement."

Charter schools -- and some regular public schools with special arrangements -- often do better by placing decisions on salary and other matters in the hands of carefully selected and trained principals who know what great teachers do because that was what they once were. They judge teacher talent on classroom technique as well as test scores. They help teachers who are struggling. They praise those who are improving. They reward staffers in ways that make sense for each individual. Paying big bonuses to stars based on a formula decided by a district committee means fewer thoughtful conversations about students and more bitter gossip about adults.

Of course, this approach requires great care in selecting and training the school leader, but what's wrong with that? The superintendent also must be ready to relieve principals whose schools do not raise achievement. Results will be measured, of course, at least in part by test scores. There is no escape from that. But it will be an assessment of gains, or lack of them, by every child in the school, not just those taught by one teacher in one classroom.

Rhee might move in this direction herself. She has left her merit pay plan publicly vague as she negotiates with the teachers union. She could just hand merit and bonus money to principals and let them divvy it up in ways that will keep their teams focused on kids.

Merit pay for teachers has been a bust in many places. It won't help anyone if it becomes in the District, as many educators fear, even worse: not just a failure, but a very expensive one.

E-mail: mathewsj@washpost.com

© 2009 The Washington Post Company