- The Washington Times - Friday, July 13, 2007

No sooner had Mayor Adrian Fenty appointed Michelle Rhee chancellor of D.C. Public Schools than critics began challenging her self-reported success in raising student performance when she taught in Baltimore. She won approval in the end, but probably not on the strength of evidence that she had succeeded where others have failed.

That’s because there isn’t any. Education’s best-kept secret is the drought of data linking a teacher’s work to individual student outcomes.

We know that the most important in-school influence on student performance is teacher quality, and the difference between the results that the best and worst teachers get in the classroom is staggering.

Some studies show that the top 15 percent of teachers are three times as effective as the bottom 15 percent. The students of the ablest teachers show about a year and a half gain in tested performance annually, compared with only about a half-year gain for kids stuck with the less competent teachers. You don’t have to be a math whiz to figure out how these effects might accumulate for students lucky enough to get a string of good teachers or unlucky enough to get one ineffective teacher after another.

Teaching stars are out there, and Mrs. Rhee may well have been one of them. If so, her students were lucky. And they got a double dose of whatever she had to offer since she stayed with them for two consecutive school years.

In any case, Mrs. Rhee probably won’t ever be able to supply bulletproof evidence of her star turn in Baltimore for the simple reason that few school districts and states link student test performance over time to individual teachers. Excuses abound. Technical difficulty is one. Often, student files are kept in one data silo and teacher files in another. Teacher resistance is no doubt another. Few people want their annual performance so closely tracked. And, of course, test scores don’t measure everything important that students learn.

But none of these “reasons” justifies largely flying blind the way school administrators must do now. The technical data-collection and management issues are solvable. And, whatever the limits of what tests can measure, it’s hard to argue that what they do measure isn’t valuable or shouldn’t be tracked. In the search for better ways to assess overall teacher effectiveness, student test performance gains belong in the toolkit.

A second secret the huge differences garnered by teachers is out of the bag now and impossible in good conscience to ignore. To raise the bar for poor performers, a school district needs to learn what about the selection, training and support for teachers makes some far better than others and to take performance into account in rewarding and promoting teachers. But neither step is possible until local and state authorities make data available on performance results teacher by teacher.

The new data systems needed can’t be set up fast enough to help Mrs. Rhee fend off challenges, but getting good information for human resource management in D.C. Public Schools so the good teachers can be appropriately rewarded and retained and the weak ones helped or ushered out may be the most important thing a newly restructured school system can do.

Jane Hannaway is director of the Education Policy Center at the Urban Institute and director of the National Center for the Analysis of Longitudinal Data in Education Research.