UFT wrong to fight Joel Klein's attempt to release teacher data, says leading education researcher

Eric A. Hanushek
Published Date
New York Daily News
New York City's schools chancellor, with the support of Mayor Bloomberg, wants to release the value-added test score results for 12,000 teachers - revealing for parents and the public the student learning gains attributable to each instructor. News organizations have requested the data; the city is ready to comply. The city's United Federation of Teachers has challenged the release, and a judge will decide next month. I've spent many years looking carefully at such data. I know it can be incendiary; I know it has flaws. Still, I strongly support its release. Two principles lie behind this view. First, parents and taxpayers have a basic right to know about the effectiveness of the teachers and schools that they support. Second, it is impossible to think of improving our schools without focusing on the productivity of the teachers. Teacher quality is the single most important factor inside a school in driving student performance. My research has shown, for example, that the best teachers can get a year and a half of achievement gain from their students, while the worst get only a half-year gain. In other words, over a single academic year, two students who start with the same knowledge can end up a full year different in learning at the end of the year. Value-added analysis is the best tool we have available to zero in on the impact of the individual teacher on student achievement gains. Using it, we can begin to distinguish between the best teachers and the worst, so we can begin rewarding the best while learning from their successes and improving - or removing - the worst. Yes, it's true: Value-added analysis is not perfect. First, it only measures performance in the areas that are tested - typically, math and reading but not science, social studies or other areas of teacher influence. Second, the statistical measures include some errors in assessing teacher performance because the tests themselves are inaccurate assessments of knowledge. New York has had recent history with test scores being recalibrated, which has led thousands of parents to start questioning the testing regime. Ideally, value-added data would be used as just one component of a multifaceted evaluation system that would also combine principal evaluations and other assessments of teacher quality to create a nuanced picture of each teacher's quality. But to say that value-added systems are imperfect is not to say that they are worthless. They are very worthwhile. Unfortunately, many in the schools, led by the teachers unions, are not particularly interested in entering a discussion of which teachers are performing poorly, because they have generally committed to defending all teachers, and such information makes their defense difficult. The unions often acknowledge that we need to develop more reliable and accurate evaluation systems - then return to prevent evaluation systems from focusing on classroom performance. Remember, New York's union leadership previously went to Albany to block the use of student test information in teacher tenure decisions. Consider what just happened in Washington, D.C. Michelle Rhee proposed a teacher evaluation and reward system that gave large bonuses to exceptional teachers but that also eliminated teachers who showed that they were ineffective. She worked to develop a thoughtful and elaborate evaluation system based on classroom observations and supervisor ratings along with incorporating value-added measures of performance where possible. The unions resisted these efforts, leading her to institute a series of unilateral personnel actions against the most ineffective teachers. Rhee is now gone from D.C., and few can look at this as anything but evidence of the difficulty that follows any attempt to evaluate performance. While I would not like to see personnel decisions made solely on the basis of value-added scores, I do think that it is an essential element. Typical evaluations today are useless, because only a minuscule number of teachers are rated anything but great and because these ratings are never used in making personnel decisions. Unless and until a broader evaluation system is in effect, I would nonetheless fall back to a value-added system. The need to improve our schools, and to ensure that there is an effective teacher in every classroom, is too great simply to dismiss such information as imperfect. The current system of evaluations is not just imperfect; it is completely broken. Unions either must participate in developing a thorough evaluation system or live with what comes out of the statistical models.