Yesterday, the Buffalo News reported that the Buffalo Public Schools and the Buffalo Teachers Federation had negotiated a new teacher evaluation system. But what is particularly significant is that the News simultaneously reported on and endorsed the contract negotiated between Washington, D.C. teachers and administration, and promoted it as a model for Buffalo. The D.C. contract — known as IMPACT but not mentioned by name in the editorial — has, according to the Buffalo News, four key components: performance-based teacher evaluation, financial incentives to raise test scores, limits on the protections of tenure, and increased ability of the district to lay off “bad teachers” without “economic cause”. But the News is either unaware or unwilling to report facts unfriendly to its position of support.
While the News editorial characterizes the contact as one where “performance and the quality of teaching, not blind seniority, will determine who is hired and who is laid off,” it downplays the fact that “performance” and “quality of teaching” are determined by student test scores. Following adoption of the contract, D.C. Schools Chancellor Michelle Rhee fired 165 teachers based on lack of improvement in student test scores over one academic year. The method is said to measure the “value added” to students by their teacher.
Student Test Scores do Not Equal Good Teaching
Despite all the rhetoric supporting the use of scientific research to guide education reform, the amount of evidence against using test scores as a basis for teacher evaluation is very strong. President Obama, U.S. Secretary of Education Arne Duncan, and a host of billionaires who support Rhee are imposing this practice across the country, despite the warnings of the scientific community.
In his video Merit Pay, Teacher Pay, and Value Added Measures, professor Daniel Willingham summarizes the problems associated with what the News is promoting. But he is not alone. A recent report by the National Center for Educational Evaluation estimating the error in using test scores to classify teachers as effective or ineffective predicts that when using only one year of data, 35% of teacher classifications will be wrong (i.e., effective teachers will be classified as ineffective, and ineffective teachers will be classified as effective). For teachers in D.C., that means as many as 57 of the 165 teachers fired in DC might have been inaccurately identified as ineffective. The National Center for Analysis of Longitudinal Data in Education Research also released a study examining the technical limits of using student test results to evaluate teachers. Among other things, the report found that different tests yield different teacher rankings.
The Degrading Affect of Incentives
But more important than the technical limitations noted above is the philosophical underpinning of the entire system based on financial incentives to pressure educators to boost student test scores. Based on past practice, this gives rise to treating students as mere conduits of cash, leading ultimately to student abuse and debasement of public education. This is what happened under a similar system in England, Ireland, Australia and elsewhere during the latter half of the nineteenth century. The negative results of what was known as Payment by Results were widely recognized by contemporaries, and the practice was eventually halted. It was precisely this system — one that abused teachers and pushed many competent ones to leave the profession — that contributed to teachers unionizing in Britain. And most interestingly, it was a method of teacher compensation rooted in an effort to reduce spending on public education during a time of great expenditures following the Crimean War.