State Teacher Evaluation Methods Getting Questioned From All Sides

Last month, Houston area teachers became the latest group to file a lawsuit over teacher evaluations using controversial value-added models, alleging that the statistical models produce a misrepresentation of teacher quality.

Value added models have seen their popularity skyrocket among state policymakers and school superintendents across the country, largely due to endorsements from the Obama administration and research commissioned by the Bill and Melinda Gates Foundation that supports their use as a tool to measure student growth. Teacher unions, however, have taken a different stance towards the models, claiming they are unfair, biased, and hard to understand.

Similar lawsuits have been filed in Tennessee, New Mexico and Florida with a federal judge ruling in favor of Florida’s evaluation system while commenting that, while not unconstitutional, he did believe it to be unfair. Recent legal battles assume a great deal of significance because although the models are widely used with 35 states using them as a significant part of Race to the Top teacher evaluations— they have recently come under fire from independent researchers. An April 8 report from the American Statistical Association cautioned against placing too much emphasis on the models, claiming that teachers can only account for 14 percent of the variability of test scores. In other words, the remaining 86 percent of factors contributing to how well a student does on a test come from outside the teacher.

Value added models took another hit in a study published this week by the American Educational Research Association, which is especially interesting as it was funded by the Gates Foundation, a well-known proponent of these types of evaluations. The study concluded there was little to no relationship between measured teacher and instructional quality and student learning. Morgan S. Polikoff, professor of education at the University of Southern California and one of the co-authors, says in a YouTube video released with the report, “These measures are not up to the task of being put in an index to make important summative decisions about teachers.”

As the discourse around the models heats up, some states are looking for alternative ways to evaluate teachers. On May 2, the Texas Education Agency submitted a teacher evaluation waiver that would give the state more leeway in how it evaluated educators. The proposed system would weigh teacher observation and self-evaluation at 80 percent of the total evaluation, leaving only 20 percent for student growth measures done by value added model or other methods.  

Meanwhile, Florida granted a waiver from the state mandated value added model to five schools in Pinellas County that proposed an alternative evaluation model to be piloted in the 2013-2014 school year. In the county’s plan, teachers would receive feedback from students and administrators throughout the school year in order to address areas in which they needed improvement before they were formally evaluated.

As teachers and researchers continue to question value added models, it will be interesting to track the changes made by states.