Report: Bias found in teacher evals
- No Child Left Behind waivers and Race To The Top grants both require states to implement teacher evaluation systems, but many states are struggling to do so. This prompted the Brookings Institution's Brown Center on Education Policy to conduct a case study on the matter.
- The Brown Center's report analyzes four urban school districts as they attempt to create meaningful teacher evaluation systems using both observation and student academic gains.
- The report found that teachers have difficulty receiving top scores if they don't have already high-performing students, that observations by outsiders are more reliable than those by school administrators, that district size unfairly impacts a teacher's "student gain" scores, and that using academic gains to calculate "value added" by a teacher hurts good teachers in bad schools and vice-versa.
While only 19 states have currently received Race To The Top funding and 43 have No Child Left Behind Waivers, only a small number have implemented effective teacher evaluation plans. The difficulty in creating an evaluation system has resulted in the U.S. Department of Education's announcement that it will no longer incorporate teacher evaluations into its NCLB waiver requirement. If it did, too many states would lose their waivers.
Given the difficulty of implementing such systems, case studies like those out of non-profit public policy organizations like the Brookings Institution provide an idea and analysis of what districts have done. Brookings makes a number of suggestions for states attempting to implement evaluations: Adjust teacher observation scores based on the demographic they are serving, have at least one observation each year performed by a trained outsider, and eliminate the value-added method (VAM) of evaluation.
VAM, which uses student scores to evaluate teacher effectiveness, received much attention earlier this year when the American Statistical Association (ASA) commented on its unreliability.
- The Brookings Institution Evaluating Teachers with Classroom Observations: Lessons Learned in Four Districts