If principals and other administrators are doing their jobs, they already know who the best and worse teachers are. One year, Brookline administered one form of the Iowa Test of Basic Skills in the fall to third graders and the state administered a different form of the same test in the spring. I calculated the mean gain score for each teacher and asked the Assistant Superintendent for Curriculum and Instruction to tell me which teachers would have the highest and lowest gain scores. She got the group of highest teachers exactly right and missed only one of the lowest. I would expect building-based administrators to do even better. More formal analysis confirms my experience (Murnane, 1975; Jacob and Lefgren, 2008).
Measurement Matters: Perspectives on Education Policy from an Economist and School Board Member (pdf file)
by Kevin Lang