I understand that. There still can be any number of reasons that entire classes of students are doing better or worse: New computer lab for instance, will only benefit the type of teachers who will make use of it. Those who are traditional will be at an disadvantage. Class sizes being reduced for disabled and English learner students, will impact classes are that are mainstreaming and those who are not mainstreaming will be at a disadvantage. Classes are not ALL the same like how you describe. How do you figure in all of these confounding variables? Dozens of schools in LA in the last year are going through increased class sizes, expanding campuses, teacher lay offs, charter schools taking the brightest students from the public schools, etc etc, how are you suppose to calculate these factors into that formula?
There are too many things going on for you to isolate teachers as THE variable. There are MANY variables and it does not effect all students, it affects different groups of students. The mere fact that your foundation is just a mere correlation, all the confounding variables destroys your entire premise.
You missed my point. Please refer to my earlier example in this thread where I talk about about my colleague who was waging a war with the administration. The administration can give a teacher 6 periods of hell, make every class horrible, all it takes is 5 or more problem students. These problem students can drive the entire class down in terms of performance, structure, and discipline. This is a tactic that the administration use to "discipline" teachers who do not cooperate and who are untouchable because of the union. Under the merit pay system, the teacher would suffer a dip in their salary because s/he was standing up to the administration (who was then implementing NCLB). Many teachers who opposed NCLB would have been silenced or driven out of their careers under the merit pay system. How well would that have turned out?
I guess we'll just disagree here.
Lots of confounding variables in a school year. Class make-up changes, you get new students, you lose a few students, some classes you only take 1 semester. How will I know whether or not I should make adjustments? How will I know that my students will be tested on the material that I teach them? I only get one chance to prove myself every year? If I get bad luck one year and get crappy classes, I'll have to wait 1 whole year to fix it?
If not every quarter, at least once a semester like final exams.