Diane Ravitch, a professor at New York University had this to say about the way we evaluate teachers. Basically, because so many different things influence the performance on testing, it is not an effective way to evaluate a teacher's effectiveness in the classroom. Furthermore, she says that teachers should be judged by "professional standards," but does little to explain exactly what this means.
Some of the others who responded to this question made suggestions such as the outcomes of the education process, but most admit there is not a fool-proof method for evaluating this.
So what is the answer? A while back either someone told me this or I read it somewhere. (It isn't my idea, but I am unsure who I need to contribute it to. However, if you know, please let me know so that I can reference your idea in my blog.) The idea was with regards to having departmental review committees from that school.
It stems from the poor practice of having an administrator evaluate teacher’s effectiveness in the classroom. This is not a valuable evaluation in my mind because those administrators are no long dealing with the day to day needs of the individual classrooms. Furthermore, for some administrators it may have been years or possibly they have never even taught in a classroom. However, having departmental reviews done by peers who are currently working in classrooms and with the curriculum appears to me to be a more sound method for evaluation. After all, in business, the CEO or President is not the one who evaluates the office personnel. Rather it is their immediate boss or peers who work alongside them in the trenches that provide the most accurate assessment of their performance.