Teaching is an omnibus profession, but each teacher is a self-sufficient individual and many yardsticks are needed to measure competence in this role. The evaluation of a teacher should, in principle, be bound to what students learn and to the attitudes and values they hold over the long haul. In practice, however, we tend to separate teaching from learning. Each of the participating schools in the National Project III uses some form of teacher evaluation by students, and from these collective experiences it is possible to be explicit about the issues and problems generated by such arrangements. These matters range from broad policy questions to technical decisions pertaining to the evaluating instrument and the handling of data. The report is a brief summary of these issues, and most of the examples are taken from Fund Associates schools. Sources for more detailed information are listed. (Author/MSE)