One aim of student evaluation of instruction is the improvement of teaching quality, but there is little evidence that student assessment of instruction alone improves teaching. This study tried to improve the effects of evaluation by combining evaluation with individual counselling in an institutional development approach. Evaluation was conducted in a private school for speech therapy (logopedia) in 35 classes (N = 16 teachers, N = 592 students). Evaluation was done twice within a period of three to twelve months using a standardized questionnaire (HILVE-II) developed for evaluation of university courses. The intervention effect on teaching quality was more than half a standard deviation on the teacher scales. Despite the fact that the counsellor had no pedagogical training, the quality of teaching not only improved quantitatively and qualitatively but also became more homogeneous although the relative rank listing of teachers did not change. (Contains 2 tables and 2 figures.) [Abstract also included in German and French.]