Two kinds of efforts at instructional improvement in higher education, student evaluation of teachers and faculty development programs on campuses, fall short of reaching their theoretical goals. Student evaluations are neither reliable nor valid measures of a teacher's instructional effectiveness. Some of the items included in such questionnaires address only symptoms of instructional problems and not underlying causes, and others ignore the fact that learning and not teacher characteristics is the most important objective. Similarly, almost anything that a faculty member does outside the classroom is now called faculty development, but in this melange, most development programs have little potential for contributing significantly to instructional improvement. Two projects have been developed using a rigorous and systematic approach to this problem: the Center for Professional Development with pilot programs on six campuses of the California State University and Colleges; and a comprehensive program at the UCLA School of Dentistry. In the multicampus project, a variety of efforts were undertaken on different campuses: narrowly-focused faculty workshops, development of self-appraisal instruments, course and curricular materials development, institutional research projects, and a program for administrators only. It is felt that little change occurred or was initiated in this project. At UCLA, however, a nonfragmented, comprehensive faculty development and instructional improvement program addressing institutional goals is felt to be a success. The key is that the school as an organization, including both faculty and administration, accepted responsibility for the consequences of its educational programs. (MSE)