In 1992, eight two-year colleges in Northern California were chosen to participate in a quantitative study of the relationship between student outcomes and the use of instructional methods recommended in the Cross/Angelo Classroom Assessment Model. The model supports the modification of instruction based on frequent, quick, and anonymous written feedback from students about their understanding of course content and their reactions to instruction. The study compared retention, grade distribution, and classroom environment data for classes in which Classroom Assessment techniques were used and those in which they were not employed. Outcomes comparisons were also made by gender and ethnicity. A first research design compared student retention for the same instructor before and after the instructor received training in and implemented Classroom Assessment techniques. In a second design, trained instructors either utilized or withheld Classroom Assessment techniques with different sections of the same course. Pilot studies were first conducted in summer 1992, with larger studies conducted in fall 1992 and spring 1993. Study findings included the following: (1) retention rates for women in Classroom Assessment classes were higher than in control classes, though retention did not change for men; (2) grade point averages in Classroom Assessment classes were not significantly higher than in control classes, though there were a higher percentage of "A's" among Classroom Assessment students; and (3) minority students felt more positive about the classroom environment in Classroom Assessment classes than in control classes. A literature review; results of student and instructor surveys; recommendations; study letters and survey instruments; and a 52-item bibliography are included. (PAA)