The results of three years of evaluative investigation of the University of Pittsburgh Learning Research and Development Center's program of individualized education are described. The study was conducted in seven Follow Through sites and three Pittsburgh area schools. Standardized tests were used as input and outcome measures; questionnaires and videotapes were used to gather information about the classroom processes. The results indicate that over the three-year period the implementation of the program continued to move toward ideal goals. Extensive data reduction procedures were used and the rationale of usage discussed. Results also point to input as the primary explanation of student end-of-year performance, though classroom processes contribute a small but consistent amount. The data indicate greater ease in identifying negative factors in successful classroom processes than in determining the positive factors. (Author/MV)