From six institutions in the Illinois Department of Corrections school district, 359 adult male inmates served as subjects for this research evaluation of basic skills. Three groups were in traditional self-paced instruction, and the other three groups were in Program Logic for Automated Training Operations (PLATO) computer managed instruction (CMI). Treatment effects were measured on the Test of Adult Basic Education and the Tennessee Self-Concept Scale. Significant pretest to posttest gains were found for all students in language, mathematics, and reading. Results may have been influenced by the loss of more than half the sample between pretests and posttests because of inmate mobility. The results showed that evidently no simple way exists to predict which students will gain the most from either classroom or CMI instruction, given information about the student's age, ethnicity, type of sentence, and security level of his institution. Also, measuring gains is difficult. Finally, it was determined that further research is needed to understand how and why the two instructional delivery systems worked and, if possible, to determine which method works best for which individual. (Author/KC)