To determine a cost effectiveness relationship between instructional costs and student achievement gains in reading, six programs in the Grand Rapids, Michigan, Public Schools were studied. Subjects were engaged in three performance contracting programs --Alpha II, Westinghouse Learning Corporation (WLC), and Combined Motivation and Educational Systems (CMES); Project Read; a traditional remedial (TRR) program; and the regular (control) program. Among the study's major conclusions are: (a) in grades 2 and 3 (all programs but CMES were in operation) WLC was least expensive per one-tenth grade gain, and TRR produced the highest overall mean gain at a considerably higher cost; (b) in grade 7 (all but WLC and Project Read were used) CMES and Alpha II were the most cost effective, and TRR was the most expensive per one-tenth achievement gain; (c) in grades 8 and 9 (only Alpha II, CMES, and the control operated) the control was most expensive per one-tenth grade gain; gains for Alpha II and CMES were slightly above one grade level, whereas the control produced about one-third that gain. (RD)