This study investigated the effect of a direct instruction program in fractions, on achievement and self-concept. A pretest-posttest, quasi-experimental, nonequivalent control group design was used to compare the two groups of upper elementary school students (N=337). Treatment group students received instruction in fractions via a teacher-directed, videodisc-based program. The control group received their normal grade-four or grade-five mathematics program. Results indicated that the treatment group mean on a criterion referenced test was higher than that of the control; differences in achievement test scores among the treatment classes varied directly with the level of program implementation; and the scores of the treatment group classes on the self-concept instrument were higher than those of the control group classes. A list of 31 references is included. (Author/CW)