Reported are the results of an investigation of the effects of programing a computer in a seventh grade mathematics class. Two treatments were conducted during two successive years. The students in the treatment group used the programing language BASIC to write computer algorithms following supplemental instruction. The mathematical content was the same for the treatment group and the control group. Measures were made on both author-produced tests and commercial standardized tests. A measure of previous achievement in mathematics determined three levels within the groups. At the end of the first year, the measure of the knowledge of numeration systems was the only one found to be significant, and favored the control group. Supplementary materials and tests were revised for use during the second year. At the end of the second year, three of the twelve tests showed significant differences, all favoring the treatment group. It was felt that the results supported computer-assisted problem solving in grade seven mathematics, particularly for the study of number theory and word problems, and particularly for students identified as high or average in achievement. (RS)