ANOVA and Profile analysis were employed to compare the computational skills of high and low ability sixth graders during a decade of efforts to individualize instruction. Comparison of lowest ability subgroups demonstrated no differences with statistical significance in their computational skills. There were statistically significant differences favoring the 1965 SMSG highest I.Q. subgroup in fraction and decimal addition, subtraction and total. Statistically significant differences favored the 1975 highest I.Q. subgroup in decimal division. Attempts to meet individual differences appear not to have been effective. (Author)