The purpose of this study was to determine the effects of using calculators on the mathematical achievement and attitude of low-income sixth-grade students when they were taught problem solving. Thirty-five sixth-grade students in two intact classes were assigned to two treatment groups. The experimental group was taught problem solving using calculators. The control group was taught using paper-and-pencil. Treatment for both groups lasted eight weeks. The results of this study reveal that there was no significant difference in mathematical achievement gain in problem solving of students who were taught using calculators and those who were not. The students who were not taught using calculators had a significantly better attitude towards problem solving than those who were using calculators. Implications for the classroom and future research were discussed. Appendices included are: (1) the mathematics test and attitude survey including scoring system; (2) instructional plans; (3) "t" calculation; and (4) raw data. (YP)