This paper reports the results of a ten-week study investigating the use of electronic calculators to perform basic operations with rational numbers in ninth-grade mathematics classes. In all, six classes and 101 students were involved. A short literature review is given and three treatments are described. Dependent variables were five mastery tests, a transfer test, a retention test, and an attitude survey. Use of an "alternative algorithm set" with the calculator appeared to be a viable alternative to use of the "conventional algorithm set" with or without the calculator. Other results, conclusions, and educational implications are presented. (LS)