This study was designed to test the hypothesis that use of a probability unit in ninth-grade general mathematics classes would yield student improvement in computational skill, arithmetic reasoning, and attitude toward mathematics. A 47-page unit on probability (developed for this study) covered basic terminology, probability of events, favorable, and unfavorable outcomes of dice tossing, counting procedures, discussion of games of chance, and other topics. Three games, also developed for the study, were played twice a week to provide students with an opportunity to apply the probability concepts learned, and practice computational skills. The experimental and control groups each consisted of sixth ninth-grade classes; control classes used their regular instructional sequence. Pretests indicated no significant differences between groups on the criterion measures. After six weeks of instruction posttests were administered, including a test of knowledge of probability. Results did not support the hypotheses of the study. The only differences obtained were superior performance on the probability test by the experimental group (p less than .05), and improvement by experimental subjects on addition. (SD)