It has been suggested that a procedure based on the logic of Bayesian probabilities would make it possible to assess individual differences in stereotyping. Given the possible advantages of using the McCauley and Stitt (1978) procedures to measure individual differences, three groups of college students were tested to see if they would use Bayes rule appropriately in responding to the measure. Results indicated that the basic assumption underlying the use of the McCauley and Stitt (1978) procedures was being met and that subjects used Bayes appropriately. The correlations of the estimated percentage of trait given sex with both base rate (the overall frequency of the trait in the population) and with representativeness (the frequency of a particular sex showing a specific trait) were all large and significant. The findings suggest that a measure of sex-role stereotyping based on estimations of conditional probabilities is viable. (JAC)