Prediction of and Recommendation for Honest Behavior

Abstract

Given a hypothetical situation in which a stimulus person was described as having picked up money dropped unknowingly by another person, 480 undergraduate Indian students at the University of Bombay predicted the likelihood of honesty (returning the money to its owner) and also indicated the strength of their recommendation for honesty. The experiment had a 4 × 2 × 2 × 2 (Role × Location × Level of Temptation × Subject Sex) factorial design. Perceived likelihood of honesty was greater and recommendations for honesty were stronger when the location was crowded rather than deserted. Female subjects perceived a lower likelihood of honesty and gave stronger recommendations for honesty than male subjects. Perceived likelihood of honesty was higher and recommendations for honesty were stronger for the roles of policeman and bystander compared with those of beggar and pickpocket. These and other findings were interpreted within the framework of the subjective expected utility model.