Evidence of Deliberate Violations of Dominance Due to Secondary Satisfaction - Attraction to Chance

This paper concerns the empirical significance of secondary satisfactions - attractions - to chance - and whether 95 successful highly educated people consider that these should influence rational decisions. It analyses their answers to a hypothetical agency decision situation. Over 50% of academics in economics and other decision sciences chose a transparently stochastically dominant act for the sake of their principals' secondary satisfactions. There was not a single participant whose answers unambiguously supported the view that people should always prefer stochastically dominant acts. The findings suggest that focussing biases underlie (i) the exclusion of secondary satisfactions from decision models and (ii) the presumption that violating the stochastic dominance principle is irrational.