AbstractWe compare and critique two measures of risk perception. We suggest that a single question — “How risky is the situation?” — captures the concept of risk perception more accurately than the multiple-item measure used by Sitkin and Weingart (1995). In fact, this latter measure inadvertently captures notions of attractiveness or expected return, rather than risk perception. We further propose that the error underlying the construction of Sitkin and Weingart’s measure is explained in terms of a top-down model of risk perception, in which perceived risk and perceived return are consequences, rather than determinants, of attractiveness. Two studies compare the validity of the two alternative measures.