Objective: The present study investigated the extent to which direct (D) and indirect (ID) measures of unrealistic optimism truly capture people's beliefs about their comparative risk status.Method: Respondents (N= 425) were contacted in a general population survey. The D measure required respondents to estimate whether their personal chance of contracting an illness was above/below the average for people of the same age/gender. The ID measure was based on separate assessments of personal/others' chances.Results: Poor concurrent validity between the measures implied that one or both was tapping into perceptions unrelated to comparative risk. This was explored by examining the relationships between the measures and various health problem attributes (e.g. perceived preventability, personal experience). A picture emerged that suggested the D measure did not seem to reflect the effects of social comparison as much as the ID measure, a crucial criterion for a measure intended to elicitcomparativerisk ratings.Conclusion: When measuring unrealistic optimism by asking ‘Compared to other men/women your age how would you rate your chances of X?’ researchers should be sensitive to the possibility that the question is being interpreted simply as ‘How would you rate your chances o