The occurrence of unrealistic optimism with regard to a genetic risk situation was investigated within a group of female adults (study 1) and a group of adolescents (study 2). In both studies, the indirect method of measurement elicited a significant optimistic bias. Contrary to Weinstein (1982, 1987) we found no relation between the direct or indirect measures of unrealistic optimism and personal experience with the risk, perceived frequency of occurrence or perceived preventability of the risk. Trait-Anxiety was significantly related to the extent of unrealistic optimism, but only when the direct measure was used. Together with the fact that the indirect measure resulted in a much stronger bias than the direct one, this suggests that there exists an important difference between both measures. Further research on the measurement of unrealistic optimism and on its determinants in specific risk situations is needed.