Experiments showed a scarcity principle in evaluative judgments such that the identical characteristic is evaluated more extremely the lower its perceived prevalence. In Study 1, Ss evaluated a fictitious medical condition that was described as either beneficial or detrimental to health and as occurring in either 30% or in one half of 1% of a test population. The condition was evaluated more extremely--as as a more positive health asset or a more negative health liability--in the low-prevalence than in the high-prevalence conditions. Study 2 demonstrated the same effect in self-evaluations and with a different manipulation of perceived prevalence. Ss were told that they actually had the fictitious medical condition, that it was either beneficial or detrimental to their health, and either that they were the only 1 of 5 Ss who had it or that 4 of the 5 did. Low-prevalence Ss exhibited more extreme evaluative, affective, and behavioral reactions to the medical condition than did high-prevalence Ss. The origins and validity of the scarcity principle are discussed, as are its implications for uniqueness theory, reactance theory, and social evaluation theories.