The degree to which people perceive that their health-relevant behaviors are also performed by their peers may influence whether they continue these practices or are susceptible to change. The present paper examined estimates of social consensus for health-relevant behaviors. It was hypothesized that respondents would perceive their own behaviors to be relatively more common than do people not performing them, the so-called False Consensus Effect. However, in terms of the accuracy of consensus estimates, people who behave in undesirable ways will tend to overestimate the actual number of others who behave similarly. In contrast, people who behave in desirable ways will underestimate the actual number of people who behave like themselves (false uniqueness). A group of college-aged males were asked to report about their performance of a series of health-relevant behaviors and to estimate the frequency of each behavior among their peers. The findings strongly supported the first two hypotheses and provided some support for the third. One practical implication of the results is that persons with unhealthy practices may resist public health campaigns and other interventions by overestimating consensus for their own behaviors. Also, individuals who think their undesirable health behaviors are relatively common may believe that they involve few health risks for them. The research suggests the need for further research on social cognitions about health, illness, and health-protective behaviors.