The illusion of control and optimism about health: on being less at risk but no more in control than others.

If unrealistic optimism is simply a manifestation of the illusion of control (McKenna, 1993), differences in estimates of the likelihood of events for the self and others (comparative optimism) should have concomitant differences in perceived capacity to control these events. This possibility was tested in the domain of health, along with Perloff & Fetzer's (1986) risk-factor version of this downward comparison hypothesis and the hitherto neglected possibility that unrealistic optimism is an artifact of the comparative procedures employed in studies to date. It was found that likelihood judgements for a range of illnesses tended to be significantly lower when made by subjects providing self-ratings than when made by those providing judgements for the 'typical' student and, to a lesser extent, those making judgements for a friend's friend, but not by those doing so for an acquaintance. This comparative optimism was associated with equivalent differences in perceptions of these targets on a selection of relevant risk factors, but not in ratings of their capacity to control these outcomes (for which there was consensus across targets). The data are thus consistent with the risk-factor version of the hypothesis rather than the control version or the artifact explanation. The theoretical implications of these findings are addressed along with their consequences for health promotion.