Unhealthy and unaware? Misjudging social comparative standing for health-relevant behavior

Abstract People sometimes modify their behavior based on whether they believe they do more or less of that behavior than others. But are people's perceptions of their social-comparative status for behaviors generally accurate? The current research assessed accuracy and bias in perceived social-comparative status for a number of health-related behaviors. In two studies, participants estimated their social-comparative percentile regarding behavior frequency for 20 behaviors—pre-classified according to a 2 (healthy or unhealthy) × 2 (generally common or uncommon) design. Participants also reported their absolute frequency of engagement for the behaviors, and these reports were used to approximate people's actual percentiles. Subjective percentile estimates were overly favorable for both healthy and unhealthy behaviors and were biased by a behavior's general commonness/rarity. People who were least healthy regarding a behavior tended to be the most miscalibrated in their percentile estimation for that behavior. There was also support for a noise-plus-bias model of people's percentile estimates.