Never…Often? Comparisons that Shape People’s Likert-Type Ratings of Behavior Frequencies

Responses to Likert-type behavioral frequency (LBF) questions often do not consistently map onto objective numerical estimates. Prior research suggests that social and other comparisons may underlie this divergence, but the relative influence of different comparison standards—and the cognitive processes supporting them—remains unclear. Across two studies, we examined how comparisons to peers, averages, experts, past selves, and conceptually irrelevant standards shape LBF responses for common health behaviors (e.g., hand washing, flossing). Participants provided LBF judgments, absolute frequency estimates, and comparative judgments for each behavior. Study 1 showed that direct comparisons predicted LBF judgments above and beyond participants’ own absolute frequency estimates, with comparisons to experts and average others being especially influential. Even when controlling for shared methodological variance, all comparison types explained unique variance in LBF responses. Study 2 replicated this pattern of results. Moreover, additional analyses in Study 2 suggest that participants were not making precise, pairwise comparisons between numeric estimates, but were instead relying on more abstract, gist-like impressions of how their behavior compared to others’. Together, these findings underscore the importance of considering the comparative and interpretive nature of self-report measures, particularly in contexts where behavioral frequency carries social, normative, or evaluative meaning.