Abstract
The ranges of response alternatives presented in frequency scales influence respondents’ behavioral estimates. This research aimed at complementing the existing cognitive interpretations of this scale effect (e.g., Schwarz, 1994). We propose a normative interpretation, according to which targets associated with generic behavioral norms, and with ingroup norms, lead participants to rely more strongly on the scale’s frequency properties. Studies 1–2 showed stronger scale effects when participants reported behaviors over an extended (vs. short) time period, specifically when they reported behaviors of themselves (vs. people in general). Studies 3–6 showed that the use of a scale’s properties in self-reports increased when participants were led to believe that the scale’s frequency alternatives mirrored typical ingroup (vs. outgroup) behaviors. Finally, Study 7 demonstrated the role of ingroup identification in the production of these scale effects. Collectively, these findings suggest that behavioral estimates based on shared norms override memory scanning when behavior is difficult to retrieve, and when relevant ingroup information is available.