How Few Can Be More Powerful Than Many: The Effect of Unbalanced Sampling on Norm Perception
(A2023-114459)

AUTHORS

Gaël Le Mens, Pompeu Fabra University; Rahil Hosseini, Universidad Carlos III de Madrid; Thomas Woiczyk, University of the Balearic Islands

ABSTRACT

We study how people learn group norms based on a sample of behaviors by group members. We hypothesized that norm perceptions could be explained by the central tendency of the observations in the sample (majority or average behavior), or by the central tendency of the typical behavior of the group members (behavior of the majority). Parsing out between these two hypotheses is important in settings where people obtain larger samples of observations about some group members than others. In three online studies with Prolific participants (N = 1,851), we found that norm perceptions were better explained by observation-level central tendencies (average or majority behavior) than by member-level central tendencies (behavior of the majority of group members). We also found that the observation-level central tendency better explained the consequential behavior of participants. Our findings cast new light on how social influence operates in social groups and on social media.