Social influence in subgroups from superordinate group polls

Abstract

Polling can influence public opinion—polls convey normative information people are motivated to follow. Polls are often presented as reflecting the larger population (superordinate group)—frequently representing high-status subgroups. Marginalized groups are treated as exceptions to the norm in need of explanation. The current work examines if low-status subgroup members perceive their group as represented in national polls and the consequences of perceived representation for following superordinate group norms (bandwagoning). I expected: 1) Low-status subgroup members would identify with the superordinate group less than high-status subgroup members, 2) Low-status subgroup members would perceive less subgroup representation in national polls than high-status subgroup members, 3) Low-status subgroup members would be less likely than high-status subgroup members to bandwagon, 4) Norm adherence would depend on perceived representation and identification, with low-status subgroup members bandwagoning less than high-status subgroups members to the extent they see their group as underrepresented and less identified with the superordinate group, and 5) Subgroups would be less likely to bandwagon when the issue was relevant to the group’s interests. Participants viewed a national poll in which the majority of respondents either supported or opposed a position (Studies 1-2). Participants rated support for the issue and perception of racial groups’ representation in the poll. In Study 3, participants viewed a similar poll, but viewed poll sample gender composition information—either women or men were a majority of respondents. In all studies, low-status (vs. high-status) subgroup members had equal (Studies 1-3) or higher (Study 2A) national identification. Participants viewed low-status subgroup members as less represented than high-status subgroups (Studies 1-2) or perceived representation as manipulated (Study 3). Reliable general bandwagon effects emerged in all three studies. Counter to expectations, low-status subgroup members bandwagoned equally to or more than high-status subgroups members (Studies 1-3; but see exceptions Study 2A/B). Perceived subgroup representation did not reliably influence bandwagoning (Studies 1-3). In Study 3, women were more likely to bandwagon and perceived relevance decreased bandwagoning compared to men for a less gender-relevant issue. Results offered mixed support for hypotheses—subgroup representation did not reliably predict bandwagoning, but relevance may help explain inconsistencies.