Group identity effects on social influence

This experiment investigated the effects of different group identities on social influence in a sample of 294 predominantly first-year college students. The primary aim of this study was to test whether forming social identities on different criteria along the trivial-important dimension of group definition has significantly different effect on individuals\u27 conformity to their group\u27s majority opinion in response to social dilemmas. The second aim of this study was to investigate whether conformity to group norms occurs in computer mediated contexts, in which a person does not know the identities of his or her group members and is not in face-to-face contact with them. For this study, participants were randomly assigned to three groups that were distributed along the trivial-important dimension of social identity and were presented with three social dilemmas in a computer-mediated context. Each scenario ended with a question regarding what the protagonist should do to solve the dilemma. Before they could give their individual response, participants were exposed to the responses of their group members that formed a consensus regarding the solution. A 3 x 2 x 2 x 3 Mixed ANOVA was conducted for the main analysis. Results indicated that participants in all three groups responded to the questions presented at the end of the dilemmas in line with their group\u27s majority opinion. This finding supported the second hypothesis by showing that individuals were influenced by their group\u27s majority opinion. However, the results also showed that conformity to the group norm was strongest in the “Minimal Group” and weakest in the “Attitude Group” contradicting the prediction made in the first hypothesis of this study. In other words, conformity to the group norm was strongest within the group that was formed based on a trivial criterion, but weakest within the group that was formed based on a seemingly important criterion. Implications of the study findings for social influence research as well as study limitations and recommendations for future studies are discussed