An experimental test of collegiate drinking norms

Social norms play a pivotal role in both explaining the development and maintenance of collegiate alcohol use and creating prevention and intervention programs targeted at reducing heavy drinking. By theoretically functioning as a model of normative and popular behavior, descriptive and injunctive norms are consistently associated with college drinking. In the current study, we endeavored to test the mechanisms through which social norms influence drinking by experimentally manipulating normative beliefs. Participants (N = 181) were assigned to one of nine conditions in a 3 (descriptive norms (DN): positive, negative, none) x 3 (injunctive norms (IN): positive, negative, none) experimental design. Norms exposure occurred within a series of three same-gender Internet-based chat room sessions. The norms manipulation was partially successful in creating groups with distinct normative beliefs, with the no norms groups failing to maintain a neutral norm for both descriptive and injunctive norms. Consequently, no descriptive norms groups were combined with positive descriptive norms groups and no injunctive norms groups were combined with negative injunctive norms groups, resulting in a 2 (DN: positive, negative) x 2 (IN: positive, negative) design for analyses. Overall findings for type (DN, IN) and valence (positive, negative) of norms indicated that participants globally reduced descriptive norms and drinking from pre-chat room to post-chat room, regardless of the type or valence of the manipulation, indicating that there were no experimental effects by condition. Whereas drinking appeared to stabilize at post-chat room, descriptive norms continued to decrease by three-month follow-up. Injunctive norms and personal attitudes about alcohol use also decreased by three-month follow-up. Although we were unsuccessful in changing normative beliefs in expected directions, these findings have important implications for college prevention and intervention programs for reducing drinking. The lack of experimental effects suggested that changing norms may be more complex than previously hypothesized and that changes in norms may not result in changes in drinking, which is the purported mechanism of change in norms-based interventions. These results further suggested that continued research is necessary to provide empirical support for a causal link between norms and drinking and that alternative explanations for the association between norms and drinking need to be considered.