A Social Comparison Analysis: Differentiating Losses and Gains

Social comparison involves considering information about others in relation to the self, whether that information is encountered incidentally or actively sought. This thesis focuses on the active pursuit of information that can provide a social comparison standard, termed social comparison seeking. The theoretical background brings together social comparison theory, research on information seeking, and gain–loss framing to motivate the central research question: Do gain and loss domains influence social comparison seeking? An online within-subjects experiment examined this question with 140 participants in a roulette game. Participants completed gain-framed roulettes starting from $0 and loss-framed roulettes starting from $60, with matched possible final monetary amounts across domains. After seeing their own outcome on each roulette, they could move on or sequentially request outcomes attributed to previous participants. These requests could not change their own monetary result. The task distinguished seeking initiation (whether any outcome was requested) from conditional extent (how many outcomes were requested once seeking had begun). The study predicted less seeking in the loss domain, proposed identification with the other participants as a mediator of this effect, and predicted greater seeking when upward rather than downward comparison was possible. Momentary happiness was lower after the loss than after the gain block, indicating the expected affective difference between domains. Nevertheless, confirmatory analyses provided no evidence of a gain–loss difference in either seeking initiation or conditional extent under the conditions examined. Identification did not differ between domains, and the proposed indirect effect through identification was not supported. Higher identification was instead associated with greater initiation, contrary to the prediction. Seeking was also more likely to begin when upward comparison was possible, providing partial support for the third hypothesis; no corresponding association was found for conditional extent. By adapting earlier comparison-selection and information-seeking approaches into a sequential task, the study addresses an unresolved question. Its findings illustrate the value of analyzing initiation and conditional extent separately, without establishing distinct psychological mechanisms.