The exclusive focus of existing research on the compromise effect has been on the individual. This paper investigates compromise effects in a dyadic choice setting. We investigate whether dyads exhibit compromise effects, factors that may moderate this effect and strategies that could be used mitigate it. We build a statistical model of dyadic choice that incorporates individual preference and influence and accounts for compromise effects-individual and dyadic-by transforming context-independent utility into a more concave context-dependent utility. We conducted two studies to test our proposed models empirically. Study 1 begins with an investigation of individual compromise effect (ICE) and its moderators. In Study 2 we test for the presence of dyadic compromise effect (DCE) in which married couples are asked to make retirement investment choices. We find strong empirical evidence in support of DCE-models incorporating DCE provide better fit than models that do not. We find that ICE tendency of a group member with a greater stake in the decision results in greater DCE. Our findings suggest that choice sets with fewer options mitigate DCE. Further, education of segments most vulnerable to compromise effects (e.g. women in the context of retirement planning) could be effective to mitigate DCE.

No abstract

Abstract