Positive-Negative Asymmetry of Disconfirmations on User Satisfaction Judgment

Past research in the area of user satisfaction has primarily adopted the conventional “key-driver analysis” approach under the implicit assumption that either positively or negatively valenced events would have similar impact on user satisfaction. Researchers in other disciplines have already found that the links in satisfaction models are more complex than originally proposed. By modeling the link between satisfaction and its antecedents as symmetric and linear, researchers run the risk of systematically misestimating the impact of the antecedent variables on user satisfaction. Thus, this study is a step forward towards incorporating the concept of positive-negative asymmetry into user satisfaction research. Specifically, the purpose of this study is to explore both the impacts of positive and negative disconfirmations on user satisfaction. Building upon previous work, we empirically tested an end-user satisfaction model in the context of e-portal usage. Consistent with the findings of other disciplines, our results support the argument that negative disconfirmation has a stronger impact on satisfaction than positive disconfirmation. Through recognizing the asymmetric cognitive responses underlying satisfaction, we believe that this study has further advanced cognitive research in general and theories in user satisfaction research in particular, and provided new insights to practitioners on design priorities. Implications and future research are addressed.