Understanding the mechanisms underlying group decision-making is a cornerstone of social psychology. In order to shed light on these mechanisms, group researchers typically conduct mediation analyses. In my talk, I will argue that almost all previous studies in group research fail to establish causality when conducting mediation analyses. This is because they typically rely on measurement-of-mediation designs which means that they measure the proposed mediating variable. This is ironic because, after all, talking about mediation is assuming causality. Thus, it is assumed that the independent variable causes a change in the mediator that causes a change in the dependent variable. However, we cannot test for causality if we only measure the mediating variable. While not being a panacea, one way to address this issue is to conduct manipulation-of-mediator designs. In my talk, I will give several examples on how to conduct manipulation-of-mediator designs in group decision-making research.