Decision making for others involving risk: A review and meta-analysis

Abstract Are choices for others riskier than choices people make for themselves? This question has been asked by economists, psychologists, and other researchers in the social sciences – which has generated a diversity of research accounts and results. For example, a number of studies have found strong instances of a risky shift in choices for others, while other studies have found no such effect or have found that choices for others instead generate a cautious shift. In a meta-analysis of 128 effects from 71 published and unpublished papers (totaling 14,443 observations), we found a significant though small effect size (d = 0.105) in favor of a risky shift when people choose for others. Moreover, we found considerable variance between studies (Q = 1106.25), suggesting that the net effect is susceptible to moderating factors or study characteristics, which we identify and discuss as well (viz. choice recipient, decision frame, decision reciprocity, theoretical perspective, study design). Thus, we document not only whether decisions for others are riskier, but when (and when such decisions are less risky). We further discuss what is distinctly unique about decision making for others – how such choices are not just different in degree from personal choices but different in kind.