Gain-loss domain and social value orientation as determinants of risk allocation decisions

AbstractPeople often make less risky decisions for themselves than others. We examined how people allocated risks (i.e., determining the ratio of uncertain outcomes to certain outcomes) between themselves and others. We also investigated gain (vs. loss) domain and social value orientation as predictors of risk allocations. The results of three experiments demonstrated that participants were more likely to share their risks equally between themselves and others than distribute risk unequally. In the gain (vs. loss) domain, participants allocated fewer risks to themselves and more risks to the other person for unequal risk allocations. Compared to proselfs, prosocials were more likely to allocate risks equally. We also found stronger domain effects on unequal risk allocations for proselfs than for prosocials. Therefore, our findings clarify the effects of risk distribution, domain, and social value orientation on interpersonal allocation decisions and highlight equal risk distribution between oneself and others.Keywords: Allocation decisionrisk distributiongain-loss domainsocial value orientationself-other Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 For the options of risk allocation, a percentage refers to the proportion of an uncertain outcome to a total outcome allocated to oneself or another individual.2 We declare that there is no conflict of interest. We also confirm that the manuscript adheres to ethical guidelines specified in the APA Code of Conduct and our national ethics guidelines. Please see the data and analysis codes at https://osf.io/zmwfc/?view_only=95de83989f50460c97477aa6cb5239b0.3 We conducted power analyses to demonstrate that our sample sizes have sufficient power in Experiments 1-3. We also ran additional analyses to demonstrate the even distribution of sample sizes across different conditions before and after filtering the data in Experiments 1 and 2. These are presented in the supplemental materials.4 We followed the incentive instructions to pair participants, calculate each participant’s final points, and awarded the three participants accordingly in each condition with the incentive instructions (in Experiments 1 and 2).5 We also ran additional analyses without the controls for the differences in the items and choices, and the results demonstrated consistent patterns between the results with and without controls, which suggests that the differences in the first-stage allocations did not significantly influence our results. Please see the relevant results in the section titled “Results without Controls for the Allocation Outcomes During the Initial Decisions in Experiments 1 and 2” in the supplemental materials.6 The percentage refers to the average difference of the selection likelihood of a specific risk-distribution option between the gain and loss conditions.