Risk taking on behalf of others - the importance of uncertainty revelation

How to cite

Sørensen, Erik Ø.; Cappelen, Alexander W.; Tungodden, Bertil; Xu, Xiaogeng. 2019. "Risk taking on behalf of others - the importance of uncertainty revelation." AEA RCT Registry. AEARCTR-0004403. https://doi.org/10.1257/rct.4403-2.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Erik Ø. Sørensen* , Alexander W. Cappelen , Bertil Tungodden , Xiaogeng Xu

* Primary investigator

Registry ID
AEARCTR-0004403
DOI
https://doi.org/10.1257/rct.4403-2.0
First registered
Last updated
Status
Completed
Trial dates
2019-07-03 – 2019-10-01
Countries
Norway (Norway)
JEL codes
D81
Secondary IDs
RCN 262675
Registry record
http://www.socialscienceregistry.org/trials/4403

Abstract

Much decision making is decision making on behalf of others, and it is not and there are competing theories of how such decisions should be made. When making decisions on behalf of others, it might also be that the decision maker will never learn the outcomes of the decision only with delay, or possibly not at all. We explore how decisions on behalf of others is made in a setting where the revelation of uncertainty is either immediate, delayed, or it will never happen. Data from a large scale online experiment is analyzed within hierarchical Bayesian model of rank-dependent utility.

Experimental design

Participants are recruited from the general population, via a commercial survey agency. On research team platform, participants choose between risky prospects and safe outcomes for random draws of "recipients" (that are recruited via an online labor market). Four treatments differ in what decision makers are told about how they will be informed about outcomes when risky alternatives are chosen (different points in time). Post decision making, a few survey questions are asked of the participants. The recipients are recruited and paid after the the final revelation of uncertainty to the decision makers (and decision makers are informed about this).

Design details

Randomization method
Individual level randomization (by computer, dynamically during survey) into one out of four treatments. Post data collection, a random subset of decisions are drawn (by computer) to determine earnings of recipients (In total, 1/5 recipients per decision maker).
Randomization unit
Individual
Clusters
2000 decision makers
Observations
2000 decision makers,
Arms
500 decision makers in each of the 4 treatments
Primary outcomes
The choice of a risky alternative vs a safe alternative. Each individual draws 4 out of a list of 10 risky alternatives, and for each risky alternative chooses between this and 7 safe alternatives, for 28 binary decisions from each individual.
Secondary outcomes
1) Subjective measures of own risk tolerance (1-7 scale), 2) beliefs about average risk tolerance in the study (1-7 scale), 3) own tendency to support social initiatives (1-7 scale). 4) Self reported feeling during experiment (subjective expression) 5) Basic demographics

Materials

Cited by

No citation data available for this registration yet.

Keywords