Rich Mixture Set, Process Preference, and Home Bias - Testing Preference for Decisiveness of the Preferred Source (PDPS)

How to cite

Kader, Gavin; Chew, Soo Hong; Wang, Wenqian. 2024. "Rich Mixture Set, Process Preference, and Home Bias - Testing Preference for Decisiveness of the Preferred Source (PDPS)." AEA RCT Registry. AEARCTR-0014036. https://doi.org/10.1257/rct.14036-1.1

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

Gavin Kader* , Soo Hong Chew , Wenqian Wang

* Primary investigator

Registry ID
AEARCTR-0014036
DOI
https://doi.org/10.1257/rct.14036-1.1
First registered
Last updated
Status
Completed
Trial dates
2024-06-15 – 2024-06-22
Countries
China (Chengdu, Sichuan)
Registry record
http://www.socialscienceregistry.org/trials/14036

Abstract

The primary purpose of this experiment is to test the theoretical predictions of our paper which are all based on rich mixture sets. Imposing standard axioms on this flexible choice domain gives rise to a parsimonious representation of source recursive expected utility (SREU). Under the resulting SREU, if a decision-maker exhibits source preference (a consistent preference for equivalent lotteries, differing only in the source of resolution), then the decision-maker prefers more uncertainty being resolved by the preferred source. We refer to this as preference for decisiveness of the preferred source (PDPS). The hypothesis is to test whether those with source preference also have PDPS.

H0: Random behaviour i.e. proportion of subjects satisfying PDPS is 50\%
H1: Proportion of subjects who satisfy PDPS is more than the proportion of subjects who do not i.e. proportion of subjects satisfying PDPS is greater than 50%

Another hypothesis we can test is to see whether we observe more PDPS for those who prefer to flip the cards themselves, compared to their partner flipping to see if this is consistent with the literature on the illusion/locus of control.

Experimental design

We test whether PDPS holds i.e. whether subjects have a preference for their preferred source serving a more decisive role. This means getting those who have consistent preference for a source (flipping themselves or their partner) and checking how many choose to have their preferred source in the first stage of the rich lotteries in the PDPS task.

Design details

Randomization method
Randomisation done by a computer in the lab
Randomization unit
Individual level randomisation
Clusters
160 subjects
Observations
160 subjects
Arms
160 subjects
Primary outcomes
Subjects are asked to complete two types of tasks: i) source preference tasks, and ii) a PDPS task. Both involve flipping face-down cards (red or green) and two sources of risk: i) flipping themselves, or ii) partner flipping. For a source preference task, there is one red card and one green card. Subjects choose whether to flip a card themselves or have their partner flip a card. The payoff distribution is identical regardless of who flips the card. However, subjects receive the higher payoff if they flip the red card themselves, and the lower payoff if their partner flips it. Subjects complete two source preference tasks, with payoffs differing between tasks. We say a subject exhibits source preference if they consistently choose the same source. For a PDPS task, there is one red card and two green cards. Subjects choose whether to flip a card themselves first (and their partner second) or have their partner flip a card first (and themselves second). The payoff distribution is identical regardless of the order in which cards are flipped. Subjects receive the higher payoff if they flip the red card themselves, and the lower payoff otherwise. If the red card is never flipped, subjects receive an intermediate payoff. Subjects complete one PDPS task. Their choices in each task serve as the variables used for the main analysis.

Cited by

No citation data available for this registration yet.

Keywords