How does choice affect beliefs?

How to cite

Hajdu, Gergely; Krusper, Balázs. 2020. "How does choice affect beliefs?." AEA RCT Registry. AEARCTR-0005974. https://doi.org/10.1257/rct.5974-1.0

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

Gergely Hajdu* , Balázs Krusper

* Primary investigator

Registry ID
AEARCTR-0005974
DOI
https://doi.org/10.1257/rct.5974-1.0
First registered
Last updated
Status
In development
Trial dates
2020-06-18 – 2020-07-18
Countries
Private
JEL codes
C91, D91
Registry record
http://www.socialscienceregistry.org/trials/5974

Abstract

People tend to be optimistic about their own future. For example, compared to rational beliefs,investors typically expect higher returns from their funds. In line with these observations, previous research has documented that owning a product leads to more optimistic beliefs about its value.However, in many important economic contexts, people choose the products for themselves (e.g.they decide to invest in a particular fund). We propose and experimentally test the explanation that, beyond the effect of ownership, making a choice can lead to additional belief distortions. Using a between-subject design, we compare a person who chooses a product to a person who receives the same product exogenously. This comparison controls for the effect of ownership on beliefs. To address the key concern of endogenous choice, we vary the choice set to create predictable variation in choice.

Intervention

The main task of participants is to estimate the payoff probabilities of made-up financial portfolios. The main intervention is that participants either receive one of the portfolios (Allocation condition) or choose one of the portfolios (Choice condition). We inform participants that they will get a reward if the portfolio they own– chosen or received – pays off. Besides the main interventions, we create two additional treatments to study the mechanism. In the Relative Choice condition, participants have to submit their estimates before the choice. In the Ego Choice condition, participants read a short text about the positive relationship between IQ and the quality of financial decision making.

Experimental design

Participants complete four rounds. In each round, they are asked to estimate payoff probabilities of made-up financial portfolios. They receive a signal that helps them to find out the relative ranking of the payoff probabilities while the absolute levels remain uncertain.

Design details

Randomization method
Randomization is done by the computer. First, individuals are assigned to one of the treatment groups. Second, in each round, one of the portfolios is randomly selected for individuals in the Allocation condition.
Randomization unit
Individual.
Clusters
1000 individuals.
Observations
8000 individual-portfolio pairs..
Arms
350 individuals Allocation, 350 individuals Choice, 150 individuals Relative Choice, 150 individuals Ego Choice.
Primary outcomes
Beliefs about payoff probabilities

Cited by

No citation data available for this registration yet.

Keywords