Investigating Motivations for Information Avoidance - The Role of Certainty, Rewards and Overconfidence

How to cite

Meißner, Stefan; Ay, Fehime Ceren. 2018. "Investigating Motivations for Information Avoidance - The Role of Certainty, Rewards and Overconfidence." AEA RCT Registry. AEARCTR-0003473. https://doi.org/10.1257/rct.3473-1.0

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

Stefan Meißner* , Fehime Ceren Ay

* Primary investigator

Registry ID
AEARCTR-0003473
DOI
https://doi.org/10.1257/rct.3473-1.0
First registered
Last updated
Status
Completed
Trial dates
2018-10-22 – 2018-10-23
Countries
Norway (Hordaland)
JEL codes
D82, D83
Registry record
http://www.socialscienceregistry.org/trials/3473

Abstract

This study investigates different motivations for information avoidance. For this purpose, participants take part in an intelligence test in which their results are compared to other participants. All participants are member of the same cohort of a Business School. Participants are asked to guess the share of others who performed the test better than them and are paid based on the accuracy of that guess. They are informed that they can find out the exact share after they made their initial estimate. If they find out their initial guess, they can revise their previous guess, ensuring full payout from the experiment but potentially also finding out unpleasant information about themselves.
We investigate the fraction of participants who actively avoid that information, thereby foregoing earnings from the experiment. We furthermore test for different motivations that these participants could have for that decision.

Experimental design

First, participants take an IQ test and they are informed that that test was taken from a longer test. After taking the test they are asked to guess their rank compared to their peers in the session and if their guess is correct they are going to win 80 NOK. After making their guess they are assigned to two treatments randomly: costly information and costly avoidance. In both treatments a Becker-DeGroot-Marschak(BDM) (Becker et al., 1964) auction takes place to implement the participants’ decision. The aim of having a BDM auction is to elicit their real preferences and willingness to pay for that. A bonus payment of 50NOK will be introduced and they are asked to submit how much they would be willing to pay for their decision (getting/avoiding information)to be implemented. The submitted price is compared to a randomly chosen game price in the next stage and if the submitted price is higher participant pays the game price and the decision is implemented. If the auction is lost, participant’s decision is not implemented and bonus payment will be added to final payoff. If the participant gets the information (with or without choosing it)there is a chance to revise the guess. At the end of the game they will receive a payment from their guess (0or80) and the rest of the bonus payment after the BDM results.

Design details

Randomization method
Randomization is performed by zTree within sessions.
Randomization unit
Individual
Clusters
400 Individuals
Observations
400 Individuals
Arms
Each treatment contains 200 individuals
Primary outcomes
We investigate two key variables in the experiment. In one treatment, participants are asked to state their willingness to pay to receive the information about the share of participants that did better than them. Here, they can avoid that information by stating a willingness of zero. In the other trial, participants can state their willingness to pay to avoid that information. They will receive a bonus earning of 50 Norwegian Kroner. If they are willing to pay this full amount, they can ensure that they will not find out the information.

Cited by

No citation data available for this registration yet.

Keywords