To the Depths of the Sunk Cost: an Experiment Revisiting the Elusive Fallacy (Prolific)

How to cite

Stalinski, Mateusz; Ichiba, Sota; Beknazar-Yuzbashev, George. 2022. "To the Depths of the Sunk Cost: an Experiment Revisiting the Elusive Fallacy (Prolific)." AEA RCT Registry. AEARCTR-0009017. https://doi.org/10.1257/rct.9017-1.0

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

Mateusz Stalinski* , Sota Ichiba , George Beknazar-Yuzbashev

* Primary investigator

Registry ID
AEARCTR-0009017
DOI
https://doi.org/10.1257/rct.9017-1.0
First registered
Last updated
Status
In development
Trial dates
2022-03-22 – 2022-04-30
Countries
United States of America ()
JEL codes
C91, D11, D12, D90
Registry record
http://www.socialscienceregistry.org/trials/9017

Abstract

Despite being often discussed both in practice and academic circles, the sunk cost effect remains empirically elusive. We develop an experimental paradigm that overcomes the deficiencies of the previous research in identifying the effect. Additionally, motivated by a model based on reference point dependence, we formulate a novel prediction that if the intervention places individuals on the same side of the reference point (for example, both in the gain region), the sunk cost effect does not manifest itself. In an online study, we randomize the price of a ticket to enter a game where the payoff depends on performance, and observe its effect on playtime. The price (low, medium, or high) is realized after the purchase of the ticket. The distribution is fully disclosed prior to the event, with the low and high prices placing the participant on the opposite sides of the reference point.

Experimental design

We randomize the price of a ticket to enter a game and observe its effect on playtime. The objective of the game is to count the number of zeros in a sequence of digits presented on the screen. The price of entry ($0.5, $1.5, or $2.5) is realized after the purchase of the ticket.

Note: In November 2021, CloudResearch unexpectedly changed its policy regarding options for screening of participants, which prevented us from conducting the study on MTurk as initially planned (pre-registration URL: https://doi.org/10.1257/rct.8595-1.1). Subsequently, we moved our study to Prolific, made adjustments to comply with the new platform’s compensation requirements and to reflect the new participant pool (recalibration). The Prolific protocol is the subject of this pre-registration.

Design details

Randomization method
We use Qualtrics built-in randomization.
Randomization unit
Individual
Clusters
1800 individuals
Observations
1800 individuals
Arms
600 individuals per treatment group (high price, medium price, low price).
Primary outcomes
The time a participant spends playing the counting game.

Cited by

No citation data available for this registration yet.

Keywords