Exogenous versus Endogenous Peer Comparisons in the Lab

How to cite

Banko-Ferran, Daniel. 2021. "Exogenous versus Endogenous Peer Comparisons in the Lab." AEA RCT Registry. AEARCTR-0008023. https://doi.org/10.1257/rct.8023-1.0

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

Daniel Banko-Ferran*

* Primary investigator

Registry ID
AEARCTR-0008023
DOI
https://doi.org/10.1257/rct.8023-1.0
First registered
Last updated
Status
On going
Trial dates
2021-09-13 – 2021-12-15
Countries
Private
JEL codes
D90, D91, C91
Registry record
http://www.socialscienceregistry.org/trials/8023

Abstract

I propose to conduct a lab experiment in which some participants will be given the choice between opting into whether they receive information about their performance relative to their peers or not, and compare their subsequent behavior to those who were randomly not given the choice. This study will seek to disentangle the effect that forcing participants to receive peer information or not may have in contributing to the adverse responses observed in the literature.

Intervention

We test participant sensitivity to feedback on a performance task and their decisions to receive feedback on the task.
Specifically we ask if participants who receive feedback perform worse on average than those that do not receive feedback, and if those who can choose to receive feedback or not do best.

Experimental design

Three treatments: No Peer Comparison (NC), Peer Comparison (PC), Flexible Comparison (FC). Participants randomly assigned to one of the three treatments at the session level. Participants enter the lab and perform a keyboard pressing task through a computer terminal simultaneously across multiple rounds. Instructions are delivered verbally, through the computer screen, and through a print out. Sessions lasted 60 minutes. Participants were paid a piece-rate payment based on the number of key presses they completed in the round randomly selected for payment. In between rounds, PC participants received information about their relative performance on the task (top 20%, top 50%, bottom 50%). FC participants could choose to receive this information or not. NC participants did not receive this relative peer information.

Design details

Randomization method
randomization done in office by a computer
Randomization unit
experimental sessions
Clusters
6 experimental sessions
Observations
126 participants
Arms
2 clusters per treatment arm
Primary outcomes
Effort (measured as number of tasks completed)

Cited by

No citation data available for this registration yet.

Keywords