The Effect of Relative Income on Conformism

How to cite

Bopp, Fabian; Wienke, Larissa. 2021. "The Effect of Relative Income on Conformism." AEA RCT Registry. AEARCTR-0007310. https://doi.org/10.1257/rct.7310-1.0

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

Fabian Bopp* , Larissa Wienke

* Primary investigator

Registry ID
AEARCTR-0007310
DOI
https://doi.org/10.1257/rct.7310-1.0
First registered
Last updated
Status
In development
Trial dates
2021-03-29 – 2021-03-31
Countries
Private
JEL codes
D72, C92, D83
Registry record
http://www.socialscienceregistry.org/trials/7310

Abstract

In this study, we test how relative income effects affect conformism behavior. We conduct an experiment, in which we vary relative income in form of an endowment and use a repeated voting structure between 2 policies to observe conformism behavior in groups. The repeated voting structure allows us to disentangle conformism behavior caused by different motives. The different high incomes could influence the uncertainty about the ability to make a good decision and thus affects the chance for social learning. Another treatment effect could be that the income effects induce general reciprocity and affect conformism behavior over the channel of social responsibility. Knowing when people do or do not conform, can help to implement measures to support the implementation of the majority´s vote.

Experimental design

In this study, we create a relative income effect. Subjects form a group of three members. Using a repeat decision between two options with different decision rules we first elicit the unbiased preferences, then create a situation in which we can observe minority behavior, and then by changing the decision rule between the repeated choices observe whether and why subjects conform.

Design details

Randomization method
randomization is done by a computer
Randomization unit
individual
Clusters
102 subjects
Observations
102 subjects* 5 choice problems=510 observations (We are only interested in minority observations) 510*0.25 (estimate of minority observations per choice problem &subject)= estimated 126 (interesting) observations. We use groups of 3 subjects but do not force them into a minority. Thus it can happen, that all 3 choose the same option and no group member ends up in the minority. This explains why estimated share is smaller them 0.33.
Arms
estimated 63 observation per treated type
Primary outcomes
Share of conformism behavior within the minority subjects

Cited by

No citation data available for this registration yet.

Keywords