Aversion to Dominance

Last registered on October 27, 2020

View Trial History

Pre-Trial

Trial Information

General Information

Title

Aversion to Dominance

RCT ID

AEARCTR-0006657

Initial registration date

October 25, 2020

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

October 27, 2020, 7:28 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

United States of America

Region

Primary Investigator

Name

Kailash Rajah

Affiliation

Massachusetts Institute of Technology

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Frank Schilbach

PI Affiliation

Massachusetts Institute of Technology

Contact Investigator

PI Name

Leonardo Bursztyn

PI Affiliation

The University of Chicago

Contact Investigator

PI Name

Daron Acemoglu

PI Affiliation

Massachusetts Institute of Technology

Contact Investigator

Additional Trial Information

Status

In development

Start date

2020-10-25

End date

2021-01-31

Keywords

Governance

Additional Keywords

Dominance, Aversion, Authoritarianism

JEL code(s)

Secondary IDs

Abstract

People’s preference to make their own decisions rather than being dominated by others is at the heart of liberal democracies. This aversion to dominance might be impeded by threats to their physical safety, which political actors may seek to exploit. We study this issue in the context of the 2020 United States presidential election, where fear of riots and urban crime has featured as a prominent election issue. We measure the extent to which individuals prefer not to be dominated using willingness to pay to avoid being “dominated,” or yielding control over some aspect of their lives to another party, via an experiment called the “Boss Game”. We examine whether people’s aversion to being dominated is weakened when they are exposed to political advertising that evokes fear.

External Link(s)

Registration Citation

Citation

Acemoglu, Daron et al. 2020. "Aversion to Dominance." AEA RCT Registry. October 27. https://doi.org/10.1257/rct.6657-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Intervention Start Date

2020-10-25

Intervention End Date

2020-11-03

Primary Outcomes

Primary Outcomes (end points)

Willingness to pay (Boss Game)

Primary Outcomes (explanation)

In the Boss Game, subjects asked to choose between i) a computer deciding the allocation of $0.50 randomly to either the subject or to another subject; ii) another person making the same allocation decision without knowledge of any information about the two subjects. Using a multiple price list, we elicit participants' willingness to pay (from an additional $1 endowment) to implement either option i) or ii).

Secondary Outcomes

Secondary Outcomes (end points)

Fear

Secondary Outcomes (explanation)

We measure fear, the hypothesized underlying mechanism affecting dominance aversion, in two ways: First, by using questions taken from the State Trait Anxiety Inventory, and second, by directly asking whether participants feel worried or concerned about rioting or crime. All participants answer the State Trait Anxiety Inventory questions, and a random subset answer the direct questions.

Experimental Design

Study participants will be recruited using Facebook advertisements targeted at users who are likely to identify as Republicans. Individuals who click on the advertisements will be taken to our survey.

The survey begins with demographic questions, including political affiliation and media consumption. Participants who do not identify as Republicans will only complete the demographic questions. This section also contains several attention checks. Participants who fail these checks will be dropped before treatment.

Following the demographics section, participants are told that they will be asked to watch a short finance news clip and answer an incentivized “comprehension” question about the video. Participants are randomized to watch one of four versions of the video.
- Pure control - the finance news clip on its own
- Placebo control - a dog food commercial will play before the finance news clip.
- Biden placebo control - a Joe Biden election commercial (that does not evoke fear) will play before the finance news clip
- Treatment - a Donald Trump election commercial that creates fear about crime and the “defund the police” movement will play before the finance news clip

After watching the video, participants are asked an incentivized comprehension question about the finance news clip. If they answer incorrectly, they are asked to watch the video a second time and are allowed to re-attempt the comprehension question.

Following treatment, the respondents play the Boss Game and answer questions from the State Trait Anxiety inventory. The order of these tasks is randomized.

In the Boss Game, subjects asked to choose between i) a computer deciding the allocation of $0.50 randomly to either the subject or to another subject; ii) another person making the same allocation decision without knowledge of any information about the two subjects. Using a multiple price list, we elicit participants' willingness to pay (from an additional $1 endowment) to implement either option i) or ii).

Finally, respondents are randomized to answer specific questions about either (i) their fear towards rioting, looting and the “defund the police” movement, or (ii) the number of ads they saw in the survey. Task (i) provides a more direct measure of fear following treatment, while (ii) measures whether respondents viewed the treatment videos as ads rather than videos that were part of the survey.

The intended sample size includes 2,500 individuals who identify as Republicans, though the final sample size may differ from this number depending on recruitment costs and sample availability.
We hypothesize that:
1. Subjects will exhibit a positive average willingness to pay to avoid being dominated as measured by the “Boss Game”
2. Subjects in the treatment group will be willing to pay less to avoid being dominated than the control groups.

In addition, we hypothesize that treatment effects are stronger among subjects who report lower levels of media consumption.

Experimental Design Details

Randomization Method

Randomization done by Qualtrics

Randomization Unit

Treatments are randomized at the individual level

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

2500 individuals

Sample size: planned number of observations

2500 individuals

Sample size (or number of clusters) by treatment arms

625 per experimental arm

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Minimum detectable effect: Willing to pay $0.05 to avoid the BOSS (dominance). We assume a control group WTP of $0.29 and standard deviation of $0.32.

Supporting Documents and Materials

IRB