How many norms are there? Behavioral Implications of Uncertainty over Multiple Social Norms

Last registered on September 20, 2023

View Trial History

Pre-Trial

Trial Information

General Information

Title

How many norms are there? Behavioral Implications of Uncertainty over Multiple Social Norms

RCT ID

AEARCTR-0012012

Initial registration date

September 18, 2023

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

September 20, 2023, 11:21 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

Folco Panizza

Affiliation

IMT School for Advanced Studies Lucca

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Eugen Dimant

PI Affiliation

University of Pennsylvania

Contact Investigator

PI Name

Erik Kimbrough

PI Affiliation

Chapman University

Contact Investigator

PI Name

Alexander Vostroknutov

PI Affiliation

University of Maastricht

Contact Investigator

Additional Trial Information

Status

In development

Start date

2023-09-13

End date

2023-09-30

Keywords

Behavior

Additional Keywords

social norms, belief elicitation, false consensus, pluralistic ignorance

JEL code(s)

Secondary IDs

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

Social norms guide human behavior in various contexts, even when violating them does not lead to immediate material costs. However, there are situations where it is unclear if a norm exists, or if different subgroups follow different norms. Existing methods of norm elicitation fail to capture this uncertainty, which can have significant consequences in both economic and social contexts. This experiment investigates the presence of multiple social norms employing a new elicitation method under two different economic games: a simple dictator game and a more complex economic problem involving trade-offs between a selfish, an equitable, a rawlsian, and an efficient option. Our design allows us to incentive-compatibly elicit multiple norms and sheds light on the extent to which people are aware of such multiplicity.

External Link(s)

Registration Citation

Citation

Dimant, Eugen et al. 2023. "How many norms are there? Behavioral Implications of Uncertainty over Multiple Social Norms." AEA RCT Registry. September 20. https://doi.org/10.1257/rct.12012-1.0

Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Experimental Details

Interventions

Intervention(s)

Treatments vary in two dimensions: the scenario to evaluate, and the agents involved. There are two possible scenarios, a Dictator Game and an allocation game, inspired by Engelmann and Strobel (2004), that includes four options (selfish, equitable, maximin, and efficient). For each scenario, there are two types of agents: all recipients are participants, or one recipient is a charitable organization and the others are participants. This manipulation is to test whether agent composition affects the beliefs elicited. We thus conduct a two (scenario) by two (charity/no charity) experiment. Participants will be randomly assigned to one of the four treatments.

Intervention Start Date

2023-09-13

Intervention End Date

2023-09-30

Primary Outcomes

Primary Outcomes (end points)

Norms, Sanctioning behavior

Primary Outcomes (explanation)

Norms are elicited using our new elicitation method. Participants are asked to guess what other people’s normative beliefs about the scenario at hand are. Participants guess norms by rating the appropriateness of each action in the scenario. Ratings for each action are expressed on a three-point scale: participants rate each possible action as either appropriate, inappropriate, or neither appropriate nor inappropriate. Participants may guess multiple norms if they believe that other people actually have multiple, diverging views about the scenario. Each norm elicited in this way is our unit of observation, although we will also cluster norms for some analyses if they share similar properties (e.g., norms that agree on which action is appropriate).

Sanctioning behavior. Sanctioning behavior is measured in a third-party punishment task in which participants can spend some of their earnings to reduce the payoffs of the player in the scenario allocating the resources (e.g., the dictator in the dictator game). Sanctioning is measured as the number of experimental currency deducted in this way.

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

In this online experiment, participants are randomly assigned to one of the four experimental conditions that differ in the scenario they are asked to evaluate and in the agents involved. Participants will first be asked to estimate which different normative beliefs are associated with the described scenario using an incentivized task. Participants will then play as a third-party punisher against the player in the previously described scenario that was tasked to allocate the resources (e.g., the dictator in the dictator game). Finally, participants will play in another scenario that will be used for other experimental treatments.

Experimental Design Details

Randomization Method

randomization into experimental groups through the online crowdsourcing platform (Prolific.co)

Randomization Unit

participants

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

No clusters are planned in this study

Sample size: planned number of observations

800 participants

Sample size (or number of clusters) by treatment arms

200 participants per experimental condition

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

Although we determined our sample size according to budget constraints, our pre-analysis plan estimates the minimum detectable effect size for our main hypothesis H1. Assuming that all participants guess the maximum possible number of norms (10 in the experiment), given all possible norms that can be guessed in the scenario with the most actions (N = 243), and a sample of 200 participants per test, then our MDE would be w = 0.20 with an alpha of 5% and a power (1 - beta) of 95%, whereas it would be w = 0.16 with an alpha of 10% and a power (1 - beta) of 80%.

Supporting Documents and Materials

IRB