Determinants of Norm Compliance: Moral Similarity and Group Identification

Last registered on March 13, 2020

View Trial History

Pre-Trial

Trial Information

General Information

Title

Group Identitiy Project

RCT ID

AEARCTR-0005394

Initial registration date

March 13, 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

March 13, 2020, 12:41 PM 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

Alexander Schneeberger

Affiliation

University of Cologne

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Erin Krupka

PI Affiliation

University of Michigan

Contact Investigator

Additional Trial Information

Status

In development

Start date

2020-03-17

End date

2020-09-30

Keywords

Other

Additional Keywords

JEL code(s)

C91, D03

Secondary IDs

Abstract

While previous research in experimental economics has established that individuals react to changes in the applicable social norms, the question to when individuals comply with social norms has been neglected. In this study, we examine how group identification affects compliance with identity-dependent social norms. In our online experiments, we manipulate group identification by varying the moral similarity between the subjects and an artificial social group, while keeping all other factors including the identity-dependent social norms fixed. Our experimental design allows us to test how moral similarity affects group identification and how group identification affects norm compliance in a decontextualized setting.

External Link(s)

Registration Citation

Citation

Krupka, Erin and Alexander Schneeberger. 2020. "Group Identitiy Project." AEA RCT Registry. March 13. https://doi.org/10.1257/rct.5394-1.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

We randomly assign participants to one of two treatment conditions. In each treatment, we introduce the participants to the moral position of a different social group. Through this treatment variation, we exogenously alter the moral similarity between our subjects and the social group (high or low moral similarity) and, consequently, the group identification towards the social group.

Intervention Start Date

2020-03-17

Intervention End Date

2020-09-30

Primary Outcomes

Primary Outcomes (end points)

Norm compliance (the number of balls in the blue bucket of the modified rule-following task in the Behavioral Experiment).

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Group identification (the rating of the "Inclusion of Ingroup in the Self Scale" in the Behavioral Experiment).

Secondary Outcomes (explanation)

In our confirmatory analysis, we utilize the distance measure of this scale rather than the overlap measure.

Experimental Design

The research project includes two surveys and two experiments:

The Pre-Screening Survey: In this survey, participants answer the Moral Foundations Questionnaire (Graham et al., 2011). Based on their responses, we calculate the progressivism index (Clark et al., 2017).

The Rule Elicitation Survey (before the registration): We re-invited ten individuals with a positive progressivism index and ten individuals with a negative progressivism index. In the survey, we asked for a description of the rule from the rule-following task (Kimbrough & Vostroknutov, 2018). Their choices were the "The rule is to put the balls in the blue bucket" or "The rule is to put the balls in the yellow bucket". We used those answers to construct our treatment variation (two social groups that have a distinct moral positions but prescribe the same behavior in the rule-following task).

The Behavioral Experiment: We re-invite participants with a positive progressivism index. In this experiment, we introduce participants to the moral position of either social group, ask for a rating in a continuous version of the "Inclusion of Ingroup in the Self Scale" (Tropp & Wright, 2001) and let them play a modified version of the rule-following task. In this modified version, the rule "The rule is to put the balls in the blue bucket" is replaced by the rule "According to the members of Group A, the rule is to put the balls in the blue bucket".

The Norm Elicitation Experiment: We re-invite participants with a positive progressivism index. In this experiment, we introduce participants to the moral position of either social group and elicit the group norms of individuals with positive progressivism in either treatment of the Behavioral Experiment. Hereby, we utilize the method of Krupka & Weber (2013).

Experimental Design Details

A full description of the experimental design can be found in our pre-analysis plan.

Randomization Method

When subjects participate in either the Behavioral or Norm Elicitation Experiment, the experimental server randomly assigns them to either treatment. To achieve this, we use the Python module NumPy.

Randomization Unit

Individual level

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

Behavioral Experiment: 2 treatments with 160 participants each.
Norm Elicitation Experiment: 2 treatments with 100 participants each.

Sample size: planned number of observations

Behavioral Experiment: 2 treatments with 160 participants each. Norm Elicitation Experiment: 2 treatments with 100 participants each.

Sample size (or number of clusters) by treatment arms

Behavioral Experiment: 2 treatments with 160 participants each.
Norm Elicitation Experiment: 2 treatments with 100 participants each.

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

The sample size in the Behavioral Experiment is based on the following power calculation: We used data from Kimbrough & Vostroknutov (2018) to calculate an expected effect size of 0.3. Subsequently, we used G-Power (Faul, Erdfelder, Lang, & Buchner, 2007) to calculate the minimal required sample size for a one-tailed Wilcoxon-Mann-Whitney test with a parent distribution min ARE, an effect size of 0.3, an alpha-error probability of 0.05, a power of 0.80 and an even allocation between treatments. Based on this power calculation, we require at least 160 subjects per treatment.

Supporting Documents and Materials

IRB