How does social identity affect network formation and social learning

Last registered on November 13, 2023
View Trial History

Pre-Trial

Trial Information

General Information

Title
How does social identity affect network formation and social learning
RCT ID
AEARCTR-0012397
Initial registration date
October 30, 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
November 13, 2023, 2:19 PM EST

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

Locations

Country
Germany
Region

Primary Investigator

Name
Xiaoyu Zhou
Affiliation
FAU Erlangen-Nurnberg
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Friederike Mengel
PI Affiliation
University of Essex
Contact Investigator
PI Name
Veronika Grimm
PI Affiliation
Friedrich-Alexander-Universität Erlangen-Nürnberg
Contact Investigator

Additional Trial Information

Status
In development
Start date
2023-11-03
End date
2023-12-31
Keywords
Behavior, Lab, Other
Additional Keywords
Social learning, Social network
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
The proposed project aims to investigate the effect of social identity on the dynamics of social learning in networks. Social learning, i.e. learning by observing or communicating with others, plays a key role in individuals, firms, and governments’ decision-making processes. A lot of casual evidence suggests that both the learners and the sources social identity play a crucial role in how information from others is processed. Despite this fact there is little research on this question. In this project we study whether and how the social identity of an information source affects the perception of credibility of information, the willingness to update based on this information and ultimately how those factors affect learning dynamics in networks.
External Link(s)

Registration Citation

Citation
Grimm, Veronika, Friederike Mengel and Xiaoyu Zhou. 2023. "How does social identity affect network formation and social learning." AEA RCT Registry. November 13. https://doi.org/10.1257/rct.12397-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2023-11-03
Intervention End Date
2023-12-31

Primary Outcomes

Primary Outcomes (end points)
Group Level Outcomes:
1. Network Characteristics: Network density and homophily.
2. Consensus.
3. Overall welfare.

Individual Level outcomes:
1. Degree centrality.
2. Accuracy
3. Confidence




Primary Outcomes (explanation)
Density: the number of links divided by the maximum number of possible links.

Homophily: the extend people form links with the same nationality. We will use the E-I index: the number in-group links minus out- group links normalized by total link.

Consensus: We will calculate the standard deviation of group members’ answers in the last period. Because some questions have a much wider range of possible answers than others, we will normalize the standard deviation to foster comparison.

Welfare: We will calculate group level payoff in the final period, which is the total points earned by answering questions less the points paid for link formation.


Accuracy: Accuracy is measured by the absolute distance between the submit- ted answer and the true answer in the final period. Because questions varies in the boundaries of answers, we will normalize the distance to allow for comparison across questions.

Confidence: we elicit their confidence, using a likert scale question, after their first and final submission of each question.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Participants are formed into groups and play a social learning game.
Experimental Design Details
In the experiment, participants are asked to answer different types of factual questions related various fields.

Participants play the game in groups of six. For each question, they need to submit their answer a total of five times. Each group member submit their initial answer independently. For subsequent submissions, based on their decisions in the link formation stage (explained in detail in the next subsection), they can observe other group members’ past submitted answers.

After reading each question, participants are prompted with an interface that present each group member as nodes. By clicking on each node, participants can form links with other group members. The link allows them to observe the selected members’ submitted answers in the social learning game. However, link is not free. Each link costs a fixed amount of their endowment.

Importantly, network formation and information flow in our setting are both one way: player A can unilaterally form a link with player B after incurring the cost and only player A can observe player B’s answers. Our choice of directed networks allows participants to concentrate on determining the sources of information they wish to access. Conversely, in an undirected network, where information can flow in both directions but only one player incurs the cost, participants’ decisions might be further influenced by the perception of the number of group members linking to them. The strategic consideration of free-riding on others to form links could substantially complicate our ability to infer the effects of social identity on network formation and social learning.

We employ a between-subjects design. In each treatment, participants will be arranged into groups of six, which is fixed throughout the session. Each group comprises people from two nationalities. In the baseline treatment, the nationality of each group member is kept confidential. Conversely, in the identity treatment, the nationality of each group member is made public to all group members.
Randomization Method
Random based on computer program
Randomization Unit
Experimental session -- between subjects design
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
We plan to have 60 independent observations/groups.
Sample size: planned number of observations
Each group consists of 6 participants, which implies 360 participants.
Sample size (or number of clusters) by treatment arms
30 independent observations/groups in baseline and independent observations/groups in identity treatment.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
German Association for Experimental Economic Research e.V.
IRB Approval Date
2023-10-24
IRB Approval Number
1qKRik4t
Analysis Plan

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

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials