How social identity affect network formation and public good provision in long run?

N/A

N/A

2025-04-02

2025-08-31

Participants' contribution level in playing the public good, and who they choose to play the public good game (the network).

The key outcome variables (endpoints) of interest in this experiment include:

1. Network Formation Metrics:
1.1 Average number of neighbors per subject per round.
1.2 Number of links initiated per subject per round.
1.3 Number of links deleted per subject per round​

2. Linking Decisions:
2.1 Probability of initiating links with in-group vs. out-group members.
2.2 Probability of deleting links with in-group vs. out-group members.

3. Contribution Behavior:
3.1 Choice of plans (Plan 1 to Plan 4) by participants in each round.
3.2 Modal plans of neighbours
3.3 Mean plans of neighbours

4. Network Segrereation:
4.1 Composition of in-group members in network

Summary of Experimental Design
This experiment investigates the role of social identity in network formation and public goods provision. It employs a laboratory experiment where participants form and adjust social networks while playing a congestible public goods game over 25 rounds. We follow the base setting of our experimental design, while provide the following 7 treatments.
1. Based on previous treatments I30 and O30, but each subject’s payoff depends on his/her own choice of plan and the mean plan chosen by all his/her neighbours; they are 2 new treatments: I30-Mean_Contribution and O30-Mean_Contribution.
2. Based on previous treatments I30/O30/I80/O80, but each group of subjects are asked to interact for 50 rounds; they 4 are new treatments: I30-50Rounds, O30-50Rounds, I80-50Rounds, and O80-50Rounds.
3. Based on previous high mobility design, I80 and O80, but each subject starts the interaction with an initial network in which he/she connects with one in-group and one out group members; this leads to a new treatment: Medium80.

Implementation Plan:
- Phase 1: Conduct I30-Mean_Contribution and O30-Mean_Contribution, and if the results show no significant effect of using the mean (compare to the original I30 and O30), we continue to Phase 2.
- Phase 2: Conduct all other 5 new treatments proposed.
However, if in Phase 1 we find results are different from the original I30 and O30, then the project will stop here; the listed treatments in Phase 2 will be designed again for further implementation. In this eventuality, we will aim to run an expanded series of new treatments using mean plans rather than modal plans as the determinants’ of subjects’ payoffs. However, we are not yet able to fully formulate these alternative plans as the number of treatments and sample sizes would be dependent on obtaining additional funding which we have not yet secured.


How to compare to old sessions:
1. I30-Mean_Contribution vs I30 and O30-Mean_Contribution VS O30: to study the effect of using mean plan in determing individual payoff on the outcome variables listed above..
2. I30-50Rounds vs I30, O30-50Rounds vs O30, I80-50Rounds vs I80, and O80-50Rounds vs O80: to study the effect of playing 50 periods on the outcome variables listed above.
3. I80 vs Medium80 vs O80: to study the effect of initial integration on the outcome variables listed above.

--------------------------------------------------------------

Our basic settings, based on the last pre-registration Chen (2023) and the corresponding project are summarized below:

1. Participants & Social Identity Assignment
1.1 12 participants per session, divided into two social identity groups (e.g., Group A and Group B).
1.2 Identities were assigned using a music preference task and reinforced through a group discussion task to enhance group identity.
1.3 Each participant received a unique ID indicating their assigned group.

2. Game Structure & Network Formation
2.1 Each round consists of two steps:
Step 1: Network Adjustment
- Participants decide whom to connect with or whom to remove from their network.
- Forming a new connection requires mutual agreement, while deleting a link is unilateral.
- Participants can see the social identity of others when forming new links but also past behavior when considering deletion.
Step 2: Contribution & Payoff Calculation
- Participants choose one of four contribution plans (Plan 1 to Plan 4).
- A participant’s payoff depends on:
- Their own contribution choice.
- The most common contribution choice among their neighbors.
- The number of neighbors they have (too few or too many is inefficient).

3. Treatment Conditions (2×2 Design)
3.1 The experiment employs a 2×2 factorial design with two key variations:
3.1.1 Integration (Initial Network Structure)
- Low Integration: Each participant starts with two in-group neighbors (segregated start).
- High Integration: Each participant starts with two out-group neighbors (mixed start).
3.1.2 Mobility (Flexibility to Change Network)
- Low Mobility: Only 30% chance of being allowed to update relationships per round.
- High Mobility: 80% chance of being allowed to update relationships per round.

Public lottery: Subjects in the pool are randomly invited to participate in the experiment. In the invitation email they are allowed to choose to participate one of the sessions, while they do not know which session belongs to which treatment.

Individual.

Yes

576 undergraduate/postgraduate students.

576 undergraduate/postgraduate students.

There are 7 new treatments:

1. I30-Mean_Contribution: 4 sessions * 12 subjects/session = 48 subjects
2. O30-Mean_Contribution: 4 sessions * 12 subjects/session = 48 subjects
3. I30-50Rounds: 8 sessions * 12 subjects/session = 96 subjects
4. O30-50Rounds: 8 sessions * 12 subjects/session = 96 subjects
5. I80-50Rounds: 8 sessions * 12 subjects/session = 96 subjects
6. O80-50Rounds: 8 sessions * 12 subjects/session = 96 subjects
7. Medium80: 8 sessions * 12 subjects/session = 96 subjects