A Dictator Game Experiment with the COVID-19 Vaccinated and Unvaccinated People

How to cite

Sasaki, Shusaku; Kurokawa, Hirofumi. 2022. "A Dictator Game Experiment with the COVID-19 Vaccinated and Unvaccinated People." AEA RCT Registry. AEARCTR-0008951. https://doi.org/10.1257/rct.8951-3.1

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Shusaku Sasaki* , Hirofumi Kurokawa

* Primary investigator

Registry ID
AEARCTR-0008951
DOI
https://doi.org/10.1257/rct.8951-3.1
First registered
Last updated
Status
On going
Trial dates
2022-01-28 – 2025-03-31
Countries
Japan ()
Registry record
http://www.socialscienceregistry.org/trials/8951

Abstract

In this study, we conduct a dictator game experiment with the COVID-19 vaccinated and unvaccinated people in Japan, and ascertain allocation tendencies between two anonymous individuals, between two vaccinated individuals or two unvaccinated individuals, and between a vaccinated individual and an unvaccinated individual. By so doing, we assess “ingroup bias,” “outgroup bias,” and “ingroup-favoritism” of vaccinated and unvaccinated people, respectively. Here, ingroup bias is defined as the difference between the money amounts allocated to an individual of the same group as the allocator and to an anonymous individual in the dictator game experiment. Outgroup bias is defined as the difference between the money amounts allocated to an individual of a different group from the allocator’s and to an anonymous one, and ingroup-favoritism is defined as the difference between the money amounts allocated to an individual of the same group and to an individual of the different group.

Specifically, we employ an experimental design that allows for both between- and within-individual comparisons to conduct the following analyses. First, in the between-analysis, we test the existence of ingroup bias, outgroup bias, and ingroup-favoritism in each group of vaccinated and unvaccinated individuals. In addition, we examine how these biases differ between vaccinated and unvaccinated individuals. We further investigate how the biases change between when the vaccination status of the allocator is not informed to the recipient and when it is informed to them.

Next, in the within-analysis, we test whether individuals' ingroup bias, outgroups bias, and ingroup favoritism are associated with their real-world attitudes and behaviors. In both the vaccinated and unvaccinated group, we examine the association between their biases and attitudes toward the COVID-19-related policies. In the vaccinated group, we also examine the association between their biases and the timing of COVID-19 vaccination.

To stably manage a society under a pandemic, it is crucial to build cooperative relationships between vaccinated and unvaccinated people. This study can contribute to the smooth construction of such cooperative relationships.

Intervention

We set the following treatment conditions in the dictator game experiment:

- Private-Ingroup: An allocator is informed that a recipient belongs to the ingroup (a vaccinated individual in the vaccinated sample, or an unvaccinated individual in the unvaccinated sample). The allocator is anonymous for the recipient.
- Private-Outgroup: An allocator is informed that a recipient belongs to the outgroup (an unvaccinated individual in the vaccinated sample, or a vaccinated individual in the unvaccinated sample). The allocator is anonymous for the recipient.
- Public-Ingroup: An allocator is informed that a recipient belongs to the ingroup. The recipient is notified of the vaccination status of the allocator.
- Public-Outgroup: An allocator is informed that a recipient belongs to the outgroup. The recipient is notified of the vaccination status of the allocator.

Experimental design

We conduct dictator game experiments in the following five conditions:

(I) Anonymous: A recipient is anonymous for an allocator. The allocator is also anonymous for the recipient.
(II) Private-Ingroup: An allocator is informed that a recipient belongs to the ingroup (a vaccinated individual in the vaccinated sample, or an unvaccinated individual in the unvaccinated sample). The allocator is anonymous for the recipient.
(III) Private-Outgroup: An allocator is informed that a recipient belongs to the outgroup (an unvaccinated individual in the vaccinated sample, or a vaccinated individual in the unvaccinated sample). The allocator is anonymous for the recipient.
(IV) Public-Ingroup: An allocator is informed that a recipient belongs to the ingroup. The recipient is notified of the vaccination status of the allocator.
(V) Public-Outgroup: An allocator is informed that a recipient belongs to the outgroup. The recipient is notified of the vaccination status of the allocator.

One subject participates in five dictator game experiments as an allocator. We randomly set the order of the above conditions to create eight groups in the vaccinated and unvaccinated samples, respectively. After we first present the dictator game in the Anonymous condition (I) in all the eight groups, we randomly set the order of the Private (II, III) and Public (IV, V) conditions. Then, within each of the Private and Public conditions, we randomly set the order of the ingroup conditions (II, IV) and outgroup conditions (III, V). Consequently, each of the vaccinated and unvaccinated samples have eight groups.

Furthermore, to address potential order effects caused by successive participation in the dictator games in the within-analysis, we create a control group where we present the anonymous dictator game (I) five times. Totally, each of the vaccinated and unvaccinated samples have nine groups, including the control group.

We explain more details in the attachment of the analysis plan.

Design details

Randomization method
Stratified randomization by a survey company. The strata are based on age, gender, and baseline generosity.
Randomization unit
Individuals
Clusters
N/A
Observations
1,600 individuals (800 vaccinated individuals and 800 unvaccinated individuals)
Arms
800 vaccinated individuals include eight treatment groups (8*80 vaccinated individuals=640 vaccinated individuals) and one control group (160 vaccinated individuals). Similarly, 800 unvaccinated individuals include eight treatment groups (8*80 unvaccinated individuals=640 unvaccinated individuals) and one control group (160 unvaccinated individuals). In the between-analysis, we combine two or more treatment groups to create a paired group. Therefore, the sample size of one paired group is more than 160.
Primary outcomes
Allocation in the dictator game experiment
Secondary outcomes
- Attitudes toward the COVID-19-related policies (Balancing infectious disease control and socio-economic activities, vaccination certification and negative test results, and financial incentives for encouraging vaccination) - Dates of the first and second COVID-19 vaccination uptakes (Vaccinated sample, only)

Cited by

No citation data available for this registration yet.

Keywords