Understanding Mechanisms Underlying the Relationship between Witchcraft and Prosocial Behavior

Last registered on November 22, 2021

Pre-Trial

Trial Information

General Information

Title
Understanding Mechanisms Underlying the Relationship between Witchcraft and Prosocial Behavior
RCT ID
AEARCTR-0004878
Initial registration date
October 19, 2019
Last updated
November 22, 2021, 8:28 PM EST

Locations

Primary Investigator

Affiliation
Stanford University

Other Primary Investigator(s)

PI Affiliation
Harvard University
PI Affiliation
CES

Additional Trial Information

Status
On going
Start date
2019-07-22
End date
2021-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
We describe the analysis plan for analysis of survey data, which is aimed at probing mechanisms as a followup to findings from a previous study that examines the relationship between prosocial behavior and others’ beliefs in witchcraft in the Democratic Republic of the Congo (Lowes and Nunn, 2018). As with the first study, participants will be individuals living in a city in Sud-Ubangi province which is in northern DRC. Participants will complete a series of incentivized questions asking about how socially appropriate most people in the city will view various actions in three experimental games that involve another player. For each hypothetical game, the other player is anonymous, but participants are given some information about them. They are informed of the other player’s age, sex, educational level, if from the same ethnic group, strength of belief in the Christian God, strength of belief in witchcraft, and whether the individual is from the city. The primary experimental treatment is that the other player is randomly assigned to have either a strong or weak belief in witchcraft. We outline the study design, the treatments, and the econometric strategy for the analysis.

We will also expand our sample to include 50 villages in rural areas. We will administer the activities described in Lowes and Nunn, 2018, and also implement the activities described above.
External Link(s)

Registration Citation

Citation
Le Rossignol, Etienne, Sara Lowes and Nathan Nunn. 2021. "Understanding Mechanisms Underlying the Relationship between Witchcraft and Prosocial Behavior." AEA RCT Registry. November 22. https://doi.org/10.1257/rct.4878-2.0
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2019-10-22
Intervention End Date
2021-12-15

Primary Outcomes

Primary Outcomes (end points)
The key outcome variables are how appropriate the various possible actions are that a hypothetical decision maker can make in the dictator game, choose your dictator game, and joy of destruction game.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The norms measurement strategy follows Krupka and Weber (2013). Individuals will be given information on how three experimental tasks are played. These tasks are a dictator game (DG), a choose your dictator game (CYD), and a joy of destruction game (JOD). The details of these experiments will be discussed shortly. These are the same experiments that participants completed in 2018.
After reviewing each experimental task, the participant will be asked to imagine that there is a hypothetical decision maker who is completing the experimental task. They participant will be given information on the identity of the person that the decision maker in the task has been paired with. For each possible choice that the decision maker in the task could make, the participant will be asked: “Is this choice very socially inappropriate, somewhat socially inappropriate, somewhat socially appropriate, very socially appropriate?.” Individuals are not given the exact identity of the player that the decision maker is paired with in the experimental tasks. However, they are given the following information about the other player: age group, sex, education level, ethnicity, strength of belief in the Christian God, strength of belief in witchcraft, and whether they grew up in Gemena. The primary experimental manipulation is the reported belief in witchcraft of the other player. Participants complete two iterations of the set of questions about each experimental activity. The assignment of player characteristics is stratified so that in one of the two iterations (randomly chosen), the participant answers questions about a decision maker who is paired with someone with either a ‘strong’ or ‘very strong’ belief in witchcraft and in the other, the participant answers questions about a decision maker who is paired with someone with a ‘weak or very weak’ or ‘neither believe or disbelieve’ belief in witchcraft.
Experimental Design Details
The norms measurement strategy follows Krupka and Weber (2013). Individuals will be given information on how three experimental tasks are played. These tasks are a dictator game (DG), a choose your dictator game (CYD), and a joy of destruction game (JOD). The details of these experiments will be discussed shortly. These are the same experiments that participants completed in 2018.
After reviewing each experimental task, the participant will be asked to imagine that there is a hypothetical decision maker who is completing the experimental task. They participant will be given information on the identity of the person that the decision maker in the task has been paired with. For each possible choice that the decision maker in the task could make, the participant will be asked: “Is this choice very socially inappropriate, somewhat socially inappropriate, somewhat socially appropriate, very socially appropriate?.” Earlier in the survey the following definition is given of socially appropriate:
“After I describe the situation and decision made by the person, I would like you to evaluate the decision and decide whether the action is ‘socially appropriate’ and ‘consistent with moral or proper social behavior’ or ‘socially inappropriate’ and ‘in- consistent with moral or proper social behavior’. By socially appropriate, I mean behavior that most people in Gemena agree is the ‘correct’ or ‘ethical’ thing to do.”
Individuals are not given the exact identity of the player that the decision maker is paired with in the experimental tasks. However, they are given the following information about the other player: age group, sex, education level, ethnicity, strength of belief in the Christian God, strength of belief in witchcraft, and whether they grew up in Gemena. The various characteristics take the following values:
1. Age group: (a) young; (b) old.
2. Sex: (a) male; (b) female.
3. Education level: (a) has not completed primary school; (b) has completed primary school; (c) has completed secondary school or higher.
4. Ethnicity: (a) same ethnic group; (b) a different ethnic group.
5. Strength of belief in Christian God: (a) strong belief in the Christian God (b) very strong
belief in the Christian God.
6. Strength of belief in witchcraft: (a) weak or very weak, (b) neither strong nor weak, (c) strong, or (d) very strong.
7. Grew up in Gemena: (a) grew up in Gemena (b) did not grow up in Gemena.
The primary experimental manipulation is the reported belief in witchcraft of the other player. Participants complete two iterations of the set of questions about each experimental activity. The assignment of player characteristics is stratified so that in one of the two iterations (randomly chosen), the participant answers questions about a decision maker who is paired with someone with either a ‘strong’ or ‘very strong’ belief in witchcraft and in the other, the participant answers questions about a decision maker who is paired with someone with a ‘weak or very weak’ or ‘neither believe or disbelieve’ belief in witchcraft.
For the other reported characteristics of the other player, they are randomly assigned with equal probability (e.g. half the time the other player is old, half the time the other player is young; a third of the time the other player had not completed primary etc.).
Each participant will respond to questions regarding the socially appropriate thing to do in three different experimental games: the Dictator Game (DG), Choose Your Dictator Game (CYD), and Joy of Destruction Game (JOD). The respondent will answer questions about socially appropriate behavior for each game two times; each time, the identity of the other player that the hypothetical decision maker has been paired with will vary. To encourage individuals to consider their answers carefully, the responses are incentivized. If all of the responses about the appropriateness of each choice are the same as the most common response in all of Gemena for the situation, then the respondent receives an extra CF 5,000.
Randomization Method
The primary experimental manipulation is whether an individual is answering questions about someone with a strong or weak belief in witchcraft. The survey will be implemented on a tablet. The tablet will randomly assign the characteristics of the individual about whom the respondent answers the question. An respondent answers the questions for two different individuals: one with a strong belief in witchcraft and one with a weak belief in witchcraft.
Randomization Unit
Randomization is done at the individual level.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
We aim to follow up with our sample from 2018, which had 520 individuals.

We will also expand our sample. We will visit 50 rural villages and select a random sample of 12 villagers from each village for a rural sample of 600 individuals.
Sample size: planned number of observations
We aim to follow up with our sample from 2018, which had 520 individuals. The rural sample will be 600 individuals from 50 randomly selected villages.
Sample size (or number of clusters) by treatment arms
Each individual is treated in the sense that they answer questions about how appropriate actions are in each game two times: one time when the other player has a strong belief in witchcraft and one time when the other player has a weak belief in witchcraft. We aim to follow up with our sample from 2018, which had 520 individuals.

We will also expand our sample to include 600 individuals from rural areas (12 villagers across 50 villages).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials

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

Request Information
IRB

Institutional Review Boards (IRBs)

IRB Name
Harvard University
IRB Approval Date
2015-06-10
IRB Approval Number
24087
IRB Name
Harvard University
IRB Approval Date
2020-05-29
IRB Approval Number
IRB19-2059
Analysis Plan

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

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

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

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