NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Preferences for Religious Giving - follow-up
Initial registration date
May 06, 2019
July 15, 2020 10:14 AM EDT
Other Primary Investigator(s)
Additional Trial Information
This study investigates why people give money to religious organizations, using a laboratory experiment in Accra, Ghana. In particular, we seek to understand to what extent churches provide informal insurance to households. Although contracts exist with quite low premiums, less than 1% of Ghanaian households are insured. With 96% of inhabitants reporting themselves to be religious, beliefs in a God play a major role in the daily lives of many Ghanaians. Much of this religious activity is backed by regular and significant contributions to the church through tithes, regular offering, and other forms of giving. We offer households a range of incentivized choices between donations to religious organizations and other financial options. We implement treatments including increasing the perceived risk of life accidents and the provision of free insurance to observe the effect on the demand for religious giving. We also compare how these treatments effects vary for individuals participating in the experiment during "revival weeks" - weeks of intense church participation, with others participating during normal weeks. This study is a follow-up for study AEARCTR-0000558.
The study will be conducted in two parts: first, a survey covering demographic
questions and religious beliefs and behaviour and second, a decision task presenting participants with binary choices on ways to spend 19GHS. Our interest is 1) to investigate the effects of increasing the salience of life accidents risk (informaton treatment) and the effect of providing individuals an insurance policy (insurance treatment) on their preferences for making donations to different causes, and 2) how these effects differ when individuals participate in the experiment during normal weeks and when participation happens during "revival weeks" - weeks of intense church participation and in which churchgoers donate much more than in an average week.
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
The key outcome variable is the difference in donations between individuals endowed with the information treatment and individuals with the insurance treatment.
Primary Outcomes (explanation)
Secondary Outcomes (end points)
The secondary outcome is the difference between treatments effects for individuals that participate in the experiment during revival and those that participate during normal weeks.
Secondary Outcomes (explanation)
We divide the subjects into three groups; 1/3 of the subjects will be allocated with a life insurance policy, 1/3 of the
participants will be told about, but not endowed with, this insurance policy, and 1/3 of the participants will form the control group. The study participants will be recruited from the Assemblies of God church and half of them will be randomly invited to participation in the study during a revival week while the other half will be invited to participate in a normal week.
Experimental Design Details
Church participants will be recruited from the Assemblies of God church population. They will be approached in advance and invited to show up to a specific session. Participants will be met at their own church premises and driven to the experiment location. Anyone above the age of 18, and the mental and physical capacity to answer questions for 1.5 hours will be eligible to participate in the experiment. The experiment room will be equipped with 10 tablets. All locations will also have a black/whiteboard. The choices to be played and their associated tags (A, B , C,…J) will be prominently displayed for the duration of the experiment. Participants will have to choose how to allocate 11GHS between 2 options. The alternatives are: keeping the money, donating to the thanksgiving offering of the National Prayer Week, donating to the Street Children Empowerment Fund, making a private donation directly to their church and making a non-anonymous donation directly to their church. All these choices are kept anonymous, except for the public choice to give to the church, for which the name of the participant and the amount he chose to give is communicated to the Pastor.
Randomization between information and insurance treatments is performed at the session level and done by the draw a sealed enveloppe by a participant.
Randomization between particpation in a normal or revival week is performed at the participant level and done by random draw from a box without replacement.
Randomization between information and insurance treatments is performed at the session level.
Randomization between particpation in a normal or revival week is performed at the participant level.
Was the treatment clustered?
Sample size: planned number of clusters
At least 64 sessions of 10 participants
Sample size: planned number of observations
Sample size (or number of clusters) by treatment arms
Approx. 100 individuals per treatment (6 treatments)
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
This round of experiments is under-powered. The primary aim is to pool the data from this follow-up with data from previous rounds for non-revival weeks and validate the main effects for non-revival weeks. The secondary objective is to provide preliminary evidence for how treatment effects differ between revival and non-revival weeks.
INSTITUTIONAL REVIEW BOARDS (IRBs)
Post Trial Information
Is the intervention completed?
Intervention Completion Date
July 01, 2019, 12:00 AM +00:00
Is data collection complete?
Data Collection Completion Date
July 01, 2019, 12:00 AM +00:00
Final Sample Size: Number of Clusters (Unit of Randomization)
80 sessions (more than 4 participants; as specified in the pre-analysis plan, we discarded sessions with less than 4 participants)
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
Final Sample Size (or Number of Clusters) by Treatment Arms
176 participants in Insurance session, 156 participants in Insurance Information session, 162 in No Insurance session
Is public data available?
This section is unavailable to the public. Use the button below
to request access to this information.
Reports, Papers & Other Materials
This paper provides experimental support for the hypothesis that insurance can be a motive
for religious donations. We randomize enrollment of members of a Pentecostal church in Ghana
into a commercial funeral insurance policy. Then church members allocate money between themselves and a set of religious goods in a series of dictator games with significant stakes. Members
enrolled in insurance give significantly less money to their own church compared to members
that only receive information about the insurance. Enrollment also reduces giving towards other
spiritual goods. We set up a model exploring different channels of religiously based insurance.
The implications of the model and the results from the dictator games suggest that adherents
perceive the church as a source of insurance and that this insurance is derived from beliefs in an
interventionist God. Survey results suggest that material insurance from the church community
is also important and we hypothesize that these two insurance channels exist in parallel.
Emmanuelle Auriol, Julie Lassébie, Amma Panin, Eva Raiber, and Paul Seabright. “God Insures Those Who Pay? Formal Insurance and Religious Offerings in Ghana,” The Quarterly Journal of Economics, forthcoming
REPORTS & OTHER MATERIALS