The value of thoughts and prayers

How to cite

Thunstrom, Linda; Noy, Shiri. 2018. "The value of thoughts and prayers." AEA RCT Registry. AEARCTR-0003571. https://doi.org/10.1257/rct.3571-1.0

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

Linda Thunstrom* , Shiri Noy

* Primary investigator

Registry ID
AEARCTR-0003571
DOI
https://doi.org/10.1257/rct.3571-1.0
First registered
Last updated
Status
Completed
Trial dates
2018-10-01 – 2018-10-20
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/3571

Abstract

Political leaders and citizens alike have been criticized in the United States for sending people thoughts and prayers, especially in the wake of natural disasters and mass shootings. However, these criticisms do not account for how recipients value such intercessory gestures. We designed an experiment to elicit people’s monetary value (willingness to pay) of intercessory thoughts and prayers, in support after a recent hardship. We find Christians positively value a prayer from a Christian stranger at $4.30, and assign an even higher value ($6.80) to a prayer from a priest. By contrast, non-religious people (Atheists and Agnostics) are “prayer averse” – they are willing to pay $3.40 to avoid receiving a prayer from a Christian stranger. The value of thoughts seems affected by in-group bias over religious belonging -- Christians positively value thoughts from a Christian, but negatively value thoughts from a non-religious stranger. Non-religious people negatively value thoughts from Christians.

Intervention

Our experiment is designed to elicit the value of being prayed for, or thought of, by strangers and a religious authority – a priest – when experiencing a personal hardship. We recruited 482 people from North Carolina to our study, during the weeks following landfall by hurricane Florence. The experiment entails four treatments. In treatment religious thought, subjects are offered to exchange some or all of their $5 endowment for a thought from a Christian stranger, and in treatment non-religious thought subjects are offered to exchange their endowment for a thought from an Atheist stranger. In treatment stranger prayer, subjects are offered to exchange their endowment for a prayer from a Christian stranger, while in treatment priest prayer subjects are offered to exchange their endowment for a prayer from a priest.

Experimental design

Our experiment is designed to elicit the value of being prayed for, or thought of, by strangers and a religious authority – a priest – when experiencing a personal hardship. We recruited 482 people from North Carolina to our study, during the weeks following landfall by hurricane Florence.
Subjects were recruited from Qualtrics consumer panel and the experiment was conducted online. We recruited religious (Catholics and Protestants who believe in God) and non-religious subjects (Atheists and Agnostics who either do not believe in God or are unsure of God’s existence).
In addition to their standard payment from Qualtrics, all subjects received an endowment of US$5 for participating in the experiment. This money could be used in the experiment to pay for a prayer or a thought, by a stranger or a priest. Specifically, the experiment entails four treatments. In treatment religious thought, subjects are offered to exchange some or all of their endowment for a thought from a Christian stranger, and in treatment non-religious thought subjects are offered to exchange their endowment for a thought from an Atheist stranger. In treatment stranger prayer, subjects are offered to exchange their endowment for a prayer from a Christian stranger, while in treatment priest prayer subjects are offered to exchange their endowment for a prayer from a priest. Importantly, in all treatments, subjects could also exchange money not to receive a thought or a prayer. In other words, it was possible for subjects to pay to avoid being prayed for or thought of.
Specifically, the experiment was conducted in the following steps:
1. Subjects were asked questions to elicit information on common demographics (gender, age, income, religion, state of residency). Some of these questions were used for subject screening (religious belonging, belief in God and state of residency), such that they needed to be at the front end of the experimental survey. To diffuse the focus on religion and reduce priming, we added a couple of questions on behavior irrelevant to the study (frequency of buying organic food, spendthriftiness).
2. All subjects conducted a training session on the instrument (a multiple price-list) used later in the survey to elicit their monetary value of a prayer/thought. The multiple price-list is a commonly used instrument in experimental economics to elicit preference revealing willingness-to-pay (WTP) measures of different goods and services. The multiple price-list used herein follows the design of Allcott and Kessler (2015), which allows subjects to state both positive and negative monetary values for energy usage reports.
3. Subjects were asked if they had been affected by hurricane Florence or not. If they had been affected, they were asked to categorize and shortly describe the hardship they experienced from Florence. If they had not, they were asked to categorize and describe another hardship they had experienced in the last 12 months.
4. Subjects were randomized into one of the four treatments. Subjects in treatment religious thought and treatment non-religious thought were informed that they would be matched with a stranger, and offered the opportunity to “receive supportive thoughts from this stranger, aimed at the positive and peaceful resolution of the hardship you described above.” To prevent altruism from impacting subjects’ value of thoughts, they were also informed the stranger would not receive any additional payment from sending such thoughts. They were told the stranger would send supportive thoughts based on the description the subject had provided of their hardship. Subjects in treatment religious thought thereafter found out that the stranger was Christian and believe in God, while subjects in treatment non-religious thought learned the stranger they were matched with was Atheist and does not believe in God.
5. Subjects in all treatments were informed they were endowed with US$5 in financial support of the positive and peaceful resolution of their hardship.
6. Subjects thereafter faced a multiple price-list designed to elicit the monetary value they would assign to a thought/prayer from the stranger/priest. The multiple price-list consisted of 13 different choice pairs, where subjects were asked to choose between a monetary amount and a thought/prayer. The monetary amount ranged from positive to negative over the choice pairs. They learned the computer would choose one of the choice pairs at random, and the choice stated by the subject in that choice pair would then be “binding,” i.e., determine the subject’s payoff from participating in the study. For instance, one of the choice pairs in treatment non-religious thought was:
“Which do you prefer?
(a) The Atheist stranger's supportive thoughts for the positive and peaceful resolution of your hardship + receive $5 in financial support of the positive and peaceful resolution of your hardship
(b) No supportive thoughts from the Atheist stranger + receive $4.50 in financial support positive of the positive and peaceful resolution of your hardship.”
If this choice pair was randomly picked as the binding one by the computer, and the subject had chosen alternative (b) as his/her preferred alternative, the subject would be paid USD 4.50 from participating in the study and receive no thoughts from an Atheist stranger. Note that if the subject chose alternative (b), s/he also indicates a negative willingness to pay for a thought – s/he is willing to abstain money to not receive a thought. Identical multiple price-lists and information was faced by subjects in the other treatments, except the wording differed minimally if the subject was instead offered a prayer, and depending on the sender (Christian stranger/Atheist stranger/priest).
7. All subjects answered a battery of religiosity questions and beliefs about the extent to which intercessory thoughts and prayers can be helpful to the recipient, as well as ethnicity, schooling and political preferences.

Design details

Randomization method
Randomization of subjects into treatments is done by a computer.
Randomization unit
Individual randomization into each treatment, until quotas for required number of subjects (Christians and Atheists/Agnostics) were met.
Clusters
No clusters.
Observations
We were aiming for 460 subjects, but due to oversampling by Qualtrics, we got 482.
Arms
115 subjects in each of our four treatments, with about an equal split between Christians and Atheists/Agnostics.
Primary outcomes
The willingness to pay (WTP) for intercessory thoughts and prayers.

Cited by

No citation data available for this registration yet.

Keywords