Do expressions of care crowd-out monetary donations?

How to cite

James, Alexander; Thunstrom, Linda; Veld, Klaas Van 't; Noy, Shiri. 2026. "Do expressions of care crowd-out monetary donations?." AEA RCT Registry. AEARCTR-0018029. https://doi.org/10.1257/rct.18029-1.0

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

Alexander James* , Linda Thunstrom , Klaas Van 't Veld , Shiri Noy

* Primary investigator

Registry ID
AEARCTR-0018029
DOI
https://doi.org/10.1257/rct.18029-1.0
First registered
Last updated
Status
On going
Trial dates
2026-02-27 – 2026-03-31
Countries
Private
JEL codes
C9; D64
Registry record
http://www.socialscienceregistry.org/trials/18029

Abstract

Existing literature suggests people give in part because it makes them feel good, or receive "warm glow". But there is also evidence that people give due to social pressure and use donations as a public signal of their pro-social type. As such expressing one's level of care for a charitable cause reduces the signaling power of additional action and so reduces donations. We test this theory using an online survey experiment in which subjects were given the opportunity to donate to the American Society for the Prevention of Cruelty to Animals (ASPCA). We hypothesize that giving people the opportunity to express their level of care for animals in need reduces their monetary donations to the ASPCA.

Intervention

Existing literature suggests people give in part because it makes them feel good, or receive "warm glow". But there is also evidence that people give due to social pressure and use donations as a public signal of their pro-social type. According to this theory, expressing one's level of care for a charitable cause reduces the signaling power of additional action and so reduces donations. We test this theory using an online survey experiment in which subjects were given the opportunity to donate to the American Society for the Prevention of Cruelty to Animals (ASPCA). All subjects are given a $5 bonus payment. Subjects assigned to the control group are simply given the opportunity to donate some amount of money (between $0 - $5) to the ASPCA. Subjects assigned to the "care only" treatment are simply asked to state their level of care for abused animals. Subjects assigned to the "decision" treatment first decide whether to i) donate only, ii) express their level of care only, or iii) donate & express their level of care. If their decision involves expressing their level of care or donating, these decisions are made on the following page.

Experimental design

The survey experiment was designed and implemented using Qualtrics. Subjects first provide their consent to participate in the study and then answer a set of demographic questions. Subjects are then told they will receive a $5 bonus payment before being randomly assigned to one of three treatments:
i) donate
ii) express care
iii) donate and/or express care
Subjects assigned to the first treatment and simply asked if they would like to donate some of their $5 bonus payment to the ASPCA. Subjects in the second treatment are simply given the opportunity to state how much they care about abused animals. Subjects in the third treatment are given the option of i) donating, ii) expressing care, iii) both, or iv) nothing. For subjects in this third treatment, if their choice includes donating or expressing care, they do so on a following page. After donation decisions are made, additional demographic information is collected.

Design details

Randomization method
Randomization is achieved using Qualtrics logic.
Randomization unit
individual
Clusters
1
Observations
800
Arms
Roughly 266 subjects in each of the 2 treatments and control.
Primary outcomes
Our primary outcomes of interest are i) the rate at which people donate to the ASPCA and ii) the amount that people donate to the ASPCA.
Secondary outcomes
In one treatment described above, we give subjects the opportunity to express their level of care for abused animals without the option to also donate. We will compare levels of care across treatments. In this case, the outcome is the reported level of care. We will explore a variety of sources of heterogeneity including subject i) reputational concerns, ii) gender, iii) age, and iv) religiosity.

Cited by

No citation data available for this registration yet.

Keywords