Choice set structure and donation decisions: a field experiment

Last registered on March 10, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Choice set structure and donation decisions: a field experiment
RCT ID
AEARCTR-0018028
Initial registration date
March 06, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
March 10, 2026, 10:33 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

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

Request Information

Primary Investigator

Name
Konstantinos Ioannidis
Affiliation
University of Cambridge
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Eriko Yamakuma
PI Affiliation
University of Cambridge
Contact Investigator
PI Name
Atiyeh Yeganloo
PI Affiliation
University of Cambridge
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-03-16
End date
2026-07-31
Keywords
Behavior
Additional Keywords
charitable giving, field experiment
JEL code(s)
C93, D01, D64, D91, H41
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
This study investigates how the structure of donation options affects how much people give. Using a field experiment with alumni donors at university Colleges, we compare three treatments varying the donation designs. We compare a Restricted treatment, where donors must select a single cause to support, against an Unrestricted treatment, where donors can select multiple causes to create a donation portfolio. Additionally, we create a Default treatment where they must pick one, but we make the decision easier by introducting and preselecting a general cause (area of greatest need). We are interested in whether removing choice constraints increases the total donation amount (intensive margin) and the likelihood of giving (extensive margin). Following the donation, participants fill in a survey to explore the underlying sociological and psychological mechanisms of alumni giving behaviors, using measures such as network-based social capital, warm glow, diversification bias, and multidimensional identity. Our results aim to provide insights into how choice architecture can be optimized to enhance charitable giving in institutional settings.
External Link(s)

Registration Citation

Citation
Ioannidis, Konstantinos, Eriko Yamakuma and Atiyeh Yeganloo. 2026. "Choice set structure and donation decisions: a field experiment." AEA RCT Registry. March 10. https://doi.org/10.1257/rct.18028-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Participants are randomly assigned to one of three donation interfaces to support university College initiatives:
(i) Restricted: Donors must choose exactly one cause from a fixed list before entering an amount.
(ii) Unrestricted: Donors may select one or more causes from the same list. If multiple causes are selected, they specify individual amounts for each in a second stage.
(iii) Default: Similar to the Restricted treatment, but an additional general initiative is added and pre-selected.
Intervention Start Date
2026-03-16
Intervention End Date
2026-06-30

Primary Outcomes

Primary Outcomes (end points)
Donation Amount: measured in GBP (intensive margin)
Donation Propensity: measured as binary (extensive margin)
Primary Outcomes (explanation)
Donation Propensity is defined as a binary with value 1 if Donation Amount is positive, and 0 otherwise.

Secondary Outcomes

Secondary Outcomes (end points)
Survey Scores: Likert-scale measures on network-based social capital within College community and psychological mechanisms underlying donation.
Number of Causes Supported: For the Unrestricted treatment, the count of unique initiatives donated to.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
It is conducted as a natural field experiment where alumni receive a donation appeal via email. They are randomly presented with one of three web interfaces, which vary as described in the Intervention section.

After making their donation, donors are invited to complete a survey. The survey is designed to map social networks within Colleges, understand the psychological factors behind donors’ choices, such as warm‑glow effects and diversification bias, and collect basic demographic information.
Experimental Design Details
Not available
Randomization Method
Computer-generated random assignment into the three treatments. The randomization is within each participating College, ensuring that approximately one third of the alumni of each College will be allocated to each treatment. Thus, when estimating treatment effects, all College specific effects are averaged.
Randomization Unit
Individual
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A
Sample size: planned number of observations
The Colleges we are collaborating with have roughly 24,000 alumni who will receive the e-appeal. While the response rate is not under the experimenter’s control, assuming a response rate of 3% to the e-appeal, we expect to have a minimum of 720 individual donors in total.
Sample size (or number of clusters) by treatment arms
Given our randomization process, we expect 240 individual donors per treatment arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations are based on a two-sample t-test for differences in means. Since we do not know the available donor budget from which donations are made in a field experiment, we benchmark the Restricted treatment to have a mean of £40 with a standard deviation of £30. Those numbers imply a Coefficient of Variation of 0.75, which is a conservative estimate for alumni donation data. With 240 observations per arm, the study is powered to detect a 16% increase in donations with ~75% power and a 20% increase with >90% power. A previous online study which motivated the current design found an increase of 32% between Restricted and Unrestricted, so we expected to have sufficient power to detect treatment effects even if they are of much smaller magnitude.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Cambridge Judge Business School, Departmental Ethics Review Group Approva
IRB Approval Date
2025-10-13
IRB Approval Number
25-51
Analysis Plan