Effect of Learning Recipient Preferences on Donations

How to cite

Hickman, Peter. 2022. "Effect of Learning Recipient Preferences on Donations." AEA RCT Registry. AEARCTR-0008974. https://doi.org/10.1257/rct.8974-1.0

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

Peter Hickman*

* Primary investigator

Registry ID
AEARCTR-0008974
DOI
https://doi.org/10.1257/rct.8974-1.0
First registered
Last updated
Status
Completed
Trial dates
2022-02-13 – 2022-02-14
Countries
United States of America (Online: Prolific)
Registry record
http://www.socialscienceregistry.org/trials/8974

Abstract

A large amount of aid money flows annually from wealthy nations to less developed nations, with roughly one quarter from private donations. A longstanding policy question is what form this aid should take. Improving the quality of aid requires not only understanding the effects of different forms of aid but also knowing why decision-makers send some forms of aid instead of others. This study aims to provide new evidence on the behavior of possible aid donors. In particular, we study the extent to which donors update their donation decisions after learning about recipient's preferences for cash vs goods as aid.

Experimental design

We assign 100 people to each of three different treatments:
A) Control
B) Preferences
C) Consideration

In the control treatment, participants allocate $50 between themselves and poor Kenyan recipients. They can choose to allocate either via a cash transfers or by purchasing goods to give to the Kenyans. We explain the five goods they can use funds to purchase. Donors are informed about the cost to purchase and transport each of the goods.

The Preferences treatment also informs participants about the fraction of Kenyans who wanted the good instead of the money it takes to purchase and transport the good. Then donors update their allocations.

The Consideration treatment does not show participants the recipients' preferences, but rather asks donors to guess the recipient's preferences.

The primary hypothesis is that the Donors respond to learning recipient preferences by increasing cash donations and decreasing goods donations. The secondary hypothesis is that simply considering what recipients' preferences might be (the consideration treatment) also leads donors to shift towards cash transfers.

Design details

Randomization method
Computer. Treatment assignments done in Stata then uploaded to SQL database. Online participants are assigned an ID when they begin survey, and each ID has a treatment variable in the database.
Randomization unit
Individual survey taker.
Clusters
300
Observations
300
Arms
100 per arm
Primary outcomes
Amount of money allocated to cash transfers
Secondary outcomes
Amount of money allocated to goods

Cited by

No citation data available for this registration yet.

Keywords