The Hidden Cost of Nudging: Experimental Evidence on Reminders and Unsubscribing

How to cite

Gravert, Christina; Damgaard, Mette Trier. 2015. "The Hidden Cost of Nudging: Experimental Evidence on Reminders and Unsubscribing." AEA RCT Registry. AEARCTR-0000762. https://doi.org/10.1257/rct.762-2.0

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

Christina Gravert* , Mette Trier Damgaard

* Primary investigator

Registry ID
AEARCTR-0000762
DOI
https://doi.org/10.1257/rct.762-2.0
First registered
Last updated
Status
In development
Trial dates
2015-07-07 – 2015-09-14
Countries
Private
JEL codes
C93, D03, D64
Registry record
http://www.socialscienceregistry.org/trials/762

Abstract

Studies of the effects and effectiveness of reminders in economics have documented that reminders can influence behavior in desired ways across a wide variety of settings. However, using a large scale field experiment with a charitable organisation, we document that this may come at a cost because reminders may push some recipients to unsubscribe. We set up a model explaining this finding and test the model's implications using another large scale field experiment.

Experimental design

We will send an e-mail asking for a charitable donation to a project targeting a reduction in food waste to 35,000 newsletter subscribers of Folkekirkens Nødhjælp. The receivers will be randomized into 3 equal sized treatment groups - a baseline to fix beliefs, a treatment to reduce the annoyance of the newsletter and a treatment to increase the option value of staying on the list.

Design details

Randomization method
The randomization is done by the computer, creating 3 equal sized groups out of all the subscribers on the mailing list.
Randomization unit
We can randomize on person level since we don't assume that there will be any spillover effects that we need to control for. No clustering is necessary in this case.
Clusters
No clusters, every person is treated individually with no spillover effects.
Observations
35,000 newsletter subscribers will be contacted, but the response rate will likely be at around 4 percent, so 1400 responses.
Arms
11,666 per treatment arm, so 466 responses per arm.
Primary outcomes
While the main outcome of interest is the decision to unsubscribe, we also measure how these nudges affect the decision to donate and the amount donated. In addition we look at the answers given to the question asked, when people decide to unsubscribe.

Cited by

No citation data available for this registration yet.

Keywords