Annoyance costs of reminders

How to cite

Gravert, Christina; Meer, Janna Ter. 2022. "Annoyance costs of reminders." AEA RCT Registry. AEARCTR-0009608. https://doi.org/10.1257/rct.9608-1.1

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

Christina Gravert* , Janna Ter Meer

* Primary investigator

Registry ID
AEARCTR-0009608
DOI
https://doi.org/10.1257/rct.9608-1.1
First registered
Last updated
Status
In development
Trial dates
2022-07-30 – 2023-07-31
Countries
Private
JEL codes
M31, C93, I19
Registry record
http://www.socialscienceregistry.org/trials/9608

Abstract

We study the impact of reminders on unsubscriptions from and engagement with The All of Us Research Program (AoURP). While reminders are generally seen as effective in encouraging individuals to complete the requested task, they may also impose an ‘annoyance cost’ on the user (Damgaard and Gravert, 2018 (D&G2018)). This in turn can increase the likelihood that users may unsubscribe from the communication or otherwise disengage from future communication efforts. In this study, we set out to replicate the finding by D&G2018 in a different setting and further analyze the mechanisms through which beliefs about future annoyance costs arise.

Intervention

After an individual registers for the All of Us Research Program (AoURP), but does not complete consent, they are defaulted into an e-mail sequence of a total of 10 reminders to complete the consent. The first email comes 5 hours after registration and the remaining emails come every 5 days (e.g., after 5 days, after 10 days, after 15 days). We vary the information given in the first reminder concerning the cadence of the reminders they will receive to investigate the effect of beliefs on unsubscriptions.

Note that one of the platforms on which this intervention will be implemented, will feature a total of 3 reminders. We currently expect that the platform will update the experiment to the 10 reminders one month after data collection begins.

Experimental design

We will vary the information people will receive about the cadence of the reminders to consent to taking part in the research study they will receive. The experiment is a natural field experiment embedded in the AoURP reminder sequence.

Design details

Randomization method
Users will be randomized to experimental conditions when creating an account. This randomization will be set up by Scripps using RKStudio and by PTSC using Iterable.
Randomization unit
The level of randomization are individuals.
Clusters
Based on platform traffic between March and May 2022 and outreach forecasts, we expect that 1,180 individuals become eligible for the evaluation each week across both CareEvolution (CE) and PTSC platforms (about ~5,100 per month). We have policy support to run the evaluation for three months. This leads to 15,300 individuals in total.
Observations
15,300
Arms
5100 per arm.
Primary outcomes
The unsubscribe rate after receiving the first reminder (before the second reminder) and the cumulative unsubscribe rate over the series of reminders. The primary consent rate after receiving the first reminder (before the second reminder) and the cumulative consent rate for the full series of reminders.
Secondary outcomes
The average number of data donations after primary consent within a 30 day period of completing primary consent. The open rate for each reminder email (as far as possible per platform) The click rate for each reminder

Cited by

No citation data available for this registration yet.

Keywords