Nudging with reminders – the role of beliefs and emotions
Last registered on November 06, 2019

Pre-Trial

Trial Information
General Information
Title
Nudging with reminders – the role of beliefs and emotions
RCT ID
AEARCTR-0004018
Initial registration date
March 18, 2019
Last updated
November 06, 2019 1:39 PM EST
Location(s)
Region
Primary Investigator
Affiliation
University of Copenhagen
Other Primary Investigator(s)
PI Affiliation
WZB
PI Affiliation
Aarhus University
Additional Trial Information
Status
In development
Start date
2019-03-18
End date
2019-12-31
Secondary IDs
Abstract
This study aims to identify the potential risks that can result from using one of the most popular nudges, reminders, to change behavior. We analyze what drives reminder response and in particular the role of reminders as a means to overcome attentional problems, informational limitations or invoke emotions such as guilt or reassurance. The method of research is a large-scale field experiment.
External Link(s)
Registration Citation
Citation
Barron, Kai , Christina Gravert and Mette Trier Damgaard. 2019. "Nudging with reminders – the role of beliefs and emotions." AEA RCT Registry. November 06. https://doi.org/10.1257/rct.4018-1.1.
Former Citation
Barron, Kai et al. 2019. "Nudging with reminders – the role of beliefs and emotions." AEA RCT Registry. November 06. https://www.socialscienceregistry.org/trials/4018/history/56578.
Sponsors & Partners

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

Request Information
Experimental Details
Interventions
Intervention(s)
We are working together with a mobile-health provider in Cape Town to study the effect of text message reminders on medication adherence for pregnant women.
Intervention Start Date
2019-03-19
Intervention End Date
2019-06-30
Primary Outcomes
Primary Outcomes (end points)
Response rates to the incentivized tasks
Unsubscription from the messages
Willingness to pay for reminders
Observed time preferences from a task
Primary Outcomes (explanation)
The time preferences will be elicited in a task in which the subjects have to write words backward to earn money, they earlier they do the task, the less work it is. We will measure how early they do the task and whether this is consistent with their stated intentions on when to do the task.
Secondary Outcomes
Secondary Outcomes (end points)
Stated adherence
Stated beliefs about the importance of iron supplements
Stated anemia symptoms
Quiz test results
Secondary Outcomes (explanation)
There will be several survey outcomes, which will serve as a complement to the main behavioral outcomes and the willingness to pay.
Experimental Design
Experimental Design
Over the course of 3 months, a randomly selected subset of pregnant women will receive text messages with health information. There will be 6 treatment arms. We will measure their engagement with the program as well as their responses to incentivized tasks and their answers to quiz and survey questions.
Experimental Design Details
The experiment will be an addition to the regular health text messages the organization sends out to expectant mothers. We will select a random subset of mothers for our experiment, who will receive additional text messages over the course of 3 months. The expectant mothers participating in the study are randomized into one of six treatment groups. 1) Control: Will receive no additional messages on iron supplements than those provided by the standard MomConnect program. 2) Promise: Will receive no additional messages on iron supplements than those provided by the standard MomConnect program, except a message during the treatment period suggesting that they make a promise to their unborn baby to take iron supplements every day. The promise permanently increases the weight on emotional utility to mP > m0 where m0 is the baseline weight on moral utility. 3) Attentional Reminder: Will receive additional messages during the treatment period about iron supplements with no additional informational content i.e. for this group the probability that MomConnect sends a message is temporarily higher i.e. pR t > p0 t during the treatment period where p0t is the baseline level in the control and promise treatments. The temporary increase in the frequency of messages is announced and after the treatment period, the probability of receiving a message about iron supplements is again at the baseline level of p0t . 4) Informational Reminder: Will receive pre-announced additional messages about iron supplements with new informational content about the health benefits of taking iron supplements. For this group, MomConnect sends a message with probability pRt during the treatment period and the messages permanently improve the accuracy of the perceptions of the health benefits to gI > g0 where g0 represents the baseline awareness level. The improvement in accuracy Deltag = gI-g0 is achieved by sending n informational messages. Each of these messages on average improves the accuracy by Dgn . After the treatment period, the probability of receiving a message about iron supplements is again at the baseline level of p0t . 5) Informational Reminder + Promise: Combines the informational reminders with the promise treatment. That is the treatment permanently raises m and g to mP and gI, respectively. The increase in g is delivered gradually over the treatment period and after the treatment period the probability of receiving a message about iron supplements is again at the baseline level of p0t . 6) Time preferences: We develop a novel time preference elicitation task that will elicit the mothers' time preferences using a simple word task. Otherwise the treatment will be identical to the control treatment.
Randomization Method
We will draw a random sample of ca 15,000 mothers from the population of all pregnant women in South Africa who are signed-up to the MomConnect program. Randomization will be done by the software RapidPro.
Randomization Unit
After consent, mothers will be randomly assigned to one of the 6 treatments. Within the treatment 6, there will be an additional sub-experiment where mothers will be randomized into 4 incentive levels for one of the sub-tasks.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
No clusters
Sample size: planned number of observations
15,000 women will be invited.
Sample size (or number of clusters) by treatment arms
2500 mothers per groups will be invited. We anticipate a response rate of 25%, so 625 mothers per treatment arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Because variation in outcomes is hard to anticipate, we did not do a formal power calculation.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Pharma Ethics
IRB Approval Date
2019-01-10
IRB Approval Number
181021588
Analysis Plan

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

Request Information
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers