x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
The Nuts and Bolts of SMS for Parental Engagement
Last registered on June 26, 2016

Pre-Trial

Trial Information
General Information
Title
The Nuts and Bolts of SMS for Parental Engagement
RCT ID
AEARCTR-0001381
Initial registration date
June 26, 2016
Last updated
June 26, 2016 3:53 PM EDT
Location(s)
Region
Primary Investigator
Affiliation
University of Zurich
Other Primary Investigator(s)
PI Affiliation
Stanford University
PI Affiliation
Stanford University
PI Affiliation
University of São Paulo
Additional Trial Information
Status
In development
Start date
2016-07-11
End date
2017-01-27
Secondary IDs
Abstract
A growing education literature suggests that supporting parents through text messages (SMS) can positively impact students’ behavior and educational attainment. While those studies highlight the potential of text messages for producing cost-effective educational results, there is limited evidence on the optimal design of SMS campaigns. What it the optimal frequency of texting, so as to most effectively capture parents’ attention without saturating it? At what time should messages be sent? Should parents get messages always at the same time? Is interactive content more effective? The answers to those questions are critical as governments and international organizations consider scaling up successful SMS interventions. This paper cross-randomizes different features of the design of a typical SMS campaign targeted at making parenting a habit among families of public schools’ 9th graders in Brazil. Those experiments assess the impacts of alternative campaign parameters: (i) frequency , (ii) time of the day, (iii) consistency of time of delivery, and (iv) interactivity, on student’s attendance, grades, and drop-out rates.
External Link(s)
Registration Citation
Citation
Bettinger, Eric et al. 2016. "The Nuts and Bolts of SMS for Parental Engagement." AEA RCT Registry. June 26. https://doi.org/10.1257/rct.1381-1.0.
Former Citation
Bettinger, Eric et al. 2016. "The Nuts and Bolts of SMS for Parental Engagement." AEA RCT Registry. June 26. https://www.socialscienceregistry.org/trials/1381/history/9055.
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)
Typical SMS campaign targeted at making parenting a habit among families of public schools’ 9th graders in Brazil.
Intervention Start Date
2016-07-11
Intervention End Date
2016-12-02
Primary Outcomes
Primary Outcomes (end points)
We will conduct surveys through automated voice calls (Interactive Voice Response, IVR) at the end of the intervention to collect self-reported parenting practices and parents’ views about their children.

At the end of the intervention, the São Paulo Education Secretariat will provide data on student attendance and grades in 2016 (per quarter), and enrollment in 2017. Moreover, the Secretary of Education of São Paulo implements annually a standardized test to all schools in the state of São Paulo, SARESP (System of School Performance Evaluation of the State of São Paulo). All students in grades 1st, 3rd, 5th, 7th, 9th of primary school and the 3rd (final) year of high school are tested on their knowledge of Mathematics and Portuguese.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
Campaign parameters are randomly assigned at the student level, comprising a sample of 6000 students within of 180 classrooms at 60 Brazilian public schools.

Assignment to each treatment branch across the four experiments is cross-randomized, except in what comes to the control group, since those receiving no messages cannot be assigned to other campaign parameters.

Experiment 1 randomly assigns the frequency at which SMS messages are delivered. The control group receives no messages. The decision to assign 1/3 of the sample to this group is based on maximing power for Experiments 2 through 4. Treatment 1A (1/3 of the remaining subject pool) receives 1 message a week, a suggestion of activity for parents to do along with their children (delivered on Wednesday). Treatment 1B (also 1/3 of the remaining subject pool) receives 2 messages a week, a ‘fact’ with information about how an activity is linked to children’s development (delivered on Monday) and a suggestion of activity for parents to do along with their children (delivered on Wednesday). Treatment 1C (also 1/3 of the remaining subject pool) receives 3 messages a week, a ‘fact’ with information about how an activity is linked to children’s development (delivered on Monday), a suggestion of activity for parents to do along with their children (delivered on Wednesday), and a reinforcement of that activity, which tries to make it a habit (delivered on Friday).

Experiment 2 randomly assigns the time of the day at which messages are delivered. Treatment 2A (1/3 of the sample) receives messages at the evening (7pm), while Treatment 2B (also 1/3 of the sample) receives messages at the afternoon (noon).

Experiment 3 randomly assigns the consistency of SMS delivery. Treatment 3A (1/3 of the sample) receives messages at always the same time of the day (either noon or 7pm), while Treatment 3B (also 1/3 of the sample) receives messages at alternating times (at the scheduled time, 1 hour before and 1 hour after, following a 3-week cycle).

Last, Experiment 4 randomly assigns whether content is interactive. Treatment 3A (1/3 of the sample) receives a follow-up message (delivered on Thursday) asking whether the parent complied with the activity suggested the day before – to which parents can reply ‘yes’ or ‘no’ –, while Treatment 3B does not receive follow-up messages.
Experimental Design Details
Campaign parameters are randomly assigned at the student level, comprising a sample of 6000 students within of 180 classrooms at 60 Brazilian public schools. While there is a concern that assigning different treatments within the same classroom may lead to contamination, we are less worried about it in this setting parents typically have no recurring interactions at this age – most of them no longer take their children to school, and parent-teacher meetings are rather infrequent in Brazilian public schools. Having said that, both potential contamination and students’ peer effects are expected to bias our estimates towards not detecting differences across the variations in the campaign parameters. The research design is outlined in Table 1. Assignment to each treatment branch across the four experiments is cross-randomized, except in what comes to the control group, since those receiving no messages cannot be assigned to other campaign parameters. Experiment 1 randomly assigns the frequency at which SMS messages are delivered. The control group receives no messages. The decision to assign 1/3 of the sample to this group is based on maximing power for Experiments 2 through 4. Treatment 1A (1/3 of the remaining subject pool) receives 1 message a week, a suggestion of activity for parents to do along with their children (delivered on Wednesday). Treatment 1B (also 1/3 of the remaining subject pool) receives 2 messages a week, a ‘fact’ with information about how an activity is linked to children’s development (delivered on Monday) and a suggestion of activity for parents to do along with their children (delivered on Wednesday). Treatment 1C (also 1/3 of the remaining subject pool) receives 3 messages a week, a ‘fact’ with information about how an activity is linked to children’s development (delivered on Monday), a suggestion of activity for parents to do along with their children (delivered on Wednesday), and a reinforcement of that activity, which tries to make it a habit (delivered on Friday). Experiment 2 randomly assigns the time of the day at which messages are delivered. Treatment 2A (1/3 of the sample) receives messages at the evening (7pm), while Treatment 2B (also 1/3 of the sample) receives messages at the afternoon (noon). Experiment 3 randomly assigns the consistency of SMS delivery. Treatment 3A (1/3 of the sample) receives messages at always the same time of the day (either noon or 7pm), while Treatment 3B (also 1/3 of the sample) receives messages at alternating times (at the scheduled time, 1 hour before and 1 hour after, following a 3-week cycle). Last, Experiment 4 randomly assigns whether content is interactive. Treatment 3A (1/3 of the sample) receives a follow-up message (delivered on Thursday) asking whether the parent complied with the activity suggested the day before – to which parents can reply ‘yes’ or ‘no’ –, while Treatment 3B does not receive follow-up messages.
Randomization Method
Randomization done in office by a computer
Randomization Unit
Parent
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
6000 parents
Sample size: planned number of observations
6000 parents
Sample size (or number of clusters) by treatment arms
About 1500 in Experiment 1, and 2000 in Experiments 2, 3 and 4.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials

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

Request Information
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Stanford University IRB
IRB Approval Date
2016-05-10
IRB Approval Number
35332
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