Nudging to Improve School Attendance

Last registered on October 29, 2018

View Trial History

Pre-Trial

Trial Information

General Information

Title

Nudging to Improve School Attendance

RCT ID

AEARCTR-0003480

Initial registration date

October 29, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

October 29, 2018, 5:06 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

United States of America

Region

Southeast

Primary Investigator

Name

Alexa Prettyman

Affiliation

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Tareena Musaddiq

PI Affiliation

Contact Investigator

PI Name

Jonathan Smith

PI Affiliation

Georgia State University

Contact Investigator

Additional Trial Information

Status

In development

Start date

2018-10-26

End date

2021-05-31

Keywords

Education

Additional Keywords

Attendance, Text Messaging Nudges

JEL code(s)

Secondary IDs

Abstract

The purpose of this study is to increase attendance among chronically absent students through monthly text messages and emails to parents of these children. The study will use existing messaging platforms in four school districts in the metro Atlanta area. The content of the text messages and emails vary slightly based on the age group of the students. Parents of children in grades K-8 will receive messages informing them about their child’s attendance and how it compares to their peers. The content for parents of high school students in grades 9-12 will contain information about their child’s attendance, how it compares to their peers, and how it relates to high school graduation.
Researchers will work with the data provided by the school districts to identify students with chronic absenteeism and assign them to treatment and control groups. The treatment group will receive the text messages while the control group will not receive any messages. At the end of the school year, researchers will analyze the impact of the monthly messages (text and email) on the treated group’s attendance relative to the control group.

External Link(s)

Registration Citation

Citation

Musaddiq, Tareena, Alexa Prettyman and Jonathan Smith. 2018. "Nudging to Improve School Attendance." AEA RCT Registry. October 29. https://doi.org/10.1257/rct.3480-1.0

Former Citation

Musaddiq, Tareena, Alexa Prettyman and Jonathan Smith. 2018. "Nudging to Improve School Attendance." AEA RCT Registry. October 29. https://www.socialscienceregistry.org/trials/3480/history/36387

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Intervention Start Date

2018-11-01

Intervention End Date

2019-05-31

Primary Outcomes

Primary Outcomes (end points)

Student attendance measured by the number of days absent.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

The study will be conducted in two phases. In phase one, the school districts will be in charge of texting parents, and in phase two, researchers will analyze the efficacy of the messages.
Working with the school districts and their existing messaging platform, researchers first receive data on students’ attendance so far in the school year. Researchers then determine who will receive the messages by identifying those students on track to be chronically absent, as measured by the number of absences to date in the school year, and randomizing them into a treatment group. Those students’ unique and de-identified IDs will be uploaded into the messaging platform, where messages are written and delivered by the school district to the parents of the students assigned to the treatment group.
Parents in the treatment group will be messaged once a month (using the existing messaging platform of the school district) until the end of the school year (around May) with the same message but an updated number of absences (provided by the school districts) as the school year unfolds. The message will be personalized with the child’s name, the number of days absent, and the relative percentile rank as compared to classmates. In the first message, parents are given the option to opt-out. They may choose to opt-out at any point during the intervention.
In the second phase, researchers will use secondary data to perform analyses. School districts send de-identified data on the number of absences and basic characteristics and demographics of the student, which is merged with the treatment and control identifier. All of these data, other than the treatment and control identifier already exist (and the number of absences is continually tracked outside of the experiment).

Experimental Design Details

The study will be conducted in two phases. In phase one, the school districts will be in charge of texting parents, and in phase two, researchers will analyze the efficacy of the messages.
Working with the school districts and their existing messaging platform, researchers first receive data on students’ attendance so far in the school year. Researchers then determine who will receive the messages by identifying those students on track to be chronically absent, as measured by the number of absences to date in the school year, and randomizing them into a treatment group. Those students’ unique and de-identified IDs will be uploaded into the messaging platform, where messages are written and delivered by the school district to the parents of the students assigned to the treatment group.
The school districts already have the emails and cell phone numbers within the messaging platform. In addition, we will offer parents an additional layer of consent by sending them an initial opt-out message. Specifically, the first message we send to parents will inform them that they are part of a study and if they do not want to receive messages, they can opt out by clicking on the "unsubscribe" link provided to them over email or text. The unsubscribe link will allow districts to identify the parents who do not wish to participate. This set of parents will then be excluded from the study and will not receive any text messages or emails.
Parents in the treatment group will be messaged once a month (using the existing messaging platform of the school district) until the end of the school year (around May) with the same message but an updated number of absences (provided by the school districts) as the school year unfolds. The message will be personalized with the child’s name and the number of days absent.
In the second phase, researchers will use secondary data to perform analyses. School districts send de-identified data on the number of absences and basic characteristics and demographics of the student, which is merged with the treatment and control identifier. All of these data, other than the treatment and control identifier already exist (and the number of absences is continually tracked outside of the experiment).

Randomization Method

Randomization was done in the office by a computer.

Randomization Unit

Randomization was done at the individual level by grade group. About 70% of the treatment students will be from grades K-8 and the other 30% will be from grades 9-12.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

4 school districts

Sample size: planned number of observations

37,535 students

Sample size (or number of clusters) by treatment arms

32,535 students in the control group
5,000 students in the treatment group, with 3,462 coming from grades K-8 and the other 1,538 coming from grades 9-12
The amount of students per district is based on relative district size. For example, a district with 12% of the students will have 5000*.12=600 students treated, 415 that will be from grades K-8 and 185 from grades 9-12.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Supporting Documents and Materials

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

Request Information

IRB