Text message intervention for student engagement

How to cite

Owen, Stephanie; Burland, Elizabeth. 2021. "Text message intervention for student engagement." AEA RCT Registry. AEARCTR-0007231. https://doi.org/10.1257/rct.7231-1.0

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

Stephanie Owen* , Elizabeth Burland

* Primary investigator

Registry ID
AEARCTR-0007231
DOI
https://doi.org/10.1257/rct.7231-1.0
First registered
Last updated
Status
On going
Trial dates
2021-01-19 – 2025-05-31
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/7231

Abstract

The 2020-21 school year is a challenging one for college students. Many students may be struggling to stay engaged and on top of their coursework in a mostly remote setting, especially those who have less experience navigating college. Even in more conventional academic settings, students often struggle with time management and fail to take advantage of course resources. Prior research has shown that nudge interventions delivered as text messages can improve student outcomes, particularly when they target concrete tasks, are time-sensitive, and are relevant for the particular student (Page, Lee, & Gelbach 2020). This project is a nudge intervention that will send strategic, tailored text messages to undergraduate students in a large introductory economics course. The intervention will take place in the context of fully remote/virtual course instruction during the COVID-19 pandemic. The messages will be customized to the structure of the course and to students' observed behavior such as attendance, performance, and engagement. We will use administrative data to examine whether the messages improve students' engagement with and performance in the course.

Intervention

The intervention consists of strategic, tailored text messages to students, sent throughout the semester. The messages will remind students of key deadlines and benchmarks and direct them to class resources. The timing of these messages will align with the schedule of the course, and they will be tailored to students’ observed behavior. For example, students who have turned in an assignment may receive a different message than those who did not, or students may receive different messages based on their class performance.

Experimental design

All students in the course (N = approx. 370) will be eligible to participate. Students who consent to participate will be randomized into treatment or control using individual level stratified randomization. Strata will be defined by student level (first year vs. sophomore, junior, or senior), gender (male vs. female) and parental education (first-generation college student vs. college-educated parent). Within each of the 8 strata, students will be assigned to treatment or control with equal probability.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual
Clusters
The sample size will depend on how many students consent to participate. The maximum number of observations is 370 students (the total enrollment of the course). The design is not clustered.
Observations
The sample size will depend on how many students consent to participate. The maximum number of observations is 370 students (the total enrollment of the course).
Arms
The sample size will depend on how many students consent to participate. Among those consenting, half will be assigned to treatment and half to control.
Primary outcomes
Academic performance in course; course participation and attendance; engagement with course resources (office hour attendance, posting to course discussion forum)

Cited by

No citation data available for this registration yet.

Keywords