Supportive text messages from coaches boosted adult high school graduation rates in a randomized study, with the biggest gains among students paired with experienced coaches.
More than 23 million U.S. adults have not earned a high school diploma, a credential that strongly shapes job prospects and earnings. Yet adult learners who try to return to school often leave before completing a program. A new randomized evaluation by Alberto Ortega, Pietro Pellerito, and Patrick S. Turner examines whether a simple, low-cost tool—supportive text messages from coaches—can help adult high school students persist long enough to graduate.
The researchers studied an adult high school setting in which students were paired with coaches. The intervention was straightforward: coaches sent text messages designed to encourage students to keep going. Because the program was evaluated using random assignment, the results allow a clear comparison between students offered the messaging support and those in the control group.
Graduation rates rose for students assigned to receive the coaching texts. The study finds that treatment assignment increased graduation by 4.9 percentage points. When the analysis focuses on “compliers”—students who actually received the messages as intended—the estimated impact is larger: a 9.3 percentage-point increase in graduation. Put differently, for those who engaged with the texting intervention, graduation rose by roughly 41% relative to the control group’s average graduation rate.
The pathway to higher completion appears to be academic progress rather than keeping students enrolled on paper. The intervention increased the number of courses students completed, indicating that students moved further through their requirements. At the same time, the researchers report no measurable effect on overall enrollment, suggesting the texts did not change whether students stayed officially registered so much as how much they accomplished while enrolled.
Not all messages—or messengers—were equally effective. The gains were concentrated among students assigned to experienced coaches. The authors attribute this pattern to the way seasoned coaches tailored their outreach: they customized messages more extensively than less experienced staff. This finding points to a key operational lesson for programs considering similar nudges: the human element, including coaching skill and personalization, may be as important as the communication channel itself.
To place the results in an economic frame, the authors compare labor market earnings of graduates versus non-graduates. Using that comparison, they estimate an earnings benefit of $149 per treated student over a four-year period—an attempt to quantify the payoff of a low-cost intervention in terms of expected income gains.
The research received financial support from the Wilson Sheehan Lab for Economic Opportunities (LEO) at the University of Notre Dame and benefited from feedback at several academic meetings and seminars, including the Southern Economics Association Annual Meeting, the MidSouth Education Policy Workshop, the Association for Education Finance and Policy (AEFP) 51st Annual Conference, and the University of Notre Dame Applied Micro Brown Bag Seminar. The study was conducted with partners at Goodwill, with the authors noting that the views in the paper are their own and not those of Goodwill Central Texas or the Texas Workforce Commission. The randomized trial is registered with the American Economic Association’s RCT Registry under ID AEARCTR-0006383, and the authors also note that their conclusions do not necessarily reflect the views of the National Bureau of Economic Research.
Why it matters
Adult high school programs serve learners balancing work, family responsibilities, and financial pressure—factors that make persistence difficult even when motivation is high. This study suggests a scalable, inexpensive coaching approach can measurably increase graduation by helping students complete more coursework, especially when delivered by experienced coaches. The next question for policymakers and providers is how to replicate the personalized support that appears to drive the strongest effects while maintaining low program costs.