Can Text Messages Mitigate Summer Melt?

Higher education officials have long been familiar with the concept of “summer melt,” where students who have paid a deposit to attend one college or university instead matriculate at a different institution, usually presumed to be of comparable quality. While melt may be a concern for individual institutions as they try to predict their fall enrollments, historically it has not been viewed as a major policy issue. But education leaders and policymakers have largely overlooked a different and more troubling form of melt: where high school graduates who have been accepted to college and chosen where to enroll as of graduation fail to matriculate anywhere in the fall semester following high school.
Melting away

In previous research, drawing on longitudinal data from various urban school districts throughout the U.S. as well as from a nationally representative survey, we document summer attrition rates ranging from 10% to 40% among high school graduates who had been accepted to college and who had indicated their intentions to enroll in college as of graduation. Summer melt is particularly pronounced among students from disadvantaged backgrounds and could explain a sizable portion of the gap in college entry by socioeconomic status among students who are college-qualified.

Why do college-intending, low-income students melt at such surprisingly high rates? A number of factors may contribute to students with strong college intentions nonetheless failing to successfully matriculate in college. Several of these factors relate to the financing of higher education. For instance, students may acquire new information about college expenses, such as mandatory health insurance charges, which alters their assessment of the costs and benefits of going to college. Other factors relate to the complexity of college and financial aid information. Students and their families may have difficulty accessing, digesting or completing various processes over the summer, such as orientation registration and academic placement tests.

A final set of considerations relates to how students respond behaviorally to the range of tasks they are expected to complete, particularly since students frequently lack access to professional guidance or support during the summer months. For instance, students may put off tasks that are particularly onerous or complex or struggle to keep track of relevant deadlines throughout the summer. As a result, students who have already surmounted many obstacles to college enrollment and who would potentially earn high returns to postsecondary education may nonetheless fail to matriculate.

The challenges that students encounter during the summer are likely familiar to many higher education administrators; a number of colleges and universities invest in summer bridge programs for first-generation college students to address just these issues. Yet, these programs are expensive to operate and, therefore, challenging to bring to scale, particularly as colleges, universities and state governments continue to grapple with budget deficits. Encouragingly, though, much lower-touch outreach and support efforts during the summer months can positively affect students’ enrollment. For example, we find that an intervention as simple as sending students 10 text messages with personalized reminders of important tasks to complete at their intended college can have a sizable impact on whether they realize their postsecondary aspirations.

Texting while aspiring

There are several reasons why text messaging is a promising approach to improve students’ access to high-quality, accurate college information and to connect them to professional support when they need assistance. First, texting is the predominant means by which young people communicate. Whereas only 6% of teens exchange emails and 39% of teens talk via mobile phones, 63% send texts on a daily basis. Moreover, it is straightforward to automate and personalize text-message content to provide students with consolidated and timely information about the tasks required by their intended institution. Finally, personalized messaging may effectively turn one of adolescents’ greatest liabilities—their impulsiveness—into an asset: By providing simplified and timely information, text messages can prompt students to complete required steps in the moment, before their attention is diverted elsewhere.

In summer 2012, we sent college-intending students from four urban school districts text reminders of important tasks to complete over the summer. For instance, the messages reminded students to log on to their intended college’s web portal (e.g. wiser.umb.edu) to access important paperwork; register for orientation and placement tests; complete housing forms; and sign up for or waive health insurance, if relevant. The messages also offered to connect students with a school counselor who could help them complete these tasks.

An additional set of messages offered to connect students to a counselor who could help them complete the federal financial aid application, if they had not done so already, and interpret their financial aid award letter and tuition bill. Most of the messages included web links that allowed students to complete tasks directly from their phone (if they had a smart phone and data plan).

We found  the text messages had a positive impact on whether students enrolled in college—particularly students with less access to quality college information and support with the application process. For instance, in Dallas, Texas, the messages increased enrollment among low-income students by over four percentage points. In Lawrence and Springfield, Mass., where fewer than two in 10 adults have a bachelor’s degree and where there are few school-based supports for college-going, students who received the text messages were over seven percentage points more likely to enroll in college than students who did not receive the messages.

Interestingly, in Boston, where there are many school- and community-based college planning supports—and students already have access to considerable college information—the text messages had zero impact on fall enrollment.

An inexpensive intervention

Perhaps the most exciting feature of the intervention is its cost-effectiveness; the summer text-messaging campaign cost $7 per student, including the expense of hiring counselors to support students when they requested assistance. By comparison, the financial aid literature has typically found that offering students an additional $1,000 in grant aid increases enrollment by three to six percentage points, according to a National Bureau of Economic Research report.

For several reasons, colleges and universities are even better-positioned than school districts to implement a similar outreach campaign.

First, high school graduates may be particularly responsive to communication from the college or university they have indicated an intention to attend, rather than the high school from which they have recently graduated. Second, higher education institutions have access to much better information about the tasks that students have and have not completed, allowing for more targeted and individualized outreach. Finally, one of the primary challenges to any broad-based text-messaging intervention is having an access point to collect students’ cell phone numbers. We relied on high school exit surveys, but obtained working cell numbers for only approximately half to three-quarters of students, depending on the district. The college application provides an optimal vehicle for gathering cell phone contact information, as well as student consent for personalized messages about important college-related tasks.

In fact, text reminders about required summer tasks could be only the midpoint of a more comprehensive, personalized messaging campaign by higher education institutions to increase college access and success among low-income students. Messages could start earlier, perhaps creating opportunities for prospective students to ask questions of current college students who attended high school in the same community. Messages could also continue later; these messages could remind students of important administrative tasks such as registering for courses or renewing the FAFSA or could provide students personalized encouragements to access available supports on campus if they are demonstrating unsatisfactory academic progress. Currently, we are developing new strategies to improve the design and delivery of important college information to students and their families and would welcome the opportunity to partner with higher education institutions on future interventions.

For full report, visit Summer Nudging: Can Personalized Text Messages and Peer Mentor Outreach Increase College Going Among Low-Income High School Graduates?

Benjamin L. Castleman is a doctoral candidate in Quantitative Policy Analysis at Harvard Graduate School of Education. Lindsay C. Page is with Harvard’s Center for Education Policy Research.

Related Posts:

Stemming Summer Learning Loss

Reaching the Connected Generation

Good Intentions: Many Mainers Plan to Go to College, but Don’t

Share This Page

Comments are closed.