Association for Public Policy Analysis & Management

In this paper, we examine the impact of providing personalized college information via text messaging to West Virginia students starting during the spring of their senior year and continuing through their freshman year of college. The goal of these messages is to increase college enrollment and persistence. Students from 14 high schools served by the West Virginia GEAR UP program received customized college and financial aid information, and had the opportunity to request, via text messages, individualized help first from the GEAR UP coordinator in the student’s high school and then from members of the appropriate office at the student’s college/university of matriculation. This innovative approach successfully transitioned student mentorship from high school counselors to college administrators, ensuring students did not suffer a gap in support during their transition into college.

The project focused on three distinct and equally challenging phases during a student’s college application and enrollment process: first, during the spring of the senior year in high school, messages focused on completing the Free Application for Federal Student Aid (FAFSA) and filing school-specific financial aid forms; second, during the summer after high school, messages reminded students about key, institution-specific tasks they need to complete prior to matriculation; and third, during the freshman year of college, messages reminded students of academic and social supports available on campus and encouraged them to re-apply for financial aid during the spring semester.

We employ two evaluation strategies to assess the impact of the texting campaign on students’ outcomes. First, we use a differences-in-differences strategy, in which we compare students at the 14 West Virginia GEAR UP high schools to students in the prior graduation year and at non-GEAR UP high schools within the state. Second, we use a propensity matching estimation strategy, in which we match students who signed up for the texting campaign to two separate groups of students: (1) observationally similar students from their high school who did not sign up for the texting campaign; and (2) observationally similar students from comparison high schools who were not eligible to receive the text messages. Our presentation will report preliminary impacts of the text campaign on students’ college application behaviors, enrollment choices, and academic performance in college.