Colleges and universities spend hundreds of millions of dollars on predictive analytics software. However, many of these products are operated by private companies that provide little if any transparency about the underlying modeling that drives the risk ratings they generate. The goal of this project is to compare the performance of predictive models when we systematically vary the methods we use to generate risk ratings and the data assumptions that underlie our modeling decisions.
Nationwide, there are 36 million “SCND” (some college, no degree) adults (Shapiro et al, 2019). The evidence we provide in this study can assist efforts to increase re-enrollment and completion among adults by: (1) informing the types of interventions and outreach to be directed at students; and (2) more efficiently targeting the students for whom the benefits of earning a credential is the highest.
Only 58% of degree-seeking students obtain a degree or certificate within six years of starting college. A promising but under-utilized avenue for increasing college completion is to focus on students who already have the majority of the credits they need to complete their degree but who are at risk of dropping out before graduating. Nudges to the Finish Line (N2FL) investigates, through an RCT, whether behaviorally informed “nudges” can increase degree attainment among this population.
Leveraging nearly a decade of employment and enrollment data, we’re partnering with the Virginia Community College System (VCCS) to examine and understand the wage and non-wage employment outcomes of their graduates, and to assess how we might improve on these post-graduation outcomes through targeted intervention.
TBR – The College System of Tennessee and the Nudge4 Solutions Lab are partnering to evaluate the impact of TN Reconnect and to develop and test an intervention that provides adult students with tailored guidance on how they can pursue courses that build on their prior experience.
We are evaluating how CollegePoint, a national initiative providing one-on-one virtual advising to high-achieving, low-and moderate-income high school students, impacts attendance at selective colleges and universities.
A machine-learning algorithm to provide transfer-intending community college students with a personalized list of courses that meet degree requirements and maximize their predicted probability of academic success.
At libraries across the country, youth patrons accumulate fines on overdue books that they struggle to pay, and which prevent them from borrowing additional materials from the library. We are partnering with the Brooklyn Public Library to leverage behavioral nudges to support youth patrons and their families to more effectively manage library circulations and maintain strong engagement with the library system.
We are developing a project that will implement and rigorously evaluate a technology-based intervention to help court-supervised youth attend their court-mandated appointments. We will send automated and personalized text messages to youth to provide them with simplified, highly-personalized information and reminders about the terms of their community release, mandatory dates when they have to appear in court, and caseworker meetings they are required to attend.
We are proud to partner with First Lady Michelle Obama’s Better Make Room campaign to design and implement Up Next, a national mobile messaging campaign that provides youth across the country with important information and reminders about college, financial aid, and loan repayment. This builds on years of large-scale text messaging campaigns that we have conducted.