Search for EdWorkingPapers here by author, title, or keywords.
College completion rates declined from the 1970s to the 1990s. We document that this trend has reversed--since the 1990s, college completion rates have increased. We investigate the reasons for the increase in college graduation rates. Collectively, student characteristics, institutional resources, and institution attended do not explain much of the change. However, we show that grade inflation can explain much of the change in graduation rates. We show that GPA is a strong predictor of graduation rates and that GPAs have been rising since the 1990s. We also find that increases in college GPAs cannot be explained by student demographics, ability, and school factors. Further, we find that at a public liberal arts college, grades have increased over time conditional on final exam performance.
Enrollment in higher education has risen dramatically in Latin America, especially in Chile. Yet graduation and persistence rates remain low. One way to improve graduation and persistence is to use data and analytics to identify students at risk of dropout, target interventions, and evaluate interventions’ effectiveness at improving student success. We illustrate the potential of this approach using data from eight Chilean universities. Results show that data available at matriculation are only weakly predictive of persistence, while prediction improves dramatically once data on university grades become available. Some predictors of persistence are under policy control. Financial aid predicts higher persistence, and being denied a first-choice major predicts lower persistence. Student success programs are ineffective at some universities; they are more effective at others, but when effective they often fail to target the highest risk students. Universities should use data regularly and systematically to identify high-risk students, target them with interventions, and evaluate those interventions’ effectiveness.
This study used college dormitory room and social group assignment data to investigate the peer effect on the probability of college students switching their major fields of study. The results revealed strong evidence of peer effects on students’ decisions to switch majors. In particular, the number of a student’s peers who have the same major significantly reduces the student’s likelihood of switching majors; however, when a same-major peer switches majors, it significantly increases a student’s probability of switching majors. This study also found that peers’ majors affected students’ choice of destination majors. Students in the same peer group are more likely to choose the same destination majors, compared to non-peers. Finally, we found that in general peer effects at the dormitory room level, both in choice and persistence of major, were stronger than were peer effects at the social group level.
Success in postsecondary education requires students to engage with their institution both academically and administratively. As with the transition to college, administrative requirements students face once enrolled can be substantial. Missteps with required processes can threaten students’ ability to persist. During the 2018-19 academic year, Georgia State University implemented an artificially intelligent text-based chatbot to provide proactive outreach and support to help undergraduates navigate administrative processes and take advantage of campus resources. A team of centralized university administrators orchestrated outreach “campaigns” to support students across three broad domains: (1) academic supports; (2) social and career supports; and (3) administrative processes. We investigate GSU’s implementation of this persistence-focused chatbot through an experimental study. Of the three message domains, outreach was most effective when focused on administrative processes, many of which were time-sensitive and for which outreach could be targeted specifically to students for whom it was relevant based on administrative data. In contrast, outreach to encourage take up of other supports had little effect on student behavior. By the end of the academic year, rates of FAFSA filing and registration for the subsequent fall semester were approximately three percentage points higher, suggesting positive effects on year-to-year college persistence. The positive effects on fall enrollment persisted into summer 2019, at which time the GSU administration judged that the study results were compelling enough to conclude the experiment and roll the chatbot system out to all students. We situate our findings in the literature on nudge-type efforts to support college access and success to draw lessons regarding their effective use.
Tens of millions of Americans have lost their jobs in the wake of the COVID-19 health and economic crisis, and a sizable share of these job losses may be permanent. Unemployment rates are particularly high among adults without a college degree. Recent state policy efforts have focused on increasing re-enrollment and credentialing among adults with some college but no degree (SCND); these efforts are likely to accelerate given the COVID-19 disruptions to the U.S. economy. Yet little is actually known about the background characteristics, academic experiences, or labor market trajectories of this population. Using data from the Virginia Community College System (VCCS), we provide the first detailed profile on the academic, employment, and earnings trajectories of the SCND population, and how these compare on key measures to VCCS graduates. We also develop a framework for prioritizing which segments of the SCND population states might target for re-enrollment and completion interventions. This framework may be particularly useful to states that need to fill critical workforce shortages in healthcare and other sectors or re-train their workforce in the wake of mass unemployment and economic disruption stemming from the COVID-19 crisis.
We provide the first estimated economic impacts of students’ access to an entire sector of public higher education in the U.S. Approximately half of Georgia high school graduates who enroll in college do so in the state’s public four-year sector, which requires minimum SAT scores for admission. Regression discontinuity estimates show enrollment in public four-year institutions boosts students’ household income around age 30 by 20 percent, and has even larger impacts for those from low income high schools. Access to this sector has little clear impact on student loan balances or other measures of financial health. For the marginal student, enrollment in such institutions has large private returns even in the short run and positive returns to state budgets in the long run.
Family and social networks are widely believed to influence important life decisions but identifying their causal effects is notoriously difficult. Using admissions thresholds that directly affect older but not younger siblings’ college options, we present evidence from the United States, Chile, Sweden and Croatia that older siblings’ college and major choices can significantly influence their younger siblings’ college and major choices. On the extensive margin, an older sibling’s enrollment in a better college increases a younger sibling’s probability of enrolling in college at all, especially for families with low predicted probabilities of enrollment. On the intensive margin, an older sibling’s choice of college or major increases the probability that a younger sibling applies to and enrolls in that same college or major. Spillovers in major choice are stronger when older siblings enroll and succeed in more selective and higher-earning majors. The observed spillovers are not well-explained by price, income, proximity or legacy effects, but are most consistent with older siblings transmitting otherwise unavailable information about the college experience and its potential returns. The importance of such personally salient information may partly explain persistent differences in college-going rates by geography, income, and other determinants of social networks.
After increasing in the 1970s and 1980s, time to bachelor’s degree has declined since the 1990s. We document this fact using data from three nationally representative surveys. We show that this pattern is occurring across school types and for all student types. Using administrative student records from 11 large universities, we confirm the finding and show that it is robust to alternative sample definitions. We discuss what might explain the decline in time to bachelor’s degree by considering trends in student preparation, state funding, student enrollment, study time, and student employment during college.
Selective college admissions are fundamentally a question of tradeoffs: Given capacity, admitting one student means rejecting another. Research to date has generally estimated average effects of college selectivity, and has been unable to distinguish between the effects on students gaining access and on those losing access under alternative admissions policies. We use the introduction of the Top Ten Percent rule and administrative data from the State of Texas to estimate the effect of access to a selective college on student graduation and earnings outcomes. We estimate separate effects on two groups of students. The first--highly ranked students at schools which previously sent few students to the flagship university--gain access due to the policy; the second--students outside the top tier at traditional "feeder" high schools--tend to lose access. We find that students in the first group see increases in college enrollment and graduation with some evidence of positive earnings gains 7-9 years after college. In contrast, students in the second group attend less selective colleges but do not see declines in overall college enrollment, graduation, or earnings. The Top Ten Percent rule, introduced for equity reasons, thus also seems to have improved efficiency.
We examine through a field experiment whether outreach and support provided through an AI-enabled chatbot can reduce summer melt and improve first-year college enrollment at a four-year university and at a community college. At the four-year college, the chatbot increased overall success with navigating financial aid processes, such that student take up of educational loans increased by four percentage points. This financial aid effect was concentrated among would-be first-generation college goers, for whom loan acceptances increased by eight percentage points. In addition, the outreach increased first-generation students’ success with course registration and fall semester enrollment each by three percentage points. For the community college, where the randomized experiment could not be robustly implemented due to limited cell phone number information, we present a qualitative analysis of organizational readiness for chatbot implementation. Together, our findings suggest that proactive outreach to students is likely to be most successful when targeted to those who may be struggling (for example, in keeping up with required administrative tasks). Yet, such targeting requires university systems to have ready access to and ability to make use of their administrative data.