Can Summer Bridge Programs Increase First-Year Completion and Second-Year Persistence in College?

Shakya (2026) asks whether participation in a pre-collegiate summer bridge program improves early college outcomes for economically disadvantaged and first-generation students. He analyzes student-level administrative data from a large U.S. public land-grant university, focusing on applicants to the Bridge Scholars Program between 2016 and 2022. Using a fuzzy regression discontinuity design based on a $16,000 Expected Family Contribution eligibility cutoff, he estimates causal effects of program participation. The results show large improvements in persistence: first-year completion increases by roughly 24–70 percentage points and second-year persistence by about 6–54 percentage points for students near the cutoff, while effects on GPA and credit accumulation are smaller and less precise.

Why This Article Was Selected for The Policy Scientist

This article examines a policy issue with broad national relevance: whether targeted interventions can improve college persistence among first-generation and economically disadvantaged students. In the United States, disparities in college completion remain substantial despite decades of expansion in access, making the transition from admission to degree attainment a central concern in higher education policy. Research on bridge and transition programs has grown in recent years, and this study contributes to that literature at a time when institutions are reassessing student success strategies and the cost-effectiveness of support programs. The analysis uses administrative student-level data from a large public university and employs a quasi-experimental regression discontinuity design, a credible causal inference approach. While the findings are institution-specific, the mechanisms examined—financial support, mentoring, and structured transition programs—are common across many universities, suggesting potential relevance beyond the study site.

Full Citation and Link to Article

Shakya, P. (2026). Bridging the gap to access? Impact of pre-collegiate summer program on college outcomes. Education Finance and Policy. https://doi.org/10.1162/EDFP.a.440 

Central Research Question

The article investigates whether participation in a structured pre-collegiate summer bridge program improves early college outcomes for economically disadvantaged and first-generation students. Specifically, the study evaluates the Bridge Scholars Program (BSP), an eight-week summer initiative at a large public land-grant university designed to support incoming students from low-income backgrounds as they transition into college. The central research question asks whether participation in this bundled intervention—combining academic preparation, mentoring, community-building, and a modest scholarship—causally improves persistence and academic progress during the first two years of college. The author focuses primarily on student persistence rather than GPA or course grades, emphasizing the importance of early retention as a critical determinant of long-term college completion. The broader goal is to determine whether programs designed to ease the transition into college can produce meaningful improvements in student retention among populations that historically experience lower completion rates.

Previous Literature

The study builds on a substantial literature examining barriers to college persistence among students from economically disadvantaged and first-generation backgrounds. Prior research consistently documents large disparities in college completion rates across income groups, with students from low-income households substantially less likely to graduate from college than their higher-income peers. Earlier work has emphasized the role of financial aid in increasing college enrollment, with studies by Dynarski and others showing that grants can increase college attendance and persistence. However, more recent scholarship suggests that financial support alone is often insufficient to address the complex challenges facing disadvantaged students.

A growing body of research has therefore examined non-financial barriers to college success, including gaps in academic preparation, limited access to advising, weak social integration, and difficulties navigating institutional systems. Scholars such as Castleman, Page, and Oreopoulos have shown that interventions combining financial aid with advising, mentoring, and behavioral supports can improve student outcomes. Within this broader literature, summer bridge programs have emerged as a common institutional strategy aimed at preparing incoming students for college life through early academic exposure and community building.

Despite the widespread adoption of bridge programs, empirical evidence on their effectiveness remains limited and mixed. Earlier evaluations have often relied on descriptive comparisons or simple regression models that struggle to account for selection bias. Reviews conducted by the What Works Clearinghouse have found that relatively few studies of bridge programs meet rigorous methodological standards. The present article contributes to this literature by applying a causal inference framework designed to overcome many of the identification challenges that have limited prior research.

Data

The analysis uses administrative student-level data from a large public land-grant university in the United States. The dataset includes undergraduate students who enrolled between Fall 2013 and Fall 2022, with academic outcomes tracked through early 2024. The primary analytical sample focuses on students who applied to the Bridge Scholars Program and subsequently enrolled at the university. This restriction allows the author to compare program participants with applicants who were not selected for the program, thereby reducing some forms of selection bias.

The dataset contains detailed demographic and academic information, including high school GPA, Expected Family Contribution (EFC), financial aid awards, race and ethnicity, gender, first-generation college status, and college performance measures. The sample includes approximately 410 BSP applicants across multiple cohorts, with 267 students participating in the program and 143 applicants not selected.

The study focuses on several outcome variables. The primary outcome is persistence, measured as whether students remain enrolled through subsequent semesters, particularly through the first and second years of college. Additional outcomes include first-year credit accumulation and grade point average. The author places less emphasis on GPA because grading standards vary across courses and departments, making GPA comparisons less reliable as a measure of academic progress.

Methods

The study employs a quasi-experimental causal inference strategy based on a fuzzy regression discontinuity design (RDD). Program participation is largely determined by students’ Expected Family Contribution, a financial aid metric derived from federal financial aid formulas. Students with EFC values below approximately $16,000 are substantially more likely to be admitted to the Bridge Scholars Program, creating a discontinuity in the probability of treatment at this threshold.

The analysis exploits this eligibility cutoff to estimate causal effects by comparing students just below and just above the threshold. Because the program does not strictly enforce the cutoff—some students above the threshold are admitted and some below it are not—the design is “fuzzy” rather than sharp. The author therefore estimates treatment effects using two-stage least squares (TSLS), which isolates the causal impact of actual program participation among students whose participation status is influenced by the eligibility threshold.

To address the relatively small sample size near the cutoff, the study adopts a local randomization framework. This approach treats students within a narrow window around the cutoff as if they were randomly assigned to treatment and control groups. Window selection is determined by testing for covariate balance in key variables such as high school GPA and financial aid levels.

The author also conducts several robustness checks and falsification tests. These include placebo tests using alternative cutoff values and analyses of predetermined covariates to confirm that no discontinuities appear where no treatment should exist. The consistency of results across multiple specifications strengthens the credibility of the causal interpretation.

Findings/Size Effects

The results indicate that participation in the Bridge Scholars Program substantially improves early college persistence for students near the financial eligibility threshold. The estimated effects on first-year completion are particularly large. The 95 percent confidence interval for first-year persistence ranges from approximately 23.8 to 69.9 percentage points. Even the lower bound of this interval represents a sizable increase in the probability of completing the first year of college.

The estimated effects on second-year persistence are also positive and meaningful. The confidence interval ranges from roughly 5.9 to 54.3 percentage points. While the point estimates vary across specifications, the lower bound of the interval consistently indicates a meaningful improvement in student retention.

The analysis finds less consistent evidence for effects on academic performance measures. Grade point average improves slightly during the first semester but the effect disappears by the second semester. This pattern suggests that the program may help students adjust to the academic environment initially, but that the advantage does not persist over time.

Results for credit accumulation are also mixed. When students who drop out are assigned zero credits, program participation appears to increase the number of credits earned. However, when the analysis focuses only on students who remain enrolled, the estimated effect becomes smaller and statistically imprecise. This pattern suggests that the observed credit differences primarily reflect increased persistence rather than changes in course loads or academic productivity among continuing students.

Additional analyses show that the program’s effects are weaker for cohorts that experienced their first year during the COVID-19 pandemic. For earlier cohorts, however, the persistence effects remain robust across multiple bandwidth choices and alternative model specifications.

Conclusion

The study concludes that participation in the Bridge Scholars Program significantly improves early college persistence for students from economically disadvantaged backgrounds. Across multiple specifications, confidence intervals for first-year and second-year persistence consistently exclude zero, indicating a robust positive effect on student retention.

The results suggest that the program’s primary impact operates through increased persistence rather than improved academic performance or accelerated credit accumulation. Students who participate in the program appear more likely to remain enrolled in college during the critical early stages of their academic careers.

The findings also indicate that the program’s bundled structure may be an important factor in its effectiveness. While participants receive a $2,500 annual scholarship, supplementary analyses suggest that financial aid alone is unlikely to explain the magnitude of the observed persistence gains. Instead, the combination of summer coursework, mentoring, community building, and structured academic support likely plays a central role in helping students navigate the transition into college.

From a methodological perspective, the study demonstrates how regression discontinuity designs can be applied to evaluate relatively small institutional programs that lack randomized experiments. Although the sample size is limited and the confidence intervals are sometimes wide, the analysis provides credible causal evidence on the effects of a widely used higher education intervention.

Overall, the article contributes to the literature on college access and persistence by providing causal evidence that structured transition programs can meaningfully improve early college retention among disadvantaged students. The findings suggest that interventions addressing both financial and non-financial barriers may play an important role in reducing disparities in higher education outcomes.