What Happens When Community Colleges Reduce Remedial Education Requirements?
Li (2026) examines whether reducing remedial education requirements improves outcomes for California community college students with different levels of academic preparation. She analyzes administrative records from California community colleges linked to public high school data, comparing students before and after the state expanded direct access to transfer-level courses. The reform substantially increased transfer-level enrollment and course completion. Among the least-prepared students, transfer-level English enrollment rose 42.6 percentage points, while earning at least six transfer-level units increased 15.3 points. However, their grades and course pass rates declined, and persistence into the following academic year fell 6.9 percentage points.
Why This Article Was Selected for The Policy Scientist
Remedial education policy addresses a central challenge in higher education: how colleges can expand access to degree-bearing coursework without sacrificing the academic support needed for long-term completion. The issue is especially timely as states continue reconsidering traditional placement and remediation systems. This study makes an important contribution by examining whether removing remedial barriers produces different effects across levels of academic preparation. Its unusually rich, linked statewide administrative data are a major strength. The quasi-experimental use of California’s policy changes also provides substantially stronger causal evidence than conventional multivariate regression. The findings should inform other states, although differences in placement systems and student populations may limit direct generalization. Publication in Education Finance and Policy, a prominent peer-reviewed journal in education policy and economics, further reflects the study’s relevance to the field.
Full Citation and Link to Article
Li, A. (2026). Remedial education reform in California and community college student outcomes. Education Finance and Policy.
Central Research Question
The study examines whether reducing traditional remedial education requirements improves academic outcomes for community college students and whether those effects vary according to students’ prior academic preparation. California provides an unusually consequential setting for this question. Beginning in 2013, the state changed how community colleges placed students into remedial courses, encouraging the use of multiple measures rather than relying primarily on standardized placement examinations. The more consequential change came with Assembly Bill 705 (AB 705), enacted in 2017. The law substantially restricted colleges’ ability to require remedial English and mathematics before students could enroll in transfer-level coursework.
The analysis therefore asks what happens when students who previously would have been directed into remediation receive substantially greater access to transfer-level courses. The outcomes include enrollment and success in transfer-level English and mathematics, credits attempted and earned, grade point averages, academic momentum, and persistence. Particular attention is given to whether students with relatively weak prior preparation experience the same benefits as better-prepared students. This distinction is central because eliminating a prerequisite can accelerate academic progress while simultaneously exposing some students to coursework for which they may need additional support.
Previous Literature
Previous research offers no consensus about whether traditional remedial education improves student outcomes. Bettinger and Long (2005, 2008), Calcagno and Long (2008), Martorell and McFarlin (2011), Boatman and Long (2018), and Kurlaender, Lusher, and Case (2020) report combinations of positive, negative, and statistically insignificant effects on credit accumulation, persistence, degree completion, transfer, and labor-market outcomes. Bettinger and Long (2005), for example, find beneficial effects of mathematics remediation on mathematics credits and transfer but no comparable effects for English remediation.
Several important studies use regression discontinuity designs based on remediation placement thresholds. Calcagno and Long (2008), Martorell and McFarlin (2011), and Duchini (2017) exploit such thresholds to obtain stronger causal estimates than simple comparisons of remedial and nonremedial students. However, these designs generally identify effects for students near placement cutoffs rather than across the complete distribution of academic preparation.
Research examining heterogeneous effects further complicates the picture. Boatman and Long (2018) find that consequences differ according to the amount of remediation required. Xu (2016), by contrast, finds particularly adverse consequences among students assigned to the lowest level of English remediation. Clotfelter et al. (2015) similarly find especially negative effects among students with low prior achievement. Another literature questions the accuracy of placement itself. Belfield and Crosta (2012) and Scott-Clayton (2012), among others, show that standardized placement tests can place students into remediation who might succeed in college-level coursework.
More recent research evaluates corequisite remediation, in which students enter transfer-level courses while receiving concurrent academic support. Studies including Logue, Watanabe-Rose, and Douglas (2016, 2019), Meiselman and Shudde (2022), Miller et al. (2022), Austin et al. (2024), and Ran and Lee (2025) generally find improvements in transfer-level course completion. California’s reform is especially informative because AB 705 increased direct access without requiring corequisite support statewide.
Data
The study combines administrative records from the California Community Colleges Chancellor’s Office with student-level records from the California Department of Education. California’s community college system contains 116 colleges and serves more than two million students, giving the analysis an unusually large institutional setting. The college records contain detailed demographic, enrollment, transcript, course, credit, grade, certificate, award, and transfer information. The analysis concentrates on college enrollment from 2011 through 2020.
These records are linked to data covering the universe of California public-school students. The school data encompass approximately 4.75 million students from 2010 through 2020, averaging about 475,000 students per cohort. They include demographic characteristics, socioeconomic disadvantage, high school attended, and earlier measures of academic achievement. The analysis focuses on recent high school graduates who immediately enroll in community college, making the population particularly relevant to policies governing initial college placement.
Methods
The empirical strategy uses the timing of California’s reforms as quasi-exogenous variation in access to transfer-level coursework. Three periods are compared: the prereform period from fall 2011 through fall 2013; the intermediate reform period from spring 2014 through fall 2017; and the period following AB 705, from spring 2018 through spring 2020. The earlier placement reforms generated comparatively modest changes, so the analysis emphasizes the effects following AB 705.
A methodological challenge is that actual placement recommendations are unavailable. The researcher observes whether students enrolled in remediation but not necessarily whether colleges initially recommended remediation. To address this limitation, she constructs a predicted treatment-intensity measure representing each student’s probability of having enrolled in remedial English or mathematics under the prereform system.
Separate adaptive lasso-logistic models predict remediation using characteristics observed before reform, including standardized achievement measures. Sixth-grade test scores are used because California subsequently changed testing systems, making later scores less comparable across cohorts. Students are then divided into quartiles of predicted remediation probability, allowing outcomes to be examined across the academic-readiness distribution rather than solely near a placement cutoff. The outcome models also incorporate student characteristics and college and high-school fixed effects.
Findings/Size Effects
AB 705 produced large changes in course placement. The reduction in remedial English enrollment was concentrated among students with weaker predicted preparation. Remedial English participation declined by 7.6 percentage points for students in the third readiness quartile and by 23 percentage points for those in the fourth, least-prepared quartile. For the latter group, enrollment in transfer-level English increased by a striking 42.6 percentage points.
Access translated into substantial gains in completing important coursework. Across all four readiness groups, the probability of passing transfer-level English with a B or higher increased by approximately 20 percentage points when students who did not take the course are included in the outcome. Mathematics gains were smaller. Transfer-level mathematics participation among the least-prepared quartile increased by 16.8 percentage points.
The results nevertheless reveal important differences between gaining access and succeeding conditional on enrollment. Among the least-prepared students, the conditional transfer-level mathematics pass rate declined from 67 percent before reform to 59 percent after AB 705. Conditional English pass rates also deteriorated for this group. Thus, more academically vulnerable students became substantially more likely to attempt and pass transfer-level courses overall, but students induced into those courses were less successful on average than comparable course takers before reform.
Students also attempted more degree-applicable coursework. The least-prepared English-readiness quartile attempted an additional 2.82 transfer-level units. This exceeds the approximately 1.28-unit increase that would be expected if students merely substituted a three-unit transfer-level English course for remediation, suggesting broader changes in course-taking behavior. However, students in the third and fourth readiness quartiles experienced declines in semester GPA.
Measures of academic momentum generally improved. The probability of earning at least six transfer-level units increased by 13 percentage points for the third quartile and 15.3 points for the least-prepared quartile. Passing both transfer-level English and mathematics with a B or higher increased across all readiness groups, with gains ranging from 15.3 percentage points among the most academically prepared students to 8.7 points among the least prepared.
These short-run gains did not translate into greater persistence. Among less-prepared students, persistence into the second consecutive semester declined, including a 3.2-percentage-point reduction for the fourth quartile. Conditional on reaching the second semester, persistence into the following academic year fell across all readiness groups. The decline ranged from 3.4 percentage points among the most prepared students to 6.9 points among the least prepared.
Conclusion
The findings distinguish between removing barriers to transfer-level coursework and improving longer-term college progression. California’s reform substantially changed the academic pathways of entering community college students. Many students who previously would have entered remediation successfully completed transfer-level English and mathematics after receiving direct access, indicating that traditional placement practices had prevented some students capable of succeeding in college-level coursework from attempting it.
The results are less favorable for students with the weakest prior academic preparation. Although these students experienced sizable increases in transfer-level enrollment, credit accumulation, and foundational-course completion, they also experienced lower conditional pass rates, lower GPAs, and reduced persistence. The evidence therefore indicates that increased access and improved short-term academic momentum do not necessarily produce comparable improvements in continued enrollment.
These findings resemble evidence from Florida and from studies of corequisite remediation: reforms frequently generate substantial short-run gains in transfer-level course participation and completion without producing corresponding improvements in persistence or longer-run outcomes. The study consequently shifts the remedial-education debate beyond the narrower question of whether students should be permitted to enter transfer-level coursework. Its results indicate that direct access can remove an important academic barrier for many students while leaving unresolved the separate problem of providing sufficient support for students to sustain progress toward a degree or transfer.
Comments