The Impact of COVID-19 on Community College Enrollment and Student Success: Evidence from California Administrative Data

Understanding the impacts of COVID-19 on community colleges is important, as they enroll nearly half of all students attending public institutions and more than half of all students of color. Because community colleges, in addition to providing workforce training, serve as an important gateway to four-year colleges and universities, they can be seen as a crucial part of the postsecondary educational system in the United States. In fact, in some states, including California, nearly half of all students attending a four-year institution previously attended a community college.¹ With recent calls for major expansions in enrollments and provision of four-year transfer courses, one can expect that community colleges will gain even greater importance.² Impacts of the pandemic on community colleges are therefore likely to have important effects on both the supply of labor in vocational careers and the higher educational system as a whole.

The net impact of the COVID-19 pandemic on community college enrollment is theoretically ambiguous, especially in comparison to the impacts on four-year colleges. The transition to remote instruction, as well as changes in health and work status, could reduce enrollment among community college students relative to their four-year college peers given their more tenuous patterns of attachment and the emphasis on hands-on technical training in many vocational programs. Additionally, community colleges serve different student populations and open enrollment does not provide an excess supply of student applicants to maintain enrollment levels. On the other hand, given the weak labor market during the pandemic, young adults might have been attracted to open-enrollment and relatively inexpensive community colleges. Nationally, data reveal a 10 percent decline in public two-year college enrollment from Fall 2019 to Fall 2020 in the beginning of the pandemic (National Student Clearinghouse 2021), and a 12 percent reduction in initial public two-year enrollment for the 2020 high school cohort relative to the prior cohort (Howell et al. 2021). In contrast, public four-year enrollment increased slightly in fall 2020 (0.2 percent), while initial enrollments decreased by less than 3 percent for the 2020 high school cohort. More recently, however, enrollment at both public two-year colleges and four-year colleges dropped by 3 percent from Fall 2020 to Fall 2021.

The effects of the pandemic on academic outcomes such as course loads, pass rates, and grades among community college students is also theoretically ambiguous. Students might have struggled with online learning given the lack of structure, requirement of more self-discipline, and technological disruptions. On the other hand, instructors might have been more lenient in grading, students were less distracted by social events, job opportunities were limited, COVID-19 relief aid was available, and options for taking courses pass/no pass increased. Thus, GPAs and other outcomes might have improved instead of worsened in the pandemic. The empirical evidence is mixed. Students at Arizona State University studied less and were more likely to withdraw from classes and delay graduation in the spring of 2020 (Aucejo et al. 2020). In contrast, although students reported challenges with online learning at Queens College in New York City, GPAs were higher and there was no effect on credits earned during the spring semester (Rodríguez-Planas 2020, 2021).³

In this paper, we use administrative college-level panel data covering the universe of students and courses from the California Community College System to analyze the impact of COVID-19 on community college enrollment and academic outcomes. The California Community College system has 116 colleges and enrolls roughly two million students per year, which represents 20 percent of all community college students in the United States and 30 percent of all Californians ages 18–24 years (CCCCO 2020). We are careful to include a consistent set of colleges throughout the time period analyzed to remove the possibility that colleges not reporting data contribute to an overstatement of enrollment losses. Consistent with national patterns, we find that community college enrollment dropped precipitously in the first four semesters of the pandemic in contrast with mostly stable enrollment in the California State University (CSU) and University of California (UC) systems. The total number of enrolled students fell year-over-year by more than 50,000 students, or 4 percent, in Spring 2020 and by more than 180,000 students or 11 percent in Fall 2020.⁴ Enrollment dropped even further in the next two semesters of the pandemic, with year-over-year (YOY) drops of 10 percent in Spring 2021 and 7 percent in Fall 2021. From Fall 2019, just prior to the start of the pandemic, to Fall 2021, California community colleges lost a total of nearly 300,000, or 17 percent, of students. All racial and ethnic groups experienced reduced enrollment in the pandemic, but black and Latinx students experienced the largest drops. Continuing students constituted the bulk of the net change in enrollment and experienced much larger enrollment drops from Fall 2019 to Fall 2021 relative to first-time students in percentage terms. Negative enrollment effects were observed in every community college in the system and, interestingly, colleges with large online presences prior to the pandemic were not less likely to lose students as courses moved online.

The administrative data also contain information on courses taken and academic outcomes among enrolled students. Reductions in enrollment were large for both vocational courses and academic courses that are eligible to transfer to four-year institutions. Fields with the largest reductions from Fall 2019 to Fall 2021 include Education, Interdisciplinary Studies, and Mathematics. Conditional on enrollment, student course loads were similar before and during the pandemic. However, there were significant changes in course completion rates and grade distributions, especially during the first semester of the pandemic (Spring 2020). From Spring 2019 to Spring 2020, course completion fell from 73 to 71 percent. The underlying cause was that a much higher percentage of students withdrew or dropped their courses, while course fail rates decreased substantially. The percentage of course grades of A increased from 40 percent in Spring 2019 to 50 percent in Spring 2020, whereas the percentage of courses with B and C grades decreased. The grade distribution remained elevated and only partially returned to pre-pandemic patterns in the next three semesters.

Our analysis builds on early findings of the effects of the pandemic on college enrollment and student success. The California Community College system provides an ideal setting to study these questions because of its size and diversity. The findings also contribute to the broader literature on the effectiveness of online education, which generally finds worse outcomes for online relative to in-person instruction prior to the pandemic, and negative effects of online instruction during the pandemic.⁵ Revealed preferences of students in community colleges, which offer substantial flexibility in course taking and enrollment, suggest that students may view online classes as inferior to in-person classes. The findings can also be interpreted in the context of a broad literature that documents increased enrollment during economic downturns (e.g., Betts and McFarland 1995; Barr and Turner 2013; Long 2015), shedding light on the unique characteristics of the pandemic and changes in instructional formats.

We first examine the impact of COVID-19 on how many students enrolled in the California Community College system. To first think about the timing of potential impacts, on 11 March 2020, the World Health Organization declared COVID-19 a pandemic. A few days later, on 16 March, the San Francisco Bay Area imposed shelter-in-place or social-distancing restrictions, and on 19 March California imposed statewide restrictions. New York State followed the next day, and by early April most states imposed social distancing restrictions that shifted in-person instruction to online. The timeline suggests that health concerns and shelter-in-place restrictions would have had their first potential effects during the spring semester of 2020 and full impacts starting in the fall of 2020.

The data on enrollment and course outcomes used in this study were downloaded through extracts from the California Community Colleges Chancellor's Office (CCCCO) DataMart system and compiled into a college-level panel dataset. Currently, three community colleges have not released their enrollment data for Fall 2021, so these colleges are excluded in prior years to maintain a balanced panel before and after the pandemic.⁶ It should be noted that California totals across years that are downloaded from the DataMart system do not make these adjustments, as noted on their Web page. In fact, the DataMart system specifically notes: “Statewide results for the most recent term should not be considered complete until all districts have submitted data.” Three community colleges operate on the quarter system so we use their fall and spring quarters to align with fall and spring semesters.

As the pandemic continued, enrollment dropped further. From Spring 2020 to Spring 2021 enrollment dropped by 158,000 students or 10 percent. And, from Fall 2020 to Fall 2021 enrollment dropped by 106,000, or 7 percent. Combining the YOY changes in the pandemic, Fall 2021 enrollment is 288,000, or 17 percent, lower than where it was in Fall 2019 just prior to the start of the pandemic. These losses were larger than the U.S. total for public two-year colleges. Estimates for public two-year colleges indicate that enrollment dropped by 13 percent from fall 2019 to fall 2021 (National Student Clearinghouse 2022).

Looking at YOY changes by semester back to 1992, the changes in student enrollment during the pandemic were large outliers. The drops in enrollment from Fall 2019 to Fall 2020, Spring 2020 to Spring 2021, and Fall 2020 to Fall 2021 were much larger than YOY changes over the last two decades. The magnitude of the YOY changes in Spring 2020 and Fall 2020 are unchanged when we estimate an event-study regression that adjusts for a time trend and includes semester dummy variables. We focus the remaining analysis on YOY comparisons for clarity.

Although all racial and ethnic groups experienced large decreases in enrollment during the pandemic, black students experienced the largest effects, followed by Latinx students. This evidence is consistent with the enrollment patterns of new high school graduates, which reveal the largest changes for black and Latinx students nationally (Howell et al. 2021).

The California Community College system classifies students into several categories related to their current or previous enrollment in a community college, another college, or high school. Students in these groups have varying levels of attachment to education and therefore may have responded differently to the pandemic and the switch to fully online courses. Table 1 reports enrollment totals and changes in enrollment for different classifications of students, focusing on changes in Fall enrollments. Interestingly, large enrollment changes were not concentrated among first-time students. First-time students in the community college system, students returning to college after taking time off, and students continuing with their educations all experienced a decrease in enrollment of 11 percent between Fall 2019 and Fall 2020. Apparently, many high school graduates chose not to start college when it became clear that classes would be held remotely, and students returning to community college after taking time off might have seen enrollment decreases for similar reasons. One would expect these students to have less attachment to college and thus to be more likely to not enroll or defer enrolling due to the pandemic and shift to online courses. However, surprisingly, community college students continuing with their educations experienced a similar magnitude of change. That is, being connected to a community college and having greater insights into the impact of moving classes online did not attenuate negative enrollment responses.

A group that might have been less likely to want to disrupt their longer-term education plans are students transferring between community colleges in the system or from a college outside of the system. Further, transferring between campuses could increase if students moved in response to the pandemic (e.g., if students moved home). We find that the number of transferring students decreased by 7 percent, which was also large (note that this is a different definition than students intending to transfer to four-year colleges). In contrast to the patterns for other types of students, special admit students, primarily high school students taking community college courses, actually increased in the fall of 2020 by 4 percent. High school students taking courses at community colleges might not have had as much flexibility to drop classes given the requirements and strict timeline for graduating from high school. Also, the trade-off for these students was between taking a high school class online or a community college class online.

The patterns differed somewhat for the drop in enrollment from Fall 2020 to Fall 2021. Specifically, enrollment among first-time students and students returning after time off decreased by 1 and 4 percent, respectively. In contrast, first-time transfer students experienced a much larger drop in enrollment rates, at 16 percent, from Fall 2020 to Fall 2021. Continuing students experienced the same enrollment drops from Fall 2020 to Fall 2021 as from Fall 2019 to Fall 2020 (11 percent for each year). Thus, in the pandemic enrollment dropped less for first-time students than for continuing students when measured as the change from Fall 2019 to Fall 2021. Finally, special admit students experienced an additional increase in Fall 2021 as many high schools continued to face difficulties with in-person versus remote learning in the pandemic. The larger reduction in enrollment of continuing students in both the fall of 2020 and 2021 drive the overall changes, and understanding why this population was disproportionately affected is an important area for future research.

Were some fields of study spared enrollment losses in Fall 2020 and others hit especially hard? In conversations with officials at various community colleges, we learned that the primary response to the pandemic was that colleges did not drop sections for classes but instead switched to teaching them remotely. The main exceptions were that programs such as nursing and medical fields were considered essential and were generally taught in-person even in Spring 2020. Colleges continued to operate a handful of allied health programs in-person that would train medical professionals to help with the community response to COVID-19. On the other hand, the small subset of courses that were impossible to convert to online instruction because of dependence on physical facilities (e.g., automotive, aviation, cosmetology, culinary arts, dental hygiene, horticulture, and welding) were cancelled.¹⁴

Table 2 reports enrollment changes by field. We focus on YOY changes from Fall 2019 to Fall 2020 and Fall 2020 to Fall 2021. Some fields experienced large enrollment losses in the pandemic. Among larger fields, Education, Engineering and Industrial Technologies, and Interdisciplinary Studies each lost more than 20 percent of enrollment from Fall 2019 to Fall 2020 and Fine and Applied Arts lost 19 percent. On the other side of the spectrum, Physical Sciences, Biological Sciences, Foreign Language, and Psychology experienced enrollment losses of 3 percent or less in Fall 2020. Enrollment losses from Fall 2020 to Fall 2021 shifted a little with some new fields experiencing large losses. Foreign Language, Mathematics, and Social Sciences each experienced losses of 15 percent or higher. Interestingly, Engineering and Industrial Technologies reversed course with an increase of 7 percent in enrollment in Fall 2021 after losing 30 percent in Fall 2020.

These results indicate that losses were felt across a wide range of fields of study, including academic subjects, and were not limited to vocational programs. But, there is some evidence that early in the pandemic the largest reductions were in fields in which online learning might be inferior (e.g., industrial technologies, teaching aides, art, and police training). Overall, the results reveal that the pandemic not only reduced the supply of labor with career technical training, but also the supply of students eligible to transfer to four-year colleges and universities.

We turn to examining how enrollment intensity and student success changed during the pandemic. We first focus on patterns of course-taking. The pandemic might have impacted the number of units taken by students, affecting their longer-term time to degree or transfer preparedness. One possibility is that students dislike online learning and might have shifted their total units downward, taking only the most important courses that they needed and postponing others. Alternatively, with higher unemployment, students may have pursued more full-time education. Table 3 reports the distribution of students taking different unit loads from Spring 2019 to Fall 2021. On average, course loads did not change substantially. The percent of students taking 15 or More units, for example, was 10 percent in Fall 2019, 10 percent in Fall 2020, and 10 percent in fall 2021. Similarly, the percent of students taking 15 or more units was 9 percent in Spring 2019 and 10 percent in both Spring 2020 and Spring 2021. Although it may be the case that students who remained enrolled tended to be those inclined to take higher course loads, the lack of significant changes in the distribution of unit totals provides some evidence that there was not a major shift toward lower or higher loads during the pandemic. The losses in the pandemic were not simply due to low-unit enrollment students.

The major cause of the decrease in course completion rates in Spring 2020 was that a much higher percentage of students withdrew from courses during the semester. Figure 7 displays course withdrawal rates and course fail rates, which are the primary reasons for not completing courses. Although course withdrawals increased substantially in Spring 2020, this was offset by a large reduction in course fail rates. In Fall 2020, both withdrawal and fail rates reversed course. However, withdrawal rates remained elevated relative to pre-pandemic levels. Partially offsetting this effect on course completion, however, fail rates remained lower in Fall 2020 than Fall 2019.

As the pandemic continued into 2021 course completion returned closer to long-term levels. Course completion rates returned to 72 percent in Spring 2021, and nearly 70 percent in Fall 2021. Withdraw rates and fail rates generally returned to long-term levels instead of the disruption found in Spring 2020 at the very beginning of the pandemic.

The decrease in fail rates in courses might be due in part to instructors being more lenient in grading during the pandemic. Table 4 reports the distribution of grades from Spring 2019 to Fall 2021. The percentage of courses with A grades increased from 40 percent in Spring 2019 to 50 percent in Spring 2020, whereas the percentage of courses with B and C grades decreased. The shift in the grade distribution is too large to be fully explained by the increased withdrawal rate among students who would have otherwise earned low grades. Thus, there is evidence that instructors were more lenient in awarding high grades during this period.¹⁷ The percentage of A grades decreased in Spring 2021 to 47 percent but not fully back to pre-pandemic levels. Additionally, in Fall 2020 and Fall 2021 the percentage with A grades remained high but lower than the 50 percent level reached in the disrupted Spring 2020 semester at the very beginning of the pandemic.

The unanticipated disruption in Spring 2020 resulted in more changes in course completion, withdrawals, and grades than during Fall 2020. It was well known among students that Fall 2020 would be held online, and thus adjustments might have already been made through enrollment decisions and by instructors in their teaching methods. Those students enrolling in online courses during Fall 2020 likely did so with full knowledge that courses would be taught remotely. Instructors were also more likely to be prepared to teach their classes remotely.

We use administrative college-level panel data for the California Community College System to evaluate COVID-19 impacts on students in the system. In contrast to the CSU and UC systems, we document large enrollment losses during the pandemic that are consistent with national trends. These losses were widespread across colleges in the system. We also find that there was a substantial drop in course completion among community college students in the rapid movement to remote instruction in Spring 2020. Colleges with larger online programs prior to the pandemic were not insulated from these effects. Reduced enrollment is evident in both four-year transferrable and career-oriented classes.

These findings, in conjunction with national evidence that enrollments at community colleges dropped by 13 percent from Fall 2019 to Fall 2021, have important policy implications and raise questions about how to reverse these effects. Future studies can shed additional light on why enrollment at community colleges was so much more sensitive to the pandemic and the shift to online courses than was enrollment at four-year colleges. In particular, the interplay between labor market opportunities, college costs, and fields of study seem especially compelling. For example, our evidence suggests that, despite the economic downturn and low cost of community college courses relative to tuition at four-year institutions, students were not systematically drawn to community colleges. Indeed, enrollment declines at California's community colleges occurred despite significant federal aid for students and institutions through the Higher Education Emergency Relief Fund in 2020 and 2021.¹⁸ This highlights the tenuous nature of enrollment patterns for the types of students who attend community colleges, as well as the potential vulnerability of the institutions themselves. It will be important to examine whether reductions in enrollment ultimately reduce the fraction of students who earn associate's degrees in technical fields or who transfer to four-year colleges, and whether short-run reduction in enrollment translate into permanent reductions in educational attainment that affect future labor market returns.

Community college enrollment might bounce back as courses return to being held in person and health and work conditions return to normal. There also might be some pent-up demand for community college as prospective students postponed entering college or current students took time off during the pandemic. However, nationally, such a return to normal was not evident in the fall of 2021, suggesting that additional policy efforts may be necessary.

To the extent that community college students are especially responsive to the cost of education, changes in tuition and grants could help to restore enrollments. Empirical studies have found that reducing community college tuition (Denning 2017) or implementing free community college through “promise” programs (Gurantz 2020) significantly increases enrollment. Thus, a shot in the arm might come from the recent donation to the California college system from the Jay Pritzker Foundation, which is the largest ever to the system and provides $100 million in scholarships to students over 20 years. The availability of financial aid might be especially helpful for black and Latinx students for whom we observed higher enrollment losses. Nationally, Democrats in Congress are proposing to make community college tuition free, which could generate a further rebound and boost. Another promising avenue for future enrollments are high school students. Our analysis reveals that enrollment only dropped by 1 percent for this group. As high school students became more familiar with remote instruction, an increasing number might take advantage of enrolling in advanced online courses at community colleges.

However, the large and persistent reductions at community colleges may also reflect shortcomings in the quality of the programs being offered and student beliefs about their economic returns. While certain programs, such as nursing, have been found to produce significant increases in earnings, others exhibit little evidence of such benefits. Further, Grosz (2019) documents that the programs offered by community colleges significantly lag changes in demand from employers, while Baker et al. (2018) find that many community college students are not well informed about the returns to different fields. Thus, building enrollments back to their prior levels could be aided by improvements in the quality of programs overall and the availability of sufficient capacity in high-demand fields. There is a precedent for initiatives that target such capacity. Between 2011 and 2018, The Department of Labor provided $1.9 billion to community colleges through the Trade Adjustment Assistance Community College and Career Training (TAACCCT) grant program. Colleges used the grants to align their programs with industry needs, expanding programs in areas such as manufacturing, information technology, construction, and health care.¹⁹ Providing potential matriculates with accurate information about earnings returns could further increase demand. Brock and Diwa (2021) note that the pandemic has highlighted the vulnerability of community college enrollments, especially for students of color, and the need for policies that improve funding and the quality of the programs they offer.

Improving community college quality and expanding offerings poses a particular challenge when revenue is down. From the college perspective, the revenue losses from missing nearly a quarter million students in the California community college system might be severe and have lingering budgetary consequences. Unlike the CSU and UC systems, California's community colleges are open enrollment and cannot simply accept a higher percentage of applicants when demand decreases. They also do not have the same buffer created by the overhead on research grants and donors. Addressing budgetary shortfalls will be crucial for ensuring future educational opportunities for low-income students and students of color. Given the unique circumstances of the pandemic, federal and state governments might have to step in with short-term gap funding to fill the void created from lost tuition and other expenditures by students. In January 2022, the U.S. Department of Education announced $198 million in American Rescue Plan funding targeting continuing enrollment and reenrollment at community colleges as well as the expansion of programs that supply in-demand fields (US DOE 2022). Similarly, the budget proposed by Governor Newsom in California includes hundreds of millions of dollars targeted to retaining and reenrolling community college students, providing additional grant aid, and expanding training for in-demand health care fields (Burke and Smith 2022). Future research can assess whether such initiatives were successful in rebuilding community college enrollments and the pipeline they provide to four-year institutions and the labor market.

We thank Mallory Newell, Jake Knapp, and Bradley Olin for helpful conversations about institutional effects within the community college system, and Valerie Lundy-Wagner and John Hetts for help understanding some of the issues with the data early in the pandemic. We thank Thomas Brock, Michal Kurlaender, and seminar participants at the University of California, Sacramento Center, College Board Forum, and UC Davis Wheelhouse: The Center for Community College Leadership and Research Workshop at UCLA for comments and suggestions.