A comprehensive outside-of-school intervention to promote school attendance and achievement during the COVID-19 pandemic

Abstract

Objectives

School absenteeism predicts poor academic achievement and long-term health inequities. Chronic absenteeism increased dramatically during the COVID-19 pandemic, and both student attendance and achievement disparities widened. This study examined the association between participation in an intensive outside-of-school intervention, Student U, and school absenteeism, as well as changes in attendee academic performance during the pandemic.

Study design

Observational cohort study.

Methods

A simple linear regression model was performed testing the association between Student U attendance and schools absences. Paired t-tests and a global test for overall improvement using generalized estimating equations were performed to examine changes in pre- and post-test academic performance.

Results

The sample included 65 students (Hispanic/Latino, 56.9 %; Non-Hispanic Black/African American, 33.8 %; 80.0 % first-generation college-bound). For every day a 7th-grade student attended Student U, school absences decreased by 0.14 days (p = 0.02). Mean differences in overall improvement for 7th- and 10th-grade student test scores were 23.65 (95 % CI: 19.91, 27.39; p < 0.001) and 22.12 (95 % CI: 15.67, 28.58; p < 0.001), respectively.

Conclusions

We demonstrated the association between participation in Student U and lower school absenteeism, as well as improved academic performance across multiple subjects for underserved youth. Findings reinforce the potential for comprehensive and intensive outside-of-school programs to support underserved youth and mitigate inequities following pandemics and other public health emergencies.

1. Introduction

School absenteeism among youth is associated with low academic performance and multiple school and behavioral health outcomes including poor school adjustment and disengagement, substance use, teen pregnancy, and juvenile crime [[1], [2], [3]]. Chronic absenteeism, defined as missing ≥10 % (about 18 days) of the school year [1,4,5], also predicts delayed graduation, school dropout, and negative long-term health and socioeconomic outcomes [[6], [7], [8], [9]]. Nationally, about 28 % of the >13 million students in pre-K through grade 12 were chronically absent in the 2022–2023 school year [1]. Chronic absenteeism rates are higher among older (high school vs. elementary) students and those of minoritized student race/ethnicity status and in higher household and area poverty [7,8,10,11]. For example, nationally, 37 % of Black and 33 % of Hispanic students compared with 22 % of White students were chronically absent in 2022–2023 [1].

Chronic absenteeism further increased dramatically during the coronavirus disease 2019 (COVID-19) pandemic. The pandemic also caused school closures all around the US, which exacerbated wide disparities in attendance and achievement for underserved and marginalized populations [12] and increased concerns for potential negative health impacts on underserved students [13]. In North Carolina, public schools were closed at different times; in Durham County, for example, schools were closed during the pandemic from March 16, 2020, to April 3, 2020 [14], during which they used remote learning tools to educate their students. The Durham Public Schools system, which serves primarily underrepresented populations, reopened their schools to staggered in-person learning in late fall 2020 [15,16]. However, a 12.1 % decrease in mathematics proficiency and an 18.1 % decrease in reading proficiency at the end of 2020–2021 compared to 2018–2019 was documented for North Carolina public school youth [17]. Despite schools reopening across the nation with similar timelines, high chronic absenteeism and disparities in attendance and achievement remained widespread [[18], [19], [20]].

A growing body of literature has reported on school-based interventions to reduce chronic absenteeism. For example, in New York City, where >200,000 youth miss at least one month of school each year, the Mayor's Interagency Task Force on Truancy, Chronic Absenteeism and School Engagement has instituted several measures to reduce absenteeism, including early warning flags, success mentors, school/student progress-monitoring systems, and collaborations with community partners [7,10]. Consistent with the socioecological model [21], school absenteeism is impacted both independently and through the interactions among intrapersonal, interpersonal, and contextual factors. This includes anxiety and depression, substance use, family structure, household poverty, and community factors [[22], [23], [24]]. To this end and taking into account student, school, and neighborhood levels, prior research also has demonstrated the impact of outside-of-school comprehensive and intensive interventions to promote student attendance and academic success [[25], [26], [27], [28], [29]]. Student U, a Durham-based non-profit organization designed to provide an intensive intervention to underserved youth outside of the regular school day, was shown to have had a positive impact on early high school outcomes, including greater accumulation of academic credits, fewer academic suspensions, and greater growth in academic achievement compared with nonparticipating youth, and particularly for those with lower baseline achievement [26]. Limited work, however, has addressed the potential for outside-of-school interventions to support student attendance and academic achievement following the COVID pandemic. This study builds on the work by Komisarow [26] and specifically addresses the association between participation in Student U and both student attendance and academic performance during the 2021–2022 school year. Findings from this work will inform comprehensive interventions outside the K–12 system to mitigate disparities in youth attendance and academic success for underserved populations related to the COVID-19 pandemic and other natural disasters.

2. Methods

2.1. Setting and participants

Student U is a community organization that leverages education, advocacy, and leadership in order to foster a just and equitable Durham [30]. Established in 2007, Student U aims to address structural racism, poverty, and other systemic injustices by helping first-generation college-bound students overcome significant personal and institutional obstacles to educational success [31]. The program uses three key elements to support this aim: 1) empowering students through direct support, 2) building leaders in education to drive systemic change, and 3) changing communities through collective impact [30]. Student U intervenes on time spent outside of school year-round through comprehensive education, nutrition, and social support services, including financial literacy, family support funds, wellness workshops, mental health services, parent programs, community events, and mentorship [32]. The program is offered at no cost Monday–Thursday after school from 3:00–7:00 p.m. at a central location with free transportation provided from schools. Participants complete homework, develop study skills, and explore extracurricular interests in clubs. All students and families who join Student U also have access to a Licensed Clinical Social Worker and wellness team, who provide support services including coping skills and healing that contribute to the community's social and emotional health. These services include individual, group, and family counseling, and social-emotional learning for students.

Student U also includes a 5-week, full-day Summer Academy for rising 6th–10th-grade students that provides academic programming, enrichment opportunities, intentional community building, and exposure to college and career options [33]. Instruction is offered in core subject areas: English Language Arts, Math, Science, and Global Connect (history, communities, and identities), along with electives such as Spanish, activism, dance, yoga, and theater. Additional emphasis is devoted to fostering a holistic student experience addressing identity, community, and justice themes. The Summer Academy also includes field trips that provide exposure to career opportunities, local colleges, and community services [32].

Each year, approximately 60 rising 6th-grade students are selected by a lottery application system for Student U. Students may join in other grades if there is availability. Enrollment positions are secure through 12th grade, with an opportunity to sign up for additional support through its College Promise program. To date, the program has served over 850 students (approximately 500 students in grades 6–12 in 2024). Student U currently employs 27 full-time staff and serves primarily youth who attend Durham Public Schools (81 %) [[30], [31], [32]]. Middle and high schoolers receive on average 495 h and 415 h of programming per student, respectively, per year.

2.2. Study population

Data for this observational cohort study were drawn from the Student U database jointly managed by Student U and Duke University. Academic, attendance, and demographic data were obtained from Durham Public Schools student records (provided with parent consent through a memorandum of understanding with the school system), Student U program academic performance assessments, and Student U enrollment demographic surveys. All analyses were performed in 2023. The main inclusion criteria for this study were being enrolled in grades 7 or 10 at a Durham public school in the fall of 2021 and enrollment in Student U during the 2021–2022 school year and 2022 Student U Summer Academy. Additional inclusion criteria for academic performance analyses were students who had both pre- and post-test data for the Summer Academy. This study was conducted in accordance with the Declaration of Helsinki and approved by the Duke Health Institutional Review Board (protocol Pro00112994).

2.3. Measures

The primary variables for this analysis were continuous variables representing individual Durham Public Schools student absenteeism during the 2021–2022 school year, Student U program attendance during the 2021–2022 school year, and individual student change in academic performance based on 2022 Summer Academy test scores. Summer Academy subjects included English Language Arts, Math, Science, and Global Connect, as well as electives (Black Studies; Community of Change; Activism; Argument and Advocacy; Spanish for Native Speakers; and Latinx Community, Power, and Identity). Durham Public Schools absenteeism was defined as the sum of days absent, in-school suspension, and out-of-school suspensions from Durham Public Schools during the 2021–2022 school year. Student U attendance was based on the sum of attended program sessions during the Summer Academy and separately during the regular school year. Academic performance was measured based on individual change in subject-specific test scores using pre- and post-test performance during the Summer Academy.

2.4. Statistical analysis

Descriptive statistics were computed to summarize sample characteristics. Categorical variables were presented as counts with percentages. We excluded missing data from analyses.

2.4.1. Student U attendance and Durham Public Schools absences association

A simple linear regression model was performed testing the association between Student U attendance during the academic calendar and Durham Public Schools absences in the concurrent school year. Regression assumptions were tested, and appropriate transformations were applied as necessary. Parameter estimates for school attendance, Pearson's correlation coefficient, and p-values were calculated. We repeated the same analysis for each grade.

2.4.2. Student U attendee change in academic performance

Each student's Summer Academy test score was converted to a percentage value, and mean pre- and post-test scores and differences were reported for each academic subject. Paired differences between post- and pre-test scores were computed, and a t-test was used to test whether the mean difference was non-zero. The 95 % confidence interval and p-value were reported. We conducted a global test for overall improvement for each grade combining all unique subjects with pre/post data and using generalized estimating equations (GEE) to account for correlation in repeated outcomes measured on the same individual. For each performance endpoint and each grade, a GEE model (SAS Proc Genmod) was fit to the data using an exchangeable (compound symmetric) correlation structure with clusters defined by unique individuals to account for repeated student's scores in each subject. We estimated the empirical sandwich variance to account for departures from the working correlation. Parameter estimates and p-values were provided.

We hypothesized that students with higher Student U program attendance (during the academic calendar) would have fewer days absent from Durham Public Schools in the concurrent school year. We also hypothesized that 7th- and 10th-grade students who attended Student U would have improved academic performance based on pre- and post-Summer Academy test scores. A p-value of < 0.05 was considered statistically significant, and all p-values were 2-tailed. All analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC), and figures were created using R version 4.4.1 (ggplot2, RColorBrewer). No imputation was performed. Missing data were reported but were excluded from all analyses.

3. Results

The final analytic sample comprised 65 youth (41 in 7th grade, 24 in 10th grade; see Table 1). Most students identified as Hispanic/Latino (56.9 %) or Non-Hispanic Black/African American (33.8 %). Most (80.0 %) of the students were first-generation college-bound students, and most (81.5 %) qualified for free or reduced school meals. The predominant language spoken by guardians was Spanish (55.4 %). Mean (SD) Student U attendance was 65.0 % (28.9) and 60.2 % (18.9) for 7th and 10th grades, respectively.

Table 1.

Student demographic information summary.

Demographics	All Students (N = 65)	7th Grade (N = 41)	10th Grade (N = 24)
Race
White/Caucasian	2 (3.1 %)	2 (4.9 %)	0 (0.0 %)
Black/African American	22 (33.8 %)	15 (36.6 %)	7 (29.2 %)
Hispanic/Latino	37 (56.9 %)	22 (53.7 %)	15 (62.5 %)
Asian/Pacific Islander	1 (1.5 %)	1 (2.4 %)	0 (0.0 %)
Mixed race	1 (1.5 %)	0 (0.0 %)	1 (4.2 %)
Did not indicate	2 (3.1 %)	1 (2.4 %)	1 (4.2 %)
First-generation college-bound
Yes	52 (80.0 %)	34 (82.9 %)	18 (75.0 %)
No	11 (16.9 %)	5 (12.2 %)	6 (25.0 %)
Did not indicate	2 (3.1 %)	2 (4.9 %)	0 (0.0 %)
Lunch
Free/reduced	53 (81.5 %)	33 (80.5 %)	20 (83.3 %)
Full price	4 (6.2 %)	2 (4.9 %)	2 (8.3 %)
Did not indicate	8 (12.3 %)	6 (14.6 %)	2 (8.3 %)
Language (English/Spanish) preferred by guardian 1
English	28 (43.1 %)	19 (46.3 %)	9 (37.5 %)
Spanish	36 (55.4 %)	21 (51.2 %)	15 (62.5 %)
Did not indicate	1 (1.5 %)	1 (2.4 %)	0 (0.0 %)
Language (English/Spanish) preferred by guardian 2
English	27 (42.2 %)	18 (45.0 %)	9 (37.5 %)
Spanish	34 (53.1 %)	20 (50.0 %)	14 (58.3 %)
Bilingual	3 (4.7 %)	2 (5.0 %)	1 (4.2 %)
Student U attendance, %, mean (SD)	63.2 (25.6)	65 (28.9)	60.2 (18.9)

3.1. Association between Student U attendance and Durham Public Schools absenteeism

Simple regression models showed that for every day a 7th-grade student attended Student U programming (during the 2021–2022 school year), the number of school absences decreased by 0.14 days (p = 0.02; see Table 2 and Fig. 1). The Pearson's correlation coefficient showed a moderate negative association, indicating that increased programming attendance was associated with decreased school absences (r = −0.48). No significant association was found for 10th-grade students (p = 0.89). The Pearson's correlation coefficient showed a weak negative association (r = −0.04).

Table 2.

Result of regression model of Student U programming attendance and Durham Public Schools absenteeism (2021–2022).

Grade	Number of Studentsa	Parameter Estimate (95 % CI)	p-value	Pearson's Correlation Coefficient
7th	25	−0.14 (−0.26, −0.03)	0.02	−0.48
10th	15	−0.26 (−4.12, 3.6)	0.89	−0.04

^aThe number of students missing Durham Public Schools attendance was 25.

Fig. 1.

Student U program attendance vs. Durham Public Schools absences for (a) 7th grade and (b) 10th grade.

3.2. Student U attendee change in pre- and post-test academic performance

3.2.1. 7th grade

Mean Math pre-test and post-test scores were 51.78 and 62.06, respectively, with a mean difference of 10.28 (95 % CI: 1.48, 19.08; p = 0.02). Mean Science 1 pre-test and post-test scores were 23.03 and 60.11, respectively, with a mean difference of 37.08 (95 % CI: 27.87, 46.29; p < 0.001). Mean Science 2 pre-test and post-test scores were 69.00 and 86.49, respectively, with a mean difference of 17.49 (95 % CI: 11.81, 23.17; p < 0.001). Mean Global Connect pre-test and post-test scores were 32.54 and 74.79, respectively, with a mean difference of 42.25 (95 % CI: 33.13, 51.37; p < 0.001). Pre- and post-test change in academic performance was not significantly different for English Language Arts, with mean pre-test and post-test scores of 58.17 and 65.71, respectively, and a mean difference of 7.54 (95 % CI: −2.51, 17.58; p = 0.13). The global test for overall improvement showed mean 7th-grade student pre-test and post-test scores were 45.72 and 68.85, respectively, with a mean difference of 23.65 (95 % CI: 19.91, 27.39; p < 0.001; see Table 3).

Table 3.

Mean difference in Student U Summer Academy test scores.

Subjects	N observations	Pre-test	Post-test	Score Difference (95 % CI)	p-value
7th grade
English Language Arts	25	58.17	65.71	7.54 (−2.51, 17.58)	0.13
Math	32	51.78	62.06	10.28 (1.48, 19.08)	0.02
Science 1	35	23.03	60.11	37.08 (27.87, 46.29)	<0.001
Science 2	26	69.00	86.49	17.49 (11.81, 23.17)	<0.001
Global Connect	24	32.54	74.79	42.25 (33.13, 51.37)	<0.001
Overall for 7th grade	142	45.72	68.85	23.65 (19.91, 27.39)	<0.001
10th grade
Math	16	18.50	50.83	32.33 (18.32, 46.35)	<0.001
Functionsa	1	N/A	N/A	N/A
Quadratics	10	9.00	40.00	31 (7.53, 54.47)	0.02
Linear, Quadratic, Exponential	5	37.20	80.00	42.8 (20.27, 65.33)	0.01
Language and Literature	17	62.35	80.00	17.65 (7.59, 27.7)	<0.001
Activism	8	77.50	87.50	10 (−7.87, 27.87)	0.23
Argument and Advocacy	3	66.67	100.00	33.33 (−15.43, 82.1)	0.10
Spanish for Native Speakers	6	40.00	60.00	20 (−2.99, 42.99)	0.08
Ethnic Studies	14	64.00	81.86	17.86 (5.21, 30.5)	0.01
Latinx Community, Power, and Identity	3	66.67	75.00	8.33 (−17.52, 34.19)	0.30
Black Studies	8	59.50	93.88	34.38 (20.79, 47.96)	<0.001
Community of Change	3	73.33	56.67	−16.67 (−31.01, −2.32)	0.04
Overall for 10th grade	47	47.91	70.62	22.12 (15.67, 28.58)	<0.001

^aScores are N/A due to sample size of 1.

3.2.2. 10th grade

Mean Math (Quadratics) pre-test and post-test scores were 9.00 and 40.00, respectively, with a mean difference of 31 (95 % CI: 7.53, 54.47; p = 0.02). Mean Math (Linear, Quadratic, Exponential) pre-test and post-test scores were 37.20 and 80.00, respectively, with a mean difference of 42.8 (95 % CI: 20.27, 65.33; p = 0.01). Mean Black Studies pre-test and post-test scores were 59.50 and 93.88, respectively, with a mean difference of 34.38 (95 % CI: 20.79, 47.96; p < 0.001). Pre- and post-test change in academic performance was not significantly different for Activism; Argument and Advocacy; Spanish for Native Speakers; or Latinx Community, Power, and Identity). The global test for overall improvement showed mean overall 10th-grade student pre-test and post-test scores were 47.91 and 70.62, respectively, with a mean difference of 22.12 (95 % CI: 15.67, 28.58; p < 0.001; see Table 3).

4. Discussion

Chronic absenteeism and associated education and health disparities have widened since the COVID-19 pandemic. Chronic absenteeism is greatest for underserved youth, and particularly among low socioeconomic and minoritized populations [10,11]. Findings from this study on an intensive outside-of-school intervention, Student U, demonstrated significant associations between program attendance and reduced public school absenteeism for 7th-grade youth during the 2021–2022 school year. Based on this study's observations, a student who was chronically absent in 6th grade (i.e., missing 18 days that year) and who attended Student U 65 % of the time in the 7th grade might be expected to improve their total school absences by on average 14 days that year, corresponding to a pattern of regular attendance (≤5 days absent per year) [6]. These findings are especially important in light of associations with improvements in academic performance in both English Language Arts (17 %) and Math (26 %) observed in prior research among underserved students who moved from chronic absenteeism to regular attendance [34]. In addition, findings from this study demonstrated significant association between Student U participation and improvements in academic performance for Student U participants in 7th-grade Science and Global Connect, and 10th-grade Math and Black Studies during the Summer Academy.

This analysis highlights the positive associations between participation in the Student U program and both higher student attendance and improved academic performance following the COVID-19 pandemic, and particularly for younger (middle school age) students. Prior research on comprehensive outside-of-school interventions has shown improvements in student outcomes similar to results presented here. Komisarow [26] found that intensive education, nutrition, and support services provided by Student U improved early high school outcomes including credit accumulation, grade point average, and suspension, and particularly for youth who entered the program with low baseline achievement levels. Leuven et al. [27] and Ford & Kwakye [25] both demonstrated similar improvements in student achievement with quasi-experimental or experimental designs with programs that offered students financial incentives. Oreopoulos et al. [28] reported that access to a comprehensive and intensive outside-of-school program for underserved youth increased high school graduation rates and postsecondary enrollment. Likewise, Rodriguez-Planas [29] presented increasing high school graduation rates with comprehensive outside-of-school services. Although these studies were not performed in relation to COVID-19, it is conceivable that they would have shown similar benefits for underserved communities following a pandemic or other natural disaster, particularly for mitigating education and health inequities that are aggravated by health emergencies. This work supports future studies and funding in support of Student U and similar programs that aim to promote accessible after-school and outside-of-school comprehensive interventions for historically marginalized populations. These organizations offer evidence-based solutions to support youth living in underserved communities to reduce chronic absenteeism, gain academic proficiency, and realize health benefits and other lifelong advantages associated with school attendance and achievement.

Strengths of this study include drawing from a diverse sample of underserved students and data resulting from a robust community-academic partnership. This is also the first paper to the authors’ knowledge that examines the specific associations between participation in an outside-of-school intensive and comprehensive intervention on school absenteeism and academic performance following the COVID-19 pandemic. Given that temporality of exposure (Student U Summer Program enrollment) and outcomes (pre- and post-test change in Student U summer academic performance) is known, this evidence supports a positive association between Student U and improved academic performance.

Limitations for this analysis include a large number of Student U participants (n = 25; 38 %) being excluded from attendance models given that student inclusion required consent from families to provide Durham Public Schools attendance data. These exclusions may have led to selection bias. Participants also were not randomized to the intervention, and no comparison group was included in the analysis; therefore, causal inference is limited. In addition, the sample size was small and may have limited reliability of findings. The small sample may also have limited generalizability to all Durham students, including those who attend charter or private schools, in addition to those who live in nonurban settings, or populations with differing sociodemographic composition. Moreover, attendance data for Durham Public Schools and Student U were cross-sectional, again limiting potential for causal inference.

Although we observed an association between participation in Student U and lower school absenteeism among 7th-grade students, no significant correlations were found for 10th-grade school absenteeism. Previous research has reported increases in chronic absenteeism from middle school through 12th grade [6]. Future studies should explore whether amending or intensifying the Student U curriculum for high school students would help to mitigate absenteeism among 10th-grade participants. Future research should also investigate whether effect estimates would be different for students excluded from this analysis by comparing findings with students who were wait-listed on the Student U lottery and not admitted for the 2021–2022 school year, and including time lags for exposure and outcomes.

School absenteeism predicts poor academic achievement and long-term health inequities. Disparities in chronic absenteeism and academic achievement for underserved youth widened during the COVID-19 pandemic. This study demonstrated the association between participation in an intensive outside-of-school intervention, Student U, and lower school absenteeism and improved academic performance across multiple subjects for underserved youth during the Summer Academy. These findings reinforce the potential for comprehensive and intensive outside-of-school programs to support underrepresented youth and mitigate school-based inequities. Future research should assess these results with randomized and quasi-experimental methods and draw from repeated-measures data to examine long-term, multi-year effects.

Ethical statement

This study was conducted in accordance with the Declaration of Helsinki and approved by the Duke Health Institutional Review Board (protocol Pro00112994).

Author contributions

ED, RM, and KZ conceptualized the study. TE and AC curated and analyzed the data. KZ acquired funding. RM and AB were responsible for project administration. KZ, RM, AP, and ED supervised the study. ED, RM, and KZ wrote the original draft of the manuscript. All authors revised the draft manuscript critically for important intellectual content and approved the final version.

Funding

Research reported in this publication was supported by the Office of the Director of the National Institutes of Health (NIH) under award number U24MD016258; NIH agreement number OT2HD107559; the National Center for Advancing Translational Sciences of the NIH under award number U24TR001608; and the National Institute of Child Health and Human Development of the NIH under contract HHSN275201000003I. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The research data include sensitive and confidential information from individual youth participants and are therefore not available for access.