The Association between Homelessness and Children’s Academic Outcomes: Longitudinal Evidence from School Administrative Data

Background and Purpose

In School Year 2022-2023 nearly 1.4 million public school students reported experiencing homelessness, a 25% increase relative to School-Year 2020-2021. Housing instability is associated with lower attendance, graduation rates, and test scores and can have persistent negative impacts on academic outcomes over time. This study used administrative data from a large, diverse, urban school district in Northeastern United States to investigate both the immediate and longer-term effects of homelessness on children’s academic outcomes.

Methods

The sample included all students enrolled in the district over a period of seven years (School Year 2017-2018 to 2023-2024) who had data on at least one of the outcomes of interest (376,285 student-year observations). These outcomes included the students’ attendance rate, chronic absenteeism rate (defined as missing 10% or more days), standardized test scores in English and Math, and graduation rates.

The independent variables of interest were whether a student was reported as experiencing McKinney-Vento homelessness (either literal homelessness or doubled up because of economic hardship) in the current year or in a previous year. For those who experienced homelessness, we also examined how outcomes varied by the number of years since their last experience.

We used linear regressions with controls for student demographics, school, residential neighborhood, year, and the number of years of data for that student up to that point. All models accounted for clustering of standard errors at the student level. For outcomes measured at multiple time points, we also conducted a longitudinal analysis that examined how changes in homelessness status within students was associated with concurrent changes in their academic outcomes.

Results:

Annual homelessness rates ranged from 5.8% to 7.9%. In cross-sectional analyses, homelessness was significantly associated with lower attendance rates, higher chronic absenteeism, and lower test scores, though the magnitudes were relatively small. Students who had any reported homelessness were also 1.8 percentage points less likely to graduate from high school (95% CI [-0.040 0.003], P=0.095).

Chronic absenteeism was positively associated not only with experiences of homelessness in the same school year [B=4.4, 95% CI 0.036 0.053] P<0.001), but also with homelessness experienced in a previous year (B=5.2, 95% CI [0.043 0.062], P<0.001). Moreover, among those who experienced any homelessness, the effects on chronic absenteeism were higher for experiences further back in time (B=0.01, 95% CI [0.004 0.017], P=0.002).

Results from the longitudinal models show that, within students, the experience of becoming homeless was associated with a 1.1 percentage point increase in chronic absenteeism (95% CI [0.00 0.022], P=0.041).

Conclusion and Implications

Our results confirm previous findings that homeless is negatively associated with academic outcomes, particularly chronic absenteeism. Moreover, chronic absenteeism actually increases with experiences of homelessness further back in time, suggesting that housing instability at younger ages may be particularly important for this outcome. This highlights the importance of early identification and prevention of homelessness. School-housing partnerships, initiatives where school systems and housing agencies work together to support families, could be a promising strategy to address housing instability and promote children’s academic success.