|What Works to Reduce Student Absenteeism? A Systematic Review of the Literature
|University of Pennsylvania
|Unsolicited and Other Awards [Program Details]
|2 years (01/01/2024 – 12/31/2025)
Co-Principal Investigator: Le, Vi-Nhuan
Purpose: Even prior to the pandemic, national data suggested that nearly 16 percent of students were chronically absent, leading to significant public concerns about students missing large numbers of school days. These rates have been exacerbated since the COVID-19 pandemic, and chronic absenteeism has increased to 30 percent in some states from the years prior to the pandemic. Given this ongoing and heightened absenteeism crisis, many different interventions have been implemented and evaluated. In this project, the researchers will conduct a meta-analysis of absenteeism interventions and strategies. The goal is to bridge the research-to-practice gap by identifying effective strategies and interventions for reducing absenteeism and examining the variation of effect sizes by urbanicity setting, grade level, and populations at-risk for chronic absence.
Project Activities: The researchers will search and identify studies that meet their selection criteria, namely studies published after 2002 that focused on preschool through the 12th grade and examined an intervention or strategy to reduce absenteeism or increase attendance. Once studies are identified, the researchers will rate the identified studies into tiers: Tier 1 for strong evidence, a well-implemented randomized control experimental design with at least 350 participants conducted in more than one district or school; Tier 2 for moderate evidence, a well-implemented quasi-experimental design with at least 350 participants conducted in more than one district or school; and Tier 3 for promising evidence, a correlational design with statistical controls for selection bias. The research team will then extract coding key information of each study and conduct the meta-analysis.
Products: The project team will identify effective interventions and strategies to reduce student absenteeism. Specific products will include materials for practitioners such as infographics and policy briefs, and materials for researchers such as working papers and journal articles.
Setting: This project is carrying out a meta-analysis of published studies conducted in U.S. school settings (preschool through the 12th grade). The project excludes studies in international settings, which have different governance and legal structures that circumscribes the generalizability or applicability of the study findings to American schools.
Sample: The sample includes published studies that examined absenteeism mitigation interventions or strategies delivered in school settings in the U.S. to students in preschool through the 12th grade published after 2002.
Intervention: The interventions examined in this meta-analysis include both programs and strategies that mitigate absenteeism such as transportation services, enhancement of school ventilation, providing breakfast to students, and mentoring programs.
Research Design and Methods: The research team will conduct a meta-analysis with the selected studies. The research team will estimate effect sizes using a random-effects model, which assumes that variation in the observed effect sizes stems from both sampling error and random variance.
Control Condition: The control conditions for the studies in the meta-analysis depend on the particular designs of each study but are typically "business as usual", meaning that the control condition in the relevant study does not have any strategies or interventions involved in reduction of absenteeism.
Key Measures: Key measures of this study include both attendance and absenteeism. For attendance, the project team will use number of attendance days and attendance rate. For absenteeism, the project will use number of days absent, absence rate, truancy rate, and chronic absence (defined in most studies as missing 10 percent or more of the enrolled school time).
Data Analytic Strategy: The researchers will use a meta-regression approach to estimate standardized coefficient regarding the effect sizes of interventions and strategies. The meta-regression model will include an indicator for whether the data were collected before or after the pandemic. The research team will use specific statistical measures to address reporting bias issues.