Résumé
In their study on the school engagement of child welfare involved children, Landers and colleagues seek to understand how school engagement is impacted by out-of-home placement. The authors define school engagement as a child’s emotional, cognitive, and behavioural motivations to be involved with school. Using data from the National Survey of Child and Adolescent Well-Being (NSCAW II) their study compares 466 children were in out-of-home placements with 251 child welfare involved children who remained in their homes. The children were between the ages of 4-15 years old at baseline, and 6-18 years old at the 36-month follow-up. School engagement was measured at 18- and 36-month follow-up intervals. Data collection was self-reported by children through the Drug Free School Outcome Study (Silvia et al., 1997). Propensity score matching was used to control for child, caregiver, and case characteristics. Out-of-home placements did not have a statistically significant impact on school engagement compared to child-welfare-involved children who remained in their homes.
Methodological Notes
Landers and colleagues present a valuable analysis of a critical yet often overlooked area in child welfare education research. While academic discussions frequently focus on performance metrics, this study shifts attention to students' enthusiasm and engagement with academics, an equally important area of study. One of the study’s key strengths is its use of propensity score matching. Drawing from a national dataset, the research benefits from a large and diverse sample, enhancing its generalizability. Additionally, the inclusion of multiple covariates, such as child, caregiver, and case characteristics, strengthens the precision and reliability of the statistical findings. These methodological choices support the conclusion that children placed out-of-home do not exhibit significantly different levels of school engagement compared to their peers who remain in-home within the child welfare system. The study also incorporates follow-up assessments at 18- and 36-month post-baseline, ensuring that the findings are long-term.
Despite its strengths, the study has several limitations. It examines placement only at baseline, without considering whether children returned home during the 36 months. As a result, it offers limited insight into the differential effects of short- versus long-term placements. Additionally, the study does not distinguish between types of out of-home placements, such as kinship care, foster care, residential care, or adoption, which could significantly influence outcomes. Several unmeasured confounding variables may also have impacted the findings, including mental health status, geographic location, peer relationships, and trauma history. Future research should consider these factors to provide a more comprehensive understanding.