Home Environment and Child Behavior Trajectories: A Longitudinal Mixed Effects Approach

Background: Healthy child development depends upon stable home environments. Environmental instability has been linked to poorer outcomes among children, but little is known about the impact of different domains of the home environment on child behavior. The present study utilized linear mixed effects modeling to assess the impact of home environment on child behavior from middle childhood to adolescence.

Methods: Data came from the Fragile Families and Child Well-being Study (FFCW), a longitudinal cohort study that followed 4,898 children born 1998-2000 in 20 large American cities. The analytic sample was limited to children who participated in the In-Home survey in the Years 3-15 interviews (N = 1,090). Home environment was assessed at each wave by interviewer observation using the interior and exterior subscales of the Home Observation for Measurement of the Environment (HOME) Scale. The interior subscale covered housing quality such as whether the home had adequate light and space, cracked windows, and exposed wiring. The exterior subscale covered the outside condition of the home and surrounding block. Items were summed to create interior and exterior scores where higher scores indicated higher quality home environment. Child internalizing, externalizing, and behavior problems were assessed using the Child Behavior Checklist (CBCL) on a standardized 0-2 scale. Time-constant covariates included child gender, primary race, and mother’s age at the child’s birth. Time-varying covariates included household income, material hardship, and whether the child’s parents were living together at each wave. Longitudinal mixed effects models tested the impact of the home environment on child behavior problems over 12 years. An iterative model-building process investigated within- and between-child change in behavior problems over time using AIC and R² values to compare model fit.

Results: An unconditional means model yielded an intraclass correlation of 0.21, indicating 21% of the variation in CBCL scores was explained by between-child differences and justifying a mixed effects approach. Mixed effects models that allowed intercepts and slopes to vary found average CBCL scores declined slightly over time, and trajectories were similar across children. Each one-unit improvement in interior home environment was associated with a 0.01 reduction in CBCL scores, but exterior home environment had no impact. Furthermore, the effect of environment on behavior was stronger for higher-quality home environments. Significant differences in behavior trajectories over time existed by gender; being male increased total CBCL scores by 0.01 units and externalizing scores by 0.08. Boys had higher initial externalizing scores on average, but their scores decreased more rapidly than girls’ as indicated by the significant interaction term Gender*Time. Finally, living with both parents had the largest effect on reducing CBCL scores (b = -0.041, p<0.001).

Conclusions: Findings indicate interior – but not exterior – home environment has a significant positive effect on child behavior, as do other indicators that reduce exposure to instability such as cohabitating parents and less material hardship. Child- and family-serving entities should screen for and address chaos in the home environment for at-risk families as a leverage point for improving child emotional and behavioral well-being.