The Longitudinal Relationship between Housing Instability, Social Support, and Child Maltreatment

Background and Purpose: Housing instability, characterized by frequent moves, evictions, and homelessness, has been linked to increased risk of child maltreatment. While previous research has established this association, limited investigation exists on the underlying mechanisms. This study examines the longitudinal relationship between housing instability and child maltreatment during early childhood, with specific attention to social support as a mediating mechanism. Using stress process theory, which suggests stressors can erode protective resources over time, this research tests whether maternal social support mediates the relationship between housing instability and different types of child maltreatment.

Methods: Data were drawn from the Future of Families and Children Wellbeing (FFCW) study, a longitudinal birth-cohort survey that oversampled unmarried parents. The sample included 4,588 mothers interviewed when their children were ages 1, 3, and 5. Housing instability was measured at year 1 through indicators of multiple moves, doubling-up, homelessness, or eviction. Social support was assessed at year 3 through five items measuring access to tangible support. Child maltreatment at year 5 was measured through maternal self-report using modified Parent-Child Conflict Tactics Scale items covering physical assault, psychological aggression, and neglect. Structural equation modeling was employed to test longitudinal mediation, with housing instability at year 1 predicting social support at year 3 and subsequently child maltreatment at year 5, while controlling for baseline measures and relevant covariates.

Results: Housing instability at year 1 was negatively associated with social support at year 3 (β = -.047, p < .01), indicating that greater housing instability predicted lower subsequent social support. Interestingly, social support at year 3 showed varying associations with different types of child maltreatment at year 5. Social support was positively associated with physical assault (β = .085, p < .001) and psychological aggression (β = .089, p < .001), but negatively associated with neglect (β = -.049, p < .05). Significant indirect effects confirmed that social support mediated the relationship between housing instability and all three types of child maltreatment, though effect sizes were small.

Conclusions and Implications: This study reveals a complex relationship between housing instability, social support, and child maltreatment. Housing instability appears to erode social support networks over time, which has differential effects on maltreatment types. While greater social support was protective against neglect, it was unexpectedly associated with increased risk for physical and psychological aggression. These findings suggest that the quality and nature of social support may be as important as its presence. Policy and practice implications include prioritizing housing stability interventions during early childhood, implementing multi-faceted approaches that address both housing instability and social support enhancement, and developing programs that help vulnerable families build diverse, high-quality support networks while setting healthy boundaries with potentially harmful influences. Future research should examine whether formal versus informal support sources have different impacts on parenting behaviors.