Bidirectional Relationships between Parenting Practices and Child Externalizing Behavior: A Cross-Lagged Panel Analysis

Introduction: Most parent-directed preventive interventions are supposing that parenting behaviors are key mechanisms between contextual adversities and children’s disruptive behaviors, and modifications on parenting behaviors can result in improvement in child behavior (Patterson et al., 1992). However, the bidirectional nature of socialization effects suggests that child behaviors can also evoke different parental responses. Few studies have examined the transactional relations between parenting practices and child behaviors with multiple reporters (Shaffer et al., 2013), particularly in high-risk population. The current study examined longitudinal, bidirectional effects between child externalizing behaviors and parenting behaviors in the context of homeless families.

Methods: The current study relied upon secondary data from a randomized controlled trial of a family-focused preventive intervention targeting youth conduct problems. The sample comprised 137 families with 223 children between 6-12 years old (51.1% boys) who were recently homeless and placed in a supportive housing context. Parenting practices were evaluated through observational ratings during Family Interaction Tasks. Child externalizing behavior problems were assessed with the Behavior Assessment System for Children, Second Edition (Reynolds & Kamphaus, 2004) by teachers. The parent–child relationship was assessed with the parent-reported Parenting Relationship Questionnaire (Kamphaus & Reynolds, 2006). Treatment status, child age and gender, parent age and SES variables were controlled in analysis. Cross-lagged longitudinal path models were tested in Mplus 8, with 215 children at baseline (T1) and 161 children 1 year post-baseline (T2).

Results: The temporal within-domain stability for child behavior problems, parenting practices, and parent–child relationship was robust across T1 and T2 (p < .001). The cross-lagged paths were significant from teacher reported externalizing behaviors at T1 to observed parenting behaviors (b = -.262, p < .01) and parent-reported parent-child relationship (b = -.150, p < .05) at T2. Observed parenting at T1 was also predictive parent-child relationship (b = .158, p < .05) at T2. No cross-lagged effect from parenting behaviors at T1 to child behavior problems at T2 was detected.

Conclusion: The current findings suggest that child externalizing behaviors negatively affect parenting behaviors and parent–child relationship over time in a sample of homeless families using multiple measures and multiple reporters. These results encourage continued investigation of the impact of child behaviors on parenting and the parent-child relationship. Future preventive intervention research in this area may better elucidate this complex transactional relationship.