Parental Monitoring and Early Adolescent Risky Behavior: A Dyadic, Latent Class Approach

Background

Low levels of parental monitoring, the extent to which parents track youth activities, has been consistently associated with high rates of youth problem behavior. Despite the strength of this association, measurement issues have made it difficult to identify which specific combinations of parent and youth monitoring behaviors may be protective against problem behavior. This study explores monitoring holistically, using latent class analysis to identify constellations of parent and youth monitoring behaviors used in families and exploring associations between these monitoring latent classes and youth problem behavior during early adolescence. Methods

Using a sample of nearly 1,000 rural 6th graders and their mothers, we fit a series of latent class models using measures of parent and youth monitoring behaviors : parental solicitation of information and control (where parents ask youth for information and set rules), parental supervision of youth (where parents observe youth activities), parental knowledge of youth activities (what parents know about the whereabouts and activities of youth), youth disclosure of information (what information youth share with their parents) and parent-youth communication. Each measure was coded “high” or “low.” We investigated if monitoring latent classes were associated with early adolescent problem behavior ( substance use, substance use expectancies, delinquency, and antisocial peer associations) using latent class analysis with covariates and classify-analyze techniques with ANOVA.

Results

A five class solution was the best-fitting model: High-All (families were likely to report high levels of all monitoring behaviors), Discrepant (youth reported a low probability of using monitoring behaviors but mothers a high probability), Low-All (both mothers and youth reported they are unlikely to use high levels of monitoring behaviors), Communicators (families were likely to use high levels of all strategies except supervision), and Supervision-Reliant (a high likelihood of using high amounts of supervision but a low or average likelihood of using all other monitoring strategies). Membership in the High All and Communicators class were associated with low levels of adolescent substance use, delinquency, and anti-social peer associations and high levels of protective substance use expectancies. In contrast, the Low All, Supervision-Reliant, and Discrepant classes were at high risk of engaging in early problem behavior. Youth in the Discrepant class had the highest mean levels of problem behavior.

Conclusions and Implications

This study sheds light on which constellations of monitoring behaviors may be most fruitful to target in family-based interventions. Many interventions encourage parents to maintain high amounts of control and solicit information from youth (Dishion et al., 2003). Yet, our study suggests that encouraging a broad array of monitoring strategies may be most effective for encouraging healthy adolescent development. Even though supervision may be helpful when combined with a diverse array of monitoring strategies, reliance only on supervision is may have iatrogenic effects. Both “protective groups” (the High All and Communicator Groups) utilized high levels of child disclosure in addition to other monitoring strategies. The poor outcomes associated with the Discrepant group suggest that helping mothers accurately gauge their knowledge and youth disclosure may be a salient prevention strategy.