Paper Summary
Share...

Direct link:

Parenting Style and Child Outcomes: Evidence From a Randomized Intervention

Sun, April 15, 8:15 to 10:15am, Crown Plaza Times Square, Floor: Fifth Floor, Room 501/502

Abstract

Parenting style refers to a set of parent’s behaviors that responds to a child’s autonomous behavior, setting an atmosphere in which the parent and the child interact (Baumrind 1966; Darling & Steinberg, 1993). Literature in psychology reports a significant correlation between the parenting style the child receives and the child’s outcomes. In economics, parenting style is treated as an input to a skill production process, and its role is estimated using panel data methods (Fiorini & Keane; 2014). Despite growing interest, there has been no clear, causal evidence linking parenting style with long-run child outcomes. Parenting style is endogenously determined with child behavior, making identification of its causal effect challenging. Randomized intervention studies that change parenting style often report outcomes measured no later than several months to a year.

We report causal evidence of the effect of parenting style during early childhood on child behavior in early adolescence. In particular, we evaluate the 10-year follow-up of a universal intervention on the parents of preschool children in Germany. Eligibility was restricted to parents of 2.6- to 6-year-old children, but no restrictions were placed on household income or child risk factors. The intervention ran for two months, teaching and training parents on how to engage with their child and how to discipline their child’s behavior in a positive way. Data is collected 1 through 4 years after the intervention, and 10 years after the intervention. Final sample size used for the analysis is 231 parent-child pairs, where approximately 60% are in the treatment group.

We report significant treatment effects in reducing child externalizing behavior and improving child mental well-being. The effects are between 0.1 and 0.3 in standard deviation units within the sample. Treatment effects are not significant on child internalizing behavior and child social relationships. The effects are stronger for girls than boys. We account for noncompliance and attrition, and find that the estimates are robust.

In order to investigate the channel in which treatment affects child outcomes, we use the linear mediation analysis of Heckman and Pinto (2015) to show that changes in parenting styles induced by the intervention indeed mediate the improvement in child outcomes. For the four years after the intervention, the intervention caused parents to change from harsh to benign discipline methods and to increase positive engagement with the child. Both of these improvements mediate improvements in child outcomes.

Our contributions are twofold. First, we provide evidence of the causal effect of parenting style in early childhood on adolescent outcomes. As far as we are aware, this is the first causal evidence of parenting style effects beyond child outcomes measured immediately after an intervention. The linear mediation analysis shows that changes in parenting style indeed mediates improvement in child outcomes induced by the intervention. Second, by analyzing a universal intervention, we show that the benefits of early childhood interventions are not necessarily confined to low-SES households.

Authors