Journal of Research in Crime and Delinquency, Ahead of Print.
ObjectivesThe purpose of this study is to examine social learning theory (SLT) and teen dating violence (TDV) perpetration. This study aims to determine which predictors have the largest effect sizes, whether they vary for males and females, and whether they differ based on research design and sample characteristics.MethodsThis study uses hierarchal meta-analytic methods to examine both within- and between-dataset differences in relationships between a variety of SLT predictors and TDV outcomes. Both bivariate and multivariate effect sizes are computed for a sample of n = 1,157 effect sizes nested in n = 116 studies that used n = 88 unique datasets.ResultsFindings indicate that a variety of SLT predictors can explain TDV perpetration. Predictors with the largest effect sizes include anticipated benefits of TDV (Bivariate: r = .254; multivariate: r = .308) and peer TDV perpetration (Bivariate: r = .279; multivariate: r = .205). While most predictors show similar effect sizes for males and females, peer TDV perpetration appears to be a stronger influence for males. Several significant moderators are found.ConclusionsSLT should continue to provide a theoretical framework for TDV research and practice. Future research should examine same-sex teen relationships and sexual minorities.