Selection and Socialization of Aggressive and Prosocial Behavior: The Moderating Role of Social-Cognitive Processes


  • The original research and analyses presented in this article were supported by grants from IES (# R305LO30003) and W. T. Grant Foundation (#1656) to Lawrence Aber (PI) and from W. T. Grant Foundation (#7520) to Joshua Brown and Stephanie Jones (PIs). The authors are very grateful to Tom Roderick and his entire staff at the Morningside Center for their talent at and commitment to developing and implementing the 4Rs (Reading, Writing, Respect, and Resolution) and to a rigorous external evaluation of the 4Rs. The authors are also enormously grateful to the many postdoctoral fellows, doctoral students, and research assistants at New York University, Fordham University, and Harvard University who have helped in every phase of the work reported here, from data collection through data analysis. Finally, we are grateful to the many thousands of students, hundreds of teachers, and dozens of schools who have made this work possible.

Requests for reprints should be sent to Andres Molano, Graduate School of Education, Harvard University, Larsen 708, 14 Appian Way, Cambridge, MA 02138. E-mail:


This study evaluated the extent to which fourth-grade students (M age = 8.62) select and are influenced by their peers' aggressive and prosocial behavior and the extent to which intra-individual social cognitions moderate these processes. Two waves of data were collected in the fall and spring of one academic year from children attending 18 New York City public elementary schools. Stochastic actor-based social network analysis was used to evaluate whether participants modify their network or behavior in response to the behavior of their peers. Findings support an average main effect of peer influence of aggression, as well as an interaction indicating that participants with high levels of hostile attributional bias have higher odds of adopting the aggressive behavior of their peers.