Dynamic Pathways between Rejection and Antisocial Behavior in Peer Networks: Confluence Model Revisited

Introduction: Despite ample attention to the role of peer dynamics for the development of antisocial behavior (AB; Dishion & Tipsord, 2011), limited research has focused on the role of peer rejection in these processes. The confluence model (Dishion, et al., 1994) posits dynamic pathways linking peer rejection and AB in peer context. Specifically, rejection is expected to increase the tendency towards deviant clustering, is found to have short and long-term amplifying effects on AB (Dishion et al., 2008). Drawing on advances in understanding network dynamics, we tested an enhanced confluence model to study co-evolving trajectories of rejection and AB as embedded in networks. We examined: peer selection on AB and rejection, interactive dynamics between rejection and AB as contributing to peer selection, peer influence on AB and rejection, individual characteristics as moderators of peer influence processes, and reciprocal associations between rejection and AB.

Methods: We used an ethnically diverse sample of adolescents attending 3 public middle schools in the northwestern US (N = 997; 52.7% boys; 42.4% European Americans, 29.2% African Americans, 6.8% Latino/a, 5.2% Asian Americans, and 16.4% other race) who completed 3 assessments in 6-8 grades. Students provided peer nominations of rejection and AB. We constructed peer affiliation networks by creating matrices of reciprocated peer ties, which were assessed by asking students: “Which children do you hang around with?” We used a stochastic actor based modeling approach implemented in RSiena (Snijders et al., 2010) to accomplish our goals.

Results: Results from RSiena revealed significant peer selection effects on AB but not rejection: youth befriended others with similar levels of AB. As expected, rejection increased the tendency to select friends on the basis of AB during the transition from 6^th to 7^th grades, and youth became more similar to their friends on AB. This is the first study to document peer contagion of rejection. However, rejection did not predict enhanced susceptibility to peer influence on AB. Finally, we observed that rejection and AB predicted reductions in the reciprocal friendships from 6^th to 7^th grades, suggesting deterioration in quality of relationships.

Conclusions: Findings support the elaborated confluence model of a joint interplay between rejection and AB as conditions that lead to self-organization into deviant peer network clusters. This study advances developmental research on peer contributions to AB and theories of peer influence. We discuss the need to design school environments that discourage marginalization processes and formation of deviant groups.