CHARACTERIZING AND VALIDATING PROACTIVE AND REACTIVE AGGRESSION CLASSES IN A PROSPECTIVE SAMPLE

Abstract

Research investigating aggressive behavior among youth is plentiful; however, the field contains mixed findings in terms of risk factors, correlates, sequelae, and treatment response, suggesting that individuals who exhibit aggressive behaviors are heterogeneous. The current project utilized a person-centered perspective to characterize youth who differ in frequency and quality of aggressive behaviors and a variable-centered approach to validate these classes. Specifically, the aims of the current study were (a) to use latent class analysis (LCA) to identify classes of youth that are characterized by qualitatively and quantitatively different types of aggressive behaviors, and (b) to examine the external validity of the identified aggression classes in a large, prospective sample. Participants included 648 children (M = 11.42 + .92 years; 76% Caucasian) assessed at five time points between the ages of 10 and 25 as part of a longitudinal project conducted through the Center for Education and Drug Abuse Research at the University of Pittsburgh. Analyses suggest five distinct aggression profiles in the sample; individuals exhibiting (1) primarily reactive aggression, (2) primarily proactive aggression (3) mixed reactive and proactive aggression, (4) loss of control (endorsement of a subset of reactive aggression items), and (5) low reactive and proactive aggression. Classes differed in their levels of executive functioning, peer processes, lack of guilt, internalizing symptoms, and provocation to aggression. Specifically, individuals who engaged in primarily proactive, primarily reactive, or mixed aggression exhibited significantly lower ECF than individuals in the low aggression class. Individuals who engaged in mixed aggression experienced greater peer rejection, associations with deviant peers, internalizing symptoms, and lack of guilt than individuals who exhibited low aggression at specific time points. Of note, individuals who engaged in primarily proactive aggression were more likely to experience internalizing symptoms than individuals who engaged in reactive or low aggression in late childhood and early adolescence and were more likely to report lack of guilt after misbehaving than members of other aggression classes across late childhood and early adulthood. Differences among classes in terms of these variables better characterize and validate the subgroups of aggressive youth identified in the latent class analysis. The study fills gaps in the literature by identifying concurrent and prospective correlates of aggression classes and decreasing the heterogeneity found in aggression-related research by considering the qualitative and quantitative ways in which individuals differ on aggression using a dimensional approach. These findings enhance understanding of risk processes related to aggression and inform interventions that focus specifically on ameliorating deficits displayed by youth with different aggression profiles.