The development of youth antisocial behavior across time and context: A systematic review and integration of person-centered and variable-centered research

Abstract

Background: Trajectories of youth antisocial behavior (ASB) are characterized by continuity and change. Although numerous longitudinal studies have examined ASB, findings from person-centered and variable-centered research have not yet been integrated. The present paper integrates findings across statistical methods for a more comprehensive understanding of the development of ASB. Neighborhood disadvantage is considered as a core moderator.

Methods: The study protocol was registered in the PROSPERO database of systematic reviews (registration number CRD42021255820). The PsycINFO and PubMed databases were examined (September 2022) to identify longitudinal studies of youth aggression and/or rule-breaking. Results from person-centered and variable-centered studies were integrated via narrative synthesis, and a systematic quality assessment was conducted.

Results: Of 8227 studies identified, 136 met inclusion criteria. Our review indicated that rule-breaking trajectories were largely distinguished by differences in rate of change (i.e., slope), whereas aggression trajectories differed more by baseline level (i.e., intercept), particularly in childhood. For adolescents in disadvantaged neighborhoods, however, aggression trajectories differed by both intercept and slope.

Conclusions: The respective importance of the intercept and slope differed across dimensions of ASB, developmental stage, and neighborhood residence. Neighborhood disadvantage was associated with trajectories of aggression, consistent with developmental theories emphasizing the role of person-environment interactions.

Keywords: Aggression; Neighborhood disadvantage; Person-centered; Rule-breaking; Variable-centered; Youth antisocial behavior.

Figure 1.

Variable-centered (a) and person-centered (b) approaches to modeling the development of ASB across middle childhood in a hypothetical sample. The variable-centered approach identified a mean-level trend across the entire sample, whereas the person-centered analysis extracted three distinct trajectory groups.

Figure 2.

Variable-centered and person-centered approaches together in a hypothetical sample. By integrating variable-centered and person-centered models, we can determine whether differences between individuals and between trajectory groups are better-represented by the intercept or the slope. Based on the variable-centered results in 2a and person-centered results in 2b, we would conclude that the intercept is the more important consideration for aggression during middle childhood. Alternatively, the results in 2c and 2d implicate the slope as the more important consideration for rule-breaking during adolescence.

Figure 3.

Results of the systematic review screening.