Pathways and Predictors of Antisocial Behaviors in African American Adolescents from Poor Neighborhoods

Background and Purpose:

Antisocial behavior is one of the most serious mental health problems in the United States. Antisocial behavior defined as “behavior that violates social norms or the rights of fellow human beings,” has been costly to individuals, families as well as to society. Although studies have reported different trajectories of antisocial behavior during adolescence (Gorman-Smith et al., 1998; Loeber et al., 1991, 1993), the focus has mostly been on indentifying the number of trajectories as opposed to predictive models of trajectories; in addition, few studies have examined homogeneous samples of ethnic minority youth. The purposes of this study were to (1) identify the shape and number of developmental trajectories in a sample of poor, inner-city African American youth, and (2) to test predictors of group membership and the developmental course of antisocial behaviors. We expected to find distinct classes of antisocial behavior trajectories and that these different trajectories would be associated with class-specific predictors and criminal involvement.

Methods:

Data included 557 African American adolescents (ages 11 to 16) that were followed over the course of six years from the ongoing Mobile Youth Survey (MYS). The MYS is designed to identify the life-course trajectories of behaviors and associated developmental outcomes in poor, inner-city youth. Antisocial behavior was measured by an additive score of the following items: physical fighting, carrying/using a deadly weapon, attacking someone, and gang engagement. Growth mixture modeling (GMM) with Mplus 5.0 was used to examine the predictors and an outcome of antisocial behavior pathways.

Results:

The simplest form of GMM classified the sample into three distinct trajectory classes of antisocial behavior over six years: Low steady (low level of antisocial behavior involvement), high steady (consistent higher level of involvement), and incremental (increased involvement over time). Next, the model was expanded to include theoretically and empirically important predictors (e.g., school failure, risky behavior, family rules, self-worth, and hopelessness) and an outcome of criminal involvement between the ages of 14 to 16 years. The data provided an improved fit to this model (loglikelihood = -5176.6, Adjusted BIC = 10617.6, Entropy = 0.88) over baseline model. The conditional probabilities for being arrested were 0.17 for the low steady class (n = 440), 0.53 for the high steady class (n = 41), and 0.65 for the incremental class (n = 76). All conditional probabilities were statistically significant at p < .01.

Conclusion and Implications for Practice:

The results of the study suggest that there are distinct developmental pathways of antisocial behavior among African American youth in poor, inner-city communities. Individual, family and psychological factors were differentially associated with the outcome across classes. These findings are useful for social workers/researchers to identify high-risk youth and provide class-specific interventions; different doses or levels of interventions may be appropriate depending on the adolescent's developmental trajectory. Intervention for high-risk adolescents at earlier ages should prevent engagement of criminal activities in late adolescence and adulthood, and thus provide savings to society at large.