Commentary: antisocial behaviour--multidetermined across cultures.

Violent and other antisocial behaviours among youth are of great public concern in many countries. Gudlaugsdottir and colleagues’1 study of Icelandic youths found that violent behaviours were associated with variables in multiple domains, including individual, family, and school. As described by the authors, these findings suggest that interventions for youth violent behaviour should target the multiple factors in the youths’ social ecology that might contribute to violent behaviour. Consistent with this premise, the most effective treatments for serious youth antisocial behaviour are those that address the multiple determinants of youth problem behaviour. The Blueprints for Violence Prevention Report2 that reviewed more than 500 programmes designed to treat and prevent youth antisocial behaviour identified only 3 that were effective at reducing criminal behaviour: Functional Family Therapy, Multisystemic Therapy, and Oregon Treatment Foster Care. These three treatments are family-based ecological models that focus on addressing the known risk factors for antisocial behaviour in youths across their social ecology (for review see ref. 3). Gudlaugsdottir and colleagues’1 findings also have implications for the generalizability of theories about the development of adolescent antisocial behaviour. They found that violent behaviour among Icelandic youths was associated with gender, difficulty of obtaining social support from parents, substance use, psychological distress, and negative life events. With the exception of the failure to document significant effects in the peer domain (which likely occurred because the peer variables in this study did not tap deviant peer associations), these findings are generally consistent with the correlates of violent behaviour documented in US studies.4 Accordingly, the key determinants of antisocial behaviour may be similar across Icelandic and American cultures. Nevertheless, it should not be generally assumed that findings generated for one cultural group apply to another. Cross-cultural and cross-ethnic studies continue to be important. A variety of factors differ across national and cultural groups, such as the social political climate in a country, expectations for children’s behaviours, social norms, the nature and meaning of interpersonal relationships and child-rearing practices. These variables have the potential to influence the relative importance of certain predictors of youth behavioural outcomes across cultural and ethnic groups. For example, a study within the US found that the predictors of substance use are considerably stronger for Caucasians than they are for African Americans in a wide variety of domains.5 Similarly, my own research has demonstrated that family and peer factors are associated with substance use and substance use disorders equally for African American and Caucasian youths, but the manner in which psychopathology is associated with substance use and substance use disorders differs between these groups.6 Thus, some associations may be similar across ethnic and cultural groups, while others may vary with between groups. Erroneous assumptions about the applicability of findings from one group to another can result in incomplete or inaccurate theories and inappropriate interventions. Accordingly, the generalizability of findings obtained in one cultural or ethnic group should continue to be explored in different sociocultural groups both within and across national borders.