|Past research suggests that youth problem behaviors, such as substance abuse, school problems and delinquency, reflect a single underlying dimension of behavior. However, few studies examine the structure of problem behavior in samples from different countries. This study explores this issue in two representative probability samples of youth (ages 12-17) from Maine and Oregon in the United States (N = 33066) and Victoria, Australia (N = 8695). We hypothesize a three-factor solution, representing three separate constructs: school antisocial behavior, delinquency, and substance use. Confirmatory factor analyses examine the degree to which these constructs represent the data for each independent sample. Data were analyzed separately for youth ages 12-14 and 15-17 to account for possible differences in the latent structure for younger versus older adolescents. Preliminary results indicate that the data may be best represented by two factors; school-related problems do not appear to be distinct from the other two constructs. Implications for understanding problem behavior across cultural and developmental groups will be discussed.
(Return to Program Resources)