This thesis examined the correlates of delinquency and substance use by focusing on the psychosocial environmental influences that lead to sibling similarities and differences in these areas. It was shaped by data from studies strongly suggesting that environmental influences played the most important role in development and adjustment, but that these influences were not shared by siblings in the same family. The evidence for the importance of nonshared environment has led to the conclusion that family variables, such as parenting style and family events, which are shared between siblings, are less important. Adopting this conclusion is highly relevant to the study of the environmental influences that place an adolescent at-risk for problem behaviours, because family variables are important in most sociological treatments of criminal and delinquent behaviour. Therefore, a key issue of this thesis was to investigate the extent of sibling similarity for delinquency and substance use, as well as the significance of a range of variables of theoretical interest in the study of adolescent problem behaviour. A second issue of interest was to determine whether there were associations between differences in sibling experiences and differences in their delinquency and substance use. These issues were examined in 361 primarily opposite-sex adolescent sibling pairs from Wave 1 of the “Sibling Study”, a longitudinal study designed to identify those factors that contribute to adolescent engagement in illegal behaviours and those that inhibit such engagement. The design of this study provided several advantages over previous investigations of within-family differences. First, the sample used in the study consisted of siblings at the extremes of the dimensions of problem behaviour rather than based solely on a sample of students, making it possible to investigate the relative influence of shared and nonshared environment at the extremes. Second, a broad set of factors across the domains of school, family, peer, and individual were assessed rather than solely focussing on parental treatment, which is where much research is concentrated. Third, the relationship between shared family context effects and differential experience was considered. Finally, the associations between the correlates of differences for delinquency and substance use were investigated, whereas the relatively small body of previous research on sibling differences has largely ignored substance use. The results indicated substantial within-family differences in experiences of the environment and problem behaviour. Although all siblings differed, sibling pairs where one sibling was identified as an offender or at-risk for offending were significantly more different from one another than siblings identified as normative. Sibling constellation variables were less important in accounting for differential experiences than were the shared family context characteristics. Almost 60% of the variance in sibling delinquency difference scores and 38% of the variance in sibling substance use difference scores could be accounted for by differences in sibling experiences. In general, extra-familial experiences, such as peer influence, school involvement, and stressful events, and individual characteristics, such as self-control, definitions, and attitudes to authority were more important sources of systematic nonshared environment than parental treatment and sibling interaction items. In addition, different patterns were found for low and high risk families, and for delinquency and substance use. These results support the importance of nonshared environmental factors although future research is needed to confirm that these sibling differential experiences are the most important for explaining problem behaviour. The empirical, theoretical, and practical implications of this thesis for the development of problem behaviour were discussed.