Background: There is considerable evidence suggesting that many children show conduct problems that are specific to a given context (home; school). What is less well understood is the extent to which children with situation-specific conduct problems show similar outcomes to those with generalised conduct problems. Methods: Data were gathered as part of the Christchurch Health and Development Study, a 25-year longitudinal study of a birth cohort of New Zealand children. Information was obtained on: (a) conduct problems during the period 7-9 years; (b) criminal offending during the period 16-25 years; (c) measures of DSM-IV mental disorders and suicidal behaviour over the interval 16-25 years; (d) measures of DSM-IV substance dependence over the interval 16-25 years; and (e) measures of relationship, pregnancy, and parenthood outcomes during the period 16-25 years. Results: Latent-class modelling suggested three distinct groups of children with conduct problems: those with mother reports; those with teacher reports; and those with both mother and teacher reports. Both situation-specific and generalised conduct disorder were associated with increased risk of criminal offending, mental health disorders, substance dependence, and relationship and parenthood issues in late adolescence and early adulthood. Conclusions: There is a need for recognition of the significance of situation-specific conduct problems in both developmental theory and in the treatment of childhood conduct disorders. A focus only on those children with generalised conduct problems is likely to overlook the features and needs of children whose conduct problems are confined to a specific context.