Background: We investigated the role of psychosocial and proximal contextual factors on nicotine dependence in adolescence. Methods: Data on a multiethnic cohort of 6th to 10th graders from the Chicago public schools were obtained from four household interviews conducted with adolescents over two years and one interview with mothers. Structural equation models were estimated on 660 youths who had smoked cigarettes by the first interview. Results: Pleasant initial sensitivity to tobacco use, parental nicotine dependence (ND), adolescent ND and extensiveness of smoking at the initial interview had the strongest total effects on adolescent ND two years later. Perceived peer smoking and adolescent conduct problems were of lesser importance. Parental ND directly impacted adolescent ND two years later and had indirect effects through pleasant initial sensitivity and initial extensiveness of smoking. Parental depression affected initial adolescent dependence and depression but adolescent depression had no effect on ND. The model had greater explanatory power for males than females due partly to the stronger effect of conduct problems on dependence for males than females. Conclusions: The findings underscore the importance of the initial drug experience and familial factors on adolescent nicotine dependence and highlight the factors to be the focus of efforts targeted toward preventing ND among adolescents.