In American criminology, crime and other forms of deviance have often been attributed to individuals' definitions or internalized attitudes toward deviance. In previous studies, however, empirical tests of the causal processes of learning attitudes toward deviance have been relatively rare. Moreover, studies examining the mediating effect of a person's attitudes on the relationship of both peer reactions and peer behavior to the person's deviance are even rarer. These studies also have been largely restricted to adolescent samples from the U.S.
The present study is the first to address hypotheses concerning the cross-cultural generalizability of differential association/social learning theory between Japan and the U.S. Specifically, we test hypotheses concerning cultural differences in the causal processes of learning attitudes toward deviance that lead to deviant behavior. Based on the literature on individualism-collectivism in the cultures of Japan and the U.S., coupled with the internal-external dimension in Tittle and Paternoster's classification of theories of deviance, we expect that, while the causal processes of learning attitudes toward deviance will be similar, the strengths of the relationships among peer reactions to deviance, peer deviance, student attitudes toward deviance, and student deviance will vary by cultural context.
Our analyses of comparable survey data from college students in Japan (n=591) and the U.S. (n=625) generated largely supportive, but somewhat mixed, evidence for the predicted similarities and differences. In both countries, peer reactions to deviance more strongly predicted student attitudes toward deviance than did peer deviance. Moreover, peer deviance strongly predicted student deviance, while student attitudes mediated the effects of peer reactions on student deviance in both countries. Contrary to the hypotheses, peer reactions and peer deviance did not more strongly predict student attitudes in Japan than in the U.S. Additionally, peer deviance more strongly predicted student deviance in the U.S. than in Japan. In agreement with the expectations, the relation between student attitudes and student deviance was stronger in the U.S. than in Japan.
The great value of this study is that the comparable data, collected in highly similar settings in Japan and the U.S., enabled determination of the extent to which the theoretical explanations and causal relationships are generalizable across these two considerably different countries. Our results suggest that future research on the cross-cultural generalizability of American theories of deviance, whether cross-sectional or longitudinal, would benefit not only from using the same methodology that we used in this study, but also from connecting the internal-external dimension in Tittle and Paternoster's classification of theories of deviance to cultural variability in individualism-collectivism.