非行的な仲間との接触, 社会的ボンドと非行行動

分化的強化仮説と社会的コントロール理論の検証

抄録

This paper examines empirically the hypotheses derived from the differential reinforcement and social control theories of delinquency, and provides new understanding for understanding issues of delinquency in contemporary Japan. Over the last three decades, relatively few empirical studies have been conducted on the relationship between juveniles' association with delinquent peers and various types of deviant behavior in Japanese society. In particular, there have been very few case-controlled studies testing the analytic validity of differential reinforcement and social control theories for official delinquency. The author hypothesizes that official delinquency, which tends to be chronic and only occurs in the presence of strong motives and impulses, is mainly caused by the mechanism of positive reinforcement (seeking rewards) and negative reinforcement (escaping punishment) through association with delinquent peers. In addition, the author hypothesizes that school-related stress is negatively related to both self-reported and official delinquency.
The following conclusions are reached.(1) The findings are very consistent with the author's hypothesis that delinquency is learned or reinforced through association with delinquent peers. Moreover, the effect of delinquent peer association holds significance, independent of other factors.(2) As was found in previous studies, attachment to teachers is in general negatively related to self-reported delinquency scales. None of the attachment scales are significantly related to official measures of delinquency.(3) Commitment has a significant negative effect only on official traffic offenses.(4) School-related stress has a negative effect on certain forms of official delinquency, which clearly weakens the argument that school-related stress promotes delinquency. It seems very important to understand the mechanism of differential reinforcement to know why some children have positive motives and impulses toward delinquent behavior.
Finally, the implications for future research are discussed.