Abstract Unstructured socializing and differential association are two key explanations of peer influence. Unstructured socializing serves as a situational context that promotes antisocial behavior, whereas differential association with delinquent peers leads to learned behavior. Building on these perspectives and prior research, the present study models persons as contexts and examines the impact of unstructured socializing and peer delinquency on solo- and co-offending. Specifically, we use generalized linear mixed models to separate within, between, and person-context effects on offending alone and with peers, and test for interaction effects. Key findings include significant within-individual and between-individual effects of unstructured socializing and peer delinquency on co-offending, significant person-context effects of peer delinquency on co-offending, and significant within, between, and person-context effects of peer delinquency on solo-offending. Among comparable significant relationships, peer delinquency effects were notably stronger than unstructured socializing effects. Finally, interaction models revealed that within-individual effects of unstructured socializing and peer delinquency were attenuated among those with high average peer delinquency over time. The findings hold important implications for future empirical tests of theories of peer influence and for policies and programs aimed at controlling peer influence processes on offending alone and with peers.

中文翻译：

---

解开非结构化社交和差异关联对单独和共同犯罪的人内和人际影响

摘要 非结构化社交和差异关联是同伴影响的两个关键解释。非结构化社交作为促进反社会行为的情境背景，而与犯罪同龄人的差异关联导致学习行为。基于这些观点和先前的研究，本研究将人建模为背景，并检查非结构化社交和同伴犯罪对单独犯罪和共同犯罪的影响。具体来说，我们使用广义线性混合模型来分离内部、之间和个人上下文对单独和与同龄人的冒犯的影响，并测试交互效果。主要发现包括非结构化社交和同伴犯罪对共同犯罪的显着的个人内部和个人之间的影响，同伴犯罪对共同犯罪的显着个人背景影响，以及同伴犯罪对单独犯罪的显着内部、之间和个人背景影响。在可比较的重要关系中，同伴犯罪效应明显强于非结构化社交效应。最后，交互模型显示，随着时间的推移，非结构化社交和同伴犯罪的个人内部影响在那些平均同伴犯罪率较高的人中减弱。这些发现对未来同伴影响理论的实证检验以及旨在控制单独和与同伴一起犯罪的同伴影响过程的政策和计划具有重要意义。同伴犯罪效应明显强于非结构化社交效应。最后，交互模型显示，随着时间的推移，非结构化社交和同伴犯罪的个人内部影响在那些平均同伴犯罪率较高的人中减弱。这些发现对未来同伴影响理论的实证检验以及旨在控制单独和与同伴一起犯罪的同伴影响过程的政策和计划具有重要意义。同伴犯罪效应明显强于非结构化社交效应。最后，交互模型显示，随着时间的推移，非结构化社交和同伴犯罪的个人内部影响在那些平均同伴犯罪率较高的人中减弱。这些发现对未来同伴影响理论的实证检验以及旨在控制单独和与同伴一起犯罪的同伴影响过程的政策和计划具有重要意义。