Traditional social learning predicts cyber deviance? Exploring the offending versatility thesis in social learning theory

Abstract

Social learning theory has been widely implemented to understand cyber deviance. Nevertheless, the antecedent scholarship homogenously nested in the perspective of offending specification, leaving the offending versatility thesis unattained. The lack of such studies may undermine the capability of comprehensively understanding the social learning patterns of online offending. Using a sample of 3741 Chinese college students, this study estimated an array of binary logistic regressions to compare the effects of traditional and online social learning in four types of online offending (online sexual harassment, cyberbullying, hacking, and digital piracy). The results suggest that offending versatility and offending specification co-exist in the social learning process of cyber deviance, while offending specification explains a marginally greater variance. Besides, online learning variables act as potential mediators in the relationships between traditional learning and cyber deviance. Furthermore, traditional social learning shows greater predictive power in cyber-enabled crimes than in cyber-dependent crimes. Our study provides fresh empirical evidence for the non-exclusive association between offending versatility and offending specification in the social learning process of cyber deviance.