作者: Jeffrey T. Ward , Megan McConaghy , Juwan Z. Bennett
关键词: Social control theory 、 Regression 、 Social learning theory 、 Psychology 、 Social control 、 Explained variation 、 Social psychology 、 Juvenile delinquency 、 Social learning 、 Differential association
摘要: The current study uses finite mixture models (FMMs) to examine whether competing theories—social learning and social control—are differentially applicable individuals. Posterior probabilities reveal that 85% of individuals are most consistent with theory (“learners”), whereas 15% bonding (“bonders”). Relative bonders, learners have significantly lower alcohol consumption use risk—as denoted by variables. Results generally stronger variable effects in the FMM as compared full-sample ordinary-least-squares (OLS) regression, particularly for differential association belief. OLS regressions among classified subsamples resulted substantial gains explained variance but no improvements bonders. Implications applicability theories assessments theoretical validity policy development discussed.