280对犯罪少年与正常少年多因素条件Logistic回归分析

Multiple Logistic Regression Analysiso f Risk Factors of Juvenile Delinquency

本文对湖北省280名犯罪少年与正常少年进行1:1配比对照研究。单因素配对分析有9项因素与少年犯罪有联系,经多因素条件Logistic回归分析,在控制其混杂因素的情况下,仍以朋友有犯罪行为、家庭经济收入低、文化程度低有显著性意义。其交互作用项本人有不良行为与朋友有犯罪行为、家庭经济收入低与父亲文化程度低亦有非常显著性的意义。

In a case-control study analyzing risk facters in juvenile delinquency, 280 cases occurring in Hubei Province and 280 matched controls were studied. The single factor matched analysis showed that nine factors may be associated with risk of juvenile delinquency. Among them having offenders among friends, lower economic level, less education of father were still significance after the confounding of other factors were controled in multiple logistic regression analysis. Furthermore, there were manifest interactions between having bad behavior oneself and having offenders among friends, and between lower economic level and less education of father. So the authors consider that the speech and behavior of friends and parents may play important role in juveniles committing offences.