基于滞后中介效应模型的谋杀率影响因素的国际比较分析

International Comparative Analysis of Factors Affecting the Homicide Rate Based on the Lag Mediation Effect Model

摘要: 基于世界银行和联合国毒品与犯罪问题办公室的相关数据，利用中介效应分析模型，分析了20个国家的收入差距和失业率对谋杀犯罪率的影响，并比较了各因素对各国犯罪率影响的差异。由于收入差距的拉大对谋杀率的影响不一定当年就显现出来，模型中对收入差距和失业率的数据做了超前或滞后的处理，并且中介变量分别设定为失业率或者收入差距，通过这两条路径来分析各国谋杀犯罪率存在的差异。分析结果表明，采用滞后数据的收入差距和失业率都与谋杀犯罪率存在显著的正相关，但是不同国家收入差距和失业率对谋杀犯罪率的影响在中介效应模型中的路径则存在差异。当国家的收入差距与犯罪率显著相关时，失业率的中介效应显著；当国家的失业率与犯罪率显著相关时，收入差距的中介效应显著。

Abstract: Based on the relevant data of the World Bank and the United Nations Office on Drugs and Crime, using the intermediary effect analysis model, this paper analyzes the effects of income gaps and the unemployment rates on murder rates of 20 countries, and compares the differences on the impact of various factors on the crime rate in different countries. Due to the impact of the widening income gap on the murder rate may not appear in the same year, the original data of income gap and unemployment rate may be dealt with by leading or lagging. Besides, two paths with mediating variables of income gap or unemployment in the model are adopted to analyze the differences in national homicide rates. The results show that the income gap and unemployment with lagging data have significant positive correlation with homicide rates, but the effect of different national income gap and the unemployment rate on homicide rate is different in the path of the mediation effect model. When the country’s income gap is significantly associated with crime, the mediating effect of unemployment is significant. On the other hand, when the country’s unemployment rate is significantly associated with crime, the mediating effect of the income gap is significant.