警用视频监控的犯罪防控效果及空间差异 ——以苏州市姑苏区为例

采用双重差分法,对苏州市姑苏区2014~2016年的警情数据进行分析,评估警用治安视频监控的犯罪防控效果。结果表明,视频监控对犯罪具有明显的抑制性作用。在案件类型上,视频监控对盗窃类案件的防控效果较好,但对盗窃电动自行车、盗窃电动车电瓶的防控效果较小。在时间维度上,与节假日相比,对工作日犯罪的抑制效果更好;与晚上相比,对白天犯罪的抑制效果更好。针对通过局部空间自相关分析所获取的重点区域,与警务人员访谈和实地调研后发现,视频监控的犯罪防控效果与周边地理环境、人流密集程度、警务情况等密切相关。

The Effects of Police CCTV Camera on Crime: A Case Study from Gusu District in Suzhou,China

The Closed-Circuit Television (CCTV) has been widely installed throughout mainland China, and is playing an increasingly important role in the routine work of police department. Through the installation of the CCTV, it is expected to increase the risk of being exposed or being arrested of potential offenders, and thus reduce the occurrence of crime events. Despite vast amount of money has been invested, however, there is no scientific evaluation of the crime reduction effect of police CCTVs in China. To fill this gap, this article takes Gusu district in Suzhou city as a study case, and uses a Difference in Differences (DID) method to evaluate the effect of police CCTVs based on the crime incidents data from April 2014 to September 2016. For the DID method, a control area should be identified for each given target area, and the crime densities for a period of seven months before and after the installation of the police CCTVs in the target and control areas are then compared to determine the specific effect of each police CCTV. In this study, the target areas are defined as the areas which are directly viewed by the police CCTVs rather than the simple buffer areas. After merging some very near police CCTVs, at last 217 target areas are obtained. In addition, the spatial pattern of police CCTV’s effect is analyzed with the Local Indicator of Spatial Association (LISA). For the local areas identified as significant spatial clusters or outliers, both field observations and interviews with police officers are carried out to understand the interactions between the environment and crime in those areas. The empirical results show that: Overall, the police CCTV has a significant crime reduction effect based on both the boxplot graph of the DID values and the regression results. The DID values for most police CCTVs are below zero. In terms of crime types, the police CCTV has equal reduction effects on the general criminal and public-order cases. The effect of police CCTV on the general theft crime is stronger than those on the specific theft of electric bicycle and the specific theft of battery of electric cars. When grouping crime events by the type of temporal period, it is found that police CCTV has a stronger reduction effect for the crime happened during the weekdays than those happened during the holidays & weekends. Also, the police CCTV is found to reduce more crimes happened during the daytime than the nighttime period. The results of local Moran’s I show that the police CCTVs with crime-reduction or crime-increase effects are concentrated in several local areas. The police CCTVs in the old city have a significant crime-reduction effect while those in the old communities with large population flow have a significant crime-increase effect. In summary, the effects of police CCTVs on crime level subject to the influence of the geographical environment in the vicinity, the volume of population flow, and the performance of policing work.