都市街頭搶奪犯罪熱點分析：日常活動理論之觀點

Analysis of Hot Spots for Urban Bag Snatching Crime - Perspectives from Routine Activity Theory

台灣都市街頭搶奪犯罪問題日益嚴重，已造成民眾外出時的安全隱憂。雖然相關研究者提出犯罪不是均勻的分布在空間上，但在實務與研究上如何界定犯罪空間群聚現象卻是一項困難的挑戰。因此，本研究以全域型空間自相關之Moran’s I值來證明犯罪是否有空間群聚現象，地域型空間自相關以LISA來定義犯罪顯著熱點及冷點分布，並以二項式邏輯斯特迴歸分析來探討社會及空間等因子對街頭搶奪犯罪熱點分布之影響。本研究以台中市西屯區2001~2005年之街頭搶奪犯罪為實證研究對象，共計1,272件。研究結果發現：(1)街頭搶奪犯罪有空間群聚現象，且犯罪空間群聚程度與分布與商業活動有關聯，白天顯著熱點集中在住商混合區，晚上顯著熱點集中在夜市商圈。(2)銀行密度和批發零售業密度對白天街頭搶奪犯罪熱點分布有顯著正向影響力，餐飲業密度、娛樂業密度和工業密度對夜晚街頭搶奪犯罪熱點分布有顯著正向影響力，道路便捷度對白天和夜晚街頭搶奪犯罪熱點分布有顯著正向影響力。

Bag snatching on city streets has become a growing problem in cities in Taiwan, creating a serious public safety issue. Although, numerous studies have concluded that crime is not evenly distributed across space, identifying spatial crime clusterings remains challenging for researchers and police officials. This study analyzes the spatial dimensions of bag snatching crimes, and utilizes Moran's I of “global spatial autocorrelation” to demonstrate whether such crimes occur in clusters. LISA, “local spatial autocorrelation” is applied to define significant hot and cold spots bag snatching crimes. Social and spatial effects on the distribution of hot spots were also analyzed via dichotomous logistic regression analysis. This study analyzed 1,272 street snatches recorded from 2001 to 2005 in Situn District of Taichung City. The results demonstrated that: (1) spatial clustering existed and its degree and time distribution were related to specific types of daily businesses activities. The main significant hot spots for bag snatching were distributed in mixed residential and business areas in the daytime and in night-market business areas at nighttime. (2) Density of banks and retailers increased clusters of bag snatching crimes in the daytime, while entertainment businesses, restaurants and industry were positively correlated with such crimes at nighttime. Road accessibility was positively correlated with the incidence of bag snatching in both the daytime and the nighttime.