EN中文

GEOGRAPHICAL RESEARCH >

2017 , Vol. 36 >Issue 12: 2479 - 2491

DOI: https://doi.org/10.11821/dlyj201712017

Crime Geography

Impacts of community environment on residential burglary based on rational choice theory

  • XIAO Luzi , 1, 2 ,
  • LIU Lin , 3, 4 ,
  • SONG Guangwen 1, 2 ,
  • ZHOU Suhong 1, 2 ,
  • LONG Dongping 1, 2 ,
  • FENG Jiaxin 1, 2
Expand
  • 1. Center of Integrated Geographic Information Analysis, School of Geography and Planning, Sun Yat-sen University, Guangzhou 510275, China
  • 2. Guangdong Key Laboratory for Urbanization and Geo-simulation, Sun Yat-Sen University, Guangzhou 510275, China
  • 3. Center of Geographic Information Analysis for Public Security, School of Geographic Sciences, Guangzhou University, Guangzhou 510006, China
  • 4. Department of Geography, University of Cincinnati, Cincinnati OH45221-0131, Ohio, USA

Received date: 2017-06-11

  Request revised date: 2017-09-21

  Online published: 2017-12-15

Copyright

《地理研究》编辑部
Fold

Abstract

Residential burglary is one of the common property crimes in China, which has attracted the attention from many scholars. Domestic literatures of residential burglary mainly focus on its spatial pattern and try to explain why some places suffer more crimes than other places do. Although some of them have explored how social environment and natural environment affect residential burglary, they failed to consider the complex characteristics of population and housing in the context of China's rapid urbanization. To address this, rational choice theory is applied to explain the relationship between community environment and residential burglary in ZG city, one of the biggest cities in the south of China. Using the data of residential burglary, census and road network from 226 police districts, a negative binomial regression model is estimated. Results show that both of the population characteristics and the housing characteristics have significant impacts on burglary. In terms of population characteristics, police districts with more highly educated intellectuals suffer less residential burglaries, while those with more young and middle-aged migrant workers suffer more. As for housing characteristics, police districts with more ordinary commercial residential buildings are easier to attract residential burglars, and those with more public-owned housing can effectively reduce the residential burglary rate. In addition to population and housing characteristics, the results also show that the density of road network has significant impact on residential burglary, while the density of bus lines has no effect on it. In conclusion, the balance of the perceived risk and reward of different environments would affect the spatial distribution of burglaries. Residential burglars roughly follow the rule of "safety first, benefit second" when committing crimes. This study can shed light on how population and housing characteristics influence the spatial pattern of residential burglary in China, and provide suggestions on crime prevention and control to the police.

Key words： criminal geography; community environment; burglary; rational choice theory

Cite this article

XIAO Luzi , LIU Lin , SONG Guangwen , ZHOU Suhong , LONG Dongping , FENG Jiaxin . Impacts of community environment on residential burglary based on rational choice theory[J]. GEOGRAPHICAL RESEARCH, 2017 , 36(12) : 2479 -2491 . DOI: 10.11821/dlyj201712017

1 引言

伴随着城镇化进程的不断推进,中国的经济获得了迅猛发展,但其中的犯罪问题也给城市管理带来了新的挑战。例如,近些年大城市的入室盗窃案件不断发生,给人们的生活带来较大的威胁,严重影响到人们的居住安全。关于入室盗窃等犯罪的研究已逐渐受到国内学者关注。
目前国外有关入室盗窃的研究较多且体系相对完善[1,2],但由于中西方国情的差异,西方的理论和规律能否应用于中国仍需进一步的实证;而国内犯罪地理学仍处于起步阶段,有关中国入室盗窃方面的研究仍较为稀缺[3,4]。此外,由于大城市的空间形态复杂,商品房、自建住房、原公有住房、保障性租房等构成了多元的物质住房形态;不同人群的住房分异明显,既有高收入、高学历人群集聚区,又有低学历外来务工人员的集聚区[5-7]。物质空间和社会空间的分异及交织,加剧了入室盗窃成因的复杂性[8,9]。中国入室盗窃的格局特征及其成因仍有待进一步分析。
犯罪地理较成熟的理论有通过环境设计预防犯罪理论（CPTED）、社会失序理论、日常活动理论及理性选择理论等[10]。CPTED通常与社区建成环境相结合,认为通过对空间环境的合理设计能够有效减少犯罪发生概率,强调微观物质空间对犯罪的影响[11];社会失序理论强调社会环境尤其是社区紊乱会导致犯罪的增加[12];日常活动理论则强调在缺乏有效监管措施的情况下,潜在的犯罪者与受害者相遇会导致犯罪的产生,突出人的出行行为对犯罪空间格局的影响[13]。入室盗窃指的是非法进入他人住所进行盗窃,犯罪者如何选择作案目标应是基于对目标本身及其所处环境的作案风险与收益的综合判断。CPTED、社会失序理论、日常活动理论各有侧重,但入室盗窃的成因复杂,与此三大理论均有一定的联系,若仅选择一个理论,较难构建系统地反映入室盗窃犯罪特点的分析框架。相比之下,理性选择理论更具包容性,它将犯罪者看作一个“理性人”,认为他们的作案动机遵循潜在收益最大化、风险及成本最小化的特征,为分析入室盗窃的机理提供了一个更加广阔的视角,在探讨社区环境与入室盗窃这一研究问题上更具有系统性[14,15]。
理性选择理论由Cornish等首次运用于犯罪者的作案目标选择,潜在收益、风险、成本是该理论的三大重要维度[14]。成本与风险越小,潜在的收益越大,越容易成为犯罪分子的作案目标。作案地可达性是衡量成本的重要指标,可达性越高,则成本越低,越容易吸引犯罪分子[16,17]。潜在收益方面,已有研究表明,平均房屋价值越高的社区,其被盗的可能性越大[18]。Nee等通过对犯罪者的采访调查发现,入室盗窃者会根据住房的外观状态,譬如房屋大小、房屋维护情况等判断其可能的价值,从而决定是否潜入作案[19]。风险与社区的社会控制存在较大的关系。社会控制包括正式社会控制和非正式社会控制,前者是指正式的防控力量,如警察、保安等;后者是指邻里守望效应或集体效能,即邻居之间的社会凝聚力以及他们为集体利益干预制止扰乱社会秩序的意愿[20]。社会控制越强,犯罪分子的作案风险越大,该社区越不容易被盗。例如,人口流动性大、种族异质性强的社区往往社会控制较弱,容易成为犯罪分子的选择目标[21]。
需要指出的是,理性选择理论虽从三个维度探讨犯罪分子如何选择目标进行作案,但是由于社区环境的复杂性,风险与收益并存,具体衡量的指标往往与这两者均存在关系。例如,较为富裕的社区潜在的收益较大,但往往安保设施相对较好,社会控制较强,犯罪风险较大,不同的犯罪者做出的判断存在差异。针对比利时的研究发现,犯罪者选择高价值房屋的概率较其他房屋更低,因为进入作案的风险较高[22];而针对荷兰的研究发现,房产价值越高,则被盗的可能性也越大,潜在收益起主导作用[23]。
中西方的住房形态存在较大的差异。中国的住房制度改革及近些年来房地产业的发展使得中国住房特征发生巨大变化[24],形成了本地人的自建住房、计划经济体制时代的单位住房、市场经济时代的商品房及政府主导建设的保障性住房等并存的格局[25]。郊区的农村自建住房及计划经济体制时期的单位住房社区由于拥有较为熟悉的邻里关系及邻里互动,往往可形成较好的非正式社会控制。而中心城区及城郊地区的自建住房主要为城中村,由于主要是外来人口的聚居区,社会控制较弱。商品房及保障房社区主要是高层住宅,邻里关系相对陌生,但是安保设施相对较好[26]。因此,不同类型的住房对应的潜在收益与风险存在较大的差异。然而,已有的西方文献并没有考虑到中国复杂的住房特点。
国内学者对入室盗窃的研究多以个体问卷调查数据为基础,研究发现,住房是否安装防盗门、住房层高、建筑的新旧程度等均会对入室盗窃有影响[27-29]。但此类研究的样本较为局限,例如杨刚斌等仅对门禁小区进行了研究[27]。而且他们并没有将房屋的微观特点与住房类型相结合,缺乏对房屋所处宏观环境的讨论。社区的人口特征是影响入室盗窃的另一重要因素。古杰等研究发现,外来人口越多的地区,入室盗窃的发生概率越高,外来人口聚居地区往往呈现出较强的社会失序状况[28];陈刚等基于全国尺度的研究同样发现流动人口对入室盗窃存在显著影响[30]。但龙冬平等在针对入室盗窃的研究中发现,流动人口对入室盗窃并不存在影响[31]。研究结论的差异可能与指标体系的选择有关。例如龙冬平等并没有将住房的类型纳入到分析框架,而住房作为入室盗窃犯罪的最终目标,其特征在犯罪者的决策过程中必然发挥重要的作用[22]。
综上,部分文献对中国的入室盗窃的格局和机理进行了研究,但它们鲜有综合考虑到不同住房类型风险收益的特点及其对犯罪的影响,衡量人口特征的指标也较单一,对西方理论在中国应用的适用性及其拓展时应注意的问题讨论得仍不够充分。为此,本研究以理性选择理论为分析框架,结合中国实际情况,丰富和细化社区住房特征、人口特征的衡量体系及其理论含义,建立定量模型探讨社区环境对入室盗窃的影响,力求对入室盗窃的形成机理有更全面的了解,以期在犯罪地理理论和指导警务实践方面能有一定的贡献。

2 概念框架

理性选择理论强调犯罪分子对作案目标的综合判断,需要权衡潜在收益、风险及成本,寻找最有利的社区环境进行作案。对于入室盗窃而言,潜在收益通常反映为犯罪者对目标房屋可被偷盗的财富的判断;风险则是指犯罪者作案后被捕的风险,主要与社会控制、安保设施相关;成本主要体现为犯罪者的出行成本。基于此,本研究的概念框架主要围绕理性选择理论的这三大要素进行展开（图1）。
Fig. 1 Conceptual framework

图1 概念框架

在表征维度方面,潜在收益和风险主要体现在社区的住房特征和人口特征上,社区的特征同时与收益和风险密切相关。例如住房条件好的社区较富裕,犯罪者能获得的潜在收益较大,但同时由于监管措施较好,犯罪者也将面临较大的风险;人口组成复杂的城中村,社会控制效应及安保力量较弱,犯罪者面临的风险较低,但同时潜在收益也可能较低。犯罪者更重视安全还是收益,仍需进一步的探讨。成本与交通可达性存在较密切联系,目标社区的交通越便捷,犯罪者的出行成本越低。在衡量指标方面,根据上文对文献的梳理和对中国情况的考虑,本研究选用住房面积、建筑层高、建筑年代、房屋租金、住房来源对目标社区的住房特征进行衡量;选用年龄、学历、户口、职业衡量社区的人口特点;选用社区的道路网密度和公交线路密度反映社区的交通可达性。
下文中,将从数据与方法、变量分析、模型结果和结论讨论等方面对这一概念框架进行更系统和深入的阐述与论证。

3 研究方法与数据来源

3.1 数据来源

选取东南沿海发达城市ZG市进行分析,研究区域包括ZG市行政边界内所有城乡区域,研究单元为派出所。派出所是中国公安部门的基层机构。宋广文等以派出所为研究尺度发现,该尺度对犯罪分析有较好的效果[32];Boessen等认为中观尺度的社区环境对犯罪有较重要的影响[33]。本研究中,去除没有居民居住的派出所（火车站派出所、会展派出所）之后,共有226个派出所纳入分析。派出所的平均面积为32.91 km2,50%的派出所面积小于10 km2,最小为0.20 km2,最大为463.53 km2。另外,每个派出所平均包含11.69个社区。其中一个派出所最少包含一个社区,最多包含34个社区。
研究数据包括2014年ZG市入室盗窃警情数据、2010年人口普查数据及2014年交通路网数据。入室盗窃警情数据包含报警时间、案发地所在派出所等信息;人口普查数据包含基于社区尺度的人口属性、年龄、住房类别、房屋租金、职业类型等。交通路网数据主要是采购导航公司的导航路网数据,能覆盖车辆通行的绝大部分地区。由于以派出所为研究单元进行分析,人口普查数据和交通路网数据均汇总到派出所进行统计。

3.2 变量与假设

3.2.1 变量提取 因变量为2014年每万人入室盗窃案件发生率,即用派出所辖区内2014年发生入室盗窃的警情总数除以辖区内的居住总人口（万人）。
自变量的选取包括交通可达性、人口特征及住房特征三个方面：
（1）交通可达性：交通可达性主要以派出所辖区为单元,包括道路网密度和公交线路密度。道路网密度为派出所内路网的总长度与派出所辖区面积之比,反映该地区与外界交流的便捷程度。公交线路密度计算方法为：通过统计一个派出所内每个站点的经过的公交线路数量之和,得到该派出所内各站点公交线路总和,再除以该派出所辖区内总人口数量,即得到公交线路密度。公交线路密集的区域,与外界交流的交通费用成本相对较低。
（2）人口特征：六普数据中含有丰富的人口特征信息。但是由于指标之间存在较强的共线性,因此有必要对它们进行降维从而更好地表征ZG市人口特征。主成分分析要求纳入分析的数据是正定矩阵,而变量间相关性太强则会导致非正定矩阵出现。因此,经过筛选,考虑各变量的因子载荷,最终选择对学历、职业类型、年龄、户籍地、户口性质相对应的13个指标进入分析。KMO值为0.536,表明该指标选取可进行因子分析。按特征值大于1的原则,共提取3个主因子。并对初始因子载荷矩阵进行Varimax正交旋转,旋转后的前3个主因子累计方差贡献率达80.3%（表1）。三大主因子包括：① 高学历知识分子。该主因子的职业结构表现为专业技术人员、办事人员和国家机关、企事业单位负责人,该群体学历高,有良好的社会经济地位。② 青壮年外来务工人员。该主因子表现出明显的青壮年外来人口特征,职业结构主要表现为生产、运输设备操作人员及有关人员,因此主要代表城镇化背景下大量涌入经济发达的大城市地区的外来务工人员。③ 农业人口。该主因子主要表现出低学历及“重农轻商”的特点,主要为农业人口。
Tab. 1 Rotated component matrix of population characteristics

表1 人口特征主因子旋转成份矩阵

因子1
高学历知识分子		 因子2
青壮年外来务工人员		 因子3
农业人员
年龄 19~30岁(%) 0.175 0.858 0.273
学历 初中及以下(%) -0.821 -0.158 0.489
大学专科及以上(%) 0.922 0.191 -0.205
户口 常住本地人口(%) 0.060 -0.922 0.234
外来人口户口在本省其他县市(%) 0.328 0.814 -0.080
外来人口户口登记为省外(%) -0.456 0.808 -0.064
农业人口(%) -0.582 0.151 0.735
职业 国家机关、党群组织、企业、事业单位负责人(%) 0.578 0.062 -0.187
专业技术人员(%) 0.761 -0.161 -0.494
办事人员和有关人员(%) 0.482 -0.242 -0.699
商业、服务业人员(%) 0.253 0.025 -0.741
农、林、牧、渔、水利业生产人员(%) -0.141 -0.437 0.842
生产、运输设备操作人员及有关人员(%) -0.678 0.578 0.145
（3）住房特征：与人口特征筛选方法相一致,从住房面积、建筑层高、建筑年代、房屋租金和住房来源中筛选出20个指标进行因子分析,KMO值为0.596,表明适合进行因子分析。共提取五大类主因子,经Varimax正交旋转后的前5个主因子累计方差贡献率达80.3%（表2）。五大主因子分别为自建住房（包括远郊农村住房和城中村）、普通商品房、高档商品房、原公有住房及保障性住房。① 自建住房：该主因子与大面积住房、低出租率、低楼层、自建住房有较强的正相关,与小面积、高出租率有较强的负相关。从空间分布看,该因子得分较高（符号为正,绝对值较大）的地区主要位于距离市中心较远的农村地区;得分较低（符号为负,绝对值较大）的地区主要分布在城中村较多的区域。因此,可认为该类因子是自建住房因子,因子得分符号为正时,代表是农村自建住房;得分符号为负时,代表的是城中自建住房,即城中村。② 普通商品房：该主因子主要表现为高购买商品房及购买二手房率,低自建住房率,且楼层以4~9层为主,一般而言,该类住房主要为楼梯式商品房。③ 高档商品房：该主因子主要表现为高租金（租金>1000元）及高楼层（楼层>10）的特点,主要表现为高层高租金的高端住宅。④ 原公有住房：该主因子表现出明显的房屋建筑年代较远（1979年以前）,且以小户型（户面积50~80 m2）为主,房屋属性主要为原公有住房。⑤ 保障性住房：该主因子主要表现为高购买经济适用房比率及高租赁廉租房比率,理解为政府为生活水平相对较低的人群提供相应住房保障。
Tab. 2 Rotated component matrix of housing characteristics

表2 住房特征主因子旋转成份矩阵

因子1
自建住房		 因子2
普通商品房		 因子3
高档商品房		 因子4
原公有住房		 因子5
保障性住房
住房面积 每户50 m2以下(%) -0.956 -0.124 -0.123 -0.112 0.025
每户50~80 m2(%) -0.149 0.340 0.254 0.722 0.007
每户80~140 m2(%) 0.838 0.285 0.054 -0.085 0.048
每户平均面积(m2) 0.854 -0.223 -0.032 -0.246 -0.072
每户平均房间数(间) 0.858 -0.329 -0.092 -0.141 -0.087
建筑层高 3层及以下(%) 0.479 -0.731 -0.393 -0.088 -0.119
4~9层(%) -0.583 0.697 0.077 0.131 0.068
10层及以上楼房(%) -0.025 0.400 0.778 -0.042 0.150
建筑年代 1979年以前(%) -0.044 -0.361 0.013 0.728 0.047
2000年以后(%) -0.170 -0.114 0.074 -0.888 0.018
房屋租金 租金500元以下(%) -0.023 -0.404 -0.716 -0.365 0.134
租金1000~2000元(%) 0.014 0.301 0.682 0.395 -0.181
租金2000元以上(%) 0.062 0.047 0.884 -0.017 0.021
住房来源 自建住房(%) 0.597 -0.664 -0.360 -0.079 -0.117
购买商品房(%) 0.233 0.812 0.292 -0.127 -0.043
购买二手房(%) 0.052 0.750 0.257 0.301 -0.043
购买原公有住房(%) -0.211 0.262 0.446 0.681 -0.048
购买经济适用房(%) 0.084 0.085 0.212 0.059 0.804
租赁廉租房(%) -0.162 -0.059 -0.274 -0.071 0.647
租赁其他住房(%) -0.895 0.055 -0.076 -0.306 -0.037
3.2.2 变量假设 本文选择的指标体系与理性选择理论的三大要素（潜在收益、风险及成本）的对应关系如表3所示。成本主要与交通可达性相对应,潜在收益和风险与社区环境（人口特征及住房特征）相对应。
Tab. 3 Theoretical hypotheses of independent variables

表3 自变量的理论假设

类别 变量/主成分 变量分析假设
潜在收益 风险 成本 理论总效应
可达性 道路网密度 - - 可进入性强 增加
公交线路密度 - - 可进入性强 增加
人口
特征		 高学历知识分子因子 目标价值较高 社会控制较强 - 无影响
青壮年外来务工人员因子 目标价值较低 社会控制较弱 - 无影响
住房
特征		 自建住房因子 目标价值较低 安保设施较差 - 无影响
普通商品房因子 目标价值较高 安保设施一般 - 增加
高档商品房因子 目标价值很高 安保设施较好 - 无影响
原公有住房因子 目标价值较低 安保设施一般,非正式社会控制较强 - 抑制
保障性住房因子 目标价值较低 非正式社会控制较弱 - 无影响
交通可达性方面,道路网密度及公交线路密度主要反映出该区域的交通便利程度,交通便利、可达性高的区域对于犯罪者而言,通勤成本相对低,可达性强,从而导致犯罪的增加。
人口特征方面：① 高学历知识分子因子取值越大的区域可盗取财物价值越大,作案收益越高,但由于该群体警惕性高,社会秩序良好,社会控制强,犯罪者被抓获风险也较高,总体而言,高收益与高风险相抵,使得该因子对入室盗窃无显著影响。② 青壮年外来务工人员因子值越大,则表明该区域居民拥有财物价值较小,潜在获益相对较低,但由于这类人群的流动性较强,邻里关系较难构建,社会控制较弱,潜在的被抓风险较低。因此,低收益与低风险结合,该因子同样对入室盗窃无显著影响。③ 农业人口因子值越大,表明该地区主要是农业户口人群集聚的地区。该地区作案潜在获益较低,但是作案风险相对也比较低。由于该因子与自建住房因子存在较强相关性,因此在后续的分析中,该因子将不再纳入分析。
住房特征方面：① 自建住房普遍楼层比较低矮,安保设施较差,因此可进入性相对较强。此外,无论是郊区住房还是城中村住房,潜在的财物价值相对较低。整体而言,该因子对入室盗窃无影响。② 普通商品房：这类住房潜在获益较高,同时,该类住房楼层相对较低,可进入性较好,安保设施一般,容易成为犯罪分子的目标。③ 高档商品房：这类住宅楼层较高,可进入性弱,安保防护系统良好,风险较大,此外,潜在的可盗取财物价值较高。因此,高风险高收益的高档商品房对入室盗窃无显著影响。④ 原公有住房主要集中在老城区,主要住户为附近机关事业工作人员或退休人员,邻里关系依然相对其他住房类型来说更融洽,非正式社会控制较强,作案风险较大,且该类住房年代较远,从住房外观特征来看潜在获益较低,因此,原公有住房成为犯罪分子目标的概率较小。⑤ 保障性住房是政府政策指导下建设的保障性住房,主要用于解决中低收入群体的住房问题,离市中心比较远,非正式社会控制较弱,住户经济收入相对低,风险小、收益低,因此对入室盗窃无明显影响。
可以看出,潜在风险和收益共同作用于社区的人口特征和住房特征因素对入室盗窃的影响。各类因素所产生的综合效应往往很难通过定性的分析去衡量,需要定量的统计模型进行进一步的分析。

3.3 模型构建

犯罪数据往往呈现过离散现象（方差大于均值）[34]。在泊松回归系列模型中,泊松回归模型其中一个重要的假设是因变量的期望值与方差相等,而过离散数据常导致系数估计的标准误被低估,进而影响T检验,造成解释变量显著性的误判。负二项回归模型属于泊松回归模型系列,它在泊松回归模型的基础上,增加了残差项的估计,认定残差服从Gamma分布,从而大大提高了对过离散数据的拟合程度。泊松回归模型的概率分布公式为：
P Y i = y i | X i = e - λ i λ i y i y i ! , i = 1,2 , 3,4 ⋯ （1）
式中：参数λi采用下述回归模型进行表征：
λ i = e β X i （2）
ln λ i = ∑ k = 0 k β k X ik （3）
式中：λi为第i个研究单元中的案件数量;参数λi取决于一系列的解释变量Xi;xi表示对入室盗窃的影响因素;β为各变量的回归系数,其中,β0是一个常数,相当于截距。
泊松回归模型因变量的方差与期望值相等,均等于λi,而负二项回归模型在增加了对残差估计的Gamma分布后,其期望值和方差值如下[35]：
E Y i | X i , β = λ i （4）
Var Y i | X i , β = λ i + α λ i 2 （5）
式中：当α等于0时,即为传统的泊松回归模型,随着α的增大,因变量的分布将越来越离散。因此,如果α显著大于零,则应采用负二项回归模型进行分析。
在构建模型过程中,受限于自变量特征及模型的准确性,往往需要采用案件率作为因变量进行分析。因此需要对公式（3）进行转换[36],转换后的公式：
ln λ i n i = ∑ k = 0 k β k X ik （6）
式中：令 λ i n i 表示人均犯罪率;ni表示对应研究单元的人口数量。
ln λ i = ln n i + ∑ k = 0 k β k X ik （7）
公式（7）由公式（6）变换得到。由公式（7）可以发现,通过增加令其系数为1的ln(ni)变量,则该泊松回归模型由对案件量的分析可转化为对人均案件率的分析。

4 变量分析与模型结果

本节首先对因变量和自变量进行描述统计分析,介绍数据的基本情况,在此基础上,建立负二项模型进行定量分析。在模型构建中,由于农业人口因子与自建住房因子的共线性较强,因此,人口特征中的农村人口因子不纳入模型中进行分析。

4.1 变量分析

4.1.1 因变量的空间分布特征 图2为入室盗窃案件基于派出所尺度的随机点分布。网格大小为5 km×5 km,每一个点代表30起入室盗窃警情。ZG市入室盗窃在空间上主要形成两条轴线：由E3-L6构成的纵轴及G2-H10构成的横轴。其中,中心城区主要区域位于网格F2-I5,这些区域既是入室盗窃案件量高的区域,同时也是人口高度集中的地区。此外,网格J4-L5及E2-E4区域也是案件集聚区,主要位于城郊地区。在中心城区东侧的H8-H10区域有一条狭长的入室盗窃聚集区,主要沿城市干道两侧分布,该区域是ZG市重要的工业基地,集中了大量的外来务工人员。
Fig. 2 Spatial distribution of residential burglary in ZG city in 2014

图2 2014年ZG市入室盗窃的空间分布特征

4.1.2 变量的描述统计 表4介绍了因变量和自变量的平均值、方差、最小值和最大值。入室盗窃率和公交线路的方差值最大,分别达到1040.06和2338.69,说明数据分布较离散,其中公交线路的均值为28.69线路/万人,而公交线路密度差值最大的两个派出所达到686.46线路/万人。需要注意的是,共有12个派出所没有公交线路途经,其中1个属于历史文物保护区,5个林场或林业派出所,6个位于距离市中心偏远的乡镇派出所,主要通过乡镇巴士与市中心联系。道路网密度值分布也较离散,派出所辖区的平均道路网密度达到8.02 km/km2,路网最密集的派出所辖区其道路网密度达到30.21 km/km2,但其中有一林业性质派出所,以低等级道路为主,由于信息采集难度大等客观原因,该派出所道路网密度为0。人口特征和住房特征的因子分析经过正交旋转处理,平均值为0,方差为1。不同因子的最小值和最大值的差距存在差异,高档商品房因子、原公有住房因子、保障性住房因子的差距最大,自建住房的差距相对较小。
Tab. 4 Descriptive statistics of dependent and independent variables

表4 变量的描述统计

变量 平均值 方差 最小值 最大值
因变量 入室盗窃率 39.970 1040.127 0.000 211.719
可达性 道路网密度 8.020 39.753 0.000 30.215
公交线路密度 28.687 2338.786 0.000 686.456
人口特征 高学历知识分子因子 0 1 -2.127 4.153
青壮年外来务工人员因子 0 1 -1.798 3.656
住房特征 自建住房因子 0 1 -2.559 1.685
普通商品房因子 0 1 -2.038 3.175
高档商品房因子 0 1 -1.695 7.035
原公有住房因子 0 1 -3.762 3.068
保障性住房因子 0 1 -1.565 5.810

4.2 模型结果

通过表5可以看到,模型alpha值显著大于0,因此应该采用负二项回归模型而非传统泊松回归模型。各自变量的VIF均值小于2,最大值也不超过3,证明了自变量的共线性较弱,模型指标选择合理。
Tab. 5 Negative binomial regression results of the impact of community environment on residential burglary

表5 社区环境对入室盗窃率影响的负二项回归结果

变量名 R IRR Robust Std. Err. VIF
常量 2.980*** 19.688 0.131
交通可达性 道路网密度 0.060*** 1.062 0.013 2.83
公交线路密度 0.003 1.003 0.004 1.07
人口特征 高学历知识分子因子 -0.419*** 0.658 0.043 1.71
青壮年外来务工人员因子 0.283*** 1.327 0.047 2.55
住房特征 自建住房因子 0.107 1.113 0.059 2.20
普通商品房因子 0.147** 1.158 0.051 1.78
高档商品房因子 -0.058 0.944 0.054 1.96
原公有住房因子 -0.222*** 0.801 0.056 2.15
保障性住房因子 -0.057 0.945 0.042 1.02
alpha 0.320*** 0.061
N 226

注：*表示P<0.05,**表示P<0.01,***表示P<0.001。

模型结果表明：交通可达性上,路网密度对入室盗窃存在显著的正向影响,与假设理论总效应（表3）一致。较高的路网密度有利于该地区与外界进行各种交流,犯罪者可达性高,从而这些地区更容易成为犯罪者的目标。而模型中公交线路密度对入室盗窃的影响并不明显,与假设理论总效应不符。对于入室盗窃者而言,公交可能并非主要的交通作案工具,公共交通工具由于固定的线路,不一定方便他们寻找目标和逃逸,电动车或其他机动车可能是他们较常用的交通工具。
人口特征方面,高学历因子及青壮年外来务工人员因子的模型表现与其理论总效应并不一致。在假设理论总效应中,高学历因子的“高风险、高收益”与青壮年外来务工人员因子的“低风险、低收益”中风险与收益相互制衡,对该地区的入室盗窃无显著影响。模型结果则显示：① 高学历因子对当地的入室盗窃率有显著负向影响。该结果反映出对于高学历因子值较高的区域,社会控制强度大,进入此地作案给犯罪者带来的风险大于潜在收益。良好的社会秩序将显著遏制犯罪行为,其影响甚至超过了预期的高收益对犯罪者的吸引。② 青壮年外来务工人员因子对入室盗窃率有显著的正向影响。同样从潜在收益和风险的角度分析,证明在青壮年外来务工人员比例较高的地区,即便潜在收益可能相对较低,但是暴露风险的可能性降低将显著促进犯罪者的作案行为,从而导致该地区入室盗窃率的上升。
住房特征相关因子与理论总效应假设亦基本一致。① 自建住房、高档商品房及保障性住房风险收益比较低,对入室盗窃的影响并不显著。② 普通商品房因子对当地入室盗窃有显著的正向影响。普通商品房相对而言可供偷盗的财物较多,而其相对一般的安保设施对于有经验的入室盗窃者而言,往往可以找到合适的地方进行隐藏及逃逸。因此,潜在的高收益及相对并不严密的安保设施对犯罪者具有明显的诱惑性,使该地的入室盗窃率显著提升。③ 原公有住房因子对入室盗窃有显著的负向影响。外观上较为老旧的住房形态及非正式的社会控制的作用均能有效降低该地区入室盗窃的发生几率。

5 结论与讨论

本研究充分考虑了中国社会发展的特点,结合理性选择理论,考虑成本、风险、收益三大维度,综合探讨了人口特征、住房特征对入室盗窃率的影响,主要有以下结论：
（1）人口特征方面,高学历知识分子聚集度较高的地区,入室盗窃率相对较低,而青壮年外来务工人员比率越高,则当地的入室盗窃率将显著增加。
（2）住房特征方面,普通商品房容易成为犯罪者的吸引目标,原公有住房则会降低入室盗窃的风险,自建住房、保障性住房、高档商品房对入室盗窃的影响并不显著。
（3）理性选择理论对于解释中国入室盗窃的成因有较好的效果。该理论从犯罪者的视角出发,从潜在收益、风险及成本三大维度解释社区环境如何影响入室盗窃。本实证研究在现有理论基础上,进一步发现,入室盗窃犯罪分子遵循“安全第一、收益第二”的选择原则,他们更倾向于选择风险较低即便潜在收益价值相对较小的区域作案。
总体而言,本研究的发现与已有的中西方研究存在异同。相同之处例如起源于西方的理性选择理论在中国仍具有较强的适用性;具有较好经济社会地位的人口组成能有效降低入室盗窃的风险[21]。此外,外来人口密集区域更容易吸入入室盗窃者,与古杰等[28]的研究结论一致。不同之处或说本研究的主要贡献在于：① 这是理性选择理论在中国入室盗窃方面的具体化应用,重构了相关概念的衡量方式。中西方的社会环境与建成环境存在较大的差异,必然造成犯罪的影响因素存在较大的差异。例如西方的研究主要关注种族异质性,而中国的研究更加侧重外来人口[21,30];住房特征方面,西方的房屋形态主要指单栋住宅、联排房屋等,与国内的各大主要住房形态存在较大的差异[22]。② 本研究较好地分析了国内典型的住房类型的特点及其对入室盗窃的影响。改革开放以来,中国的住房市场经历了不同的阶段,在空间形成了多种制度背景和建筑形态并存的格局。本文首次采用定量方法探讨不同住房类型对入室盗窃的影响,初步构建了适合中国社会情况的指标分析体系与方法,是对已有犯罪地理文献的有效补充。
理清入室盗窃的影响因素,有助于认识入室盗窃与社区环境关系,从而能采取相应措施进行城市改造或警务防控,减少犯罪的发生。例如根据研究结论,城市管理部门可采取通过提升社区居民融入的方式,增强人群尤其是外来人口的凝聚力,使其更具备责任感,提升非正式社会控制的效果;也可针对不同类型的住房性质,有所侧重地分配警力,提高警务效率;此外可通过城市规划,对城市社会空间结构进行合理布局,因地制宜开展混居策略,抑制犯罪发生的风险。受限于数据和文章篇幅,本研究也存在一定的不足,例如由于数据的原因,无法具体衡量警力布置、巡逻路线等要素;对各社区环境要素对入室盗窃影响的时间变化考虑不足;以不同尺度为研究单元对分析结果的影响,即可塑性面积单元问题（Modifiable areal unit problem）[37,38]等。这些将在日后的研究当中不断完善。

The authors have declared that no competing interests exist.

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
严小兵, 焦华富. 犯罪地理国际研究进展. 地理科学进展, 2012, 31(10): 1390-1398.犯罪地理学是犯罪学和地理学研究的交叉学科。近30 年来, 随着GIS 技术的发展, 国外犯罪地理学在研究内容和研究方法上发生了许多转变。从5 个方面简述了国外犯罪地理研究的发展:①解释理论, 简述了不同理论的不足及发展;②犯罪活动空间分布, 全局关系和局域关系上犯罪活动空间分布的判别和可视化;③计量模型, 空间依赖和空间异质效应的计量;④虚拟犯罪, 犯罪活动预测的地理计算;⑤研究尺度, 从大尺度到小尺度的转变。最后总结了地理学对犯罪学研究的贡献, 并提出了国外的发展对国内研究的借鉴。

DOI

[Yan Xiaobing, Jiao Huafu.A review on the foreign criminal geography research. Progress in Geography, 2012, 31(10): 1390-1398.]

[2]
Lebeau J L, Leitner M.Introduction: Progress in research on the geography of crime. The Professional Geographer, 2011, 63(PII 9337286922): 161-173.The research presented in this article provides an accurate narrative of the historical evolution of the geography of crime, including its major ideological controversies. It further gives credit to geographers whose works have gained considerable respect in other academic communities. Additionally, it aims to convince geographers that they have potentially very valuable contributions to make to the study of crime. The final and most important point that this narrative is trying to make is that geography's intellectual territory has become attractive to other disciplines and fields embracing the geography and spatial analysis of crime.

DOI

[3]
姜超, 柳林. 中国犯罪地理与警务GIS研究进展. 地球信息科学学报, 2013, 15(6): 801-808.近30年来，随着公安部门犯罪防控需求的增加，中国犯罪地理学正处于新的发展机遇期。为了能较客观、全面地把握我国犯罪地理学的研究现状与动态，本文从中国知网、万方和维普3大电子资源数据库中筛选出与犯罪地理研究主题相关的期刊论文，以此为样本，对研究内容与研究实力等情况进行统计分析。结果表明，自1980s中期以来，中国犯罪地理学在论文数量、研究方法、实证区域、研究机构与核心作者等方面都日益增多，已经在城市犯罪地理、警用地理信息系统、规划设计预防犯罪、情境犯罪预防等4个主要研究方向出现了导向作用明显的核心作者，呈现出良好的发展态势。进入21世纪后，新的发展机遇在促进犯罪地理学论文数量与研究实力快速增加的同时，也导致了案例研究、量化研究的所占比重显著减少，研究区域的空间范围也大幅收敛，集中在长三角、京津、珠三角地区。新的研究阵营不断涌现，核心作者比重不断下降，犯罪地理学进入了研究阵营争鸣角力的重构期。未来我国犯罪地理学的发展任重而道远，不仅亟需完善相关的理论体系、加强国内的案例实证，而且还需注意研究规范的统一、核心作者群的代际更替等诸多问题。

DOI

[Jiang Chao, Liu Lin.Research progress of crime geography studies in China. Journal of Geo-Information Science, 2013, 15(6): 801-808.]

[4]
姜超, 唐焕丽, 柳林. 中国犯罪地理研究述评. 地理科学进展, 2014, 33(4): 561-573.当前社会犯罪问题严峻，以空间为视角的犯罪地理研究因其独特的综合性在犯罪防控中的作用日益凸显。以中国知网、维普和万方三大数据库中获得的期刊文献为基础，对中国犯罪地理的研究状况进行综合评述。在对国内五大研究主题进行回顾的基础上，围绕“研究主题”和“研究领域”两方面，分析了中国犯罪地理学的发展动态。总体来看，在过去的25 年中，中国犯罪地理研究逐渐起步，发展趋势良好，已经吸引了来自地理学、法学与警务管理、城市规划、应用数学等诸多领域的学者关注。尽管研究焦点各有侧重，但始终注重对国外研究和实践经验的引入，并提出了基于综合分析、情境预防、CPTED、PGIS的诸多防控对策。在犯罪时空分布特征、犯罪发生诱因等方面，研究范围宽广，但研究深度有待加强。未来中国犯罪地理学发展任重而道远。学界应充分重视城市犯罪与农村犯罪、犯罪与微观环境的互动关系、犯罪主体的行为空间等研究，并积极探索虚拟犯罪模拟新技术的应用。

DOI

[Jiang Chao, Tang Huanli, Liu Lin.Review of crime geography in China. Progress in Geography, 2014, 33(4): 561-573.]

[5]
陈宏胜, 李志刚. 快速城镇化下中国大城市房权分异研究: 对广州的实证. 地理学报, 2014, 69(12): 1821-1832.利用广州市第六次人口普查的微观数据,通过房权指数T、房权分异指数ID和房权集中指数II三指标对广州市城市社区的房权分异进行测算,并使用＂六普＂数据从人口、制度、市场、住房四项维度对影响广州城市房权指数的因素进行分析。研究表明,广州城市房权指数的平均水平为0.71,属较高水平;各区房权指数存在明显差异,其分布由城市核心到边缘可分为＂中—低—高＂三类,中心区较为混合,而郊区则呈现较为极化的格局;广州中心区城市房权分布以越秀、荔湾、海珠三区交接处为中心,房权指数最高,老城区外围房权指数最低,呈现一种＂差序格局＂。第二,就空间分异而言,广州不同房权住房的空间分异度为0.45,集中指数为0.68,属中等水平,说明广州的房权分异并不明显,不同房权类型的住房呈现相对混合的空间分布;但各区则呈现差异化格局：近郊区的房权分异强（空间极化）、远郊区房权分异弱（空间混合）,近郊区房权分异最为突出。在机制上,婚姻状况、年龄构成、学历状况、户籍制度、社区居住条件等5个要素对社区房权有显著影响;总体模型表明,婚姻和户籍是影响社区房权分异的决定性因素,说明当前中国城市的住房房权的分布格局主要由家庭和制度因素驱动,而非由市场因素驱动,以此展现和强调了中国城市社会地理与西方的差别。

DOI

[Chen Hongsheng, Li Zhigang.Tenure-based housing segregation under rapid urbanization in post-reform urban China: A case study of Guangzhou. Acta Geographica Sinica, 2014, 69(12): 1821-1832.]

[6]
Logan J R, Bian Y, Bian F.Housing inequality in urban China in the 1990s. International Journal of Urban and Regional Research, 1999, 23(1): 7-25.Abstract Access to housing of sufficient space and quality has been a central element in social stratification in urban China. We examine the sources of housing inequality in 1993 in Shanghai and Tianjin, when a market reform process had been underway in the national economy for nearly fifteen years. The Chinese housing allocation system favors people with political position and connections, those of higher socio-economic background, and those whose work units have greater organizational authority. There is only slight evidence that market reform has undermined this stratification order. To the contrary, there are reasons to believe that in some respects inequalities rooted in socialism are strengthened by institutional changes. These conclusions are reinforced by comparison to results of analyses of income inequality in the same cities. L'accès aux logements suffisamment spacieux et de bonne qualité est un élément central de la stratification sociale en Chine urbaine. Nous examinons les origines de l'inégalité du logement en 1993 à Shangai?? et Tianjin alors qu'un processus de réforme du marché dans l'économie nationale est en cours depuis presque quinze ans. Le système d'allocation de logements favorise les personnes qui ont une position et des contacts politiques, les personnes d'un milieu socio-économique élevé et celles des unités de travail qui ont une plus grande autorité d'organisation. Il y a peu de preuves que les réformes du marché aient ébranlé cette stratification. Au contraire, il y a des raisons de penser que certaines inégalités dont l'origine remonte au socialisme sont fortifiés par les changements institutionnels. Ces conclusions sont renforcées par la comparaison des résultats des analyses des inégalités des revenus dans les mêmes villes.

DOI

[7]
Li S M.Housing tenure and residential mobility in urban China a study of commodity housing development in Beijing and Guangzhou. Urban Affairs Review, 2003, 38(4): 510-534.

DOI

[8]
Liu J H.Crime patterns during the market transition in China. British Journal of Criminology, 2005, 45(5): 613-633.

DOI

[9]
Ye X, Xu X, Lee J, et al.Space-time interaction of residential burglaries in Wuhan, China. Applied Geography, 2015, 60: 210-216.61We use a recent urban burglary dataset in China to analyze the spatiotemporal changes of hotspots.61We apply the kernel density estimation and the space–time K-function methods.61Both spatial and spatiotemporal clustering exist.61The hotspots are stable over time.61The spatiotemporal clustering shows significant effects of distance decay.

DOI

[10]
Wortley R, Mazerolle L.Environmental Criminology and Crime Analysis. Cullompton: Willan Publishing, 2008.

[11]
Jeffery C R.Crime Prevention through Environmental Design. Beverly Hills: Sage Publications, 1971.Overview This entry examines an approach to crime reduction which differs from many others in that it focuses not on the offender or their reasoning for committing an offense but upon the environment in which an offense takes place. This approach also differs in its consideration of who should be held responsible for the reduction of crime, with a focus not solely upon the traditional criminal justice system agencies but also upon planners, architects, developers, and managers of public space. The approach is based on the presumption that offenders will maximize crime opportunities, and therefore, those opportunities must be avoided (in the first place) or removed (following the emergence of a crime problem). In the 2001 publication “Cracking Crime Through Design,” Pease introduces the concept of design as a means of reducing crime, but more importantly the premise that it is the moral responsibility of many different actors and agencies to improve the lives of those who may fal ...

DOI

[12]
Shaw C R, Mckay H D.Juvenile Delinquency And Urban Areas. Chicago: University of Chicago Press, 1942.

[13]
Cohen L E, Felson M.Social change and crime rate trends: A routine activity approach. American Sociological Review, 1979, 44(4): 588-608.In this paper we present a "routine activity approach" for analyzing crime rate trends and cycles. Rather than emphasizing the characteristics of offenders, with this approach we concentrate upon the circumstances in which they carry out predatory criminal acts. Most criminal acts require convergence in space and time of likely offenders, suitable targets and the absence of capable guardians against crime. Human ecological theory facilitates an investigation into the way in which social structure produces this convergence, hence allowing illegal activities to feed upon the legal activities of everyday life. In particular, we hypothesize that the dispersion of activities away from households and families increases the opportunity for crime and thus generates higher crime rates. A variety of data is presented in support of the hypothesis, which helps explain crime rate trends in the United States 1947-1974 as a byproduct of changes in such variables as labor force participation and single-adult households.

DOI

[14]
Cornish D B, Clarke R V.The Reasoning Criminal: Rational Choice Perspectives on Offending. New York: Springer-Verlag, 1986.The assumption that rewards and punishments influence our choices between different courses of action underlies economic, sociological, psychological, and legal thinking about human action. Hence, the notion of a reasoning criminal--one who employs the same sorts of cognitive strategies when contemplating offending as they and the rest of us use when making other decisions--might seem a small contribution to crime control. This conclusion would be mistaken. This volume develops an alternative approach, termed the "rational choice perspective," to explain criminal behavior. Instead of emphasizing the differences between criminals and non-criminals, it stresses some of the similarities. In particular, while the contributors do not deny the existence of irrational and pathological components in crimes, they suggest that the rational aspects of offending should be explored. An international group of researchers in criminology, psychology, and economics provide a comprehensive review of original research on the criminal offender as a reasoning decision maker. While recognizing the crucial influence of situational factors, the rational choice perspective provides a framework within which to incorporate and locate existing theories about crime. In doing so it also provides both a new agenda for research and sheds a fresh light on deterrent and prevention policies.

DOI

[15]
Loughran T A, Paternoster R, Chalfin A, et al.Can rational choice be considered a general theory of crime? Evidence from individual-level panel data. Criminology, 2016, 54(1): 86-112.In the last few decades, rational choice theory has emerged as a bedrock theory in the fields of economics, sociology, psychology, and political science. Although rational choice theory has been available to criminologists for many years now, the field has not embraced it as other disciplines have. Moreover, rational choice scholars have fueled this skepticism of the theory's generality by modeling offender decision making that is one‐sided—large on the costs of crime (sanction threats), short on the benefits of crime. In this article, we directly assess the generality of rational choice theory by examining a fully specified model in a population that is often presumed to be less rational—adolescents from lower socioeconomic families who commit both instrumental (property) and expressive crimes (violence/drugs). By using a panel of N = .

DOI

[16]
Davies T, Johnson S D.Examining the relationship between road structure and burglary risk via quantitative network analysis. Journal of Quantitative Criminology, 2015, 31(3): 481-507.To test the hypothesis that the spatial distribution of residential burglary is shaped by the configuration of the street network, as predicted by, for example, crime pattern theory. In particular, th

DOI

[17]
Wu L, Liu X, Ye X, et al.Permeability, space syntax, and the patterning of residential burglaries in urban China. Applied Geography, 2015, 60: 261-265.61We focus on the influence of the permeability of the street network on the location of residential burglaries in China.61Cul-de-sacs do not appear to attract more residential burglaries than other road types.61Road type, length and Household are statistically significant in relation to residential burglary risk.61The residential burglary risk on branch way is higher than on segments of other types.61Sinuosity of street segments, education and unemployment influence residential burglary distribution.

DOI

[18]
Bernasco W, Nieuwbeerta P.How do residential burglars select target areas? A new approach to the analysis of criminal location choice. British Journal of Criminology, 2005, 45(3): 296-315.

DOI

[19]
Nee C, Taylor M.Examining burglars' target selection: Interview, experiment or ethnomethodology?. Psychology, Crime & Law, 2000, 6(1): 45-59.Ethnomethodologists in the field of offender-based research have recently criticised the earlier use of prison-based samples in research on residential burglary. They claim that interviewing burglars in their natural environment has produced findings of greater validity and reliability. By describing further analysis of data from earlier experimental research on burglars in prison, and drawing on findings from other work on residential burglary, this article sets out to highlight the striking similarity between findings from interview, experimental and ethnographic studies in this area. Far from discounting earlier experimental and interview studies, the recent ethnographic works have served to build on and complement earlier work. The value of using a variety of methods in offender-based research is then discussed.

DOI

[20]
Jiang S, Lambert E, Wang J.Correlates of formal and informal social/crime control in China: An exploratory study. Journal of Criminal Justice, 2007, 35(3): 261-271.There are two major ways for a society to control its members, formal and informal. A major goal of both forms of control is to curb criminal behavior. Formal criminal justice control uses the law and official government agencies (e.g., police, courts, and corrections) to ensure compliance. Informal criminal justice control uses morals and social institutions (e.g., family, peers, and neighbors). China has a long history of using informal criminal justice controls. Nevertheless, there has been a movement during the past several decades towards use of formal criminal justice controls. This study examined the level of agreement with both forms of control and the correlates of each form using a survey of Chinese college students. Findings from multivariate analysis indicated that those who held a Confucian belief in law and punishment, those who had a higher distrust of strangers, and those who grew up in rural areas were more supportive of informal control, while those who were majoring in the area of law and male respondents were more supportive of formal control.

DOI

[21]
Sampson R J, Groves W B.Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 1989, 94(4): 774-802.Shaw and McKay's influential theory of community social disorganization has never been directly tested. To address this, a community-level theory that builds on Shaw and McKay's original model is formulated and tested. The general hypothesis is that low economic status, ethnic heterogeneity, residential mobility, and family disruption lead to community social disorganization, which, in turn, increases crime and delinquency rates. A community's level of social organization is measured in terms of local friendship networks, control of street-corner teenage peer groups, and prevalence of organizational participation. The model is first tested by analyzing data for 238 localities in Great Britain constructed from a 1982 national survey of 10,905 residents. The model is then replicated on an independent national sample of 11,030 residents of 300 British localities in 1984. Results from both surveys support the theory and show that between-community variations in social disorganization transmit much of the effect of community structural characteristics on rates of both criminal victimization and criminal offending.

DOI

[22]
Vandeviver C, Neutens T, Van Daele S, et al.A discrete spatial choice model of burglary target selection at the house-level. Applied Geography, 2015, 64: 24-34.This article studies how burglars select a house to burglarize. We draw on the rational choice perspective to investigate how burglars select a target by relying on house-related attributes to optimize a combination of perceived rewards, efforts and risk. It extends current applications of the discrete spatial choice framework to burglary target selection by adopting the house as the spatial unit of analysis and studies burglars' target selection process in a larger and more diverse study area than that of earlier studies. Using data on 650 residential burglaries and on approximately 500,000 residential properties in the Belgian province East Flanders, we consider a discrete spatial choice model of burglary target selection to establish which house-related attributes influence burglars' target selection process. Our findings demonstrate that terraced houses, houses without a garage, houses that have not been outfitted with a central heating and/or air-conditioning system and houses nearby burglars' residences are more likely to be selected. Overall, our analysis suggests that burglars rely on effort-related attributes to distinguish between targets while higher perceived rewards actually decrease the odds of a house being burglarized. Risk-related attributes are unimportant for burglars' target choice.

DOI

[23]
Bernasco W, Luykx F.Effects of attractiveness, opportunity and accessibility to burglars on residential burglary rates of urban neighborhoods. Criminology, 2003, 41(3): 981-1001.This study assesses the effects of attractiveness, opportunity and accessibility to burglars on the residential burglary rates of urban neighborhoods, combining two complementary lines of investigation that have been following separate tracks in the literature. As a complement to standard measures of attractiveness and opportunity, we introduce and specify a spatial measure of the accessibility of neighborhoods to burglars. Using data on about 25, 000 attempted and completed residential burglaries committed in the period 1996鈥2001 in the city of The Hague, the Netherlands, we study the variation in burglary rates across its 89 residential neighborhoods. Our results suggest that all three factors, attractiveness, opportunity and accessibility to burglars, pull burglars to their target neighborhoods.

DOI

[24]
Wang Y P.Housing reform and its impacts on the urban poor in China. Housing Studies, 2000, 15(6): 845-864.Housing provision in Chinese cities has experienced many changes since 1979 when the country embarked on major economic reform. During the late 1980s and the 1990s many publicly owned houses were sold to their existing tenants or other public sector employees. Large numbers of new houses were built by commercial property developers for the emerging urban housing market. As a result, new patterns of residents have emerged. Housing areas of different standards for different social groups became a dominant feature of large Chinese cities at the end of the 1990s. Will the urban poor benefit from these changes and what is their housing situation like under the reformed system? This paper addresses these questions by examining the nature of the emerging urban poor and their accessibility to housing. It involves an assessment of the implications of recent housing reform policies for the disadvantaged groups. The paper identifies two major groups of urban poor in Chinese cities: the poor among the official urban residents and the poor rural to urban migrants. It concludes that while housing problems of the official urban poor have been recognised, there is no formal policy in relation to housing provision for the unofficial poor.

DOI

[25]
陈果. 基于广州实证的后改革时代中国城市住房权问题. 地理学报, 2015, 70(12): 1973-1986.社会公正是当前地理学的一个热点也是传统研究课题.住房问题往往是社会公正研究和实践的重点案例.中国住房制度改革之后,中国住房拥有率及人均住房面积均大幅度提高.然而在住房分配的过程中,不平等和不公正现象时有发生.如何区别平均和公正,如何看待中国住房改革中的社会公平与不公共存的现象,是了解中国新城市现象的一个重要途径.本文从近期西方政治学和地理学关于住房权(housing rights)、社会公正以及城市权利等文献的讨论出发,通过在广州的实证研究,运用结构方程模型和Pratt指数对涉及住房权主要方面的住房分异的发生途径及其公正性进行综合分析.实证研究说明在住房分配领域,存在着制度因素延续下来所导致的分异.一些制度因素已经转化为市场因素,直接或间接影响个人居住条件和住房所有权的获得,并通过后者进一步影响住房分异.与此同时,纯粹市场因素的影响则相对薄弱.上述结论有助于了解当前中国城市住房权的核心问题,认识社会公正的发展前景,以及探寻减少社会不公正的制度利器.

DOI

[Chen Guo.Housing rights in post-reform urban China: A case study of housing differentiation and justice in Guangzhou, China. Acta Geographica Sinica, 2015, 70(12): 1973-1986.]

[26]
Jiang S, Wang J, Lambert E.Correlates of informal social control in Guangzhou, China neighborhoods. Journal of Criminal Justice, 2010, 38(4): 460-469.Informal social control is a central part of contemporary social disorganization theory. While there are a few studies examining how community factors affect informal social control, they are limited to Western societies, particularly the United States. This study examined whether correlates of informal social control found in the West are applicable to China, where there is a long history of informal social control. Using data from 1,196 respondents from Guangzhou, China, this study found that residential stability increased informal social control at the neighborhood level, with residents who lived in a more stable neighborhood being more willing to take action to prevent deviant behavior. Furthermore, satisfaction with police was found to be related to informal social control at the neighborhood level.

DOI

[27]
杨刚斌, 柳林, 何深静, 等. 基于多层次模型的广州门禁小区家庭入室盗窃环境影响分析. 热带地理, 2016, 36(4): 610-618.lt;p>基于环境设计预防犯罪理论（CPTED）和社会失序理论，利用2012年10―12月在广州门禁小区进行的大规模入户调查数据，采用多层次回归模型从社区内部环境角度对广州门禁小区家庭的入室盗窃受害情况及其发生的环境机理进行研究。通过5个多层次回归模型的对比分析发现：家庭被盗情况主要受到家庭的住房特征和人口特征的影响，小区变量对家庭变量有非常重要的调节作用。住房特征、公共空间使用率、家庭人口特征和小区的正式控制机构失效度对家庭是否被盗有着非常显著的影响；环境形象和人口流动性、小区开放度和功能混合度通过对家庭特征的间接影响而影响家庭的被盗情况。</p>

DOI

[Yang Gangbin, Liu Lin, He Shenjing, et al.Environmental impacts on burglary victimization in gated communities: A multi-level analysis in Guangzhou. Tropical Geography, 2016, 36(4): 610-618.]

[28]
古杰, 周素红, 闫小培. 基于家庭的广州市入室盗窃的特征分析. 人文地理, 2016, 31(5): 87-92.家庭是社会的细胞,以犯罪地理学的理论和方法作为依据,从家庭的视角研究入室盗窃的特征及其影响因素。选取广州市18个社区作为案例,以1332个家庭的问卷调查作为数据基础,分别分析不同类型家庭遭遇入室盗窃的频次特征。结果表明：不同类型家庭在遭遇入室盗窃的频次方面存在明显的差异,邻里之间是否愿意帮忙、是否是本地家庭、居住条件和家庭收入等因素都会对入室盗窃发生的频次产生影响。在此基础上把影响入室盗窃的影响因素划分为根本因素和非根本因素两大类,并分析了两类因素对入室盗窃的影响特征。

[Gu Jie, Zhou Suhong, Yan Xiaopei.Research on characteristics of urban burglary based on family in Guangzhou. Human Geography, 2016, 31(5): 87-92.]

[29]
杨刚斌, 柳林, 何深静, 等.广州门禁小区入室盗窃受害率与内部环境分析. 人文地理. 2016, 31(3): 45-51.基于CPTED（crime prevention through environmental design）和社会失序（social disorganization）理论,本文利用2012年在广州进行的入户调查数据,从内部环境角度对广州门禁小区入室盗窃受害情况进行研究。通过描述性分析和回归分析,本文发现小区的社会环境和建成环境都对小区入室盗窃受害情况都有影响,小区正式控制机制、小区开放度和功能混合度对小区入室盗窃受害率有非常显著的影响,小区环境形象和流动性对小区入室盗窃受害率有一定的影响但是并不显著。通过实证分析我们认为CPTED和社会失序理论的监视性、小区组织结构、出入控制等概念对门禁小区入室盗窃受害情况有非常强的解释力;由于门禁小区的特殊性,以及中国经济社会文化背景差异,流动性、领属感、异质性等思想未体现出与理论假设一致的结论。

[Yang Gangbin, Liu Lin, He Shenjing, et al.Community environmental factor analysis of burglary victimization in gated community in Guangzhou. Human Geography, 2016, 31(3): 45-51.]

[30]
陈刚, 李树, 陈屹立. 人口流动对犯罪率的影响研究. 中国人口科学, 2009, (4): 52-61.

[Chen Gang, Li Shu, Chen Yili.Population mobility and crime: An empirical analysis based on China's observation. Chinese Journal of Population Science, 2009, (4): 52-61.]

[31]
龙冬平, 柳林, 冯嘉欣, 等. 社区环境对入室盗窃和室外盗窃影响的对比分析: 以ZG市ZH半岛为例. 地理学报, 2017, 72(2): 341-355.社区环境的基本属性是影响城市犯罪的重要因素,而这两类犯罪相似的空间形态受社区环境哪些因素的影响及作用方式如何未探寻及验证.因此,本文结合日常活动理论、CPTED理论、可防卫空间等理论,以中国东南沿海地区ZG市ZH半岛为例,融合多源异构时空数据,利用偏最小二乘法,检验社区尺度下的建成环境和社会环境,对入室盗窃和室外盗窃的影响进行对比分析.研究发现:①ZH半岛内具有一定的规律性特征,即在建成环境背景下,入室盗窃的影响因素相对集中,而室外盗窃的影响因素较为分散;②社会环境和建成环境对两类犯罪的影响及作用力存在着相对的共性和差异性,其共性主要表现在人口密度(-)、青少年人口比重(+)、公交站点密度(-)以及到主中心的距离(-)对入室盗窃和室外盗窃的发生都有显著的影响,且作用力方向一致;差异性主要体现在道路网络密度(-)、零售商业密度(+)和餐饮住宿密度(+)是影响室外盗窃发生的主要因子,而居民文化程度(-)显著影响入室盗窃的发生.研究结果的引申,可为社区层面的入室盗窃和室外盗窃的联合防控提供一些参考.

DOI

[Long Dongping, Liu Lin, Feng Jiaxin, et al.Comparisons of the community environment effects on burglary and outdoor-theft: A case study of ZH peninsula in ZG city. Acta Geographica Sinica, 2017, 72(2): 341-355.]

[32]
宋广文, 肖露子, 周素红, 等.居民日常活动对扒窃警情时空格局的影响. 地理学报. 2017, 72(2): 356-367.扒窃现象与居民的日常活动有较强的关系,已有文献研究了不同活动场所对扒窃的影响,但鲜有考虑不同类型居民活动出行量及其场所对扒窃影响的时间差异。本文基于日常活动理论,以南方某大城市为例,以派出所为分析单元,利用居民入户调查问卷、兴趣点(Point of Interest,POI)、扒窃报警数据,检验不同类型活动出行量与扒窃的时间相关性,并划分时间段,建立多个空间滞后负二项回归模型分析不同时段居民日常活动对应场所与扒窃的关系。研究发现,扒窃警情存在明显的时空集聚性,其中,凌晨和早上的集聚性最强,白天和晚上相对较弱;不同类型的日常活动对扒窃时空格局的作用存在差异。买菜活动出行量与扒窃的时间变化存在正相关关系,买菜场所在凌晨和早上容易吸引扒窃者;购物活动与扒窃亦存在较强的时间正相关,白天和晚上在购物场所附近被盗的风险较高;外出吃饭的出行量与扒窃行为无直接关系,但在凌晨和早上餐馆周边容易吸引扒窃者;娱乐活动与扒窃同样存在较强的时间正相关,不同类型的娱乐场所对扒窃的影响存在差异,电影院、酒吧在大部分时间段均会吸引犯罪者,凌晨时段KTV附近区域的被盗风险较高。场所对扒窃的影响主要与犯罪机会吸引(犯罪目标的丰富性、暴露性)及场所自身属性有关(是否为犯罪者的集聚地)。文中最后对时间地理学和犯罪地理学结合的可能性、研究的应用意义等方面进行了探讨。

DOI

[Song Guangwen, Xiao Luzi, Zhou Suhong, et al.Impact of residents' routine activities on the spatial-temporal pattern of theft from person. Acta Geographica Sinica, 2017, 72(2): 356-367.]

[33]
Boessen A, Hipp J R.Close-ups and the scale of ecology: Land uses and the geography of social context and crime. Criminology, 2015, 53(3): 399-426.2015 by the American Society of Criminology. Whereas one line of recent neighborhood research has placed an emphasis on zooming into smaller units of analysis such as street blocks, another line of research has suggested that even the meso-area of neighborhoods is too narrow and that the area surrounding the neighborhood is also important. Thus, there is a need to examine the scale at which the social ecology impacts crime. We use data from seven cities from around the year 2000 to test our research questions using multilevel negative binomial regression models (N = 73,010 blocks and 8,231 block groups). Our results suggest that although many neighborhood factors seem to operate on the microscale of blocks, others seem to have a much broader impact. In addition, we find that racially and ethnically homogenous blocks within heterogeneous block groups have the most crime. Our findings also show the strongest results for a multitude of land-use measures and that these measures sharpen some of the associations from social characteristics. Thus, we find that accounting for multiple scales simultaneously is important in ecological studies of crime.

DOI

[34]
Berk R, Macdonald J M.Overdispersion and Poisson regression. Journal of Quantitative Criminology, 2008, 24(3): 269-284.

DOI

[35]
Land K C, Mccall P L, Nagin D S.A comparison of Poisson, negative binomial, and semiparametric mixed Poisson regression models with empirical applications to criminal careers data. Sociological Methods & Research. 1996, 24(4): 387-442.

[36]
Osgood D W.Poisson-based regression analysis of aggregate crime rates. Journal of Quantitative Criminology. 2000, 16(1): 21-43.This article introduces the use of regression models based on the Poissondistribution as a tool for resolving common problems in analyzing aggregatecrime rates. When the population size of an aggregate unit is small relativeto the offense rate, crime rates must be computed from a small number ofoffenses. Such data are ill-suited to least-squares analysis. Poisson-basedregression models of counts of offenses are preferable because they arebuilt on assumptions about error distributions that are consistent withthe nature of event counts. A simple elaboration transforms the Poissonmodel of offense counts to a model of per capita offense rates. Todemonstrate the use and advantages of this method, this article presentsanalyses of juvenile arrest rates for robbery in 264 nonmetropolitancounties in four states. The negative binomial variant of Poisson regressioneffectively resolved difficulties that arise in ordinary least-squaresanalyses.

DOI

[37]
Hipp J R.Block, tract, and levels of aggregation: Neighborhood structure and crime and disorder as a case in point. American Sociological Review, 2007, 72(5): 659-680.This article highlights the importance of considering the proper level of aggregation when estimating neighborhood effects. Using a unique nonrural subsample from a large national survey (the American Housing Survey) at three time points that allows placing respondents in blocks and census tracts, this study tests the appropriate level of aggregation of the structural characteristics hypothesized to affect block-level perceptions of crime and disorder. I find that structural characteristics differ in their effects based on the level of aggregation employed. While the effects of racial/ethnic heterogeneity are fairly robust to the geographical level of aggregation, the stronger effects, when measured at the level of the surrounding census tract, suggest more dispersed networks are important for perceived crime and disorder. In contrast, economic resources only show a localized effect when aggregating to the block-level and differ based on the outcome; higher average income reduces disorder but increases crime, most likely by increasing the number of attractive targets. Additionally, the presence of broken households has a localized effect for social disorder but a more diffuse effect for perceived crime. These findings suggest the need for neighborhood studies of crime rates, as well as the broader neighborhood effects literature, to consider the mechanisms involved when aggregating various structural characteristics.

DOI

[38]
王增利, 刘学军, 陆娟.入室盗窃多尺度地理因子分析. 地理学报. 2017, 72(2): 329-340.

[Wang Zengli, Liu Xuejun, Lu Juan.Multiscale geographic analysis of burglary. Acta Geographica Sinica, 2017, 72(2): 329-340.]

Options
Outlines

/

〈 〉