Understanding Crime and Community: Feasibility Proposal for A National Neighborhood Crime Study

本项目研究从美国各地的警察部门获取社区一级（区域）犯罪数据的可行性。目前，在社区一级还没有大规模的代表性犯罪数据样本。如果可行，调查人员将开发一个更大的项目，收集2000年美国多个城市（最多110个）的犯罪数据，并将这些犯罪数据与来自美国人口普查和其他公开数据来源的额外信息结合起来。全国社区犯罪项目的一般目的是更系统地研究不同类型城市内的社区结构如何与不同种族/民族和阶级组成的社区的犯罪联系起来。先前的研究人员已经探索了城市内社区犯罪模式的预测因素，但由于缺乏全国城市社区层面的现成犯罪数据，我们对社区犯罪变化的理解有限。这种数据的缺乏使得无法解决以下问题：(1)由于研究只考察了一个或少数环境中的社区，因此以前关于社区犯罪模式的研究结果的概括性；(2)不同类型城市所存在的条件是否影响邻里特征与犯罪之间的关系；(3)当地种族和阶级构成对犯罪率的单独影响。鉴于我们对社区犯罪的理解存在这些缺陷，本研究的目标是为美国各地在社会人口和经济结构（包括种族和阶级组成）方面差异很大的城市样本汇编区域级犯罪数据，并将这些数据与同一时期的一组一致的社区、城市和大都市社会人口和经济变量联系起来。然后，这些数据将用于解决上述问题和有关的实质性问题。为了评估获取必要的社区犯罪数据的可行性，调查人员将：(1)与要求这些数据的城市警察部门随机抽样联系；(2)利用所建立的联系获取有关现有数据的形式和性质的信息；(3)对多个部门进行实地考察，了解不同类型城市的部门所面临的制约因素，并找出合适的策略来获得不同部门的合作；(4)通过尝试从实际获得的数据中编码和构建犯罪变量，并通过将基于地址的犯罪数据地理编码到人口普查区，评估在尝试使用来自不同部门的数据时将遇到的问题的种类和程度。

This project studies the feasibility of obtaining neighborhood-level (tract) crime data from police departments across the U.S. Currently, there is no large representative sample of crime data available at the community level. If feasible, the investigators will develop a larger project to collect such crime data for multiple U. S. cities (up to 110) for year 2000, and combine these crime data with additional information from the U.S. Census and other publicly available data sources. The general purpose of a national neighborhood crime project is to more systematically examine how community structures within different types of cities are linked to crime in neighborhoods of varying racial/ethnic and class composition. Prior researchers have explored the predictors of neighborhood crime patterns within cities, but our understanding of neighborhood variation in crime is limited due to the lack of readily available crime data at the neighborhood level for cities across the nation. This lack of data has made it impossible to address issues of: (1) the generalizability of previous findings regarding neighborhood crime patterns since studies only examine neighborhoods in one or a small number of settings; (2) whether the conditions present in different types of cities influence the relationships between neighborhood characteristics and crime; and (3) the separate effects of racial and class composition of local areas on crime rates. In light of these shortcomings in our understanding of neighborhood crime, the goal of this research is to compile tract-level crime data for a large sample of cities throughout the U.S. that vary substantially in sociodemographic and economic structure, including race and class composition, and to link these data with a consistent set of neighborhood, city, and metropolitan sociodemographic and economic variables for the same time period. These data will then be used to address the above questions and related substantive issues. To assess the feasibility of acquiring the requisite data on neighborhood crime, the investigators will: (1) make contact with random samples of city police departments requesting such data; (2) use the contacts made to gain information on the form and nature of available data; (3) make site visits to several departments to gain insights into the constraints departments in different types of cities face, and to identify appropriate strategies for gaining cooperation from different departments; and (4) assess the kinds and degrees of problems that will be encountered in trying to use data from different departments by attempting to code and construct crime variables from data actually acquired, and by having the address-based crime data geocoded to census tracts.