The purposes of this study are to define the relationship between the spatial configuration and the urban crime on two cities with different spatial layout, to find the difference in the spatial-temporal distribution patterns of the cities, the factor...
The purposes of this study are to define the relationship between the spatial configuration and the urban crime on two cities with different spatial layout, to find the difference in the spatial-temporal distribution patterns of the cities, the factors to explain the city crimes in them and the specific effects on the factors.
For these purposes, the study was carried out as following methods:
First, the study suggested its background, purpose and scope and methods. In addition to them, prior studies were largely divided into three categories under the city crime study: a city crime study based on literature & some analyses about actual states, another one based on the space syntax theory, and the other one based on the geo-statistical analysis. And the trends in existing studies and their indications were drawn out.
Second, in the theoretical review on the city crime and the spatial analysis methodology, the factors causing city crimes and theoretical background related to the their prevention were explained. And, focusing on spatial analysis and space-time analysis relative to this study model, its conceptual definition and main contents were summarized.
Third, the study introduced the data construction for analyzing the result of this study and its process, and set up the analytical variables to be used in the analysis model. The variables were drawn out as 4 categories of the demographical feature, the social-economic feature, the physical-spatial feature, the defence mechanism and the approachment, which include main variables used prior studies and newly added ones in this study. And this study suggested the conceptual frame of the Integrated Model of Spatial Crime Analysis to be applied to the case analysis.
Fourth, this study found the spatio-temporal crime distribution and influencing factors on the crimes respectively in two case cities (City Changwon, City Masan).
The specific analysis methods by the Integrated Model of Spatial Crime Analysis and the results are as follows:
1. General state of two cities and their urban structures were briefly researched, summarizing their features, respectively.
2. As a basic analysis on the crime occurrence, the study searched general crime occurrence and the crime rate according to population per the administrative district. Then the study analyzed the crime densities on the basis of the area and the land usage pattern, separately. To know the relationship between the crime rate and the land usage pattern in using the aggregated data per an administrative district, the multi regression analysis was carried out.
3. For analyzing the relationship of the crime occurrence and the urban configuration, the study performed the space syntax analysis to express spatial layout with some quantity, and the spatial syntax variables from the analysis was used in the correlation analysis and the multi regression analysis to specify its impact on the city crime. And the study also did the crime-hotspot analysis on the basis of the axis space.
4. To find the spatio-temporal distribution of the crime occurrence, the study performed the standard deviational ellipse analysis which can grasp general(whole) feature in the crime distribution, and found the features in the crime distribution per time, day, season and the crime type by using Kernel density analysis and the nearest neighbor index analysis.
5. For finding the hotspot area in terms of local level, the study carried out the k-means clustering analysis and nearest neighbor hierarchical clustering analysis and the space-time permutation analysis by the space-time scan statistics analysis. The result from these analysis verified if the criminal hotspot area drawn out from the general spatial based clustering analysis coincides the criminal hotspot area based on space-time in spatial-temporal term. Contrary to the hotspot analysis based on the prior point event, the spatial autocorrelation analysis based on the polygon data proved the spatial effects about the existence of the spatial interaction between near areas in the crime occurrence.
6. The study covered the demographic variable, the social-economical variable, the defence mechanism variable, approachment variable and physical environment variable, and performed both the general regression analysis and the spatial auto-regression analysis so as to find the causal relationship of factors influencing on the crime rate and the spatio-temporal distribution features in crime occurrence. Compared to the general regression analysis, the spatial regression analysis has the advantage to include the spatial effects between near areas as the concept of the weighted value in the analytical model and can produce more reliable analytical result.
Fifth, the study compared some similarities and differences in terms of the crime distribution features and the influential factors on the crimes based on the studies of the two case cities, and suggested some political alternatives for preventing the city crimes.
Finally, Summarizing the whole research, the study explored the indications and meanings and the limits, and suggested the limits and future task.
The results from this study procedure and political alternatives are as follows:
First, by objectively quantifying the spatial constructions of two different cities by using the space syntax analysis model, their relationships with the crimes could be analyzed per the unit space. As an example, in case of City Masan, it was appeared that the discordance between the morphological-centering space and functional-centering space worsens the decline phenomenon in urban center, and its city crime distribution pattern follows this pattern that is, the crime density was appeared more focused on the sub-centers, not on the downtown of City Masan. And it was analyzed that the hotspot areas shown in the commercial & business district were Hapseongdong, not Dongseodong and Odongdong, and Hapseongdong had stronger clustering crimes and crime transition. This phenomenon was the result from for the downtown not acquiring the spatial topology as centering space in a city as well as the decline as downtown function. Therefore it is judged that some urban regeneration projects should be actively proceed for Masan's sustainable growth & development and its crime prevention.
On the other hand, concerning on the crime preventative activities (patrol action, etc.) and the establishment of CCTVs, a new local allocation strategy through the space syntax analysis will be more rational than the existing local allocation strategy based on point events.
Second, the study identified that the crime occurrence not only appears repetitive pattern change between the dispersion and the cluster with time-series according to the citizen's life cycle and the crime type but also has the clustered distribution feature in space. And the study found the space-time cluster area of crime by using the space-time statistics analysis, which haven't almost tried to do in existing researches related to urban crime in Korea, and identified some differences from the criminal hot spots considering only space. These spatio-temporal clustering areas can be considered as the most primary places for police administration's quicker response to many crimes, more effective human resources allocation and stronger activities for public security. And they are areas with priority in applying to CPTED (Crime Prevention Through Environmental Design) in the viewpoint of city planning, and the ones absolutely needing to the planned site about the crime-induced facilities and a systematical management. Therefore the crime prevention activities will be developed as the differentiated responding strategy reflecting on this crime distribution feature.
Finally, in case of City Changwon, the variables associated with a physical environment as influencing factors to explain the crime occurrence in the city affects strongly on the city crime rate. But it was identified that more various factors influencing on crimes in City Masan could explain about the crime occurrence. And, despite the degree difference, regardless of naturally created city or planned city, the physical environmental variable was analyzed to have close relationship with the city crime rate. And this fact supports that CPTED approach and the environmental criminology theory to prevent some city crimes through designing the city and architecture environment were still effective.