spatio-statistical analysis of urban crime; a case study...

12
American Journal of Environmental Policy and Management 2018; 4(1): 9-20 http://www.aascit.org/journal/ajepm ISSN: 2472-971X (Print); ISSN: 2472-9728 (Online) Keywords Spatio-Statistical, Urban Crime Management, GSCMS, GIS Received: July 30, 2017 Accepted: November 23, 2017 Published: January 11, 2018 Spatio-Statistical Analysis of Urban Crime; A Case Study of Developing Country, Kaduna Metropolis, Nigeria Nzelibe Tobenna Nnaemeka * , Bello Ashiru Department of Urban and Regional Planning, Ahmadu Bello University, Zaria, Nigeria Email address [email protected] (N. T. Nnaemeka) * Corresponding author Citation Nzelibe Tobenna Nnaemeka, Bello Ashiru. Spatio-Statistical Analysis of Urban Crime; A Case Study of Developing Country, Kaduna Metropolis, Nigeria. American Journal of Environmental Policy and Management. Vol. 4, No. 1, 2018, pp. 9-20. Abstract Geographical Information Systems (GIS) and statistical methods are implored in developing a Geospatial Crime Management System (GSCMS) to facilitate for an in- depth crime pattern analysis. Exploring the spatial and statistical dimensions of crime is optimized by the GIS capabilities to relate spatial and attribute data effectively. While the statistical analysis techniques facilitates for handling of the computation and processing aspects of such attribute data for a cogent database. However, the effectiveness of these quantitative approaches is been proliferated. This paper examines the spatial pattern of urban crime in Kaduna Metropolis (KdM), Nigeria; using computer based spatio- statistical techniques of crime analysis with a view to offering possible options, for effective urban crime management. The analysis highlights harnessing innovative capabilities for an effective urban crime management system in cities of the developing countries. 1. Introduction The rate of crime and its implications on urban areas have led to the development of the emerging urban crime management as a major issue in urban studies. The realization that crime could be explained and understood in more depth surfaced in the 1970s. Recently emerged, are the capabilities to identify patterns and concentrations of crimes; the relationships between crime and environmental characteristics as well as assessing the effectiveness of law enforcement agencies and crime reduction programs on urban crime, for a more informed and proactive urban crime management and decision making on urban crime policing, policy formulation, evaluation, and reform. The study of crime has, not surprisingly, been dominated by research in criminology, sociology and law, but spatial and ecological perspectives on crime by mostly criminologists, did preceded the ‘recent past’ after which professional geographers entered the crime research arena. The results of the questions are lightning how the environmental factors provide the opportunities for crime. There are three crime theories in environmental criminology that have interest in easy and appealing opportunities which are driving people to crime [2]: Rational choice theory, Crime pattern theory, Routine activities theory. This feature of crime events distribution was described as an ‘inherent geographical quality’ by [4] and was explained by theories such as the ecology of crime [3] or routine activities [5], amongst others. A more eloquent identification of patterns and concentrations of crime emphasize the spatial dimentions of crime.

Upload: others

Post on 07-Mar-2021

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

American Journal of Environmental Policy and Management

2018; 4(1): 9-20

http://www.aascit.org/journal/ajepm

ISSN: 2472-971X (Print); ISSN: 2472-9728 (Online)

Keywords Spatio-Statistical,

Urban Crime Management,

GSCMS,

GIS

Received: July 30, 2017

Accepted: November 23, 2017

Published: January 11, 2018

Spatio-Statistical Analysis of Urban Crime; A Case Study of Developing Country, Kaduna Metropolis, Nigeria

Nzelibe Tobenna Nnaemeka*, Bello Ashiru

Department of Urban and Regional Planning, Ahmadu Bello University, Zaria, Nigeria

Email address [email protected] (N. T. Nnaemeka) *Corresponding author

Citation Nzelibe Tobenna Nnaemeka, Bello Ashiru. Spatio-Statistical Analysis of Urban Crime; A Case

Study of Developing Country, Kaduna Metropolis, Nigeria. American Journal of Environmental

Policy and Management. Vol. 4, No. 1, 2018, pp. 9-20.

Abstract Geographical Information Systems (GIS) and statistical methods are implored in

developing a Geospatial Crime Management System (GSCMS) to facilitate for an in-

depth crime pattern analysis. Exploring the spatial and statistical dimensions of crime is

optimized by the GIS capabilities to relate spatial and attribute data effectively. While the

statistical analysis techniques facilitates for handling of the computation and processing

aspects of such attribute data for a cogent database. However, the effectiveness of these

quantitative approaches is been proliferated. This paper examines the spatial pattern of

urban crime in Kaduna Metropolis (KdM), Nigeria; using computer based spatio-

statistical techniques of crime analysis with a view to offering possible options, for

effective urban crime management. The analysis highlights harnessing innovative

capabilities for an effective urban crime management system in cities of the developing

countries.

1. Introduction

The rate of crime and its implications on urban areas have led to the development of

the emerging urban crime management as a major issue in urban studies. The realization

that crime could be explained and understood in more depth surfaced in the 1970s.

Recently emerged, are the capabilities to identify patterns and concentrations of crimes;

the relationships between crime and environmental characteristics as well as assessing

the effectiveness of law enforcement agencies and crime reduction programs on urban

crime, for a more informed and proactive urban crime management and decision making

on urban crime policing, policy formulation, evaluation, and reform.

The study of crime has, not surprisingly, been dominated by research in criminology,

sociology and law, but spatial and ecological perspectives on crime by mostly

criminologists, did preceded the ‘recent past’ after which professional geographers

entered the crime research arena. The results of the questions are lightning how the

environmental factors provide the opportunities for crime. There are three crime theories

in environmental criminology that have interest in easy and appealing opportunities

which are driving people to crime [2]: Rational choice theory, Crime pattern theory,

Routine activities theory.

This feature of crime events distribution was described as an ‘inherent geographical

quality’ by [4] and was explained by theories such as the ecology of crime [3] or routine

activities [5], amongst others. A more eloquent identification of patterns and

concentrations of crime emphasize the spatial dimentions of crime.

Page 2: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

10 Nzelibe Tobenna Nnaemeka and Bello Ashiru: Spatio-Statistical Analysis of Urban Crime; A Case Study of

Developing Country, Kaduna Metropolis, Nigeria

Complete, consistent, and reliable sets of crime data, up-

to-date information, skilled staff, appropriate geographical

information systems (GIS) background and related statistical

software are requirements to utilize the high technology

advances in crime. Data is always one of the most important

parts of the analysis. Sufficient, clear and utilizable data is

the result of severely, carefully collected and manipulated

data. Geographical information system (GIS) is a computer

based technology that requires being applied by professional

to obtain satisfactory results [8].

Crime mapping has long been a subset of the process today

known as crime analysis. Before the development of

computerized crime mapping, incidents were represented by

traditional maps with pins stuck in it. Since the traditional pin

maps have serious limitations, crime maps are now supported

with computer technology. Continual development in

computer technology innovated geographical information

systems (GIS) for the studies where the geography should be

concerned. Many industries and organizations are the users of

GIS. Crime maps started to be created with GIS to archive,

manipulate and query the crime data; to update crime

patterns; to make spatial analysis and to develop crime

pattern prediction and prevention models.

Crime mapping plays an important role in proactive

policing and crime prevention in stages of data collection,

data evaluation and data analysis. The application areas of

crime mapping are recording and mapping crime activities,

predicting crime, identifying crime hot spots and patterns,

monitoring the impact of crime reduction measures and

communicating with stakeholders [4]. Although a wide

variety of statistical and analytic techniques exist to examine

crime problems, analysts are increasingly using geographical

information systems (GIS) and mapping software to identify

areas of crime concentration. Spatial crime analysis softwares

are capable of data entry, data manipulation, pattern

identification, clustering, data mining, and geographic

profiling. Analyzing crime with related softwares is complex

but an advanced and reliable method to reach satisfactory

results [2].

Spatial patterns are identified to detect hot spots that are

explained by [4], as a geographical area of higher than

average crime. It is an area where crime incidents densely

populated compared to the average. Researchers and crime

analysts are concerned with identifying a crime hot spot

reliably and objectively. Spatial statistical techniques with

geographic information systems are combined to detect real

crime hot spots. Hot spot analysis using crime mapping is

classified into five general techniques: [14]. Which are Visual

interpretation, Crime areas as a hot spot, Choropleth

mapping, Grid-cell analysis, point pattern or cluster analysis,

spatial autocorrelation

Although, factors that explain crime have been widely

studied, its dynamics and implications on the management

of cities remained cumbersome to determine and quite

subjective. This is because of issues surrounding readily

available useful database in formats pre-developed for

crime analysis. The traditional and age-old system of

intelligence and criminal record maintenance has failed to

live up to the requirements of the existing crime scenario.

The solution to this ever-increasing problem lies in the

effective use of Information Technology. GIS can be used to

plan effectively for emergency response, determine

mitigation priorities, analyse historical events, and predict

future events. Response capabilities of GIS often rely on a

variety of data from multiple agencies and sources. The

ability to access and process information quickly while

displaying it in a spatial and visual medium allows agencies

to allocate resources quickly and more effectively. GIS

software helps coordinate vast amounts of location-based

data from multiple sources. It enables the user to create

layers for the data and view the data most critical to the

particular issue or mission [9].

The recognition of the lack of GSCMS in Kdm instigated

developing a standard model that enables rapid appraisal of

crime levels over multiple spatial scales, for developing

countries to cater for subjectivity surrounding locational

attributes and policy adequacy, using Kdm as a case in this

research. The selected precinct being centrally placed in

north western Nigeria, serves as a confluence point, while its

industrial nature has attracted a massive influx of population

interested in economic strive, estimated to be about

1,769,032 in 2014. Evidently the level of reported crime in

KdM, from the annual crime records in locations such as

Rigasa, ungwan sanusi, and narayi amongst others is

alarming [12].

Despite efforts for a more effective urban crime

management in developing countries there is still a lack in

develop stringent modalities to curb the dynamic crime

situation which is due to lack of a GeoSpatial Crime

Management System (GSCMS) with the potentials to process

and build the usual voluminous data associated with crime

into a systematic database with spatial attributes, that can be

adopted for measuring the direct implication of urban crime

and its management effectively. The GSCMS development as

a robust system that enables rapid appraisal of urban crime

occurrences over multiple spatial scales, will facilitate for a

better understanding of urban crime pattern and their spatial

tendencies. It will provide readily available useful database

with high precision and accuracy in formats pre-developed

for crime analysis, to enable envisioning and modeling urban

crime and management scenario, in a manner that is easily

visualized and understood for an informed decision making

on urban crime management policies. Most cities with

efficient crime management system such as Cape Town have

designed an integrated tool of crime analysis for combating

crime more proactively by creating a better understanding of

the patterns of crime. It was established that a large

proportion of the men of the Nigerian Police Force can

hardly ascertain the areas under the jurisdiction of their

stations or define the shortest route from their station) to

specific crime areas. He concluded that the police stations in

Ikeja LGA are far from being distributed according to

Page 3: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

American Journal of Environmental Policy and Management 2018; 4(1): 9-20 11

geographical spread, population characteristics or crime

incidence. Currently, GIS is not being used for crime control

and management in Nigeria. This is probably due to the lack

awareness of the benefits offered by GIS in crime control and

management in the country.

Applications of GIS to crime mapping and management

have been successful in many developed countries.

Information associated with crime in Lima and Columbus

(Ohio) for instance was acquired and integrated in a GIS

environment [11]. Analysis in Lima has spanned crime from

1999 to the present. As a result, the work informed policy

and decision making in Lima Police Department activities.

Also [15] has advised throughout crime analysis that

patterns are identified and relationships of crime and law

enforcement information with different sorts of info are

studied. Such information embodies Socio-demographic

and Spatial (location). [13] has proposed effective crime

analysis employs data mining, crime mapping, statistics,

fresh methods, charting and a solid understanding of

criminal behavior. In this sense, a crime analyst serves as a

combination of an information systems specialist, a

statistician, a researcher and a planner for law enforcement

agency. [7] has proposed that the use the Location

Quotients of crime (LQC) is a method of mapping the

prevalent types of crime across urban neighborhood and

adopting crime density to investigate the relationships

between neighborhood characteristics and crime. Their

research used three aspects (1) the rationale of the

employment of official crime rates for neighborhood. (2)

An enhanced understanding of the effectiveness and

statistical properties and (3) Models of crime measured by

density, crime rates and LQCS.

The conventional tools for the mapping and analysis of

urban crime are based on manual methods that are slow,

tedious and expensive. The process is also dependent on

qualitative descriptions and subjective analysis.

Advancement in technology such as the computer

technology; spatial data visualization in geographic

information systems and an enhanced spatial statistical

analysis techniques, has largely been adopted in this research

as a modern techniques in crime analysis in the field of

environmental criminology.

In reacting to what the magnitude and typology of crime in

Kaduna metropolis is, as well what is their spatial pattern

take up in recent times and implications? For the purpose of

this paper, the stipulated intent is limited to examining the

existing strategies for urban crime management; Process

retrospective crime data on Kaduna metropolis, using

computer based spatio-statistical technics; and as well

analyse and identify the spatial pattern and implications of

identified crime pattern in Kaduna metropolis. This is as to

recommend adequate measures for managing urban crime in

the frame of urban management.

2. Study Area

Kaduna’s urban centre is one of the largest cities in

northern Nigeria, created by the colonial government. It is

located on latitude 10° 30’N and longitude 7° 28’E with

height of about 600-650m above mean sea level. Spatially, it

spans to about 7.7 hectares, from Katabu in the north to the

oil refinery in the south. Figure 1 below shows kaduna

metropolis in context.

The Google earth imagery streaming, maintaining 1 metre

altitude provided for a birdeye view of the study area, using

the Google earth pro 4.2 package. This served as a means to

enhancing the basemap. Figure 2 is the Google earth image

as downloaded from the Google earth software.

Basemap of Kaduna Metropolis was collected from the

Kaduna State Urban Planning and Development Authority

(KASUPDA). It provided a platform, on which crime data

and other related datasets were overlaid. It equally provided

insight on administrative boundaries for delineating the study

area (Kaduna Metropolis) as well as the 34 administrative

wards as defined at the time of the 2006 Census, and road

layer amongst others. The basemap was collected in a hard

copy format, thus, it was scanned, georeferenced and

vectorised in the ArcGIS environment. Ground truthing was

also carried out in order to enhance the information provided

by the basemap.

The demographic data from the National Population

Commission projection from 2006 census puts the population

of Kaduna Metropolis at 2014 to about 1,769,032. The

population were derived at ward level to serve as derivative

in computing ‘population Density’, and equally facilitated

computing for crime intensities in Kaduna metropolis.

Page 4: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

12 Nzelibe Tobenna Nnaemeka and Bello Ashiru: Spatio-Statistical Analysis of Urban Crime; A Case Study of

Developing Country, Kaduna Metropolis, Nigeria

Figure 1. Kaduna metropolis in context of Nigeria [10].

Page 5: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

American Journal of Environmental Policy and Management 2018; 4(1): 9-20 13

Figure 2. Google Image of Kaduna Metropolis [6].

Page 6: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

14 Nzelibe Tobenna Nnaemeka and Bello Ashiru: Spatio-Statistical Analysis of Urban Crime; A Case Study of

Developing Country, Kaduna Metropolis, Nigeria

3. Materials and Methods

3.1. Method of Data Acquisition

Crime data used in this research were obtained from

Kaduna Police Command. Crime data was collected free of

charge from Kaduna Police Command, through the Kaduna

Police commissioner by submitting an open record request.

Acquiring the crime data through an open record request

results in a more complete and accurate dataset. The crime

dataset includes all reported crimes classified according to

the Nigerian Police (NP) Uniform Crime Reporting (UCR)

program. This research examines seven of category A Crimes

which include murder and non-negligent manslaughter,

forcible rape, robbery, aggravated assault, burglary, larceny-

theft and auto theft. Only Category A Crimes are included in

this research because these crimes are taken as more serious

than others in crime analysis and the data sources are more

reliable. The police are usually on the scene or visit the scene

to record these types of crimes in most cases, as obtained

from Nigerian Police, (2012) Report. Table 1 shows the UCR

codes for the seven Category A Crimes adopted in

classification of offenses in this research.

Table 1. UCR classification offenses codes and Impact Factor for Category A Crimes.

Category A Crimes (Category A crime, included in the Crime Index.)

Code Violent Crimes Percentage Impact

0 Murder And Non-negligent Manslaughter 25

2 Forcible Rape 14

3 Robbery 21

4 Aggravated Assault (Class I) 7

Non-Violent Crimes

5 Burglary 4

6 Larceny – Theft (Includes Burglary of Motor Vehicles) 11

7 Auto Theft 18

Total 100

[12]

In addition to Offenses type by classification the dataset

includes the offense date and time, police beat, and address,

where the offense took place. A complete set of crime data

for the selected seven Category A Crimes were obtained for

Three fiscal years from records of the Kaduna Metropolis

Police jurisdictional division (PJD) for the study period

(July, 2011 - June, 2014) were used in this research, The

crime data were compiled in a Microsoft Excel format and

limited to those crime Incidence recorded by the police,

ranging from: July 2011 – June 2012, July 2012 – June

2013, July 2013 – June 2011. The impacts of the various

category A crime types on the other hand were factored into

determine crime weight so as to facilitate conducting an

unbiased analysis

3.2. Method of Data Processing

In the collection of data for this research, data mining

methodology was adopted. This enabled crime analysis in

multidimensional space and allows the integration of

methods from other disciplines by considering the

semantic ties among data objects. This will enable the

searching of interesting patterns among multiple

combinations of dimensions at varying levels of

abstraction. The data from the crime incidence in this

article was being entered into an Oracle 10g relational

database and then analyzed using statistical techniques,

and ArchGIS. The data from the case study will be

managed using Oracle 10g, which allows for unified

integration into MATLAB and ArchGIS. The Oracle 10g

platform allows for efficient data management and

analysis through its unique architecture system. This

allows integration of the spatial data into a GIS type

environment contained completely within the database

software. The software will be used to store spatial data in

tabular form to allow output queries based on a

determined field to other software for graphic display.

Geocoding was performed with function to geocode to

the next level of precision, The ability to assign confidence

codes that describe the accuracy and precision of

geographic coordinates that have been determined for each

crime record was a function included in the geocoding

process. Each incidence with a valid known address will be

assigned a latitude/longitude coordinate on the WGS84

datum coded using the Administrative blocks and street

database. The geocoding process will assigns an “x and y”

value for each spatial point representing an incident. This

“x and y” value will then be placed on a map using a

predetermined coordinate in a specified unit. These features

will then be displayed graphically on a base map. Once, the

data is geocoded, it is migrated back to the Oracle 10g

database and joined with associated address records. Once

brought into ArchGIS environment, the files will then be

displayed in a variety of formats. The base map will then be

used to overlay layers that contained crime information.

This will allow for the presentation of the data making it

possible to recognize the location where an incident of

crime took place. This processes facilitated performing

Page 7: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

American Journal of Environmental Policy and Management 2018; 4(1): 9-20 15

spatial queries on the crime data. The queries will be

performed using a standard database language called

Structured Query Language (SQL) that builds logical

expressions to select data of interest. The types of queries

that will be performed in this research include Selecting by

attribute information, grouping by attribute information and

Selecting by geographic area

3.3. Method of Application of the GSCMS in

Crime Analysis

GSCMS will automate crime data into a unified platform

that permits for manipulating spatial data in various scales. It

will equally facilitate to describe and visualize spatial

distributions; identify typical locations or spatial outliers;

discover patterns of spatial associations, clusters, or hot

spots; and suggest spatial regimes or other forms of spatial

heterogeneity. GSCMS will be used in the analysis of crime

patterns in spatial variance in KdM. Through the use of

GSCMS, distribution of crime incidents across KdM will

yield vital information regarding concentrations of crime

spatially.

GSCMS is a robust crime management and analysis tool

that served as a unified medium, with capability to

automate crime data collected on varying spatial scales,

quantify crime and combine the various factors of crime,

into spatial crime pattern analysis. Its ability to manipulate

spatial data in one medium is the second logic in the tool

design. The demand for a unified tool is based on the

limitations of individual existing techniques being limited

to the treatment of one or at most two aspects of a spatial

phenomenon at a particular instance. The use of statistical

techniques for data processing and computational analysis

for instance computes for crime pattern, but always falls

short in visually presenting crime pattern analysis spatially.

This is because it falls to combine other layers as to

facilitate visualizing crime pattern in context. This implies

that it cannot be used in absolute isolation to achieve the

level of the analysis presented in this paper, for decision

makers to easily visualize and project crime implication.

The shortcomings of the techniques of GIS and Remote

Sensing lie with the lack of versed computation capacity to

determine crime indicators such as population in the case of

this research. The GSCMS enables cost effective

acquisition of data to enable crime mapping, pattern

analysis and computation of crime indicators. In such an

environment it is easy to undertake crime pattern analysis

and computation of correlation for the different units of

analysis. The crime Pattern model adapted in this research

is from a component of the model generated [1].

4. Spatio-Statistical Analysis of

Crime in Kaduna Metropolis

This section presents the results and discussion from a

descriptive spatial-statistical stand point, by aggregating the

crime incidence from administrative wards comprising the

identified 14 Police Jurisdictional Division (PJD) within

Kaduna metropolis area command Jurisdictional Division

(KDMACJD).

The presentation of findings for this spatial analysis is as

crime density maps indicating the spatial distributions (SD)

of crime per 10,000 population (C/10,000 pop.). This

facilitated exploring as to identifying for the spatial pattern of

crime in Kaduna Metropolis (KDM). Crime risk prone areas

through the study period were then discussed, as to

evaluating for risk prone locations in space.

All Figures presented in this section were produced from

police records within KDMACJD with elaborate details on

each crime types, author’s fieldwork and related data drawn

from archives of independent agencies within KDM. While

the spatial dimensions was to PJD resolution as delineated by

aggregating Administrative Wards (AW) as defined by The

Nigerian Police (NP).

4.1. Crime Rate in Kaduna Metropolis by

Administrative Words Constituting PJD

As indicated in the map in figure 3 Rigasa has the highest

recorded Crime incidence with 391 (10.46%) report cases,

among the 34 administrative wards constituting Kaduna

metropolis having Crime incidence of 3738 recorded. This is

followed by Ungwan_Sanusi 360 (9.63%), Ungwan_Rimi

263 (7.04%), Kabala 250 (6.69%), Tudun_Nupawa 237

(6.34%), Hayin_Banki 231 (6.18%), Badarawa 179 (4.79%),

Sardauna 176 (4.71%), Barnawa 136 (3.64%), Shaba 132

(3.53%), Ungwan_Shanu 124 (3.32%) and Badiko 114

(3.05%), ranking 2nd to 12th place respectively which are all

located above the Crime mean of 109.94 (2.94%),

constituting 2593 (69.37%). The 22 other wards with the

least Crime incidence below the stated mean constituted

about 1145 (30.63%).

Page 8: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

16 Nzelibe Tobenna Nnaemeka and Bello Ashiru: Spatio-Statistical Analysis of Urban Crime; A Case Study of

Developing Country, Kaduna Metropolis, Nigeria

Figure 3. Crime Intensity in Kaduna Metropolis for the Study Period (July, 2011 – June, 2014) [10].

Page 9: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

American Journal of Environmental Policy and Management 2018; 4(1): 9-20 17

4.2. Crime Intensity in Kaduna Metropolis by

Administrative Wards Constituting PJD

Per 10,000 Pop

This has led to drawn a beneficial deduction such as the

spatial pattern of recorded crime incidence in Kdm varied

from that of its intensity per 10,000 population by location

(ward). As deduced from the findings of this analysis

presented in the map in figure 3 and Table 2 shows Kaduna

metropolis crime intensity for the study period and details on

crime numbers, rates per 10.000 population by wards

constituting PJD’s; Details on recorded crime in Kaduna

metropolis for the study period, indicating that Rigasa has the

highest recorded Crime incidence with (10.46%) report

cases, which is followed by Ungwan_Sanusi (9.63%),

Ungwan_Rimi (7.04%), Kabala (6.69%), Tudun_Nupawa

(6.34%), Hayin_Banki (6.18%), Badarawa (4.79%),

Sardauna (4.71%), Barnawa (3.64%), Shaba (3.53%),

Ungwan_Shanu (3.32%) and Badiko (3.05%), ranking 2nd to

12th place respectively are all located above the local

statistically calculated mean crime in Kdm of (2.94%), and

constituting (69.37%) of the total reported crime in Kdm.

While Sardauna (18.60%), Shaba (15.69%), Ungwan_Sanusi

(8.71%), Kabala (7.93%), Ungwan_Rimi (6.58%),

Ungwan_Sarki (4.76%), Tudun_Nupawa (4.38%), Maiburji

(3.75%), Hayin_Banki (3.41%), are 9 wards identified to be

all located above the local statistically calculated mean crime

in Kdm of (2.94%), constituting (73.81%) of the total

reported crime in Kdm per 10,000 population. This is to say

in relation to population size, crime intensity varies widely

within each ward.

Table 2. Crime Intensity in Kaduna Metropolis per 10,000 pop.

Wards

Aggravated Assault in

Kdm (per 10000 pop)

Auto Theft in Kdm (per

10000 pop)

Burglary in Kdm (per

10000 pop)

Forcible Rape in Kdm (per

10000 pop)

intensity % intensity % intensity % intensity %

Sardauna 240.5154 22.01% 309.2341 24.80% 97.35147 15.60% 180.3865 17.71%

Shaba 149.4586 13.68% 192.161 15.41% 97.60561 15.64% 149.4586 14.68%

Ungwan_Sanusi 48.72107 4.46% 93.96207 7.54% 65.42544 10.48% 77.95371 7.65%

Kabala 58.86779 5.39% 47.80242 3.83% 54.88426 8.79% 99.14575 9.74%

Ungwan_Rimi 94.58769 8.66% 46.01563 3.69% 40.90278 6.55% 97.14411 9.54%

Ungwan_Sarki 72.30782 6.62% 15.49453 1.24% 34.4323 5.52% 48.20522 4.73%

Tudun_Nupawa 50.81733 4.65% 76.6015 6.14% 27.03582 4.33% 56.0743 5.51%

Maiburji 42.39855 3.88% 36.34161 2.92% 40.37957 6.47% 28.2657 2.78%

Hayin_Banki 25.43361 2.33% 72.28501 5.80% 21.41778 3.43% 16.06334 1.58%

Badarawa 29.41235 2.69% 18.00756 1.44% 14.40605 2.31% 25.21059 2.48%

Barnawa 10.49251 0.96% 61.67024 4.95% 15.41756 2.47% 20.98501 2.06%

Sabon_Gari_North 34.67635 3.17% 24.76882 1.99% 7.705856 1.23% 23.11757 2.27%

Dadi_Riba 32.6408 2.99% 9.325942 0.75% 12.43459 1.99% 29.01404 2.85%

Ungwan_Shanu 33.36738 3.05% 40.85802 3.28% 9.07956 1.45% 19.06708 1.87%

Sabon_Gari_South 19.4955 1.78% 14.32323 1.15% 12.73176 2.04% 16.71043 1.64%

Badiko 12.56665 1.15% 22.61997 1.81% 10.77141 1.73% 20.10664 1.97%

Sabon_Gari_West 21.35248 1.95% 12.6707 1.02% 7.508564 1.20% 9.854991 0.97%

Sabon_Gari 18.13505 1.66% 16.65464 1.34% 5.181443 0.83% 10.36289 1.02%

Ungwan_Dosa 7.48543 0.68% 9.624124 0.77% 4.277389 0.69% 14.97086 1.47%

Rigasa 12.31193 1.13% 13.07664 1.05% 6.576558 1.05% 11.77663 1.16%

Tudun_Wada_South 3.664921 0.34% 14.13613 1.13% 4.188482 0.67% 0 0.00%

Kawo 4.615435 0.42% 11.86826 0.95% 1.318696 0.21% 11.53859 1.13%

Mekera 5.77882 0.53% 8.915893 0.72% 6.604365 1.06% 13.86917 1.36%

Narayi 2.535528 0.23% 23.9064 1.92% 3.380703 0.54% 11.83246 1.16%

Kakuri_Hausa 2.282733 0.21% 5.869884 0.47% 6.522094 1.05% 0 0.00%

Tudun_Wada_North 19.05873 1.74% 0 0.00% 0 0.00% 0 0.00%

Nasarawa 10.45044 0.96% 0 0.00% 5.11858 0.82% 0 0.00%

Kakuri_Gwari 2.591777 0.24% 9.996853 0.80% 2.221523 0.36% 2.591777 0.25%

Kakau 3.020008 0.28% 9.707167 0.78% 2.157148 0.35% 3.020008 0.30%

Kujama 7.249979 0.66% 11.18568 0.90% 2.485707 0.40% 8.699975 0.85%

Yelwa 6.188301 0.57% 5.304258 0.43% 0.589362 0.09% 4.125534 0.41%

Sabon_Tasha 3.812784 0.35% 5.602459 0.45% 2.801229 0.45% 0 0.00%

Rido 4.723878 0.43% 6.748397 0.54% 1.199715 0.19% 5.248753 0.52%

Television 1.811735 0.17% 0 0.00% 0 0.00% 3.62347 0.36%

Total 1092.829 100.00% 1246.739 100.00% 624.1134 100.00% 1018.424 100.00%

Average 32.14204 2.94% 36.6688 2.94% 18.35628 2.94% 29.95364 2.94%

Page 10: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

18 Nzelibe Tobenna Nnaemeka and Bello Ashiru: Spatio-Statistical Analysis of Urban Crime; A Case Study of

Developing Country, Kaduna Metropolis, Nigeria

Table 2. Continued.

Wards

Larceny-Theft in Kdm

(per 10000 pop)

Murder in Kdm (per 10000

pop)

Robbery in Kdm (per

10000 pop)

Crime in Kdm (per 10000

pop)

intensity % intensity % intensity % intensity %

Sardauna 1354.331 17.56% 465.2827 21.53% 450.9664 16.08% 3098.067 18.60%

Shaba 939.454 12.18% 381.2719 17.65% 704.5905 25.13% 2614 15.69%

Ungwan_Sanusi 849.8347 11.02% 139.2031 6.44% 175.3959 6.25% 1450.496 8.71%

Kabala 681.6271 8.84% 154.9152 7.17% 223.0779 7.96% 1320.32 7.93%

Ungwan_Rimi 570.4477 7.40% 109.561 5.07% 138.0469 4.92% 1096.706 6.58%

Ungwan_Sarki 407.1619 5.28% 107.6009 4.98% 108.4617 3.87% 793.6645 4.76%

Tudun_Nupawa 305.655 3.96% 118.9076 5.50% 94.62538 3.37% 729.7169 4.38%

Maiburji 299.8183 3.89% 50.47446 2.34% 127.1956 4.54% 624.8738 3.75%

Hayin_Banki 227.1815 2.95% 52.5883 2.43% 152.6017 5.44% 567.5712 3.41%

Badarawa 224.4943 2.91% 40.01681 1.85% 67.22824 2.40% 418.7759 2.51%

Barnawa 150.7495 1.95% 48.17987 2.23% 67.45182 2.41% 374.9465 2.25%

Sabon_Gari_North 160.4469 2.08% 55.04183 2.55% 52.01453 1.85% 357.7719 2.15%

Dadi_Riba 148.1789 1.92% 51.81079 2.40% 43.52106 1.55% 326.9261 1.96%

Ungwan_Shanu 167.2909 2.17% 22.6989 1.05% 28.60061 1.02% 320.9624 1.93%

Sabon_Gari_South 131.2963 1.70% 39.78674 1.84% 50.1313 1.79% 284.4752 1.71%

Badiko 108.6117 1.41% 49.36897 2.28% 30.15996 1.08% 254.2053 1.53%

Sabon_Gari_West 108.4049 1.41% 41.06246 1.90% 29.56497 1.05% 230.4191 1.38%

Sabon_Gari 105.8495 1.37% 32.38402 1.50% 15.54433 0.55% 204.1118 1.23%

Ungwan_Dosa 82.33973 1.07% 40.10052 1.86% 22.45629 0.80% 181.2543 1.09%

Rigasa 90.00707 1.17% 14.33843 0.66% 25.69446 0.92% 173.7817 1.04%

Tudun_Wada_South 77.74869 1.01% 19.63351 0.91% 32.98429 1.18% 152.356 0.91%

Kawo 68.90186 0.89% 12.36277 0.57% 38.07734 1.36% 148.683 0.89%

Mekera 74.46422 0.97% 16.51091 0.76% 17.33646 0.62% 143.4798 0.86%

Narayi 51.79721 0.67% 15.09243 0.70% 20.28422 0.72% 128.8289 0.77%

Kakuri_Hausa 78.91733 1.02% 8.152617 0.38% 6.848198 0.24% 108.5929 0.65%

Tudun_Wada_North 55.62038 0.72% 9.723843 0.45% 16.33606 0.58% 100.739 0.60%

Nasarawa 51.61235 0.67% 21.32742 0.99% 8.957516 0.32% 97.4663 0.59%

Kakuri_Gwari 30.54594 0.40% 9.256345 0.43% 7.77533 0.28% 64.97954 0.39%

Kakau 24.91506 0.32% 2.696435 0.12% 18.12005 0.65% 63.63587 0.38%

Kujama 27.34278 0.35% 5.178557 0.24% 0 0.00% 62.14268 0.37%

Yelwa 14.58671 0.19% 7.367025 0.34% 12.3766 0.44% 50.53779 0.30%

Sabon_Tasha 14.55083 0.19% 3.890596 0.18% 13.0724 0.47% 43.7303 0.26%

Rido 14.02167 0.18% 1.874555 0.09% 4.723878 0.17% 38.54085 0.23%

Television 14.23506 0.18% 12.94096 0.60% 0 0.00% 32.61123 0.20%

Total 7712.441 100.00% 2160.603 100.00% 2804.222 100.00% 16659.37 100.00%

Average 226.8365 2.94% 63.54713 2.94% 82.47712 2.94% 489.9815 2.94%

Note: Rate is expressed in per 10,000 population

Total Population of Kaduna metropolis = 1,769,032

4.3. Implications of Crime Pattern in Kaduna

Metropolis

The findings contained in this study demonstrate the

importance of incorporating spatial and demographic effects

into empirical models of crime. A related implication is that

global theories of crime may need to be further modified or

expanded in order to take spatial patterns and spatial dynamics

more explicitly into account. Indicated results as in Table 2.

Given recent developments in GIS technology and spatial

analysis applications, there is now available a rich array of

tools that can be applied to the study of crime in its spatial

context. This opens the door for new ways to explore,

visualize, and understand hot spots and clusters of crime,

spatial diffusion processes, and differences based on spatial

scale or location which is what was accomplished using

ITCA to modelling crime pattern in Kdm.

The crime rate by police jurisdictional division for the

Study Period (July, 2011 – June, 2014), in figure 4 reflects

clearly that Kawo, Rigasa, Tudun Wada and Unguwan Sanusi

PJD’s have a crime report rate above local average and

accounts for about 37.1% of the total of all Category A

Crimes with 9.4%, 10.1%, 8.4% and 9.2% respectively.

Page 11: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

American Journal of Environmental Policy and Management 2018; 4(1): 9-20 19

Source: Derived from Nigerian Police on Authors Field Work, (2015)

Figure 4. Spatial Distribution of Crime by Police Jurisdictional Divisions in Kaduna Metropolis for selected (seven Category A) Crime Type for the Study

Period (July, 2011 – June, 2014).

Apart from Gabasawa and Sabon Tasha PJD with the

lowest reported crime rate conversely takes up about 10.6%

of the total of all Category A Crimes with 5.2% and 5.4%

respectively other PJD’s are near average crime rate of all

Category A Crimes.

This may be explained by the fact that Gabasawa PJD have

a very high security levels Sabon Tasha on the other hand

seemingly low security awareness while Kawo, Rigasa,

Tudun Wada and Unguwan Sanusi PJD’s have very high

population.

The results of this study thus have both theoretical and

methodological implications and point to several directions

for future research. First, the results indicate that different

processes may be operating in Kaduna metropolis. In fact, it

could be said that one of the least understood topics in the

field of criminology is that of spatial and spatio-temporal

crime pattern. Thus, there is a need for further research on

spatio-temporal crime pattern which takes location and

geographic context seriously. Future more these findings

should address the spatial dynamics of crime as a product of

social, economic, and demographic factors.

Secondly, these findings also point to the need for

spatially-informed theory construction in the field of

Criminology. Ecological approaches to the study of crime

may provide fruitful theoretical directions for studying the

spatial dynamics of crime. Testing the relative merits of the

stratification and social control perspectives from a more

spatially informed model building approach should therefore

prove to be a promising direction for future research as well.

Based on the implications of the present study, it may well be

that a spatially informed stratification model of crime would

be more appropriately applied in urban context.

Finally, the present research shows the value of applying

Geographic Information System (GIS) technologies and

spatial analytic procedures to the study of aggregate crime

patterns. The main advantage of using GIS and related

technologies is that it enables the researcher to look more

rigorously at the spatial patterns and ecological contexts of

crime. Furthermore, the analytical applications of GIS can be

used in either an exploratory or confirmatory capacity. As an

exploratory data analysis tool, GIS can be used to examine

data visually as a way of generating new hypotheses from the

data or as a way of identifying unexpected spatial patterns.

As a confirmatory data analysis tool, GIS has been given

increased analytical power with the introduction and

development of Programmable spatial statistical packages

such as MATLAB. Thus, future studies could benefit

substantially by systematically investigating factors

associated with crime from a spatial perspective utilizing the

contributions that GIS and geographic information analysis

can provide. By employing spatial analytic procedures within

a GIS environment, contextual and ecological factors

identified as theoretically relevant in studies of crime.

Overall, the findings of the present study show how

important spatial and contextual analysis can be in the study of

urban crime across various levels of geography. By combining

graphical, analytic and statistical tools in a GIS environment,

researchers can explore spatio-temporal patterns, which may

warrant further empirical examination as well as formally test

spatially informed theoretical models for their applicability at

different spatial and temporal scales and locations.

5. Conclusions

The pattern of urban crime in Kaduna metropolis was

examined using GSCMS, with a view to offering possible

options for effective urban crime management. However this

was successful in accordance to the set objective of this

paper, the following conclusions where then ascertained.

From extensive review carried out for this research it was

established that crime is an increasingly serious problem in

cities all over the world, especially in developing countries or

countries in transition. However, the study of crime is still

scanty. Advancement in technology such as the computer

Page 12: Spatio-Statistical Analysis of Urban Crime; A Case Study ...article.aascit.org/file/pdf/8970762.pdfKdM, from the annual crime records in locations such as Rigasa, ungwan sanusi, and

20 Nzelibe Tobenna Nnaemeka and Bello Ashiru: Spatio-Statistical Analysis of Urban Crime; A Case Study of

Developing Country, Kaduna Metropolis, Nigeria

technology, data visualization and geographic information

systems (GIS), has enhanced spatial statistical analysis

techniques. The field of environmental criminology has

largely adopted these techniques in crime analysis. Indeed,

city officials and policy makers have recognized that crime

patterns could be of greater importance when their spatial

dimensions are taken into account. However, in practice the

lack of geo-spatial awareness and thinking has resulted in an

un-urgent sense of developing GSCMS demonstrated as a

model in Kaduna Metropolis.

In spite of the current crime prevention strategies in

Kaduna Metropolis there has not been one which reference

analysis with a spatio-statistical inclination which implies

policy makers and police departments cannot fight or take

precautions against the symptoms of crime actively, for a

more proactively policing. The finding of this research imply

that crime treatment and prevention strategies should be re-

examined carefully to incorporate a spatio-statistical

dimensions of crime so as to reach a result that can serve as a

useful reference for policy makers and police departments for

crime prevention and related decision making process.

Finally it cannot be overemphasized that the long-term

solution to the crime problem will rely greatly on good

governance, with policies that will foster inclusiveness equity

and social justice as geared towards reduction of poverty.

References

[1] Al-Madfai, H., Ivaha, C., Higgs, G., Ware, A., Corcoran, J. (2006). “The Spatial Disaggregation Approach to Spatio-Temporal Crime Forecasting,” International Journal of Innovative Computing, Information and Control, Vol. 3.3.

[2] Boba, R. (2005). “Crime Analysis and Crime Mapping”, Sage, USA.

[3] Brantingham, P. L., & Brantingham, P. J. (1984). “Patterns in Crime”, New York: MacMillan.

[4] Chainey, S. And Ratcliffe, J. (2005). “GIS and Crime Mapping”: John Wiley, & Sons. England.

[5] Cohen, L. E. and Felson, M. (1979). “Social change and crime rate trends: A routine activity approach”, American Sociological Review, vol. 44, pp. 588-608.

[6] Google Earth Inc (2014).

[7] Haifeng Zhang, and Michael Peterson, P. (2007). “A spatial analysis of neighbourhood crime in Omaha nebbraska using alternative measures of crime rates”, Internet Journal of Criminology.

[8] Hirschfield, A. and Bowers, K. (2001). “Mapping and Analysing Crime Data”: Lessons from Research and Practice, Taylor and Francis, New York.

[9] Johnson, SD, Summers, L, & Pease, K. (2009). “Offender as forager? a direct test of the boost account of victimization”: Journal of Quantitative Criminology, 25 (2), 181–200.

[10] Kaduna State Urban Planning and Development Authority (2014).

[11] Murray CJL, Ferguson B, Lopez AD, Guillot M, Salomon J, Ahmad O. Modified-logit life table system: principles, empirical validation and application. Geneva, World Health Organization (GPE Discussion Paper No. 39), 2001.

[12] Nigerian police force (NPF,) (2012). Report from the Kaduna police command retrieved from www.Npf.gove.ng/zonedetails?id=kaduna on 14/01/2015.

[13] Osborne, D. A. and Wernicke, S. C. (2003). Introduction to Crime Analysis: Basic Resources for Criminal Justice Practice. New York: Haworth Press.

[14] Vann, B. I. and Garson, D. (2003). “Crime Mapping”: Peter Lang Publishing, New York.

[15] Weisel, D. L. (2003). “The sequence of Analysis in solving problems”, Crime Prevention Studies.