University of Southern California Dissertations and Theses

Testing social disorganization theory on violent crime: a case study on Pueblo, Colorado

(USC Thesis Other)

Testing social disorganization theory on violent crime: a case study on Pueblo, Colorado

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Testing Social Disorganization Theory on Violent Crime:
A Case Study on Pueblo, Colorado

by

Dane Francis Stanley Cornell

A Thesis Presented to the
Faculty of the USC Graduate School
University of Southern California
In Partial Fulfillment of the
Requirements for the Degree
Master of Science
(Geographic Information Science and Technology)

August 2019

Copyright © 2019 by Dane Francis Stanley Cornell

To my family, without their support and love this dream would not have been possible. And a
special thanks to my loving girlfriend who stood by me through this journey.

i

Table of Contents
List of Figures ................................................................................................................................ iii
List of Tables .................................................................................................................................. v
Acknowledgement ......................................................................................................................... vi
List of Abbreviations .................................................................................................................... vii
Abstract ........................................................................................................................................ viii
Chapter 1 Introduction .................................................................................................................... 1
1.1 A Brief History of Pueblo, Colorado ...................................................................................2
1.2 Crime in Pueblo ...................................................................................................................5
1.2.1 Police Quadrants and Staffing ....................................................................................7
1.3 Social Disorganization Theory ............................................................................................9
1.4 Research Question .............................................................................................................10
1.5 Thesis Structure .................................................................................................................10
Chapter 2 Literature Review and Crime Theory........................................................................... 11
2.1 Criminology Theories ........................................................................................................11
2.1.1 The Chicago School and Concentric Ring Theory ...................................................13
2.1.2 Evolution of Social Disorganization Theory ............................................................16
2.1.3 Application of Social Disorganization Theory In this Research ..............................25
2.2 Spatial Analysis and Crime Data .......................................................................................26
2.2.1 Spatial Analysis Techniques .....................................................................................26
2.3 Social Disorganization Theory and Socioeconomic Status ...............................................29
Chapter 3 Methods ........................................................................................................................ 31
3.1 Data Sets ............................................................................................................................32
3.1.1 Crime Data Processing and Aggregation ..................................................................33
3.1.1.1 Geocode .................................................................................................................39
ii

3.1.2 Housing Data ............................................................................................................41
3.2 Data Visualization and Analysis ........................................................................................44
Chapter 4: Results ......................................................................................................................... 46
4.1 Violent Crime Temporal Trends ........................................................................................46
4.2 Directional Distribution .....................................................................................................50
4.3 Hot Spot Analysis ..............................................................................................................52
4.4 Crime Hot Spots and Housing Values ...............................................................................60
4.5 Summary Findings .............................................................................................................65
Chapter 5: Discussion and Conclusion ......................................................................................... 67
5.1 Discussion ..........................................................................................................................67
5.1.1 Hot Spot Analysis and Social Disorganization Theory Conclusions ........................67
5.2 Future Research and Analysis ............................................................................................69
5.3 Closing Thoughts ...............................................................................................................70
References ..................................................................................................................................... 71

iii

List of Figures
Figure 1-City Limits of Pueblo, Colorado ...................................................................................... 2
Figure 2-Pueblo Police Department Quadrants .............................................................................. 8
Figure 3-Concentric Zone Example from “The City” by Park and Burgess (1925) ..................... 15
Figure 4-Map of Chicago, IL Median Rental Prices of Housing .................................................. 18
Figure 5-Map of Chicago, IL Percentage of Families Using Government Assistance ................. 19
Figure 6-Workflow Diagram for Data Processing and Analysis .................................................. 32
Figure 7-Number of Crimes by Category ..................................................................................... 37
Figure 8-Total Counts of Top Violent Crime Types .................................................................... 37
Figure 9-Crime Data CSV Sheet ................................................................................................... 41
Figure 10-Geocoded Crime Data Displayed in ArcGIS Pro ......................................................... 41
Figure 11-Web-Based Data View of Housing Variables .............................................................. 42
Figure 12-Example of Collection of Housing Data Through the Pueblo County Assessor ......... 43
Figure 13-Violent Crimes by Year ............................................................................................... 46
Figure 14-Top Violent Crimes by Year ........................................................................................ 49
Figure 15-Top Violent Crime Locations for 2006-2016 with PPD Quadrants ............................. 50
Figure 16-Directional Distribution of Crimes by Type and by Year a) All Violent Crime, b) Top
Violent Crimes, c) Aggravated Robbery, and d) Second-Degree Aggravated Motor
Vehicle Theft Value $1000 to $20,000 ................................................................................... 51

Figure 17-Directional Distribution of Crimes by Type and by Year: a) Second-Degree Assault,
b) Robbery, c) Second-Degree Robbery, and d) First-Degree Aggravated Motor
Vehicle Theft .......................................................................................................................... 52

Figure 18-Top Violent Crimes Hot Spot Analysis for 2006-2016 ............................................... 54
Figure 19-Second-Degree Burglary Hot Spot Analysis for 2006-2016 ........................................ 55
Figure 20-Second-Degree Assault Hot Spot Analysis for 2006-2016 .......................................... 56
iv

Figure 21-Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000 Hot
Spot Analysis for 2006-2016 .................................................................................................. 57

Figure 22-First-Degree Aggravated Motor Vehicle Theft Hot Spot Analysis for 2006-2016 ..... 58
Figure 23-Aggravated Robbery Hot Spot Analysis for 2006-2016 .............................................. 59
Figure 24-Robbery Hot Spot Analysis for 2006-2016 .................................................................. 59
Figure 25-Example of Housing Values for a Hot Spot in Pueblo, CO ......................................... 61
Figure 26-Housing Values Overlaid with Hot Spot Analysis of the Top Violent Crimes ........... 61
Figure 27-Housing Values overlaid with Hot Spot Analysis of Second-Degree Burglary .......... 62
Figure 28-Housing Values Overlaid with Hot Spot Analysis of Second-Degree Aggravated
Motor Vehicle Theft Value $1000 to $20,000 ........................................................................ 62

Figure 29-Housing Values Overlaid with Hot Spot Analysis of Aggravated Robbery ................ 63
Figure 30-Average Housing Value for a Sample Hot Spot Area of Pueblo, Colorado ................ 64
Figure 31-Average Housing Value for Sample Non-Hot Spot Area of Pueblo, Colorado ........... 65

v

List of Tables
Table 1-Police Staffing for Pueblo, Colorado ................................................................................. 8
Table 2-Data Sets .......................................................................................................................... 34
Table 3-Crimes by Category ......................................................................................................... 36

vi

Acknowledgement
I would like to start by thanking Dr. Laura Loyola for pushing me to my limits and
helping me every step of the way in finishing my thesis. Her continued support and guidance
gave me the drive to cross the finish line. I would also like to thank the other members of my
thesis committee Drs. Jennifer Bernstein and John Wilson for their support and help. Lastly, I
would like to thank all of the other professors who helped me to better understand the world of
GIS through my path of earning a Masters degree.
vii

List of Abbreviations
BCS British Crime Survey
CHPD Columbia Heights Police Department
CSV Comma Separated Value
EDGIS (Pueblo County) Economic Development and Geographic Information System
GIS Geographical Information Systems
IMS Interactive Mapping System
KDE Kernel Density Estimation
MAUP Modifiable Areal Unit Problem
PPD Pueblo Police Department
SES Socioeconomic Status
SDT Social Disorganization Theory
STAC Spatial and Temporal Analysis of Crime
USDL-BLS United States Department of Labor-Bureau of Labor Statistics

viii

Abstract
According to social disorganization theory, crime is caused by social and economic variables at
the neighborhood level. Coined in 1942 by Shaw and McKay, their research utilized the city of
Chicago as a natural laboratory to examine how social and economic variables affected crime. It
was decided to test this hypothesis using Pueblo, Colorado because of the high crime rate. To test
if the theory of social disorganization applies to Pueblo, violent crime and socioeconomic status
were analyzed spatially to answer the following questions: 1) have crime rates changed over
time? 2) do the changes in crime rates have a spatial pattern? and 3) does the change in crime
rates mirror housing values?
Data on violent crimes was determined with the assistance of the Pueblo Police
Department, who provided the location of 4,500 individual violent crimes across the city from
2006 to 2016. Statistical analysis showed that many of the counts of individual crime types were
too low to be statistically significant, so the five crimes with the highest occurrence were used
for further analysis. Socioeconomic status was determined using the housing values within the
City of Pueblo.
Hot spot analysis using the GI* statistic, which uses high and low z-scores to determine
clusters of high values (hot spots) and clusters of low values (cold spots), was used to determine
statistically significant high crime areas within Pueblo. These hot spots were used to determine
where housing values would be analyzed. Statistical analysis showed that 2016 housing values
sampled from within a hot spot area were lower than those samples from a non-hot spot area.
Additionally, the average housing value in the sampled hot spot areas progressively decreased
over the 10-year period, while those sampled in non-hot spots rebounded after the 2008
recession.
1

Chapter 1 Introduction

Crime in the City of Pueblo has been a major burden for its citizens for many years. Violent
crime is especially concerning, as there has been a steady increase in violent crimes from 2006 to
2016 throughout the city (Colorado Department of Public Safety 2018). Violent crime, according
to the Federal Bureau of Investigations Uniformed Crime Reporting Program, is defined as
“offenses that involve force or threat of force.” The relationship between crime and the economy
is one that is interconnected; according to the United Nations Office on Drugs and Crime (2010),
high violent crime rates affect economic levels, leading to a lack of investment which in turn
puts pressure on trends of crime. Meaning that individuals or groups are less willing to
economically invest in areas with high crime rates, additionally due to social disorganization of
the community, it is unable to solve chronic issues, such as crime. Social disorganization theory
attempts to explain this underlying pressure on crime based on economic factors, an example of
this is the housing values of a high crime area. Social disorganization theory (SDT) adopted early
use of geographical information systems (GIS) and the City of Chicago as a natural laboratory to
study the effects of socioeconomic status on crime rates. By focusing specifically on violent
crime (as defined by state and national standards) as a proxy for socioeconomic status in the City
of Pueblo, this project will test social disorganization theory as a means to help explain why the
city has witnessed an increase in these crimes. Specifically, this thesis aims to answer the
following questions: 1) have crime rates changed over time? 2) do the changes in crime rates
have a spatial pattern? and 3) does the change in crime rates mirror housing values?
2

1.1 A Brief History of Pueblo, Colorado
Pueblo, Colorado is located on the Interstate-25 corridor in Southern Colorado, 112 miles
south from Denver, the state’s capital. Located at 38°16′1″N 104°37′13″W the city of Pueblo has
an area of 45.5 square miles (Figure 1). Pueblo sits at the foothills of the Rocky Mountains
where the Great Plains stop, and the mountain range begins. The climate in Pueblo is mild, with
annual precipitation averages around 12 inches. The winters see snow and the summers see highs
in the low 100’s with low humidity. While Pueblo, population ~111,000 and Colorado Springs
are considered larger cities in Southern Colorado, spread across the area are even smaller cities.
The confluence of I-25 and Highway 50 are the two major roadways, with I-25 traversing north
to south and HWY 50 traversing the east to west.

Figure 1-City Limits of Pueblo, Colorado

3

Pueblo was nicknamed the Steel City (Pittsburg of the West) in the late 19
th
century due
to its roots as a large steel producer (Broadhead 2019) which was the main economic driver from
the 1880’s to the mid 1990’s. The enterprise that was responsible for steel production in Pueblo
was the Colorado Fuel and Iron Company (CF&I) (Broadhead 2019). CF&I’s steel production
facilities were the largest west of the Mississippi and was the largest employer in the State of
Colorado, employing roughly 15,000 people at its peak (Rees 2017). The steel crash in the late
1970’s hurt CF&I and the production in Pueblo, from which CF&I would never fully recover.
The facility changed ownership several times during the 1990’s and in 2006 was purchased by its
current owner EVRAZ Corporation. EVRAZ currently produces steel rail, seamless pipe, rod,
and coiled reinforcing bar (Ezez 2019).
Pueblo’s current economic status within Colorado is below the state average, with a
population of 109,122 (American Factfinder 2017) and an unemployment rate of 5.1% (United
States Department of Labor-Bureau of Labor Statistics 2018), which is above the state average
unemployment rate of only 2.9% (USDL-BLS 2018). This is one economic indicator that
demonstrates Pueblo’s economy is struggling. The median household income for Pueblo is
$36,280, nearly half the state average (American Factfinder 2017). Additionally, housing values
are low, with the estimated house or condo value of $126,200 in 2016 being less than half state
average of $314,200 (American Factfinder 2017). This last socioeconomic status indicator of
housing value is further examined in this thesis.
Furthermore, the large corporate employers for Pueblo tend to lack key economic drivers
such as manufacturing and technology. The current high-end employers in the City of Pueblo
include Vestas Towers America, Trane-Ingersoll Rand, United Technologies Corporation,
Parkview Regional Medical Center, and DOSS Aviation (Hickenlooper 2018). These five-
4

businesses fall within one of four categories: 1) Health and Wellness, 2) Advanced
Manufacturing, 3) Infrastructure Engineering, and 4) Transportation/Logistics. While these
industries make up the higher end employment in Pueblo and many of the jobs require
specialized training or education, these jobs are not abundant. Skilled verses unskilled labor is
defined as skilled workers having specialized training or certification to complete a job, such as
pipefitting, welding or computer programming; unskilled labor on the other hand does not
require any type of specialized training or certification to be completed, such as a janitorial
service, fast-food work, or sales. According to the United States Department of Labor-Bureau of
Labor Statistics (USDL-BLS) estimates of employment per 1000 jobs, Industrial Production
Managers account for 0.706 of the 1000, whereas Sales and Related Occupations comprise
114.349 jobs per 1000 (United States Department of Labor-Bureau of Labor Statistics 2018).
This is not to say that the employers listed above employ more unskilled labor than skilled, it is a
representation that employment as a whole in Pueblo contains much more unskilled laborers than
skilled laborers. This is just one example of the discrepancy in high verses low skilled labor jobs
in Pueblo, the entire list from the USDL-BLS is quite large and will not be included here. The
lack of economic opportunity can be traced back to an uneducated workforce (Markus 2014).
Education rates in Pueblo are below the state average as well, as of 2017 the high school
graduation rate in Pueblo’s School District 60 is 79%, of those graduating students 56.7% attend
post-secondary schools (i.e. college or university) located in or out of the state. In total, only
19% of the population of Pueblo has attained any post-secondary education (American
Factfinder 2017). The combination of below average education rates and low skilled workforce
create an environment where the population of Pueblo is not resilient against economic
fluctuations.
5

1.2 Crime in Pueblo
Given the size of Pueblo and the rather small police force (see Section 1.2.1), the City of
Pueblo utilizes community policing, a technique that is:
…a collaboration between the police and the community that identifies and solves
community problems. With the police no longer the sole guardians of law and
order, all members of the community become active allies in the effort to enhance
the safety and quality of neighborhoods. (Bureau of Justice Assistance 1994)

This technique, which has its roots in the early days of the city police in London, England
(Bureau of Justice Assistance 1994), has been used to address auto theft and other problems
throughout the City of Pueblo. It has been a common way for police departments to interact with
communities since its inception to build trust and a sense of ownership for safety in a
community. However, the efficacy of this technique relies on the ability for a community to be
involved, which can be influenced by multiple social factors, as discussed below, and economic
factors, as previously mentioned.
Based on the ideal that both the police and the public act as law enforcement, community
policing connects the police force back to the people and local stakeholders in the community.
This type of policing was used on and off for many years but, was displaced by the advent of the
automobile and radio systems in the early 1900’s (Bureau of Justice Assistance 1994). These
technological advances moved police officers off the streets and foot patrols and placed them in
vehicles and offices, which removed them from the community; thus, creating a divide from Sir
Peel’s original idea of what the police stood for. Another argument that researchers have made
for the disconnect between police and the community occurred from the reform of policies in the
early 1900’s that moved individual police officers to different areas of the city to avoid
corruption. This corruption stemmed from the close relationship that the police and policy
6

makers shared. Because these politicians had control of the police and their departments, they
were thus able to influence policy within the police in favor of the political elite (Walker 1999).
During the 1970’s the Rand Corporation looked at the role of detectives and found that
they only solved a small number of crimes without the assistance of the officers that were on
patrol in the community (Bureau of Justice Assistance 1994). This once again changed the
perspective and role of patrol officers within many police departments. Subsequently, officers
received more training on solving crimes and working within the community (Bureau of Justice
Assistance 1994), a move back to community policing. Several experimental programs were
created in cities such as Newark, New Jersey and Flint, Michigan that put officers back on foot
patrols in different neighborhoods. It was concluded that these foot patrols created a better
relationship between the police and the community and at the same time giving the population a
feeling of safer streets and less fear of crime (Bureau of Justice Assistance 1994).
Community policing has had success in many areas, one such area is the Columbia
Heights Police Department, located in Columbia Heights, Minnesota. A case study of the CHPD
found that although the police force lacked the ethnic diversity present in the community there
was a decrease in crime (DeMeester, LaMagdeleine, and Norton 2011). The authors claim that
this drop-in crime and high satisfaction of the police force by the population, was directly related
to community policing techniques that were applied, they note:
…the change from controlling crime through arrests, enforcement contacts and
threats of consequences to focusing on connections and relationship building
contacts has helped the officers know and understand the community they serve in
new ways. Rather than reacting to crime, true crime prevention has begun in the
community. (DeMeester, LaMagdeleine, and Norton 2011)

Although similar techniques have been implemented in Pueblo, it is difficult to determine if this
technique is working or not. Within the annual reports provided by the Pueblo Police Department
7

(PPD) there was a discussion about programs that have been implemented to connect the patrol
officers to the community, public, and stakeholders; however, there is little data to show that
these efforts have been successful, nor is there data on the possible barriers to successful
implementation of community policing. One possibility, further examined in this thesis, is that
SDT is at play in Pueblo, inhibiting community organization and involvement in policing efforts.
Crime, in general, in the City of Pueblo is higher than the state and national averages.
The instances of violent crimes in the United States and Colorado during 2016 were 397.5 and
338.1 per 100,000 people (Colorado Department of Public Safety 2018). For the City of Pueblo
during that same time the instance of violent crimes was 688.7 per 100,000 people (Colorado
Department of Public Safety 2018), much higher than either the national or states averages,
which is troubling for both the residents and police force. This brings into question when and
why the rate of crime has increased so dramatically, as well as the policing methods used
throughout the city from 2006 to 2016. The following section discusses the current spatial
allocation of policing resources in Pueblo.
1.2.1 Police Quadrants and Staffing
In addition to the PPD using community policing techniques, the city has been divided
into four different quadrants, varying in size from 11.24 mi
2
to 20.70 mi
2
, that are each patrolled
by 3 or 4 officers during a shift, comprising a total of seven officers for the north and south parts
of Pueblo (Figure 2). In short, Corporal David Jacober of the PPD (pers. comm. 2018) explains
the quadrants and staffing as two “crews” a Blue and a Green, each of which is staffed with 14
officers. The two crews alternate time on and off and both cover the same quadrants when the
other is not on duty. Of note are the temporal assignments of the crews by quad with the day shift
on duty from 0730 to 1730, afternoons 1200 to 2200, and overnight from 2200 to 0800 (Table 1).
8

Figure 2-Pueblo Police Department Quadrants
Table 1. Police Staffing for Pueblo, Colorado

Blue Crew Green Crew
14 Officers 14 Officers
North - Quads 1 and 2 North - Quads 1 and 2
South - Quads 3 and 4 South - Quads 3 and 4
Quad 1 (16.94 mi
2
): 4 Officers
Quad 2 (20.70 mi
2
): 3 Officers
Quad 3 (11.24 mi
2
): 4 Officers
Quad 4 (11.67 mi
2
): 3 Officers
Quad 1: 4 Officers
Quad 2: 3 Officers
Quad 3: 4 Officers
Quad 4: 3 Officers

Corporal Jacober explained that the assignment of patrol officers to quadrants is roughly based
on the number of calls for service the PPD receives. He also mentioned that even though they
may have assigned officers to a certain quad, it was not uncommon for officers to respond to a
call in another quad to provide backup or assistance. The first part of this thesis examines the
overall spatial pattern of violent crime in Pueblo, as well as the spatial distribution of these
crimes in with respect to police quadrants (Chapters 3 and 4).
9

1.3 Social Disorganization Theory
The theory of social disorganization is a criminology term that has evolved since its
creation from the Chicago School in the early 1920’s, where crime theory switched from looking
at genetic patterns of criminals to societal factors affecting crime (Williams and McShane 2004).
Social disorganization theory’s (SDT) roots are founded in urban research conducted by Robert
Park and Ernest Burgess on urban concentric ring theory, which shows how the ideal city or
town should be built. This model is discussed further in Section 2.1.2, but briefly it is described
as rings moving outward from the city center, which is considered the business district, and is
followed by the transition zone or commercial zone. Next is the workingman’s homes zone,
followed by the residential zone or the upper-class area, and lastly the commuters’ zone or the
suburbs (Park and Burgess 1925). SDT grew out of the concentric theory by looking at each ring
and determining that the inner most ring would have more crime than the outer rings because the
inner rings contain the less desirable aspects of society, less desirable housing and more
commercial business, while pressure is placed on the outer rings and crime inevitably follows the
expansion and invades the outer rings. When this crime moves into residential areas in the outer
rings the population moves out as quickly as possible, creating an environment where
reconstruction or community betterment is not possible (Shaw and McKay 1969). Thus, creating
a residential area that has low economic value and high crime. The Shaw-McKay model of social
disorganization has been a standard in criminology studies for many years, however this theory
was not tested until the late 1980’s. This research tested the hypothesis that an increase in crime
rates is due to a variety of factors, such as economic status, ethnic makeup of the area, and
family disruption. They predicted that each of these social factors negatively influence crime
through the lack of the community to accomplish common goals (Sampson and Groves 1988).
10

SDT, the findings of Sampson and Groves, and later works will be discussed further in detail in
Chapter 2. This thesis examines the rate of crime in Pueblo, CO and socioeconomic variables
through the lens of social disorganization theory to answer the stated research questions and to
provide additional information to the PPD in terms of efficacy of community policing.
1.4 Research Question
According to Sampson and Groves (1988) as well as Shaw and McKay (1969),
socioeconomic status (SES) has been an ecological correlate of crime in some urban areas. This
thesis examines the correlation between SES and crime rates, using housing value as an indicator
of SES (see Section 2.3). This project aims to answer three questions: have crime rates changed
in Pueblo from 2006-2016, do these crimes have a spatial pattern, and does the change in crime
rate reflect the change in housing values in Pueblo, Colorado? This will be accomplished by a
temporal analysis of crime from 2006-2016, then by examining the spatial distribution of crime
in Pueblo, and finally by a spatial analysis of the SES indicator. Given the small geographic area
of Pueblo, small areas of both crime and home values are examined at the scale of the city block.
1.5 Thesis Structure
The remainder of this thesis has four chapters. Chapter two covers SDT crime theory as
well as a literature review covering the analysis techniques used for determining spatial patterns
of crime. Chapter three includes the data sets utilized as well as the spatial visualization and
analysis techniques used. Chapter four presents the results of the analyses. Chapter five
addresses the limitations to this study, the conclusion, and suggestions for future research and
analysis.
11

Chapter 2 Literature Review and Crime Theory
This chapter reviews the literature that builds the foundation for the theoretical background and
methodological choices made for this geospatial crime analysis. The literature examines previous
studies to determine what quantitative spatial analyses were used, how these types of analyses
have been applied to crime mapping, and case studies that have applied these analyses to violent
and non-violent crime. Background is also provided on how the housing variable was selected to
represent economic status and how high crime and low economic opportunity are related.
2.1 Criminology Theories
In the study of criminology there are many theories as to why crime happens. While
determining the spatial distribution of crime is a key element of this thesis, the potential to
further examine crime within a theoretical framework is also relevant. Therefore, I present here a
brief summary, and by no means a comprehensive list of all the criminology theories and
literature, before expounding on SDT.
The aim of criminology theory is to examine a variety of factors in an attempt to explain
why people commit crimes. Some theories focus more on the individual, while others focus on
external and societal forces. The following summaries of each theory are all based on definitions
provided by the National Criminal Justice Reference Service (2019) and the work of Frank P.
Williams and Marilyn D. McShane (2004).
In the first group of theories that focus on the individual are: rational choice theory, life
course theory, and routine activity theory; while social disorganization theory, conflict theory,
labeling theory, and social control theory are in the second group. Rational choice theory says
that people will act in their own best interests and commit crimes after looking at the risk versus
the reward. This theory was introduced in the late 18
th
century (Williams and McShane 2004)
12

and has since been intermixed with many other criminology theories listed. Life course theory
examines one’s life and early life events, and how said events affect the outcome of that person’s
choices to commit crime. In the 1920’s, theorist Karl Mannheim thought that one’s experiences
as a child would build the foundation for adulthood, and that these experiences passed from
parent to child over many generations (1935). Mannheim’s original life course theory was not
focused on crime, and as time progressed other theorists built on his theory and applied their
findings to criminology. It was found that a lack of parental oversight as a child and one-parent
households led to higher instances of not only child offenders but adult offenders as well
(Williams and McShane 2004). A theory that removes the developmental aspect, still focuses on
the individual but includes an aspect of external forces, is the routine activity theory.
There are three aspects of this theory. First, is a person who is willing and motivated to
commit crime, second is a target, and third is the absence of some type of deterrent such as a
police officer or other form of active or passive category of security measures. This theory has
been tested for many years and it has been found that if an individual has been exposed to crime
they are more likely to commit crime (Williams and McShane 2004).
Criminology theories that focus more on external factors include conflict theory, which
pits one social class against another and states that laws are created to protect those in power,
thus creating further conflict between the two groups of society. This theory has roots in
sociological work performed in the 1950’s from Lewis Coser, who did not agree with the current
and past theories of the nature of societies (Coser 1957). Much like Marxist Social theory,
conflict theory claims that those in power create laws to protect themselves and their way of life.
This theory is not popular, mainly because it assumes that the wealthy have created a biased
13

justice system (Williams and McShane 2004). In contrast, labeling theory looks to the process of
labeling individuals.
Labeling theory states that once a person has been labeled by society as a criminal or
delinquent, they inevitably become that label. This theory was heavily used from the early
1960’s to the 1980’s and was considered to be a product of societal changes that happened
during that time. Labeling theory focuses on society’s labeling of young offenders, and these
labels following the individual for the remainder of their life, turning them from a juvenile
offender to a repeat adult offender (Williams and McShane 2004). Another theory that focuses
on societal responsibility is social control theory, which places the responsibility of crime in the
hands of society through maintaining law and order. This theory relies on clear concise
guidelines to halt crime and that without them, crime will occur because people are born to be
delinquents (Hirschi 1969). This theory is the closest relation to SDT in the fact that it looks to
society, rather than the individual to explain why crime is committed. However, SDT includes an
important factor of the community’s ability to organize, as determined by socio-economic
factors. Social disorganization theory is the best to test in this thesis, as it has its roots in GIS and
uses socioeconomic factors to explain why crime happens. The following subsections will
discuss this theory in detail.
2.1.1 The Chicago School and Concentric Ring Theory
Concentric Ring and SDT used the City of Chicago as a natural laboratory for testing and
used the population as test subjects through observation only, there was little to no manipulation
of the people or economy. The Chicago School scholars used both individual case studies as well
as population statistics to build the sociological and criminological theories that are still used
14

today. They believed that “human behavior is developed and changed by the social and physical
environment of the person rather than by genetic structure” (Williams and McShane 2004, 46).
Social disorganization theory started with Robert Park and Ernest Burgess’s idea that
human ecology was much like that of plant and animals, sharing a symbiotic interdependence
with one another through the time and space that they naturally exist in (Williams and McShane
2004). Park and Burgess first began with the Concentric Zone Theory (Figure 3), which was
briefly explained in the introduction section of this text. This breaks a city down into zones
which represent different areas of socioeconomic status as related to crime occurrence.
Concentric zone theory explains the inner zones have a propensity to grow larger in population,
and eventually said population moves to the outer zones (Park and Burgess 1925). Thus, taking
the woes of the inner zones outward to the next zone. This relates to social organization and
disorganization to explain that the relationship between the two is reciprocal and “so far as
disorganization points to reorganization and makes for more efficient adjustment,
disorganization must be conceived not as pathological, but as normal” (Park and Burgess 1925,
54). To better understand this theory, it is necessary to look at each individual zone and the
ecology of the population per zone.
15

Figure 3-Concentric Zone Example from “The City” by Park and Burgess (1925)

The first zone that is to be examined is Zone I or “The Loop,” within this zone we would
find the central business district from which the city is expanding. According to Park and
Burgess this area is where the larger homeless population exists (Park and Burgess 1925). Zone
II or “Zone in Transition”, encompasses the downtown area and is an area to where business and
light manufacturing are migrating (Park and Burgess 1925). Park and Burgess note that Zone II
is in a state of decay and this is where we will find large populations of poor, and one might find
rundown neighborhoods that see high crime rates and decay. This area is also where Park and
Burgess claim that we would find newly immigrated families and colonies. Zone III or “Zone of
Workingmen’s Homes” is where the working class would be found and those who have “escaped
the area of deterioration” of Zone II, but still want to live close to their places of employment.
Zone III sees a mix of old and new development and contains a large amount of rental properties;
in addition, this zone is where second-generation immigration families would live. The second to
16

last area is Zone IV, or the “Residential Zone,” this is where one would find “high-class
apartment buildings or of exclusive ‘restricted’ districts of single-family dwellings” (Park and
Burgess 1925, 50). And the last area of the city is Zone V or the “Commuters Zone” and is
considered the suburbs. Park and Burgess state that this area is within a half hour or hour drive to
the city center (Zone I) and is where the populations from the inner zones would like to live.
These zones comprise the makeup of the “urban” area of an American city, according to Park
and Burgess.
This makeup of the American city is where one starts to understand how SDT became
such an important criminological theory. Cities are an ever-evolving ecosystem, and within this
ecosystem groups of similar individuals are formed. In a city, these groups tend to be of the same
economic status or cultural background and are what give a city personality. This process of
separating out groups, gives not only the group but also the individual a position and function
within the coalition of the city construct (Park and Burgess 1925). Burgess and Park constructed
the foundation that SDT was built on. In the following section, the evolution from Concentric
Zone Theory to SDT and beyond is explained in more detail.
2.1.2 Evolution of Social Disorganization Theory
From Park and Burgess in “The City” (1925), we move to a set of theories completed by
Clifford R. Shaw and Henry D. McKay titled “Juvenile Delinquency and Urban Areas.” Within
this text the authors take the Concentric Zone Theory and apply it to the City of Chicago. Shaw
and McKay speak to the ever-evolving city in such a way as to explain the migration of people
from the inner city to the outer rings. This migration creates a situation where the residential
areas near the business and industrial areas are far less appealing than those of the zones that are
further out (Shaw and McKay 1942). This pattern of migration is also central to SDT, in that as
17

the population moves from the inner zones the issues that made the inner zones less desirable for
occupation eventually follow the migration, thus the migration will continually be evolving.
They claim, that as a city grows there are seven parameters that are evidences of differentiation
resulting from city growth (Shaw and McKay 1942). These 7 evidences are: demolition of
substandard housing, increase and decrease of population, segregation of population on an
economic basis, families on relief, median rentals (median rental and home value from the 1930
United States Census), occupation groups (types of jobs the population work), and economic
segregation. Each of these variables were determined for use by the authors to indicate city
growth and characterize areas of the city through their population density (total number of
population and demographics including race and economic status) and allowed for measuring
different rates of delinquency in the different areas. Shaw and McKay created maps (Figures 4
and 5) to display these variables through an early Geographical Information System (GIS). These
maps both use a square mile grid system as the unit of analysis for the city, based on main streets
as well as previous government surveys of Chicago, these grid cells represent SES variables.
These maps also display the “Concentric Zones” through first increasing rental prices as one
moves further from the central business district (Figure 4), as well as the decreasing number of
families on relief (Figure 5). Through this early adoption of GIS, the authors were able to better
understand the social construct of the city.
Shaw and McKay conclude that although social dynamics may be at play, there are many
reasons for juvenile delinquency in the City of Chicago. These conclusions state that delinquency
rates vary between both the inner zones and the outer zones, yet geographic location does not
solely determine delinquency rates. Increased delinquency is also related to lack of institutions
such as schools and social group settings, and rapid changes in populations, while ethnic and
18

racial factors are not characteristic of delinquency rates and do not show a direct causation to
higher crime. Delinquency rates are correlated with socioeconomic status, and lastly some
communities have higher and consistent delinquency rates because of the lack of community
groups. This research on city structure propelled SDT, which was then built on by other scholars,
specifically Robert J. Sampson and W. Byron Groves.

Figure 4-Map of Chicago, IL Showing the Median Rental Prices of Housing
(Shaw and McKay 1942, 35)
Department of Sociology
Illinois Institute for Juvenile Research
Prepared with the Assistance of the
Works Progress Administration
19

Figure 5-Map of Chicago, IL Showing the Percentage of Families Using Government Assistance
(Shaw and McKay 1942, 34)

Sampson and Groves (1989) claim that the lack of individual/community level data and
relying only on census data is the weak link in previous studies. In the years between the
research by Shaw and McKay and Sampson and Groves (1942-1989) there were many scholars
(Sutherland 1947; Suttles 1968; and Kornhauser 1978) that attempted to better refine SDT yet
continued to use the same type of census data as Shaw and McKay. Sampson and Groves’ goal
was to go after the two limiting factors in past research: limited data and lack of direct testing.
These were addressed through use of a large national survey of British citizens’, the inaugural
Department of Sociology
Illinois Institute for Juvenile Research
Prepared with the Assistance of the
Works Progress Administration
20

British Crime Survey (BCS). The BCS consisted of a national survey of Wales and England
during 1982 that looked at crime through a “macro-level community analysis” and sampled sixty
addresses from each of the 238 communities within England and Wales and had an 80%
response rate. The final survey included 13,702 households of randomly selected persons above
the age of 16.
To test the theory of social disorganization, the authors created a model that included
measures of three areas of social organization. These three areas are socioeconomic status (SES),
residential stability, and ethnic heterogeneity. SES was measured by designing a “scale” from
summarized z-scores of the statistics containing the major dimensions of social class, education,
occupation, and income. In this “scale” education was defined by the percent of those who are
college educated, occupation was defined by the percent of people in professional and
management positions, and income was defined by the those with high incomes (amount not
mentioned). Residential stability referred to “the percentage of residents brought up in the area
within a 15-minute walk from home” (Sampson and Groves 1989, 785). To find the
heterogeneity range of each group an index (1-Ʃ𝑝 ⅈ
2
) was used where pi was a fraction of the
population of a given group. Ethnic heterogeneity included five categories within the BCS:
White, West Indian or African Black, Pakistani or Bangladeshi Indian, other non-white, and
mixed, these population groups do not match the populations used in the United States Census
because they are from a sample of British citizens (Sampson and Groves 1989). Both the use of
the better data and the three indicators allowed for Sampson and Groves to better define social
disorganization. According to Sampson and Groves, social disorganization “speaks not only to
the ability of a community to achieve common values (e.g., to defend itself against predatory
victimization), but also to community processes that produce offenders” (1989, 786). This
21

research took previous work further by examining data that was self-reported and focused on
violent acts, allowing for a more micro-level analysis of violence in a community, and not
merely juvenile delinquency.
Sampson and Groves concluded that SDT continues to play a role in crime analysis
through the interpretation of micro-level crime rates, and that it can be applied to not only
England and the United States but other nations as well (Sampson and Groves 1989). Sampson
and Groves’ findings provide evidence that communities showing low organizational
participation, which is the participation of a population in activities in which they are not
obligated to be active in such as a book club or church group, had much higher crime rates than
those with a high percent of participation within the community. It was also concluded that
differences in participation rates in the community had a central effect on community structural
characteristics (low SES, residential mobility, ethnic heterogeneity, and family disruption)
meaning that there is a strong negative correlation to community involvement, the structures that
bind the community, and high rates of crime (Sampson and Groves 1989). The authors also note
that their study was not without limitations as well as not a “definitive test of Social
Disorganization Theory” (Sampson and Groves 1989, 799). Some of these limitations include
small data sets in some of the areas of community organization and organizational participation
rates were not precise in measure. Regardless of these limitations this study was a better test of
Shaw and McKay’s original research.
The work by Sampson and Groves was completed in the late 1980’s, therefore to better
understand the application of SDT in the present it is necessary to focus on more current works
that examine SDT. Kubrin and Weitzer (2003) take the previous works on SDT and add several
new approaches to examining it. First the shortcomings of previous works are addressed, these
22

deficiencies arise in the variables that have been decided upon for examination. The authors note
“among the substantive issues addressed are the explanatory power of certain variables
hypothesized to mediate the relationship between exogenous structural conditions and
neighborhood crime (i.e., informal control, social ties, social capital, and collective efficacy)”
(2003, 375). They address these issues by looking at variables that had not previously been used.
These variables are: neighborhood culture, formal social control, and the urban political
economy (Kubrin and Weitzer 2003). The authors also apply a new methodology to the
examination of SDT, first they use new dynamic models that provide change over time for
ecological structures and crime in a given neighborhood. They also look at the reciprocal effects
between SDT and crime, meaning how the relationship between crime and community
organization are connected. Next spatial interdependence between neighboring areas was
examined. And lastly the authors examine the contextual effects on neighborhoods through
individual level outcomes, meaning that the individual was looked at versus the neighborhood as
a whole (Kubrin and Weitzer 2003). The authors find that with the added methodology and
variables, SDT is now more than ever a viable theory to help explain why crime happens. And
although they conclude that SDT should be applied to non-urban areas as well, they note that
further research should be performed.
Continuing with current research that examines SDT, “Assessing Neighborhood Effects:
Social Processes and New Directions in Research” (Sampson, Moreenoff, and Gannon-Rowley
2002) took 40 peer-reviewed journal articles from the mid 1990’s to 2001 and incorporated the
results to provide a better understanding of social processes that lead to delinquency in youth. To
accomplish this the authors synthesize the results of the 40 articles on neighborhood studies. This
synthetization happened by creating a classification system based on the following:
23

(a) neighborhood-level studies with neighborhood process measures, in which
both the dependent and independent variables are expressed as aggregate scales,
counts, or rates across ecologically defined areas that are akin to neighborhoods;
(b) multilevel studies with neighborhood process measures, in which sample
members are nested within ecologically defined neighborhoods, the dependent
variable is measured at the individual level, and the independent variables include
both individual-level factors and aggregate level measures of neighborhood
characteristics (both structure and process); and (c) multilevel studies with pseudo
or proxy neighborhood-process measures, identical to the previous category
except that social processes are actually measured at the individual level.
(Sampson, Moreenoff, and Gannon-Rowley 2002, 448)

The authors conclude that more research and analysis need to be completed to better understand
how social processes affect crime. However, they do have two takeaways from their research,
first the methods used to look at the neighborhood-level foundations, is an acceptable form of
measurement through survey and observation. Lastly, they find that “extra-local neighborhood
mechanisms appear with considerable strength, suggesting that spatial externalities operate
above and beyond the internal neighborhood characteristics of traditional concern” (Sampson,
Moreenoff, and Gannon-Rowley 2002, p. 473). Noting that the problems within an individual
neighborhood have the potential to cross into neighboring areas. This work was an amalgamation
of 40 peer reviewed journal entries and was used as a starting point to look at more current
research compiled on SDT.
More recent research that has examined SDT is “Extending Social Disorganization
Theory: Modeling the Relationships between Cohesion, Disorder, and Fear*" (Markowitz 2001)
which discusses a feedback loop within SDT. This loop, as described by the authors, starts with
how crime and disorder affect fear of crime and how that fear interacts with neighborhood
cohesion. They feel that the loop is ever escalating meaning that “decreases in cohesion increase
crime and disorder, which increase fear, which in turn further decrease cohesion” (Markowitz
2001, 297). The data used for this study was taken from the BCS (the same survey that Sampson
24

and Groves used in 1989) in the years 1984, 1988, and 1992. The measures for disorder were
calculated through five categories: noisy neighbors, teens hanging out on the streets, drunk
people in public view, littering, and vandalism (Markowitz 2001). Within the survey the
respondents that classified these as a “very big problem” were recorded and added across the five
years to calculate a score (Markowitz 2001). Neighborhood cohesion was not always included in
the BCS, to account for this three metrics were taken from the BCS that were contained in all
three years of data used (1984,1988, and 1992) (Markowitz 2001). These measures included
percent of people who went to a community meeting within the last week, percent of people who
stated that they got along with their neighbors, and the percent of people who were very happy
living in the area. These BCS responses were then used to calculate a z-score and added together
to represent parts of neighborhood control and interaction (Markowitz 2001). Lastly fear was
measured through three categories: percent of people who felt very unsafe when walking alone
after dark, the percent of people who were very concerned about burglary, and the percent of
people who were very concerned about being robbed (Markowitz 2001). It is of note that the
SDT variable of SES was measured by neighborhood median income based on the British pound.
The findings from the analysis show that “[c]ohesion shows a consistent negative effect, adding
7% to 10% to the explained variance in disorder” (Markowitz 2001,305-306). They also find that
cohesion increases with median income.
Violence is a key aspect of this thesis and therefore “Extending Social Disorganization
Theory: A Multilevel Approach to the Study of Violence among Persons with Mental Illnesses"
(Silver 2000) is an important work to review. In this work Silver uses SDT to better understand
violence in populations with mental illness. This is accomplished by assuming that community
cohesion and social organization are a necessary part of protecting the neighborhood from the
25

violent behavior that may arise with mentally ill persons living in the area (Silver 2000).
Meaning that family and friends of the mentally ill person or persons would be held accountable
for controlling the actions of the mentally ill. Silver states “residents of socially organized
neighborhoods are likely to act as guardians in attempting to control the behavior of teenaged
peer groups, so too are such residents motivated to control the threatening or otherwise
disruptive behaviors of persons with mental illnesses” (2000,1048). Silver uses a sample of
people who were part of the MacArthur study, which took patients from the Western Psychiatric
Institute and Clinic that had a major mental disorder. In this study each patient was interviewed
several times to compile background data, the hospital charts were also used to verify their
mental illness through official diagnosis (Silver 2000). The last part of the data gathering was
completed by looking at official records to compile data about the patient’s interaction within the
community, such as arrest records or psychiatric hospitalizations. Silver found that patients that
were released into neighborhoods with higher disorganization would be more likely to engage in
violent activities, whereas a patient living in a more social organized neighborhood would be less
likely to commit a violent act.
2.1.3 Application of Social Disorganization Theory In this Research
Pueblo, CO was chosen as a case study for SDT for two reasons. First, the use of one
variable within the theory, SES. And secondly the fact that this theory applied early GIS
techniques to test crime theory. The SES variable for this thesis was applied through housing
costs, which will be discussed further in this chapter. SDT uses location to aid in the
determination that it is in fact the “where” that is important in crime analysis. This “where” is
determined by the concentric zone theory of city construction.
26

2.2 Spatial Analysis and Crime Data
Spatial analysis of crime data is a vast agglomeration of techniques used to determine
where crime happens and attempts to explain why. This section discusses different techniques
and methods and ultimately clarifies the spatial analysis approach to this research.
2.2.1 Spatial Analysis Techniques
The spatial analysis techniques that were decided upon for this thesis are Kernel Density
Estimation and hot spot analysis. These are only two of techniques that can be used for crime
mapping, and the following section discusses the literature that helps define these types of crime
analysis, their applications, and uses.
Kernel Density Estimation (KDE) is a method that computes the density of features in an
area surrounded by similar features (Esri 2018a). For crime mapping, KDE provides the density
of crime instances in a given area (Ahmadi 2003). KDE works by placing a smooth surface over
point or line features. The values are highest at an individual point and decrease as they move
away from the point until zero is reached at the set search radius (Esri 2018a). This means that
on a given map of crime one would see high values or “peaks” in areas that have many instances
of crime and would see low values or “valleys” where there are few instances of crime. One
main advantage to using KDE is the ease of use as well as resulting visualizations. A limitation
to KDE is that the user may have the propensity to ignore the statistical values by being caught in
its “visual lure” (Eck et al. 2005). The authors note that the KDE is considered to be the best
interpolation method of crime data; however, the parameters, data type, boundaries, and other
factors of analysis should lead the user to determine what type of analysis should be used
(Chainey, Tompson, and Uhlig 2008).
27

KDE was determined not to be a suitable technique for this thesis because of the
limitations that arise with Pueblo as a study area via the size of the city. KDE smoothed maps
would have been visually appealing however, because of the spatial scale used for the study of
Pueblo the data would not be a true representation of the real world. This problem of spatial scale
of the crimes and study area required more precise technique to be used.
The techniques that are used in crime mapping include Point Mapping, Spatial Ellipses,
Thematic Mapping of Geographic Boundaries, Quadrant Thematic Mapping, and Interpolation
and Continuous Surface Smoothing. Point mapping is a digitized pin map that uses points to
indicate an event that happened in a given area. Finding spatial patterns within the point data is
difficult. This type of map is good for small scale analysis however, is not very viable for hot
spot analysis without a spatial unit of analysis and reference (Eck et al. 2005).
Another technique is the “thematic mapping of geographic boundary areas,” which uses a
choropleth map to display crime data though aggregated statistical data in a given boundary such
as a census block or police precinct. This type of crime map is best suited as a starting point for
further analysis at the microlevel. Also, depending on the choice of the bounding geographic unit
for analysis, the visualization and therefore resultant interpretation may vary greatly, hence
representing an issue of the Modifiable Areal Unit Problem (MAUP) (Chainey, Tompson, and
Uhlig 2008). The third technique “Grid Thematic Mapping” is a hot spot analysis that uses grids
to adjust the different sizes of boundaries and hence address the MAUP that occur with thematic
mapping. This technique, also referred to as “Quadrant Thematic Mapping” (Eck et al. 2005),
uses a grid placed over the study area to create equal areas for analysis. The limitations to this
technique include loss of spatial detail, and the visualization of the map has been described as
blocky (Chainey, Tompson, and Uhlig 2008). According to John Eck and colleagues, a crime hot
28

spot is defined as “an area that has greater than the average number of criminal or disorder
events” (2005, 2).
An additional method includes spatial ellipses (Chainey, Tompson, and Uhlig 2008).
Spatial ellipses, while not statistical analysis, were used by an early crime mapping software
called “Spatial and Temporal Analysis of Crime (STAC)” which in itself was not a GIS but
rather a program to run in addition to a GIS. STAC looks to the densest area of point data and
then places a standard deviational ellipse to these areas. This is similar to the Directional
Distribution tool now available in ArcGIS Pro, which was utilized in this thesis. Early spatial
ellipses indicate the size and orientation of the crime clusters. A main attraction of this type of
analysis is that there is not a requirement of a boundary or unit of analysis such as police
precincts, census tracks, or census blocks (Chainey, Tompson, and Uhlig 2008).
Getis Ord Gi* is a distance statistic used in hot spot analysis (Getis and Ord 1992). The
statistic that the hot spot operation uses is the Getis-Ord-Gi*(G-i-Star), which as explained from
Esri produces p-scores and z-scores that inform a user of high and low clusters of spatial data
(Esri 2018b). By looking at neighboring features or crime points as in this thesis, the G-i-Star
will determine if a point contains a high value, and if that high value point is surrounded by other
high value points a hot spot is marked. These calculations (Equation 1-3) uses high positive z-
scores to denote hot spots and negative z-scores to denote cold spots, both of which would be
considered statistically significant.
𝐺 𝑖 ∗
=
∑ 𝜔 𝑖 ,𝑗 𝒳 𝑗 𝑛 𝑗 =1
−𝑥 ̅ ∑ 𝜔 𝑖 ,𝑗 𝑛 𝑗 =1
𝑆 √[𝑛 ∑ 𝜔 𝑖 ,𝑗 2
𝑛 𝑗 =1
−(∑ 𝜔 𝑖 ,𝑗 𝑛 𝑗 =1
)
2
]
____________________________
𝑛 −1
(Eq 1)
29

In this equation 𝒳𝑗 is the attribute value for feature 𝑗 , 𝜔 𝑖 ,𝑗 is the spatial weight between
feature ⅈ and 𝑗 , 𝑛 is equal to the total number of features, s^2 is the variance and:
𝑋 ̅
=
∑ 𝑥 𝑗 𝑛 𝑗 =1
𝑛 (Eq 2)
𝑠 =
√
∑ 𝑥 𝑗 2
𝑛 𝑗 =1
𝑛 − ( 𝑥 ̅ )
2
(Eq 3)
The Gi* Statistic requires no further calculations because it is a z-score (Esri, 2018b).
A case study that utilized the Getis Ord Gi* statistic was “Dwelling Unit Prices in San
Diego County by Zip-Code Area, September 1989.” This case study examined housing prices
within San Diego County and assigned high and low z-score values based on these housing
values (Getis and Ord 1992). They found that housing values did not disperse in a uniform way
from the downtown area.
2.3 Social Disorganization Theory and Socioeconomic Status
Using housing values as an indicator of socioeconomic status (SES) was a logical
decision based on its connection to SDT. It can be concluded that areas within a city that have
low or very low housing values would be considered areas of poverty and less desirable in which
to live. This section will discuss literature that covers why housing values are a natural indicator
of SES.
Property value is an important part of SES, partly because homes are the most valuable
asset a person or family can attain and the location and cost of said home are largely based on the
individual or family income (Vernez Moudon et al. 2011); and higher income individuals or
families can buy more expensive homes in much nicer areas of a city or suburb. Vernez Moudon
30

et al. (2011) use housing values as a way to track SES at the individual level instead of at
neighborhood levels. In one study examining how crime impacts the housing market, the authors
found that homebuyers are willing to spend more money on housing if there is less violent crime
(Ihlanfeldt and Mayock 2010). This theory is further discussed in a study conducted in
Barcelona, Spain which also examines housing values and crime (Buonanno et al. 2012). This
study found that a single positive standard deviation in perceived security gives an increase in
housing value of 0.55-0.76% of the actual value. The authors do mention that “the results here
consistently indicate that crime perception negatively affects housing prices” (Buonanno et al.
319).
And in the last study that was examined “The Property Wealth Metric as a Measure of
Socieo-Economic Status” (Coffee et al. 2018) the authors examine health and property values in
a small area within South Australia, Adelaide. The housing values were calculated using the
sales of homes within the study area. This study built on the work by Vernez Moudon et al. and
found additional evidence that housing values have potential to indicate SES.
These three studies add credence to the fact that SES and housing values are connected to
crime rates. Thus, housing value will be used as an indicator of SES status in Pueblo, CO. These
housing values will be discussed further in Sections 3.1.3 and results provided in 4.4.
31

Chapter 3 Methods
This chapter summarizes the data used in this study, the methods used in processing prior to
incorporating into the GIS, and the analyses and visualization methods. The aim of this research
was to test for a spatial relationship between low housing values (low SES) and high violent
crime rates in the City of Pueblo and specifically to examine if: 1) crime rates have changed over
time; 2) the changes in crime rates have a spatial pattern; and 3) the change in crime rates mirror
housing values.
The overall workflow (Figure 6) consisted of first importing the shapefiles and point data
into ArcGIS Pro, the shapefile of the city limits was then used to create the four PPD quadrants
based on the diagram that was provided by the PPD. The address points were imported and used
for the address locator in the geocoding of the crime data. Prior to geocoding, the crime data set
was cleaned by removing any non-supported characters (such as punctuation and parentheses
symbols) in the data, as ArcGIS Pro will not import the data if they are included. The data were
also organized by year and separated into individual spreadsheets prior to the aggregation of the
crimes by type. Next, the data were analyzed through histograms and summary statistics. The
data were then imported into ArcGIS Pro, geocoded, and visually represented. Next a Fishnet
was created for each of the top violent crimes both for the total years and annually. The fishnet
grid cells were then spatially joined to the crime point data that they corresponded with. The
final step was to then run the hot spot analysis for the total crimes, annual crime, and top crime
types. Housing value data was the next set of variables that needed to be processed. Each of
these steps are described further in the sections below. Results are presented in Chapter 4.
32

Figure 6-Workflow Diagram for Data Processing and Analysis
3.1 Data Sets
Many different data sources were needed to analyze crime occurrence and economics in
Pueblo, CO. Table 2 consists of a complete list of the data sets that were collected, created, and
analyzed for this thesis, along with processing required for each dataset used. Crime data were
provided by the PPD’s crime analyst. Shapefiles for Pueblo city limits were downloaded from
the city’s GIS web portal. The address points for the geocodes were also taken from this portal.
Acquire Data
Create Police
Quadrant Shapefile
Build Address
Locator File
Exploratory
Visualization,
Analysis of Data,
and Summary
Statistics
Pre-Processing and
Data Cleaning
Geocode Crime
Data
Determine Top
Violent Crimes via
Summary Statistics
Hot Spot Analysis
for Top Violent
Crimes
Annual Directional
Distribution for
Top Violent
Crimes
Aggregate by Crime
Type
Use Crime Hot
Spots to Determine
Locations to Gather
Housing Value Data
Analyze
Data/Histogram of
Historical Housing
Values
Visualize Selected
Housing Values
Overlay Crime Hot
Spots with Housing
Value Data
Join Housing Value
Data to Address
Points
33

The fourth dataset was the housing values data, which are discussed at length later in this section,
the housing data were obtained from the County of Pueblo GIS department’s interactive mapping
system (IMS).
3.1.1 Crime Data Processing and Aggregation
To determine the locations of high crime rates, it was necessary to attain
individual crime data records across the eleven-year period (2006-2016). This was
achieved via a partnership with the City of Pueblo Police Department’s Crime Analyst
Micaela Leffler. Ms. Leffler provided this project with crime location (except for
reported rape, which was removed from the analysis), date, time, and offence type, which
was then used to visualize the spatial distribution of violent crimes. This crime dataset
included more than 13,000 incidents of crime, which included all property and violent
crimes that were recorded over the given time span.
34

Table 2. Data Sets
Data: File-Type Created By Date
Created
Last
Modified
Acquired From Use
County Limits Shape File-
Polygon
U.S. Department
of Commerce,
U.S. Census
Bureau,
Geography
Division,
Geographic
Products Branch
2015 27 June
2018
U.S. Census
Bureau
Used as a spatial
reference and
clipping
City Limits Shape File-
Polygon
Kyle Good February
2018
27 June
2018
City of Pueblo
GIS Office
Used to help
create police
department
quadrant shape file
Police
Quadrants
Shape File-
Polygon
Dane Cornell 01 July
2018
01 July
2018
Pueblo City
Limits Polygon

Neighborhoods Shape File-
Polygon
Debi Romines 06 October
2017
27 June
2018
City of Pueblo
GIS Office

Crime Instances Excel
Spreadsheet
Micaela Leffler 19 March
2018
20 March
2018
Pueblo Police
Department’s
Crime Analyst
Converted to
crime point data
Address Points Shape File-
Point
Kyle Good 13
December
2012
04 March
2018
City of Pueblo
GIS Office
Reference for
geocoding crime
points
Housing Values CSV
Spreadsheet
Dane Cornell 01 October
2018
28
September
2018
Pueblo County
Assessors
Interactive Web
Map
Used as a
socioeconomic
indicator through
conversion to
point data

35

The two major types of crime that are defined are property crime and violent crime.
Property crime is “a category of crime in which the person who commits the crime seeks to do
damage to or derive an unlawful benefit or interest from another’s property without using force
or threat of force” (US Legal, Inc. 2018), and violent crime is “a behavior by persons, against
persons or property that intentionally threatens attempts, or actually inflicts physical harm” (US
Legal, Inc. 2018). Because violent crime has a much larger impact on society than property or
non-violent crimes and is less likely to be correlated with commercial areas where property
crimes occur, only violent crimes are analyzed in this thesis. Additionally, taking violent crime
and separating it from crime as a whole provides a better picture of the connection of SES to
violent crime and allows for the micro-level analysis. Violent crimes, in general, will be
discussed in detail in the next few paragraphs to explain how they affect the safety of the
population. The violent crimes from the data initially were broken down into over 100 individual
types of crime. Due to improper and duplicate labeling, several of the crime types were simply
labeled slightly differently but were the same crime and could therefore be collapsed together.
This left 33 distinct violent crimes differentiated between first-degree (1
st
Deg) and second-
degree (2
nd
Deg) offenses. The crime data that was provided by the PPD was lacking a unique
identifier (code) that is needed for processing in a GIS, so this was added without ranking and
irrespective of internal police codes (See Table 3). Crimes will be referenced by their assigned
code (identifier).

36

Table 3. Crimes by Category
Category Crime
Code
Crime Type Total
Number
of
Crimes
Violent Crime 1 1ST DEG BURGLARY 127
Violent Crime 2 1ST DEG KIDNAPPING 19
Violent Crime 3 1ST DEG MURDER 38
Violent Crime 4 1ST DEG SEXUAL ASSAULT- SEX ASSAULT - SODOMY -
BOY - GUN 3
Violent Crime 5 1ST DEG AGG MOTOR VEHICLE THEFT 240
Violent Crime 6 1ST DEG ASSAULT 138
Violent Crime 7 1ST DEG CRIMINAL ATTEMPT MURDER 87
Violent Crime 8 1ST DEG FORCIBLE RAPE 13
Violent Crime 9 2ND DEG BURGLARY 1399
Violent Crime 10 2ND DEG KIDNAPPING 88
Violent Crime 11 2ND DEG MURDER 15
Violent Crime 12 2ND DEG AGG MOTOR VEHICLE THEFT VALUE $1,000 TO
$20,000 472
Violent Crime 13 2ND DEG ASSAULT 790
Violent Crime 14 2ND DEG CRIMINAL ATTEMPT MURDER 60
Violent Crime 15 2ND DEG ASSAULT HEALTH WORKER 12
Violent Crime 16 AGG ROBBERY 273
Violent Crime 17 AGG ASSAULT 119
Violent Crime 18 ASSAULT 2
Violent Crime 19 BURGLARY 44
Violent Crime 20 DUI - ACCIDENT 10
Violent Crime 21 HOMICIDE 15
Violent Crime 22 INTIMIDATE WITNESS/VICTIM 37
Violent Crime 23 KIDNAPPING 11
Violent Crime 24 KIDNAPPING - KIDNAP ADULT FOR RANSOM 1
Violent Crime 25 LARCENY/THEFT - PURSE-SNATCHING 4
Violent Crime 26 ROBBERY 240
Violent Crime 27 SEX OFFENSE 42
Violent Crime 28 SEXUAL ABUSE ON A CHILD 1
Violent Crime 29 SEXUAL ASSAULT 44
Violent Crime 30 SEXUAL ASSAULT-VICTIM BETWEEN 15 AND 17 YRS. OLD 8
Violent Crime 31 UNLAWFUL SEXUAL CONTACT 63
Violent Crime 32 VEHICULAR ASSAULT-DUI 16
Violent Crime 33 VEHICULAR HOMICIDE 10
DEG= Degree; AGG=Aggravated; DUI=Driving Under the Influence
From an initial data exploration, it was determined that this subset of 33 crimes was still
too large to find statistical significance, due to the large discrepancy in the total number of
crimes year to year (Figure 7). Many crime types had a very low total count, creating outliers in
the normal distribution of the data. Because of this loss of statistical significance, the top violent
37

crimes were used for analysis. It is necessary to note that there were two crime types tied in fifth
place with 240 total counts, Robbery and First-Degree Aggravated Motor Vehicle Theft. The
five crime types with the highest occurrence rates were: Robbery (26, N=240) and First-Degree
Aggravated Motor Vehicle Theft (5, N=240), Aggravated Robbery (16, N=273), Second Degree
Aggravated Motor Vehicle Theft Value $1000 to $20,000 (12, N=472), Second-Degree Assault
(13, N=790), and Second-Degree Burglary (9, N=1399) (Figure 8).

Figure 7-Number of Crimes by Category

Figure 8-Total Counts of Top Violent Crime Types
127
19
38
3
240
138
87
13
1399
88
15
472
790
60
12
273
119
2
44
10 15
37
11 1 4
240
42
1
44
8
63
16 10
0
200
400
600
800
1000
1200
1400
1600
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
Number of Crimes
Crime Identification Number
0
200
400
600
800
1000
1200
1400
1600
9-2ND DEG
BURGLARY
13-2ND DEG
ASSAULT
12-2ND DEG
AGG MOTOR
VEHICLE
THEFT
16-AGG
ROBBERY
26-ROBBERY 5-1ST DEG
AGG MOTOR
VEHICLE
THEFT
Number of Crimes
Crime Type
38

It is necessary to define each of these crimes to understand why they are considered
violent crimes. The FBI defines Robbery as “the taking of or attempting to take anything of
value from the care, custody, or control of a person or persons by force threat of force or
violence and/or putting the victim in fear” (Federal Bureau of Investigation 2018). Aggravated
Robbery is, according to the Colorado Revised Statutes (CRS) 18-4-302:
when a person commits Robbery and also with a weapon with intent-if faced with
resistance- to kill, maim, or wound the person robbed; or knowingly wounds or
strikes another with a deadly weapon; or uses force, threats, or intimidation with a
deadly weapon and knowingly put another person in reasonable fear of death or
bodily injury; or represents that he or she is armed with a deadly weapon or
possesses an object meant to look like a weapon. (2019)

First-Degree Aggravated Motor Vehicle Theft is, according to CRS 18-4-409:

A person commits Aggravated Motor Vehicle Theft in the first-degree if he or she
knowingly obtains or exercises control over the motor vehicle of another without
authorization or by threat or deception and retains possession or control of the motor
vehicle for more than twenty-four hours; or attempts to alter or disguise or alters or
disguises the appearance of the motor vehicle; or attempts to alter or remove or alters or
removes the vehicle identification number; or uses the motor vehicle in the commission
of a crime other than a traffic offense; or causes five hundred dollars or more property
damage, including but not limited to property damage to the motor vehicle involved, in
the course of obtaining control over or in the exercise of control of the motor vehicle; or
causes bodily injury to another person while he or she is in the exercise of control of the
motor vehicle; or Removes the motor vehicle from this state for a period of time in excess
of twelve hours; or unlawfully attaches or otherwise displays in or upon the motor vehicle
license plates other than those officially issued for the motor vehicle. (2019)
Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000 is, according to
CRS 18-4-409 (2019) “as knowingly obtain or exercise control over someone else’s car, truck,
motorcycle or other vehicle either without authorization or by threat or deception and the
aggravating factors of First-Degree Aggravated Motor Vehicle Theft are not present in the act of
committing the crime.” Second-Degree Assault is, according to CRS 18-3-203:
a person "intentionally" or "recklessly" causes bodily injury to another by means
of a deadly weapon, with intent to cause bodily injury, you cause serious bodily
39

injury, cause bodily injury to anyone while intentionally trying to prevent a police
officer or firefighter from doing their duties, "knowingly" apply "physical, violent
force" to a police officer, firefighter, prison guard, or a judge while they are in the
performance of their duties, or intentionally drug someone without their consent.
(2019)
Lastly, CSR 18-4-203 (2019) defines the crime of Second-Degree Burglary as “if the
person knowingly breaks an entrance into, enters unlawfully in, or remains unlawfully after a
lawful or unlawful entry in a building or occupied structure with intent to commit therein a crime
against another person or property.” These crimes all have a possibility to be considered violent,
or they are directly a violent crime such as assault. Because of how a person can be charged with
any of these six crimes, for the purposes of this analysis all six are considered violent crime.
3.1.1.1 Geocode
The Excel spreadsheets of crime data were imported into Esri’s ArcGIS Pro and a spatial
reference was added using the Geocode tool with in ArcGIS Pro. The original crime data only
contained physical address locations of crime (Figure 9), which was composed of street address
and zip codes, which served as a reference point for the Geocoding. The Geocode tool adds a
spatial reference to data that previously had no coordinates to do so and changes physical
addresses to X and Y locations. To successfully add latitude and longitude to the crime data it
was necessary to build an “Address Locator File.” This address locator served as the reference
point for the Geocoding. There are many approaches to creating the address locator, for this
analysis it was determined that the “US Address-One Range” style would be best, because this
type of address locator would be the correct match with the information contained in the violent
crime data. The Geocoded addresses for Pueblo were available from Pueblo’s GIS department’s
web portal. Once this file was downloaded it was imported into the File Geodatabase. The
40

Address Locator file was created from the Geocoded address for the city, by running the “Create
Address Locator” tool.
To geocode the crime data the “input table” was selected as the top violent crimes CSV
sheet, the “input address locator” was selected as the Pueblo Address (created previously), the
“input address fields” were selected as single field and the FULLADDR attribute from the crime
data was selected, and then the tool was ran. Results were returned with an accuracy rate of 76%,
leaving roughly 700 addresses that needed to be manually located and given X and Y
coordinates. This manual process was completed with the “Rematch Address” tool, which allows
the user to go to the map and manually place points at their proper locations. Google Maps was
used to reference the unmatched address for location accuracy; the unmatched address would be
looked up on the Google system then a satellite image base map was loaded in ArcGIS Pro. Once
the base map was loaded the location on Google Maps would be found and finally a point would
be added in ArcGIS Pro for that unmatched point. This process was done for all the unmatched
address points. An example of the pre and post geocoded data can be viewed in Figures 9 and 10.
41

Figure 9-Crime Data CSV Sheet

Figure 10-Geocoded Crime Data Displayed in ArcGIS Pro
3.1.2 Housing Data
Housing data for Pueblo was collected from the Pueblo Counties Economic Development
and Geographic Information Systems (EDGIS) online interactive mapping system (IMS). This
system was accessed through Pueblo County’s web page where the Pueblo County Assessor has
property and parcel information, including several housing value options. The four variables that
were of use to this project were Land Assessed Value, Land Actual Value, Improvements
Assessed Values, and Improvements Actual Values. These variables were determined to be the
only values needed because they represent the housing values. The difference between the
42

assessed and actual land and improvements are that the assessed value is what the property tax is
based on and the actual values are what the property would sell for. For the purposes of this
project the sale value was decided on because it represents the SES better than tax value.
The areas for which data were needed were determined by the results of the hot spot
analysis results for crime (Section 4.3). Once assessed, the hot spot 25m by 25m grid was
overlaid on the IMS map of Pueblo and two hot spot areas for each crime type were determined.
Major cross streets were used as boundaries in the IMS for selecting housing value areas and
from the aerial map individual city blocks were identified within the hot spot of the given crime
type. Each area of housing data contained four individual grid cells from the hot spot. Once these
areas were determined the IMS was used to select housing value data by city blocks. After this
data was selected a CSV file was created for each set of 4 grid cell hot spot areas. At this time
areas that were not statistically significant for hot spots were determined as well and a sample of
housing values for these areas was taken for reference.
The IMS provided all of the data for each city block in an exportable CSV file as well as
a web-based view (Figures 11 and 12), these files were the data used for analysis.

Figure 11-Web-Based Data View of Housing Variables
43

Figure 12-Example of Collection of Housing Data Through the Pueblo County Assessors
Interactive Map

After the data for each city block within the hot spot was collected and downloaded as a CSV
file, they were combined into one file. This was done for two hot spots within each crime type,
for a result of ten sample areas for which housing value data was collected. These CSV files
were then imported in to ArcGIS Pro to be visualized.
Historical housing values were also collected at this time. To collect the historical values
dating from 2006 to 2016 it was necessary to take a random 10% sample of the individual
properties in both the hot spots and the non-significant areas. The 10% threshold was decided
upon through the total number of properties in the dataset. This was completed by creating an
Excel spreadsheet with the parcel number and corresponding years property values. At this point
the historical housing data was analyzed via histogram to show the year-to-year change. Next the
housing data was imported into ArcGIS Pro for visualization.
44

3.2 Data Visualization and Analysis
To determine the spatial pattern, if any, of crime distribution in Pueblo, first a Directional
Distribution was run and then Hot Spot Analysis was conducted. Several Directional
Distributions were created from the top violent crime point data to examine the distribution of
the crime instances year to year. These Directional Distributions were created using the
Directional Distribution (Standard Deviational Ellipse) tool within ArcGIS Pro. This tool
created individual ellipses for each year (2006-2010) and were displayed on the base map. Each
year was represented by different colored ellipses to allow analysis of the distribution of violent
crimes. The data for the hot spot analysis was processed through ArcGIS Pro’s Hot Spot Tool.
First it was necessary to ensure that the data was complete and had spatial relationships. This
was accomplished by first running directional distributions on the violent crime data. These
directional distributions showed a spatial relationship between each year of violent crime data.
Next it was necessary to create a grid system to divide the study area into equal areas, this was
done with the “Create Fishnet” option within ArcGIS Pro. The “Create Fishnet” operation
requires several inputs from the user. There are several different ways to create this grid, first is
to determine the origin coordinates and then pick the cell size. A second option is to pick the
number of columns and rows based on the current map projection. Several row and column
combinations were tried and finally a 25m by 25m cell size was decided on based on the size of
Pueblo. After the Fishnet was created the crime points were spatially joined to the fishnet for
further analysis. This was done by copying the violent crime point file and spatially joining it to
the Fishnet. Spatial joins work by taking the target feature, in this case the Fishnet layer, and
“joining” it to another layer through the “join features” option which would be the crime point
layer in this case, lastly a one-to-one join was used for the crime data. The one to one join was
45

used because the locations of the crimes are based on single address points within the Fishnet.
The address points are X and Y coordinates within the Fishnet to allow for the join to happen.
The hot spot operation within ArcGIS Pro requires an input class- the joined fishnet, an input
field- Join_Count (the total number of incidents that are contained within each cell of the
fishnet), the output file name, the distance method (which was left as the default “Euclidean”,
and the conceptualization of Spatial relationships (also left as the default “Fixed Distance
Band”). The final step in the hot spot analysis was to run the hot spot operation within ArcGIS
Pro.
The hot spot operation analyzes hot and cold spots within the created fishnet grid. These
hot and or cold spots represent statistically significant areas in the data through clustering of
incidents, which in this case would be the individual violent crime locations. There were no cold
spots in the analysis preformed on any of the years or crime locations, this will be discussed
further in Chapter 5.
In ArcGIS Pro, the housing value data was joined to the previously used address point layer
to give the housing values a spatial reference. The two files were joined on the parcel number
attribute. This join allowed for the housing data to be visualized. The final step to this process
was to overlay the hot spots with the housing data to visualize the locations of both the hot spots
and the housing values.

46

Chapter 4: Results
After examining the locations of violent crimes in Pueblo, it was clear that there were hot spots
of crimes. Housing values were gathered for select hot spots for each crime type to determine the
“economic value” of the areas. It was found that the houses in these areas hold much less value
than that of the median home value for Pueblo as a whole. These housing variables aided in
determining what would be considered high and low economic areas for the city.
4.1 Violent Crime Temporal Trends
Initial exploration of the dataset showed a general upward trend in all violent crimes
2006-2016 (Figure 13). The trend shows that the top four violent crime types have a percent
change of more than 200% with First-Degree Aggravated Motor Vehicle Theft having a rate of
change at 816% and Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000 at
757%. The bottom two types have a much lower rate of change with Robbery decreasing by 4%.
These large increases in crime rates beg the question as to where they are happening, and
ultimately a possibility of why they are happening.

Figure 13-Violent Crimes by Year

0
100
200
300
400
500
600
700
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
Total Number of Violent
Crimes
Years 2006-2016
Number of Violent Crimes
47

Second-Degree Burglary had the highest count of incidents over the 11 year time frame.
With a count of 1399 incidents across the city Second-Degree Burglary was reported zero times
in 2006 and 2007; and increased year over year with a large jump between 2009 and 2010
increasing from 88 to 135 instances. The next several years show a slow increase until 2013 to
2014 which jumped from 167 to 233 instances of this crime (Figure 14). The following two years
saw a slight decrease to 231 and 205 respectively.
The second highest violent crime type for Pueblo from 2006 to 2016 was Second-Degree
Assault with a total count of 790. This violent crime was extremely low in 2006 and 2007 with
only 4 and 8 instances respectively. 2008 witnessed an exponential jump to 58 instances
followed by 84 for 2009. In 2010 there is a small increase to 94, followed by a drop to 78 in
2011. 2012 to 2016 the instances are as follows 90, 101, 93, 88, and 92 (Figure 14).
Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000 is the third
highest violent crime with a count of 472 instances over the 11-year span. 2006 and 2007 both
saw zero reported instances of this crime. Second-Degree Aggravated Motor Vehicle Theft
Value $1000 to $20,000 was then reported 14 times in 2008 and 24 times in 2009. There was a
drop to 17 instances in 2010 and then again, an increase to 23 reported crimes in 2011. In 2012
there is a large jump to 40 followed by an increase every year after. In 2013 there were 64
instances, 2014 had 74, 2015 had 96, and finally 2016 witnessed a jump to 120 (Figure 14). It
should be noted here that Pueblo has not only had an Aggravated Motor Vehicle Theft problem
but a Motor Vehicle Theft problem as well. A 2017 report by the PPD shows that between 2013
and 2017 4,493 vehicles were stolen from in and around the city largely due to people leaving
their vehicles running on cold mornings (Second-Degree Aggravated Motor Vehicle Theft Value
$1000 to $20,000 is not included in these numbers) (City of Pueblo Police Department 2018).
48

Aggravated Robbery is fourth, with total instances of crime at 273 with 2006 and 2007
having zero instances. Starting in 2008 there were 10 crimes of this type followed by 23 in 2009,
a decrease to 21 in 2010, back up to 26 in 2011, 29 in 2012, 39 in 2013, 40 in 2014, down three
to 73 in 2015, and up to 48 in 2016 (Figure 14).
Robbery and First-Degree Aggravated Motor Vehicle Theft tie for lowest in the rankings
of violent crime for Pueblo, each with a total instance count of 240.
Overall the top violent crimes for Pueblo fluctuated in number of instances year to year,
with some violent crimes increasing much more than others. When looking at the percent change
for the top violent crimes for the City of Pueblo from 2006 to 2016 there is an interesting trend
that comes forward, there is a large increase in three, and moderate increase in one, and a
decrease in one. The breakdown of the crimes is as follows: First-Degree Aggravated Motor
Vehicle Theft 816%, Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000
757%, Aggravated Robbery 380%, Second-Degree Burglary 205%, Second-Degree Assault
58.6%, and Robbery -4%.
49

Figure 14-Top Violent Crimes by Year

The following sections visualize the data to display the results from the spatial analysis.
Section 4.1 examines the actual locations of the top violent crimes. Section 4.2 shows the results
of the Directional Distribution of the top violent crimes for 2006-2016. Section 4.3 examines the
results of the hot spot analysis. And lastly Section 4.4 shows the results of overlay of the crime
hot spots and housing values.
Figure 15 displays the top violent crime locations for the years 2006 to 2016. Looking at
this figure it can be deduced that violent crime happens across the city. However, once broken
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95
100
105
110
115
120
125
130
135
140
145
150
155
160
165
170
175
180
185
190
195
200
205
210
215
220
225
230
235
240
245
250
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016
9-2ND DEG BURGLARY
13-2ND DEG ASSAULT
12-2ND DEG AGG MOTOR VEHICLE THEFT
16-AGG ROBBERY
26-ROBBERY
5-1ST DEG AGG MOTOR VEHICLE THEFT
50

down into individual crime types these crimes seem to be further separated and clustered in
different areas. This clustering will be discussed in further detail in Section 4.3. These crime
points represent individual crime locations over an eleven-year period within the city limits of
Pueblo, CO. The city limits are broken down further into PPD Quadrants, which have been
discussed in detail in Section 1.1.1 of this thesis. Figure 15 is a map of PPD quads and all violent
crime locations and does not show any clustering or spatial patterns. This map has been included
in this text for display purposes only.

Figure 15-Top Violent Crime Locations for 2006-2016 with PPD Quadrants

4.2 Directional Distribution
The Directional Distributions for the violent crimes (Figure 16 and 17) in Pueblo do not
indicate much change year to year with the exception to Second-Degree Assault and Robbery.
Second-Degree Assault seems to constrict in the latter years of the study, (2006 and 2007 are
51

outliers when examined visually). 2006 is very much in the middle of the distribution, one cause
of this may be the low number of instances that occurred in that year, also the locations of these
crimes are very much centered in the downtown area of Pueblo. 2007’s difference can also be
explained by the low instances of crimes; however, these crimes were much more spread across
the city. Robbery sees a shift to the north in 2014, where the previous and later years see a much
tighter and similar distribution throughout the city. Aggravated Robbery also shows a slight shift
to the north in its distribution, much the same as Robbery. The shift in both of these crime types
moves in the direction of the major shopping area located on the Northside of Pueblo.

Figure 16-Directional Distribution of Crimes by Type and by Year a) All Violent Crime, b) Top
Violent Crimes, c) Aggravated Robbery, and d) Second-Degree Aggravated Motor Vehicle Theft
Value $1000 to $20,000
b) a)
c) d)
52

Figure 17-Directional Distribution of Crimes by Type and by Year: a) Second-Degree Assault, b)
Robbery, c) Second-Degree Robbery, and d) First-Degree Aggravated Motor Vehicle Theft

4.3 Hot Spot Analysis
The hot spot analysis shows areas that have the highest accumulations of violent crime.
There are two areas of the city that stood out from the rest, specifically the Downtown/Eastside
and Central/South area of Pueblo which fall into several police quadrants. There is one hot spot
that falls outside of these areas, which is around a shopping area. For the purposes of this thesis,
only the areas that fall within residential areas were examined in conjunction with the housing
values. This decision was made because the property values in these areas are not home values,
they are commercial properties which for the purposes here do not indicate SES. In Section 4.4
this housing variable will be discussed in more detail.
a)
b)
c) d)
53

The first hot spot examined is the total of the top violent crimes for 2006-2016 (Figure
18) and shows several hot spots throughout the city. These hot spots cross police quads and are
spread across the city. There are several areas that need to be noted. First the two small clusters
that are in quad one and partially into quad two are in the north of town. These areas contain
mostly commercial buildings and shopping areas, although perpetrators of these crimes may live
in close proximity, there is no way of knowing this and therefore no way of linking criminals to a
specific area. One could potentially examine surrounding housing values; however, some would
argue that this does not test SDT, and is beyond the scope of this thesis. The next hot spot splits
three police quads (1, 2, and 3) and is located in the Central/Downtown area as well as the east of
town, these areas contain some commercial properties but are primarily residential. The last area
of hot spot was located in the Southwest area of Pueblo, again this area contains some
commercial properties but is mostly residential. It is then necessary to look at each violent crime
type individually, which the following figures provide. It should be noted that the violent crime
hot spot analysis did not produce any cold spots for the study area. This is because the lack of
low negative Z-scores and small P-values within the data. The Z-scores for the hot spots showed
high values, indicating clustering of crimes in these areas. Again, the lack of low Z-scores
indicate that there are no cold spots. An example of the P-values and Z-scores were taken from
the top violent crime types. The Z-score mean for this hot spot was 0.011718, a median of
0.702828, with a standard deviation of 1.421116. The Z-score had a high of 0.509096 with a
count of 386 incidents of crime within the hotspot, the low for the Z-score was 6.667434. As for
the P-value for the same top violent crime types had a mean of 0.430262, a median of 0.419023,
with a standard deviation of 0.215471.
54

Figure 18-Top Violent Crimes Hot Spot Analysis for 2006-2016

The first violent crime that was analyzed was Second-Degree Burglary. There were many
hot spots in the 90% confidence range throughout the city, these we are less concerned about, the
99% confidence areas are where further analysis was performed. Hot spots that fall within the
99% range are the most statically significant of the group. For this particular hot spot map two
areas within police quads two, three, and four were used as an index for where housing value
data would be collected (discussed further in Section 4.4). This analysis provided interesting
analysis, initially it was hypothesized that this crime would have hot spots in the 99% confidence
range in mostly residential areas. However, the hot spots were split between both residential and
commercial areas. The largest concentration of hot spots did however fall within a residential
area, which would ultimately be used to gather housing data. This hot spot analysis is visualized
in Figure 19.
55

Figure 19-Second-Degree Burglary Hot Spot Analysis for 2006-2016

The next crime type analyzed was Second-Degree Assault. The results were spread
across the Downtown and Eastside of Pueblo as well as the South-Central area. These areas
contain a large number of bars and drinking establishments mixed into residential areas. This
crime is spread across all 4 police quadrants (Figure 20).
56

Figure 20-Second-Degree Assault Hot Spot Analysis for 2006-2016

Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000 is the next
violent crime examined, and this crime type by far had the most hot spots at the 99% confidence
throughout the city. The majority of the cells that are displayed fall into the 99% range, with only
one cell falling below at a 95% range. This crime happened within all 4 police quadrants and saw
a very high concentration in the Eastside of town. This hot spot analysis is displayed in Figure 21
located below.
57

Figure 21-Second-Degree Aggravated Motor Vehicle Theft Value $1000 to $20,000 Hot Spot
Analysis for 2006-2016

First-Degree Aggravated Motor Vehicle Theft was the next violent crime analyzed with
the hot spot function. This crime was spread across all 4 police quadrants, with areas clustered on
the Eastside of Pueblo all in the 99% range. One thing of note with this analysis is the cluster of
four grid cells in quadrant one, this area is home to the majority of auto dealerships in Pueblo.
Making it a prime area for this type of crime to occur.
58

Figure 22-First-Degree Aggravated Motor Vehicle Theft Hot Spot Analysis for 2006-2016

The next two crime types are Aggravated Robbery and Robbery (Figures 23 and 24).
These crime type hot spot analyses provided similar results to many of the hot spots at the 99%
confidence range within commercial and shopping areas. There were a few hot spots that do fall
within residential areas. Because of the small number of hot spots within residential areas no
housing values were collected for these two crime types. It was determined that because of the
value of commercial buildings and the fact that SDT does not center around commercial property
as a variable that these areas would not have data collected on property value. Overall the hot
spot analysis provided insight as to where and which violent crimes were statistically
significantly clustered in Pueblo, Colorado.

59

Figure 23-Aggravated Robbery Hot Spot Analysis for 2006-2016

Figure 24-Robbery Hot Spot Analysis for 2006-2016
60

4.4 Crime Hot Spots and Housing Values
Housing values are a key variable for this thesis, as they represent the SES variable
within SDT and act as an indicator for poverty. For the purposes of this thesis the hot spots
discussed in Section 4.3 were used as a designation for where to collect the housing value data.
For each crime type four adjoining grid cells were selected for two hot spots, and the 2016
housing values for individual properties were selected for these areas. Figure 25 is an example of
one cell of the hot spot where housing values were collected. Figures 26 through 29 provide a
visualization of the hot spots overlaid with the collected housing data. It should be noted that two
of the violent crimes, Second-Degree Assault and Robbery, do not have a visualized overlay
because of the lack of hot spots that were located in residential areas, this is because the majority
of the hot spots contained commercial areas. Also, Aggravated Robbery only has one section of
housing values because a second hot spot of four cells was not available for selection primarily
because many of these hot spots fall within commercial areas.

61

Figure 25-Example of Housing Values for a Hot Spot in Pueblo, CO one Cell of the Fishnet Grid

Figure 26-Housing Values Overlaid with Hot Spot Analysis of the Top Violent Crimes
62

Figure 27-Housing Values overlaid with Hot Spot Analysis of Second-Degree Burglary

Figure 28-Housing Values Overlaid with Hot Spot Analysis of Second-Degree Aggravated
Motor Vehicle Theft Value $1000 to $20,000
63

Figure 29-Housing Values Overlaid with Hot Spot Analysis of Aggravated Robbery

After gathering the housing values for two hot spots for each crime it was necessary to
look at the historical value of the homes in these areas. This was accomplished through the use of
Pueblo County’s IMS. A 10% sample of properties were gathered from an area that all five crime
types shared hot spots or were very close to one another. This sample was comprised of 50
individual properties, from these properties a value for 2006, 2008, 2010, 2012, 2014, and 2016
were taken. After all the values had been collected, each year was averaged to build the graph in
Figure 30. The results provided the evidence to show that housing values decreased in the same
time frame that violent crimes increased.
In contrast to the decreasing home values in hot spot areas, home values in non-
statistically significant areas increased. The same process was followed for the non-statistically
significant areas. The historical home values were collected, summed, and averaged to provide a
graph that indicates the increase in home value. This graph is provided as Figure 31. The graph
64

shows that home values did decrease over the same time, but the averages show an overall
increase in value. Also, the homes in the non-statistically significant areas have a much higher
overall value than the homes in the hot spot areas. As shown in both figures the highest price of a
home in a crime hot spot area is approximately $82,000 in 2008 where a home in a non-
statistically significant area is much higher at approximately $158,000. This gap is even more
profound if we look at the 2016 home values of both areas, where we see a hot spot area with a
value of approximately $64,000 and a non-statistically significant area with a value of
approximately $150,000 over double the value of a home in a hot spot area. In the non-
statistically significant areas we also see a large rise in value from 2012 to 2016 where the hot
spot area sees nowhere near as large of jump in value.

Figure 30-Average Housing Value for a Sample Hot Spot Area of Pueblo, Colorado
$0.00
$10,000.00
$20,000.00
$30,000.00
$40,000.00
$50,000.00
$60,000.00
$70,000.00
$80,000.00
$90,000.00
2006 2008 2010 2012 2014 2016
Home Value
Year
Average Price of Homes
Linear (Average Price of
Homes)
65

Figure 31-Average Housing Value for Sample Non-Hot Spot Area of Pueblo, Colorado

From the housing value analysis preformed on the hot spots it can be deduced that
housing values decreased, and violent crimes increased. The results of both the hot spots and the
home values provided that SDT is a theory that can be applied to Pueblo, CO when analyzing
crime.
4.5 Summary Findings
This thesis set out to answer three questions about crime and housing value within Pueblo. These
questions were: 1) have crime rates changed over time? 2) do the changes in crime rates have a
spatial pattern? and 3) does the change in crime rates mirror housing values? The following will
explain how these questions were answered.
$130,000.00
$135,000.00
$140,000.00
$145,000.00
$150,000.00
$155,000.00
$160,000.00
2006 2008 2010 2012 2014 2016
Home Value
Year
Average Price of Homes
Linear (Average Price of
Homes)
66

Have crime rates changed over time in Pueblo? Yes, they have as viewed in Figure 14,
there is an increase in crime from 2006 to 2016. Do the changes in crime rates have a spatial
pattern? Yes, the crimes did show a spatial pattern, which can be viewed in Figures 18-24. And
lastly, does the change in crime rates mirror housing values? This question is a bit harder to give
a definite yes or no. Housing values that were sampled in both areas of crime hot spots and areas
that were not hot spots for crime show volatility over the 2006-2016 time frame. There is not a
direct trend of decreasing values year to year, for example for homes in the hot spot area during
2008 values increased, during 2008 violent crime also increase. However, the following years
2009-2016 did show decreases in value year to year subsequent to and in conjunction with an
increase in the number of occurrences of violent crime, but also a leveling off of the housing
values from 2014 to 2016. After taking this into account, the answer to question 3 is more
difficult to answer and will be further explored in the next chapter.
67

Chapter 5: Discussion and Conclusion
5.1 Discussion
SDT is a theory that has been applied to the study of crime for quite some time. It has been tested
through studies performed by Sampson and Groves (1989) as well as others such as Kubrin and
Weitzer (2003) more recently. These studies have concluded that SES has some effect on crime
rates. This thesis used the City of Pueblo, Colorado as a study area to apply SDT through the
analysis of violent crime in conjunction with housing values for areas affected by violent crime
to address three questions: 1) how have crime rates changed over time? 2) do the changes in
crime rates have a spatial pattern? and 3) does the change in crime rates mirror housing values?
This chapter will cover the conclusions that were found through the analysis and study of SDT in
Pueblo when examining violent crime in relation to housing values (as an indicator of SES).
Also, a section is devoted to future research and analysis, and lastly a section on closing thoughts
from the author.
5.1.1 Hot Spot Analysis and Social Disorganization Theory Conclusions
This study found that areas experiencing higher incidents of violent crime, including
Second-Degree Burglary, First-Degree Aggravated Motor Vehicle Theft, Second-Degree
Aggravated Motor Vehicle Theft Value and Aggravated Robbery are also areas with lower
housing values. Hot spots of these three crime types, on the Eastside and Central Southside, are
mainly in PPD Quadrants 2 and 3. Additionally, the instances of these crimes increased over the
study period of 2006-2016 and the housing values for these areas declined in the same time
period. The decline of housing values for areas that show hot spots of violent crime contrast
areas that are non-hot spots. The housing values in the sampled non-hot spot areas increased in
68

value over the eleven-year time span. Even though these housing values did see an initial decline
in the years following the Great Recession (Figure 31), overall the trend shows an increase in
value.
Social disorganization theory provided the foundational framework for this thesis through
the use of socioeconomic status via housing values as an indicator of areas that would be
considered low on the socioeconomic ladder (Coffee et al. 2018). SDT was determined to be the
best criminological theory to use because of its history of using early GIS to help explain why
crime happens. This theory was applied to Pueblo, Colorado and the findings add credence to the
theory. When examining the housing values for the sampled hot spot areas, there is an overall
decrease (Figure 30). This trend appears to follow the increase in violent crime in these areas
(Figures 14). However, there is not enough evidence to parse out the causality and directionality
of this relationship, and to affirmatively state that lower SES (as expressed through housing
values) led to an in increase in violent crime in an area, or vice versa. There is a relationship
between the two even if it is a simple location-based relationship, which had not previously been
shown at this scale for Pueblo, Colorado.
Additionally, during this time there was a down turn in the economy (the 2008 recession),
which could have had an effect on certain housing values with already lower price homes
dropping in value faster and at relatively higher percentages than others. Overall, further study is
needed to parse out the spatial influence of the 2008 recession, and lag time (if any) between the
decreasing housing values and increasing violent crime occurrences. Therefore, it cannot be
determined at this time whether low housing value had an effect on violent crime in Pueblo.
Without further analysis, it cannot be fact that housing value and crime are causally connected
and which is the driver.
69

5.2 Future Research and Analysis
This thesis explored data and analysis techniques that could be further utilized to test
SDT and crime in the City of Pueblo. It is suggested that future research be done on all crimes in
Pueblo, including non-violent and property crimes using the similar methodology that was used
for this thesis, it should also be noted that in future research the crime counts in an area could be
normalized by population. This normalization could be performed using United States Census
Block groups as the level of spatial analysis. Using block groups would provide the tools needed
to normalize the crime instances by population, providing a more detailed view of the rate of
crime per capita. Additionally, a deeper examination of nonviolent crimes, including property
crime, could be incorporated.
Given that the housing values had to be individually acquired, only sample areas were
analyzed. Working with the assessor’s office, one could possibly attain all necessary data by
parcel thereby allowing for analysis on across all of Pueblo. Also, it is suggested that additional
economic variables and demographic variables be added. This thesis used only housing values as
an indicator of economic status, therefor the results were limited. This limitation can be avoided
by showing that the areas that have high crime rates also exhibit other indicators of low SES. The
demographics suggested include education, median household income and individual income,
employment rates, incarceration rates, recidivism rates, and school attendance rates. These
variables could be incorporated into a similar analysis determine if the violent crime hot spots
are also hot spots for these variables.
Lastly, one other area of examination that would provide some insight into why Pueblo
has a high violent crime rate would be to look at how the legalization of recreational marijuana
has affected the violent and nonviolent crime instances. One could examine various community
70

structures such as access to after school programs, church attendance, and other social activities
that are available for the youth and young adults throughout the community. These areas of study
would all work well for the methodology used for this thesis and would add a benefit to the
defense that social disorganization theory can be applied to Pueblo, CO. These topics would also
have a positive impact on the community and how the police interact with the citizens.
5.3 Closing Thoughts
Pueblo, Colorado is a place that the author of this thesis calls home, so this project was
very close to his heart. In an attempt to aid the city in looking at crime through a new lens he
feels that more questions were created than answered. With that being said, he does feel that
identifying the areas where violent crime has happened in the past can pave a path to identifying
a solution to lowering these types of crime. Through future study in the areas that were
mentioned in Section 5.2 the community of Pueblo could see a positive impact and be better able
to design policy and ordinances to benefit the positive growth of the city and its population. It is
the opinion of the author that if the City of Pueblo increased economic opportunity for its
citizens through economic development and attracting higher paying jobs, there would be a drop-
in crime rates.
Social disorganization theory helps to identify the problem of why crime happens, and it
is a theory that city government and law enforcement can apply to the City of Pueblo. Applying
this theory in conjunction with the variables mentioned above in Section 5.2, it may be possible
to address the underlying reasons for the high crime rates in Pueblo and thus find a solution to
lowering it permanently through economic and social development programs.

71

References
Ahmadi, Mostafa. 2003. "Crime Mapping and Spatial Analysis." Master’s thesis., International
Institute for Geo-Information Science and Earth Observation, Netherlands.
American Factfinder. 2017. [online] Factfinder.census.gov. Available at:
https://factfinder.census.gov/faces/nav/jsf/pages/index.xhtml.
Broadhead, Wade. 2019. [online] "History of Pueblo | Pueblo, Colorado." Pueblo.org. Available
at: http://pueblo.org/History.
Buonanno, Paolo, Daniel Montolio, and Josep Maria Raya-Vílchez. 2012. "Housing Prices and
Crime Perception.” Empirical Economics 45 (1): 305-321.
Bureau of Justice Assistance 1994. Understanding Community Policing: A Framework for
Action. Washington DC: Bureau of Justice Assistance Response Center. pg. 3.
Chainey, Spencer, Lisa Tompson, and Sebastian Uhlig. 2008. "The Utility of Hot Spot Mapping
for Predicting Spatial Patterns of Crime.” Security Journal 21 (12): 4-28.
City of Pueblo Police Department 2018. 2017 Motor Vehicle Theft Report. Pueblo, Colorado:
Pueblo Police Department.
Coffee, Neil T, Tony Lockwood, Peter Rossini, Theo Niyonsenga, and Stanley McGreal. 2018.
“Composition and Context Drivers of Residential Property Location Value as a
Socioeconomic Status Measure.” Environment and Planning B: Urban Analytics and
City Science (October).
Colorado Department of Public Safety. 2018. Colorado Crime Statistics. [online] Available at:
https://coloradocrimestats.state.co.us/tops/.
Colorado Revised Statutes. 2019. [online] Available at:
https://advance.lexis.com/container?config=0345494EJAA5ZjE0MDIyYy1kNzZkLTRk
NzktYTkxMS04YmJhNjBlNWUwYzYKAFBvZENhdGFsb2e4CaPI4cak6laXLCWyLB
O9&crid=7a1f2831-0023-4461-9ad3-e25e68ea3663
Coser, Lewis A. 1957. "Social Conflict and the Theory of Social Change.” The British Journal of
Sociology 8 (3): 197.
DeMeester, Deborah S., Donald R. LaMagdeleine, and Cari Norton. 2011. The Rebirth of
Community Policing: A Case Study of Success. Minneapolis, MN: Columbia Heights
Police Department.
72

Eck, John E., Spencer Chainey, James G. Cameron, Michael Leitner and Ronald E. Wilson.
2005. Mapping Crime: Understanding Hot Spots. Washington, DC: Office of Justice
Programs National Institute of Justice.
Esri. 2018a. “Kernel Density.” [online] Available at: https://pro.arcgis.com/en/pro-app/tool-
reference/spatial-analyst/kernel-density.htm
--- 2018b. “How Hot Spot Analysis (Getis-Ord Gi*) Works.” [online] Available at:
https://pro.arcgis.com/en/pro-app/tool-reference/spatial-statistics/h-how-hot-spot-
analysis-getis-ord-gi-spatial-stati.htm
Ezez. 2019. "Overview - EVRAZ North America." Evrazna.com.
http://evrazna.com/Company/Overview/tabid/57/Default.asp.
Federal Bureau of Investigation. 2018. Uniform Crime Reporting (UCR) Program. [online]
Available at: https://www.fbi.gov/services/cjis/ucr.
Getis, Arthur, and Keith Ord. 1992. "The Analysis of Spatial Association by Use of Distance
Statistics.” Geographical Analysis 24 (3): 189-206.
Hickenlooper, John. 2018. Pueblo Region | Colorado Office of Economic Development and
International Trade. [online] Available at: https://choosecolorado.com/regional-
profile/pueblo-region/.
Hirschi, Travis. 1969. Causes of Delinquency. Berkley, CA: University of California Press.
Ihlanfeldt, Keith, and Tom Mayock. 2010. "Panel Data Estimates of the Effects of Different
Types of Crime on Housing Prices.” Regional Science and Urban Economics 40 (2-3):
161-172.
Jacober, Cpl. David. 2018. "Pueblo Policing." Corporal David Jacober Interview by Dane
Cornell. In person. Pueblo Ice Arena.
Kornhauser, Ruth. 1978. Social Sources of Delinquency: An Appraisal of Analytic Models.
Chicago, IL: University of Chicago Press.
Kubrin, Charis E., and Ronald Weitzer. 2003. "New Directions in Social Disorganization
Theory." Journal of Research in Crime and Delinquency 40 (4): 374-402.
Markowitz, Fred E., Paul E. Bellair, Allen E. Liska, and Jianhong Liu. 2001. “Extending Social
Disorganization Theory: Modeling the Relationships Between Cohesion, Disorder, and
Fear” Criminology 39 (2): 293–319.
Markus, Ben. 2014. "Decades Later, Pueblo Still Reeling from Steel Exodus." Colorado Public
Radio https://www.cpr.org/news/story/decades-later-pueblo-still-reeling-steel-exodus.
73

Mannheim, Karl. 1935. Collected works of Karl Mannheim. London, UK: Routledge.
National Criminal Justice Reference Service. Department of Justice. Office of Justice Programs.
2019. [online] Available at: https://www.ncjrs.gov/
Park, Robert E., and Ernest W. Burgess. 1925. The City. London, UK: The University of
Chicago Press.
Rees, Jonathan H. 2017. "Colorado Fuel & Iron." Colorado Encyclopedia last modified
December 28, 2017, http://coloradoencyclopedia.org/article/colorado-fuel-iron.
Sampson, Robert, and W. Byron Groves. 1989. “Community Structure and Crime: Testing
Social-Disorganization theory.” American Journal of Sociology 94 (4): 774-802.
Sampson, Robert J., Jeffrey D Morenoff, and Thomas Gannon-Rowley. 2002. "Assessing
“Neighborhood Effects: Social Processes and New Directions in Research." Annual
Review of Sociology 28 (1): 443-478.
Silver, Eric. 2000. "Extending Social Disorganization Theory: A Multilevel Approach to the
Study of Violence among Persons with Mental Illnesses" Criminology 38 (4): 1043-1074.
Shaw, Clifford R., and Henry D. McKay. 1969. Juvenile Delinquency and Urban Areas.
Chicago, IL: University of Chicago Press.
Sutherland, Edwin. 1947. Principles of Criminology. Philadelphia, PA: J.P. Lippincott.
Suttles, Geraldo. 1968. The Social Order of the Slum. Chicago, IL: Univ. of Chicago Press.
United Nations Office on Drugs and Crime. 2010. Monitoring the Impact of Economic Crisis on
Crime. Vienna, Austria: UNODC Statistics and Surveys Section.
United States Department of Labor. Bureau of Labor Statistics. Colorado at a Glance. 2018.
[online] Available at: https://www.bls.gov/eag/eag.co.htm
US Legal, Inc. 2018. "Legal Definitions Legal Terms Dictionary | US Legal, Inc.”
Definitions.uslegal.com. https://definitions.uslegal.com/.
Vernez Moudon, Anne, Andrea J. Cook, Jared Ulmer, Philip M. Hurvitz, and Adam
Drewnowski. 2011. "A Neighborhood Wealth Metric for Use in Health Studies."
American Journal of Preventive Medicine 41 (1): 88-97.
Walker, Samuel. 1999. The Police in America. 3d ed. Boston, MA: McGraw-Hill College.
Williams, Frank P., and Marilyn D. McShane. 2004. Criminological theory. 6th ed. Columbus,
OH: Upper Saddle River.

Abstract (if available)

Abstract According to social disorganization theory, crime is caused by social and economic variables at the neighborhood level. Coined in 1942 by Shaw and McKay, their research utilized the city of Chicago as a natural laboratory to examine how social and economic variables affected crime. It was decided to test this hypothesis using Pueblo, Colorado because of the high crime rate. To test if the theory of social disorganization applies to Pueblo, violent crime and socioeconomic status were analyzed spatially to answer the following questions: 1) have crime rates changed over time? 2) do the changes in crime rates have a spatial pattern? and 3) does the change in crime rates mirror housing values? ❧ Data on violent crimes was determined with the assistance of the Pueblo Police Department, who provided the location of 4,500 individual violent crimes across the city from 2006 to 2016. Statistical analysis showed that many of the counts of individual crime types were too low to be statistically significant, so the five crimes with the highest occurrence were used for further analysis. Socioeconomic status was determined using the housing values within the City of Pueblo. ❧ Hot spot analysis using the GI* statistic, which uses high and low z-scores to determine clusters of high values (hot spots) and clusters of low values (cold spots), was used to determine statistically significant high crime areas within Pueblo. These hot spots were used to determine where housing values would be analyzed. Statistical analysis of the housing values within a hot spot showed that the value of a house was much lower than that of a house in non-hot spot areas. The value of houses in high crime areas decreased during the 10-year span, whereas the value of houses in areas with little to no crime increased in value over the same time frame.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Creating Hot Streets: developing an automated approach using ModelBuilder

PDF

The spatial effect of AB 109 (Public Safety Realignment) on crime rates in San Diego County

PDF

A spatial analysis of violent crime cold spots: testing the capable guardian component of the routine activity theory

PDF

A spatiotemporal analysis of environmental risk factors of Lyme disease in the Northeastern United States

PDF

Using aerial imagery to assess tropical forest cover surrounding restoration sites in Costa Rica

PDF

Obesity and healthy food accessibility: case study of Minnesota, USA

PDF

An exploration of the spatiotemporal distribution of snow crab (Chionoecetes opilio) in the eastern Bering Sea: 1982 – 2018

PDF

Spatial and temporal patterns of long-term temperature change in Southern California from 1935 to 2014

PDF

Evaluating the MAUP scale effects on property crime in San Francisco, California

PDF

Spatiotemporal hotspots of 2018-2020 crime in Houston, Texas

PDF

Wetland mapping and restoration decision making using remote sensing and spatial analysis: a case study at the Kawainui Marsh

PDF

Defining neighborhood for health research in Arizona

PDF

An accessibility analysis of the homeless populations' potential access to healthcare facilities in the Los Angeles Continuum of Care

PDF

Classifying the alpines: developing a methodology to track environmental changes in the alpines utilizing remote sensing and ecozone vegetation patterns

PDF

Using GIS to identify potential dynamic marine protected areas: a case study using shortfin mako shark tagging data in New Zealand

PDF

Harnessing GIST-enabled resources in the classroom: developing a Story Map for use with secondary students

PDF

The role of precision in spatial narratives: using a modified discourse quality index to measure the quality of deliberative spatial data

PDF

Suitable California Opportunity Zone (COZ) locations for affordable housing development in the cities of Bakersfield, Los Angeles, and Palmdale

PDF

Creating a flood vulnerability index for Houston, Texas

PDF

Risk assessment to wildlife from Ohio on-shore wind farm development: a landscape model approach

Asset Metadata

Creator Cornell, Dane Francis Stanley (author)

Core Title Testing social disorganization theory on violent crime: a case study on Pueblo, Colorado

School College of Letters, Arts and Sciences

Degree Master of Science

Degree Program Geographic Information Science and Technology

Publication Date 06/17/2019

Defense Date 03/03/2019

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Colorado,crime theory,OAI-PMH Harvest,Pueblo,social disorganization theory,violent crime

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Loyola, Laura (committee chair), Bernstein, Jennifer (committee member), Wilson, John (committee member)

Creator Email Dane.Cornell@gmail.com,dfcornel@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-174513

Unique identifier UC11660064

Identifier etd-CornellDan-7491.pdf (filename),usctheses-c89-174513 (legacy record id)

Legacy Identifier etd-CornellDan-7491.pdf

Dmrecord 174513

Document Type Thesis

Format application/pdf (imt)

Rights Cornell, Dane Francis Stanley

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA