2 28 11 SCTF Work Session_NIU Executive SummaryDubuque 2010 Study on
Crime & Poverty
Summary Report
Published January 14, 2011
for the City of Dubuque
Prepared by
Northern Illinois University
Center for Governmental Studies
Table of Contents
Executive Summary and Overview of the Summary Report 3
Summary of Study Scope and Research Plan 4
Organization & Approach to the Study and Report 5
Summary 73
Report Narrative 81
Background of the City 83
Overview of Recent Activity Contributing to the Current Study on Crime & Poverty 86
Summary of Study Scope and Research Plan 88
Organization of the Study and Methodology 89
Part I: Literature Review & Section 8 Assisted Housing Program Background 91
Section I -A: Literature Review 91
Section I -B: Section 8 Housing Program Background 117
Part II: Community Survey and Perception Analysis 131
Part III: Quantitative Analysis 188
Section III -A: Comparative & Trend Analysis of Dubuque's & Iowa's Department of Public Safety Data 188
Section III -B: Geo- coding: Mapping Incidents, Arrests and Housing Data 216
Section III -C: Quantitative Statistical Modeling, Spatial and Hot Spot Analysis 248
Loglinear Analysis of Section 8 Housing and Crime in Dubuque 248
Analysis of Areal Spatial Correlation of Crimes in Dubuque 270
Hot Spot Analysis 295
Summary Discussion of Findings & Implications for Strategies and Policies 317
2010 Dubuque Project Study Team 325
Appendix 2: Comparative Graphs of Crime 335
Appendix 2a: Comparative Graphs of Crime — Number of Incidents within Each Type of Crime 336
Appendix 2b: Comparative Graphs of Crime — Crime Rates within Each Category of Crime Type 342
Appendix 3: Documentation of Data Processing 370
Appendix 4: Documentation of Data Analysis Design 380
Appendix 5: Additional POL Perception Survey Information 385
Appendix 6: Maps and Graphs 482
Appendix 7: Literature Review 561
Appendix 8: Arrest & Incident Frequencies by Type 755
Appendix 9: September 2010 Statewide Comparison of Assisted Units 766
Appendix 10: Glossary 776
2010 Dubuque Crime and Poverty Study Summary Report Page 2 of 779 *' cenr ertor
IY�UGov Studies
kN b birch
Executive Summary and Overview of the Summary Report
In April of 2010, the City of Dubuque initiated the 2010 Quantitative Research Study on Crime
and Poverty (referred to as The Study going forward) in order to provide an objective assessment of
trends and perceptions of crime in conjunction with poverty, specifically poverty related elements
pertaining to Section 8 housing. Over 350,000 records from various departments related to the Study's
focus have been used in this analysis. The study team wishes to acknowledge the assistance and efforts
of City staff in providing the raw data to the researchers for analysis.
A study of this type and scope is highly dependent on two important elements: 1) the
technological software and hardware that make such an analysis feasible, and 2) the quality of the data
that is available. Absent the advanced technologies in use at both organizations, a study of this depth
and scope would be impractical if not impossible. Additionally, recognition of the limits of the analysis
and the limitations of source data must also be taken into account by the reader. The limitations of
specific analytical processes, and the raw data available, are noted in both the Executive Summary and
the Report Narrative. It should be further noted that the source data used in this analysis were working
"operational" data files and records that are not only fluid working documents, but are also data sets
that have been created and used by multiple users, including the transfer, over time, to new and
updated platforms.
Background
In an effort to specify and analyze the relationship between crime and poverty (specifically
Section 8 housing) through evidence -based information for decision - making, the City sought proposals
for a robust research methodology to study recent patterns of local crime, the community's perception
of crime, and the relationship (if any) of crime to other community factors.
As communicated to the study team through the RFP process and pre -study meetings,
community members in Dubuque have expressed concern about crime in the community in recent
years. Were crime patterns changing or was this simply a perception? What were the trends? Were
there contributing factors? Were there strategies to be explored that might mitigate any changes in
crime patterns? Some community members specifically shared their perception that increased crime is
related to the City's Section 8 housing program, in particular.
As framed by the City in its RFP, this Study was commissioned to help clarify the actual state of
crime in Dubuque and the characteristics most strongly related to the types of crimes, if any, that have
2010 Dubuque Crime and Poverty Study Summary Report Page 3 of 779 N !I icenierro
q Y Y Governm
eniaiSiudles
ti� ��
increased significantly in the past few years. Further, the Study would compare these data with the
public's perception of the nature of crime and criminals in Dubuque. The findings from the Study would
help guide policymakers as they decide what preventative programming might effectively prevent crime
and what educational programming might provide the most useful information to the citizens of
Dubuque.
General Research Questions Posed by the City
Dubuque summarized the project's scope into the following broad research questions:
• Do the perception of criminal activity and its causes in Dubuque match what is actually
happening?
• Within categories of crime with significant increases in arrests, what policies or
strategies can effectively decrease crime?
• If there are cases where there are community perceptions of increased criminal activity
but no evidence to support the perception, what policies or strategies can effectively
address the concerns?
In accordance with the City's expressed desire to explore perceptions of increased criminal
activity against actual incidents of crime reported as it may or may not relate to certain community
sectors or programs, the Study's primary aim was to investigate, assess and report on evidence -based
findings.
Summary of Study Scope and Research Plan
The Study focused on a quantitative analysis of local crime and the community's perceptions of
crime. As part of the scope of the Study, to evaluate perceptions of increased criminal activity against
actual incidents of crime reported and recommend related strategies where necessary, the study team's
primary tasks included the methodical collection, organization and analysis of key data that came
directly from the City's records and data bases related to crime, housing and geographic information.
The Study integrates both quantitative analysis with qualitative research and includes a survey
of community perceptions, plus a broad review of existing literature on the related topics with the aim
of providing decision makers with the information necessary to investigate, assess and recommend
revisions to strategies and policies, where necessary or supported by evidence -based rationales. In
addition to contributing to better understandings, the analysis commissioned by the City serves to
establish a mechanism for obtaining and providing meaningful information on crime and poverty to the
2010 Dubuque Crime and Poverty Study Summary Report Page 4 of 779 N! C I:Wer for
q Gov Studies
wu a«er
City of Dubuque to corroborate or dispel perceptions of crime. Ultimately, this Study is designed to
educate, engage and empower the community, as a whole, to actively combat crime in order to achieve
or maintain (if already in place) the ultimate outcome — a real and perceived safe community.
As stated in the Study plan, the research and analysis encompasses elements of crime and
poverty, particularly the possible impact of assisted housing programs, most notably Section 8 housing
units. However, it should be noted by readers of this analysis that the core focus of the Study centers
around crime in Dubuque. Hence, it was not the intent of this Study to reach beyond the core focus and
attempt to analyze causes of poverty or any other aspects of poverty other than how poverty and
related variables may or may not relate to crime in Dubuque.
Organization & Approach to the Study and Report
The work plan for the Study is presented in later sections of the report and is organized around
the three general areas within the "Scope" section of the City's Request for Proposal. The research plan
was organized around the following categories of investigation:
➢ Part I: Literature Review & Section 8 Assisted Housing Program Background
> Part II: Community Survey & Perception Analysis
➢ Part III: Quantitative Analysis
• Section I - Comparative & Trend Analysis
• Section II - Local Data Analysis, Spatial & Statistical Analysis
• Section III - Synthesis of Quantitative & Perception Analyses
• Analysis Of Arrests, Crime Incidents And Housing
• Contextual Analysis
• Assisted /Section 8 Housing Program
• Demographics
• Quantitative Analysis
> Discussion of Findings and Recommended Strategies & Policies
The Study report and executive summary follow this general presentation of information as
well. The report makes extensive use of appendices, exhibits and other forms of supporting documents
that, together, represent the totality of the research effort for this analysis.
2010 Dubuque Crime and Poverty Study Report Page 5 of 779 N � Centertnr
q y Y r _ Governmental Studies
SUMMARY OF NOTABLE FINDINGS
Part I: Literature Review & Section 8 Assisted Housing Program Background
Dubuque Crime and Housing Study: Synopsis of Findings from Literature Review
The study team condensed the topics presented by Dubuque in the RFP into the following major
subject areas:
Public Housing, Section 8 and Crime
1. The relationship between the mobility of Section 8 housing vouchers and crime displacement
2. General relationships between Section 8 housing and crime
Crime in General and Crime in Mid -sized Communities
3. Crime and Social Disorganization
4. General causes of crime in mid -sized communities
5. Contributing factors to the perception of increased crime
Effective Strategies Preventing Crime /Addressing Perception of Crime
6. Effective strategies /policies for preventing crime displacement from one community to another
7. Effective strategies /policies for preventing crime related to Section 8 housing in general
8. General effective strategies /policies for preventing crime in mid -sized communities
9. Effective strategies /policies to alleviate the perception of crime
Summary of Literature Review Results
Currently there is no consensus about the effect of public housing on crime and crime
diffusion in the literature. This often depends on the concentration, location, and physical design
characteristics of a city's projects. Section 8 housing projects that are smaller, more dispersed, garden
style, have defensible space, and are located in less resource poor neighborhoods, have not been found
to be linked to crime or crime diffusion (Roncek et al 1981; Galster et al 2003; Griffiths and Tita 2009;
Holzman et al 2001; Berry and Jones 1995; Galster et al 2002). Large, high rise towers that are
concentrated in resource poor neighborhoods do affect crime rates, which diffuse outwardly (Davies
2006; Suresh and Vito 2009).
Community policing, problem- oriented policing, and hot spots policing are effective crime
fighting strategies (Sherman 1986; Xu et al 2005; Mazerolle et al 2000; Braga et al 1999; Braga 2006;
2010 u
Dubu e Dime and Study Summary Re ort Page 6 of 779 N Center for
Dubuque Poverty -_ -- Y d Y rY p g N'� Governmental Studies
Braga and Bond 2008). Community policing has been found to reduce fear through improving police
relations with citizens and reducing social and physical incivilities (Roh and Oliver 2005; Adams et al
2005; Hinkle and Weisburd 2008).
Reducing social and physical disorder is the most reoccurring theme in the literature about
lessening fear of crime. That is why community policing is important. Other important disorder and fear
reduction strategies include problem oriented policing (Mazerolle et al 2000; Braga et al 1999), hot
spots policing (Braga 2006; Braga and Bond 2008), and broken windows policing (provided it is
combined with community policing).
Crime correlates in mid -sized cities are similar to those in larger cities. Areas with high levels of
concentrated disadvantages (high levels of poverty, unemployment, female headed households, and
minority concentrations) tend to also have high violent crime rates (Ackerman and Murray 2004).
Another cause of mid -size city crime is geographic based racial segregation (Shihadeh 2004). Goodman
(1997) and Goodman and Mann (2005) found that as the number of police per 10,000 people increased,
mid -sized city crime decreased. They also found that allowing off -duty officers to drive squad cars
provides an added deterrent effect, thereby lowering crime rates.
Research has shown Closed Circuit Television (CCTV) cameras to be moderately effective at
crime control. Welsh and Farrington (2009) conducted a meta - analysis of 44 studies. The Study's
combined effect size revealed that CCTV reduced overall crime by 16 %, which is a modest but
statistically significant decline. Much of the overall crime reduction is caused by CCTV's effectiveness in
parking lots. The authors found that using CCTV in parking lots resulted in a 51% decrease in crime. The
use of CCTV in city /town centers, public transportation systems, and in public housing areas was not
found to cause a statistically significant reduction in crime.
Curfews have long been a popular crime control method. However, there is little evidence to
suggest that curfews effectively reduce juvenile crime. Adams (2003) conducted a systematic literature
review to examine the effect of curfews on juvenile crime rates. He uncovered ten studies relating to
the topic. The results were mixed, with studies reporting a combination of no crime change, crime
decreases, and some even reported crime increases. Given the inconsistent results, the author
concludes that the research fails to support the use of curfews as a crime control strategy. However,
Adams did find some evidence to suggest that driving curfews decrease the rate of automobile crashes
by reducing teenage licensure, because driving becomes less appealing to teenagers.
2010 Dubuque Crime and Poverty Study Summary Report Page 7 of 779 N'ucenter ro<
Gov Studies
. _ kV 0 tr«er
Though the literature provides for multiple strategies that may be appropriate for mid -sized
communities such as Dubuque, it is important to note that no two communities are alike, nor have any
of the authors been able to generalize their Study's findings. Each municipality's existing landscape of
crime, demographics and policies must be taken into consideration, including the legal, fiscal, and
technical feasibility as well as the political acceptance of each strategy and /or combination of strategies.
The study team strongly encourages anyone interested in the topic of crime, perceptions of
crime, contributing factors to crime, crime and public housing or community policing and policing
strategies to read through entire summary of the literature review section of this report, and the
accompanying bibliography.
Part II: Community Survey & Perception Analysis
Public Opinion Results
The City expressed a desire for a direct assessment of community attitudes and perceptions of
safety in their community along with other features relevant to the present analysis. To obtain
empirically reliable data from the community, NIU, through the services of its Public Opinion Laboratory
(POL), developed and undertook a random sample of the community. To ensure that the results were
representative, NIU's POL used a random - digit- dialed telephone sample of City of Dubuque residents.
Only adults age 18 or older were interviewed. With ^'25,000 households in the City, the margin of error
for these results will be + 4.3 percentage points for a statistically valid sample of +/- 500 households.
Also, the survey responses within the sample were weighted to remove biases pertaining to age
of responders to better match the target population. The first column in the following table reflects the
2000 Census demographics for Dubuque, whereas the second column reflects the actual responders and
the third column indicates the responses adjusted to remove the bias of the uneven distribution
(demographics versus respondents) of responders. The majority of the responders were comparable in
terms of other City 2000 Census demographics (i.e., gender and race), yet the age of responders was
significantly different than the age demographics of Dubuque. "As a simple case, imagine that you
survey 100 Americans ages 15 to 64: 60 males and 40 females. For that age range, it should be 50/50. To
correct for this, in your analysis you weight each male's answers to be worth 0.83 of a response and
2010 ue Dubu Q Crime and Povert y Study Summary Report Srt Pa e 8 of 779 NIU i center for
_ - -- p - _ g Governmental Studies
kV B_eaM
each female's answers to be worth 1.25 responses. Now the survey results should, in theory, better
reflect the target population. "
Furthermore, City staff members worked with POL researchers to refine and review the
questionnaire /interview questions used for the survey. Questions were designed in line with the City's
particular interests on the subject matter under study. A copy of the survey questions can be found as
an Appendix to this report.
ES - Image 1: Demographic Characteristics of the 2000 Population and the Survey Respondents
2000 Census Survey Respondents Survey Respondent
(Unweighted) (Weighted for Age)
Gender
Male 46.3% 43.8% 46.1%
Female 53.7% 56.2% 53.9%
Age
18 - 30 years 24.4% 6.6% 24.3%
31 -50 years 36.7% 26.1% 36.7%
51 -70 years 23.3% 39.2% 23.3%
71 years and older 15.6% 28.1% 15.6%
Racea
Caucasian /White 97.0% 97.4% 98.4%
African American /Black 1.0% 1.2% 1.4%
Native American 0.2% 0.4% 0.2%
Asian American 0.7% 0.6% 1.1%
Latino 1.4% 0.4% 0.2%
Multiracial 0.5% 0.2% 0.1%
Home Ownershipb
Own current housing 67.5% 83.2% 80.2%
Rent current housing 32.5% 13.8% 14.8%
Overall N 44,049 502 502
b
Note - Four respondents did not report their race. One respondent did not report their home
ownership status.
The telephone survey of residents of the City of Dubuque was conducted between June 17 and
July 21 of 2010. The survey was designed to determine residents' opinions and experiences regarding
crime and safety issues within the city limits. Respondents were also asked for their suggestions for
improving safety in the city. In total, the POL conducted telephone interviews with 502 residents of
Dubuque with the average length of the interviews being 19.1 minutes.
' http: / /blog.vovici.com /blog/ bid / 28030 /Survey- Weighting - Weight - Weight- Don -t- Tell -Me
2010 Dubuque Crime and Poverty Study Summary Report Page 9 of 779 N center for
t Y t
y ry e p g Nl Governmental Studies
�a�
ES-Map 1
Dubuque POL Survey Respondent Locations
,-------- ,, ,.-,
I,. (' -‘,.
'''' \ \,:1---e---,1
1 -t
))-- — - - '
Th- ,L1-,,,LI%---,,-
1 ,. . Ly - ', . , ,, `-)..) _ \
i
, ., ,, ,,.
, . "_, 1 -,,
_) -,,,
-- I -- -4 1 , , .. „ -- , _ iii_.
"
" V 1 1
C„,... ---- 'ill N., P gr '
--\--i =I, . ,, - • . ' fty 04, ,,c).
qOr % '''' ...'
.__. „/ „----- ,,,,., _1,.. ,,-, 4 ve.,0
L'e , °- IL
...- .. ,..... .,,,,,,, ..ti I'
-:
y i i r Fig , w ;1-,
::,4- ----: ' I 4, \ 4' I A 1 1.:1,.., P.
i II 4 .=...i I' ,,
rl : ( „... ' E:-. , ,, 4 , . „ ,,,. ,,_ :,„__
k - — .f.___, .,ELNI '''
'') ' Ar '- ' ■ii M 1 A ''
rE 1:, „ .. .. V,1 1 A .'..:_'4-, 1
- -p.,„ tiety:-a_A‘.4.,,
r ------1 ---1 6 Li -' ---= ., te ,
qa- ...— • ....., ,
.---- - -
_ ., 4 , i ..„ 4
r.,-*....,orlfr „ ” ''''j ,
4 \
''.--,-
..b....t.f. ..,
....-.----- ----- . -...7 - ,
— i I
'1: --.- .• )
...
I
-- .1J I
/ f
1 c I
t'-_,,, 'i II
■■■...25 ,
r r°11
. .1
Survey Respondent Locations „ , , , ,e , i ..0, - ,
,
i - T - 4 - ,,, , '
[1 Mississippi River - ',.
f ' "-) 2
_
Miles
1 I Dubuque City Boundary ,_
005 1 2
Data Source: Suriey Data
MI Public Opinion Lab
NIU Cente”" I _ CTernmen a
!Similes
2010 Dubuque Crime and Poverty Study Summary Report Page 10 of 779
The following results represent findings of interest resulting from an analysis of the responses
from the community survey. The findings of interest are presented in two categories: 1) simple
frequencies of responses represented as a percentage of respondents sharing the same opinion; and, 2)
cross tabulations that represent the inter - relationship of two variables within the survey data but do
not reveal causal relationships.
1. Findings of Interest: Frequency Tables
Findings of Interest: Frequency Tables
• The majority of the sample (84 %) had lived in Dubuque for 11 or more years, with nearly
four out of ten (42 %) residing in the city for 35 or more years.
• Nearly nine out of ten respondents (86 %) characterized Dubuque as an excellent or good
place to live.
• Among the respondents who had lived in the City for five or more years, nearly four out of
ten (38 %) indicated that Dubuque had become a much better or somewhat better place to
live within the past five years.
• Nearly six out of ten respondents (57 %) indicated that the City of Dubuque was doing an
excellent or good job in addressing safety.
• Two - thirds (68 %) said the Dubuque police department was doing an excellent or good job
in addressing crime.
• The vast majority (88 %) of respondents reported that their neighborhood was an excellent
or good place to live. A small minority of respondents (2 %) said they felt very or somewhat
unsafe in their neighborhood during the daytime, whereas 11% indicated their
neighborhood was very or somewhat unsafe at night.
• The most common safety issues rated as being a "major problem" by more than 10% of
respondents were:
o Drugs (33 %)
o Violent Crimes (23 %)
o Gang Activity (22 %)
o Unsupervised Children (18 %)
o Property Crimes (17 %)
o Domestic Violence (14 %)
o Vandalism (14 %)
o Public Drinking (14 %)
2010 Dubuque Crime and Poverty Study Summary Report Page 11 of 779 NIU Center tor
Govemmental Studies
Findings of Interest. Frequency Tables
• Only 19% of respondents felt very or somewhat unsafe in downtown Dubuque during the
daytime; however, 71% indicated feeling very or somewhat unsafe downtown at night.
• More than three - quarters of respondents (76 %) indicated that crime was a major or
moderate problem in Dubuque. Among the respondents who had lived in the City for five
or more years, 89% said that crime had increased significantly or somewhat within the past
five years.
• Approximately one out of ten respondents (12 %) indicated that crime was a major or
moderate problem in their neighborhood.
• Among the respondents who had lived in the City for five or more years, 25% said that
crime in their neighborhood had increased significantly or somewhat within the past five
years.
• Two out of ten respondents (22 %) indicated that they were the victims of a property crime
(or threatened with a property crime) within the past year. More than half (55 %) of these
incidents were reported to the police.
• One out of ten respondents (10 %) indicated that they had been attacked (or threatened
with attack) in the past year. Only one - quarter (25 %) of these incidents were reported to
the police.
2010 Dubuque Crime and Poverty Study Summary Report Page 12 of 779 Center fo r
--- - -- - - -- —__ - -- y Y n[ p -- g Govemm entalSludlea
2. Findings of Interest: Cross -Tab Analysis Significant at the .05 level
_ Education
▪ People with higher levels of education tend to feel higher levels of safety and
lower levels of concern about crime and disorder
• People with bachelor's degrees are less likely to rate crime as a moderate to
major problem in Dubuque
O Those with at least a bachelor's degree are more likely to feel safe in
downtown Dubuque at night.
O Those with at least a bachelor's degree have lower levels of concern
about being physically attacked or mugged /robbed. They are also
less concerned about having their homes and cars broken into.
O Social disorder problems, such as drugs, loitering, disruptive teens,
noise (significant negative ordinal relationship) and public drinking
are less likely to be rated as moderate to major problems by those
with at least a bachelor's degree.
�.� Gender.
• Women in Dubuque feel lower levels of safety and higher levels of concern
about crime and disorder than men do.
o Women are more likely to rate crime as a moderate to major
problem in Dubuque.
o Women are less likely to feel safe in downtown Dubuque in the day
and at night.
o Women are more likely than men to rate social disorder problems,
such as drugs and unsupervised children as moderate to major
problems in Dubuque.
o Women have higher levels of concern about having one's car broken
into and being sexually assaulted.
Age Ca #egory ,
• There is a positive ordinal relationship between age and respondent ratings
of how bad the loitering problem is in Dubuque. The likelihood that
respondents will rate loitering as a major problem increases with age.
and Dubuque Crime nd Povert Study Summary Report Page 13 of 779 NIU cFiZior
ntal Studies
q -- __ Poverty Y Governme
r�u a�
• People who lived in Dubuque at leas 21 years were significantly more likely
to rate crime as a moderate to major problem. People who have lived in
Dubuque longer are less likely to feel safe downtown in the day.
• People who have lived in Dubuque longer are less likely to feel safe
downtown at night.
• There is a positive ordinal association between length of residence and
perceiving loitering and unsupervised children as moderate to major
problems in Dubuque.
• There is a positive ordinal association between length of residence and
higher levels of concern about being physically attacked, being
mugged/robbed and having one's house broken into.
• However, not all the relationships between length of residence and fear of
crime/disorder are linear.
o People who have lived in Dubuque at least 21 years are more likely
to rate drugs and disruptive teens as moderate to major problems.
o People who have lived in Dubuque less than 11 years, or at least 35
years, are more likely to rate noise as a moderate to major problem.
They are also more likely to be moderately to extremely concerned
about being sexually assaulted.
• Previous physical attack victims are less likely to feel safe in downtown
Dubuque in the day and at night.
• Social disorder problems, such as drugs, public drinking, unsupervised
children and disruptive teens are more likely to be rated as moderate to
major problems in Dubuque among previous attack victims than non-victims.
• Previous victims have higher levels of concern about being physically
attacked, being mugged/robbed, and having one's car broken into.
• Residents who have had their property damaged or vandalized in the past
year are more likely than those who have not to rate crime as a moderate to
major problem in Dubuque.
• Previous property damage victims are less likely to feel safe downtown at
night.
• Previous property damage victims are more likely than non-victims to rate
public drinking, loitering, disruptive teens, and unsupervised children as
moderate to major problems in Dubuque.
• Previous property damage victims have higher levels of concern about being
physically attacked, being mugged/robbed, having one's car broken into, and
having one's home broken into.
• Non-homeowners tend to have more fear and concern about crime and
disorder than homeowners do.
o Non-homeowners are more likely to rate disruptive teens, public
drinking and noise as moderate to major problems in Dubuque.
�
2Vz0 ou�qyeoime and p»«e Report Page � ���K� Studies
Part Ill: Quantitative Analysis
Comparative & Trend Analysis
Dubuque's UCR property and violent crime rates were analyzed in the context of nine other
Iowa cities with comparable populations: Ames, Cedar Rapids, Council Bluffs, Davenport, Des Moines,
Iowa City, Sioux City, Waterloo, and West Des Moines. Crime rates for the 6 -year period of 2004 -2009
were included in the analysis.
Approximately 17,000 agencies from 50 states contribute to the FBI's UCR data each year.
Though the data is useful for analyzing crime statistic trends, the FBI discourages using the data to rank
agencies to measure law enforcement effectiveness. Caution in reporting and describing crime
statistics of multiple agencies' UCR data is required because this data is derived from agencies that may
have local offense definitions that differ from the FBI's UCR offense descriptions and must be adjusted in
order to integrate with the FBI's UCR reports. This study team acknowledges the rationale for not
ranking and the need to proceed in examining Dubuque's comparables on a descriptive basis only.
Thus, charts and figures illustrate Dubuque's and its comparables' offense statistics together in order to
prompt the City to further investigate areas where they are significantly different than its peers. The
purpose of these types of reviews is to encourage the City to ask "how (in which categories or years)
might we be different ?" and "why might we be different ?" Addressing these questions can lead to
changes in policies that can enhance law enforcement techniques for Dubuque.
As noted in a later section of this comparative analysis, the inspection of comparative UCR crime
rates during 2004 to 2009 for the ten comparably sized Iowa cities reveals that, overall; Dubuque's crime
profile is not uniquely different from the other cities. However, the analysis did point to some
differences with regard to offenses classified as violent crime, using the standard UCR coding approach.
2 Source: FBI Uniform Crime Reports http: / /www.fbi.gov /about- us /cjis /ucr /word
2010 Dubuque Crime and Povert y Study Summary Report Page Stud S Rt P 15 of 779 N!1 Center for
—_ _ Governmental Studies
gg - Ames Cedar Rapids Council Bluffs Davenport
0 0 - �s
0
o �_•- • •
.
Des Moines Dubuque Iowa City Sioux City
s-
c 2004 2008 2008. 20102004 2008 2008 2010
Waterloo s .;West Des Moines
U 8 _
Q S
U
m - - �•�. -� -�
0
H zoom. 2006 20oe 2o102o04 •, 2006 zoos. 2010
YEAR
Total UCR Crimes Including Simp. Assault per 10000 2004 -2009
Source File: city crime rate_longpanel_10 5 2010
Uniform Crime Reports (UCR) Property Crime
Dubuque's level of UCR property crime rate was below the average of the ten cities, ranking it
7th lowest among the cities. Enough variation in these average rates across cities exists to be statistically
significant. Simple measures of city income, such as white per capita income and percent families living
below the poverty level, can account for some of this variation in average rates.
When examining the trends in UCR property crimes within and across the ten cities, there was
no statistically significant difference in the linear trend lines summarizing the overall declining rates of
UCR property crime from 2004 to 2009.
2010 Dubuque Crime and Poverty Study Summary Report Page 16 of 779 N!Ucenrer
Y Y T - --- --- --- Governme Studies
wu ash
UCR Violent Crime
The 6 -year averages for the ten cities were compared on UCR violent crime rates, making an
adjustment by adding simple assault along with aggravated assault due to the apparent anomalous
incidents of those two crimes in Dubuque. This anomaly involving Dubuque's simple assaults and .
aggravated assaults is discussed on pages 22 and 23. When UCR violent crime is analyzed in the
standard method, adding in only aggravated assault, Dubuque reported the third highest average UCR
violent crime rate (93.17 per 10,000) among the ten cities. This higher level than average was large
enough to be considered statistically significant, even with this small sample size. Dubuque's UCR
violent crime rate is similar to those of Davenport (101.44 per 10,000) and Council Bluffs (94.42 per
10,000) using the standard definition of UCR violent crimes.
Ames _ Ceder Rapids Council Slufts Davenport
O
0—
` SC Des Moines Dubuque Iowa City - Sioun Clty =: -_-
O
Q1 °_
L' �.�
re
O 4 8 8 10 4 B 8 50
f ? Waterloo West Des Moines
C
5
4 8 8 1 4 8 8 10
Year
Figure 9: 6 -Year Trends of UCR Violent Crimes per 10000 2004 -09
Source File city core _rate ong_panel_11_5 2010.dta
The varying 6 -year average UCR violent crime rates are strongly associated with the usual
indicators of city income and poverty. Thus, Dubuque's higher levels of violent crime can be generally
understood in terms of Dubuque's relative positions on median family income, and its complementary
measure of poverty, and percent of families living in poverty relative to more advantaged cities used in
this comparison. Dubuque's higher average UCR crime rate is attributable to the different levels of
simple and aggravated assault when compared to other cities. All other cities have higher rates of
simple assault than aggravated assault; Dubuque's distribution is the reverse: it reports higher levels of
aggravated than simple assault. When simple assault is added as a violent crime, Dubuque's 6 -year
average violent crime places it in the middle of the comparison cities rather than in the top three.
2010 Dubuque Crime and Poverty Study Summary Report Page 1 o f 7 79 N1 lcenter for
q Y Y Governmental Studies
Ames Cedar Rapids ?. Council Bluffs Davenport
tn -
0
n
--- ti. _� 4 �� - -•
as 8
V Des Moines Dubuque Iowa City j]j Sioux City
0
°
w
°
•—`
2 4 6 8. 10 4 ... 8 '. B. 10
m
Waterloo -' : West Des Moines
E
v
0
F.
4 e 8 10 4 6 8 .'.. 10
Year
Figure 10: Violent Crime Rates per 10000 including Simple Assault, 2004 -2009
Source File: city crime_rate_longcanel_11_5_2010 dta
Furthermore, statistical estimates of linear trend lines of changes within each city revealed that
the changes in UCR violent crime over the period of 2004 to 2009 followed different patterns. These
estimated linear trend lines showed Dubuque to have one of the larger upward trending lines (5.7
violent crimes per 10,000 per year) over this period (second to Waterloo, 8.4). The variation in the
estimated trend lines across cities was large enough to be considered non - random.
Lacking yearly estimates of median income and poverty measures, the analysis attempted to
model the variation in trend lines, hence trying to explain the increasing and decreasing crime patterns
using static measures of city household and family income and poverty measures. None of these static
measures could account for the varying trend lines consistently, but for the more inclusive measure of
violent crime that included simple assault, poorer cities (on two measures, percent families in poverty
and median household income) were estimated to have slightly higher rates of increase in violent crime.
The findings associated with the assault data generated a dialogue with City staff in an effort to
better understand or explain the anomaly found relative to the reporting of simple and aggravated
assaults. The dialogue resulted in a recent analysis by the City of the historical treatment of these
offenses when records were coded and reported in the Police Department's crime record data base
system. The analysis by City staff and Police Department management revealed the following finding
that sheds light on the anomalies found in the assault data. The following finding was shared with the
study team:
2010 and Poverty Crime Study Summary Report Page 18 of 779 NI I lcenterlor
_- - -- --- --- -- -- -y Y u
ry p g -- - -_— _ -- Govemmental
"A review of our data that is reported to UCR shows that ALL cases of Assault
with Injury under Iowa Code 708.2(2) are reported to UCR as Aggravated Assaults. A
review and comparison of the definitions in the handbook and the listed codes shows
that in most instances we are over reporting this classification. While some cases of
Assault with Injury cases may fit the definition of Aggravated Assault as described in the
UCR Handbook, minor injuries such as scratches, bumps, bruises, etc. which make up
the majority of our Assault with Injury cases, would not fit the UCR definition, yet they
are reported to UCR as Aggravated Assaults." Chief Dalsing research memorandum, Nov. 10,
2010.
The results of this analysis add understanding and context to the pattern of differences the
study team found when comparing Dubuque's statistics for aggravated and simple assaults. The analysis
of reporting protocols within the Police Department provided by the City indicate that the "over
reporting" of incidents as aggravated could be substantial. Staff provided an analysis that indicated that
reported 2009 aggravated assaults totaled 265, but when applying UCR criteria for these offenses, 250
would be classified as simple assaults rather than aggravated assaults. This represents a substantial
difference and is likely impacting the analysis of data that was provided to the study team for
comparative /UCR analysis.
Statistical Analysis Of UCR Violent Crime Rates Per 10,000 Across 10 Iowa Cities
Figures 1 — 5 of this section describe the 6 -year average of UCR violent crimes across the 10
Iowa cities. As the data indicate, Dubuque does not have uniquely high rates of violent crime, with the
only exception being aggravated assault, the most common violent crime. This led us to examine the
rates of simple assault as well, a crime usually excluded from the UCR counts of assault. Dubuque has
relatively high aggravated assault rates and relatively low simple assault rates. This is the only feature
visibly distinguishing Dubuque's violent crime rates in the graphs. Assaults are by far the more
commonly occurring type of violent crime.
t d P
b
D
1
200 Dubuque Crime and Study Summary Report Page of �r Center tor
— - q Poverty Y rY a 19 f 779 p g ___ N�� Governmental Studies
o 84 o
71
E3
m 66
m __. 2 31. .33
Q211 °a' 34 _ .2i
v N- fe
m 23 31 2< Q -
! ": N- 27- i 12
a a 2 s 5 .f a ; Z u e - a
- a e - 3 0 $ o C a il a o
s - z 3 a c z 1 3 a
- a if
Figure 1: Average Aggrevated Assaults per 10000 2004-2009. 10 Iowa Cities Figure 2: Average Homicides per 100002004 -2003 10 Iowa Cities
Solace Fib: cty_crne ng- Pan411 6201 4. Boe . Fh cry cnn ate bng_panel 11_5_2010 dt.
E S 15 o o N '.
O N
d2 _
II. .2 a O
dN_ 14
13 K
iii 8.7 6.3 K
Q h _ 4.8 5 . 4 4.7 5.1 5.7
0 3.1 32 ¢ 5.6
'tor m 44
T 2.8 2.4
i r, @ a 0 6 m u 1 e i o I T o i 6 i a
m p n
y n m
9 � o
Figure 3: Average Rapes per 10000 2004-2009: 10 Iowa Cities Figure 4: Average Robberies per 100002004-2009: 10 Iowa Cities
Source 404 C y_crbre b g_panel 11 5 201.dae Source He c crtreJ446_1,8481_11_5_201.02
ry_
In every city but Dubuque, the rates of simple assaults exceed those of aggravated assaults by
factors ranging from roughly 2:1 to just over 5:1 (compare the rates in Figure 1 to those in Figure 5
below). However, in Dubuque, the opposite is the case: aggravated assault rates exceed simple assault
rates, a pattern anomalous from the usual case. On face value, Dubuque has higher rates of aggravated
assault and lower rates of simple assaults than the comparison cities.
While UCR coding classifications are well defined in the Uniform Crime Reporting Handbook
(Federal Bureau of Investigation 2004), it is possible that rules used to charge and classify assaults in
Dubuque vary as well as the actual crime levels of simple and aggravated assaults.
2010 Dubuque Crime and Poverty Study Summary Report Page 20 of 779 N' 1 lcenter f
Govern Sludles
- - - -- _ kv sera,
84
'2- 238 o
71 -�
0 2 66
179
167 Y! m- - --
51
118 1 20 ¢ 0 41 44
m
CO - 77 Q = a
j 21 24
m fg 45 12
Oo
II I 11
3 ° g o _ c g o 0 0 0 a E
a ; B A a a 3 o
Figure 5: Average Simple Assaults per 10000 2004 -2009: 10 Iowa Cities Figure 1: Average Aggrevated Assaults per 10000 2004 -2009: 10 Iowa Cities
Source F7e: oy_ctlne_rate_bng_pnnel_11 5 2010 Ma Source File : cOy crere_rete_bng_panel 11 s_20t0.Me
The anomaly of the different rates of aggravated and simple assaults warranted treating the
assault category with two different approaches in the subsequent statistical analysis. We first analyzed
the violent crime rates using the standard UCR definition of including aggravated assault along with the
other measures, and then we repeated the analysis adding in the simple assaults as well as the
aggravated assaults to obtain the UCR violent crime rates for each city. The varying levels of the total
assault rates for the ten cities are graphed in Figure 6. When considering total assault rates, Dubuque's
rate (161) is closer to the average total assault rates for the ten cities (163).
0 308
0 0
4 206 213
a o - - -. -186 -- -. -. 195
N
161
139
m �68 ro I i P PI , Figure 6: Average Total Assaults per 10000 2004-2009:10 Iowa Cities
Source He cry orirre_brg_par0 115_X81.80
Summary Findings Regarding the Rate and Trend of Violent Crime Across 10 Iowa Cities
Six -year average UCR violent crime rates across these ten Iowa cities are statistically dependent
upon income and poverty levels of the cities. As the percent of families with incomes below the poverty
level increase, so does the average violent crime rate. Just as the average violent crime rates depend on
2010 Dubuque Crime and Poverty Summary Report Page c en t er r or
Y Stud -__ Y S ry R a 21 of 779 p g �G ovarnmen tl Studies
kv 3e..eh
•
poverty levels, so do the starting levels of violent crime rates in 2004 depend on the poverty levels of
the ten cities.
However, the analysis was unable to find a statistically significant city level correlate with the
varying estimates of the trend line slopes summarizing changing UCR violent crimes rates across the
cities. The only potentially emergent poverty or income indicator was the level of household income
inequality in cities estimated using a measure of household income inequality, the GINI coefficient. The
effect is in the expected direction; cities with higher levels of household income inequality have
somewhat steeper increasing slopes of UCR violent crime rates over this 6 -year period, but the effect
was not statistically significant at the usual .05 or .10 levels. When simple assault is added as a violent
crime, the median household income and the percent of families in poverty affect the estimates of the
linear trends: poorer cities had somewhat higher rates of increases in violent crime.
Statistical Analysis Of UCR Property Crime Rates Per 10,000 Across 10 Iowa Cities
UCR property crimes include arson, auto theft, burglary, and larceny. Our category of larceny
includes all of the theft and larceny categories provided in the Iowa Dept. of Public Safety crime data
reports. Larcenies comprise the largest category of property crime. The 6 -year average rates of the four
property crimes are graphed in Figures 12 — 15. These graphs show Dubuque as having close to average
property crime rates except for the high level of arsons, but arsons are a low frequency crime, so small
numeric differences can translate into misleading rate comparisons. There appears to be nothing
notable about Dubuque's relative position among the ten cities with respect to the four UCR property
crime categories — it is in the middle of the pack.
0
00
m m .
4
c 5.1
o
0 39
o m
2B Q
= 2.6 U V _ - - - -. 34 - -..-
N 25 24 27
1.6 1.6 - 1.7 - 1.9 U
m o II I7ii ii i Q m N
} 1 n n a o c
q e
Figure 12: Average UCR Arson Crimes per 10000 20042009: 10 Iowa Cities Figure 13 Average UCR Auto Thefts per 10000 20042009: 10 Iowa Cities
Source f9e: cy crime bngpanei 11 5201.d1e source He cy crine_bngyanel 11 5_201 daa
2010 Dubuque Crime and Poverty Study ort Page 22 of 779 Center for
Y Summary ! p g ��Governmental
o 0 151 o
134 o
o N- o 5E9
126 491
08 a 44J
m o 79 317
257 268 290 267
U H_ 54 44 U - 213
m m N
>11' m
FEZ
s — a a
A C ; n ° 7J ; n
. .. E s g g
Figure 14 Average UCR Burglaries per 10000 20042009: 10 Iowa Cities ' Figure 15 Average UCR Larcenies per 100002004 -2009 10 !ma Cities
Source Fli s: cry cr roe bngyane111 5_201. de Source FM : cry_cr4re_bngya 11 5_271.�1a
The table and figure below describe the UCR property crime rates for the ten Iowa cities in this
analysis. The property crimes range from a high of 830 property crimes per 10,000 population for
Council Bluffs to a low of 280 property crimes per 10,000 population for Iowa City. Dubuque's UCR
property crime rate (381) ranks seventh lowest among the ten cities, below the overall average (480).
While there appears to be significant variation among the cities, Dubuque is less burdened by property
crime than its comparative cities.
The coefficient of variation indicates relatively small year -to -year fluctuations around the 6 -year
averages for the cities. Dubuque's yearly deviations amount to just over 5.6% of its average value; one
of the lowest observed variations relative to the average rate seen across the ten cities. Even the cities
with the largest relative variations have fluctuations amounting to only fifteen to seventeen percent of
the average 6 -year values. These results indicate fairly small year -to -year variations in property crime
rates, relative to their averages, for all of the cities examined.
2010 Dubuque Crime and Poverty Study Summary Report Page 23 of 779 Studies
N'Ucenterror
Governmental
w ri�n
ES -Image 2: Comparative 6 -Year Averages in Property Crime Rates per 10,000 Population - 10 Iowa Cities for Years 2004 to
2009
Avg. 6 year Standard Coefficient of
Iowa City & Place Code Property Crime Rate Deviation Variation
Ames (1901855) 349.22244 10.794683 3.09112
Cedar Rapids (1912000) 489.42982 38.973297 7.96300
Council Bluffs (1916860) 830.19975 80.527680 9.69980
Davenport (1919000) 664.06699 101.923860 15.34843
Des Moines (1921000) 604.29028 107.460520 17.78293
Dubuque (7) (1922395) 381.30791 21.467907 5.63007
Iowa City (1938595) 280.36037 25.998742 9.27333
Sioux City (1973335) 400.63828 69.547577 17.35919
Waterloo (1982425) 474.55089 37.109896 7.82000
West Des Moines (1983910) 324.33218 20.717293 6.38768
Overall 10 Cities 479.83989 174.91313 36.45239
F- Statistic (47.45, df =9; p -value <.0001)
Coefficient of Variation = (Std. Dev. / mean) *100. It reflects the yearly variation in the rate as a percentage of the average rate.
o .. 830
0
v 684
a
m o 604
w o .
E co
V 489 475
0 381 401
o-
ai 349 324
a o ;...� 280
o
� f
a7 n o o O
, o . " 2 o O s b
a m m 5 m o
b N O
. m
Figure 16: Average UCR Property Crimes per 10000 2004 -2009: 10 Iowa Cities
Source File : city crime _ongcanel 11 5_201.dta
Summary Findings Regarding the Rate and Trend of UCR Property Crime Across 10 Iowa Cities
The six -year average UCR property crime rates showed significant variation across the ten cities:
but Dubuque was not the focus of the differences. Rather, Council Bluffs, Davenport, and Des Moines
differed from the other seven cities due to their higher levels of property crime.
These statistical differences in UCR property crime rates were associated with two measures of
city level income and poverty: negatively associated with white per capita income (as white per capita
2010 t
v
Dubuque Crime and Poverty Study Summary Report Page 24 of 779 N'Ucenterror
- -- .- - - -- _ -.- q _ -- -- Governmental Studies
income increased, property crime rates decreased), and positively associated with the percent of
families below the poverty level (as family poverty levels were higher, so were the property crime rates).
Neither any substantively notable nor statistically significant variation in the trends of property
crime over this period across the cities emerged in this analysis. Dubuque was the sole city estimated to
have a very slightly increasing property crime rate while all other cities had decreasing trend lines, but
the different slope estimated for Dubuque is small enough to be a random fluctuation around the
generally decreasing trend line for all the cities.
Figure 17: Six Year, City by City Patterns of UCR Property Crime Rates, 2004 -2009
Ames - Cedar Rapids Council Bluffs Davenport
•
C
O N
Y Des Moines Dubuque Iowa City Sioux City
o
N
:'E l 2004 2006 2008 20102004 2006 2008 2010
`Waterloo ? West Des Moines
o
0
Q.
,
2004 2006 2008 20102004 2006 2008 2010
YEAR
Graphs by City
Geo - Coding and Spatial Analysis Results
Spatial analysis, at its core, is an analytical technique that identifies the spatial distribution of
particular variables or pieces of information selected for examination. The study team proposed, as part
of the research effort, the geo- coding of relevant data and variables as part of the overall investigation
of the research questions posed by the City. The summary maps that follow depict some of the key
results from the geo- coding exercise. Geo- coding is the process of assigning latitude /longitude to a data
point or address from a particular information source. The maps that follow present a summarization of
Crime (Arrest and Incident) and Section 8 Housing locations for the years that are under study. Detailed
2010 Dubuque Crime and Poverty Study Summary Report Page 25 of 779 NlUcenterlor
G overnmen t al Studies
wv a+.,en
maps broken down by year and aggregate maps using Census Block Groups can be found in the
Appendix of this Summary Report.
The input of over 168,000 records, including crime incidents, arrestees, arrest locations, housing
program locations and survey respondents, was the first step required in the spatial analysis before
moving to statistical modeling and measurement. Crime mapping, using GIS analysis techniques,
facilitates the tracking and understanding of all these data by producing density and cluster maps
showing patterns.
Going forward there will be a few key terms used repeatedly that should be briefly
defined /described up front. The table below highlights a few terms that are worth of introducing before
proceeding through the analytical sections of this report. The remaining definitions and terms can be
found in the final Appendix labeled "Glossary."
Term Definition = — _
Arrest A seizure or forcible restraint; an exercise of the power to deprive a person of his or
her liberty; the taking or keeping of a person in custody by legal authority, especially,
in response to a criminal charge
Authorized Includes individuals contained within the Dubuque Housing data file. Voucher holders
Section 8 as well as dependents within this file are considered "authorized" residents or
participants of Section 8.
Complainant The party who makes the complaint in a legal action or proceeding
Incident Incidents include events with associated arrestees, victims or victims' businesses,
complainants, complainants' businesses and other involvements (i.e., accident driver,
business owners or vehicle passengers associated with the accident, businesses, cited
persons, deceased persons, missing persons, owners of involved vehicle or property,
pedestrians and other individuals impacted by the event) — referred to as involvement
records. Individuals such as adult or juvenile suspects and witnesses are not included
in this data. For example, incident A has one arrestee, two victims and one witness.
The record of this witness is dropped from this data set and Incident A is counted as
ONE incident for this incident count table. Given another example, Incident B has one
Juvenile suspect and has no arrestees, no victims and no other individuals involved.
3 http: // legal- dictionary.thefreedictionary,com /arrest (accessed December 9, 2010).
4 http: / /www.merriam- webster.com /dictionary /complainant (accessed December 9, 2010).
2010 Dubuque Crime and Poverty Study Summary Report Page 26 of 779 center for
--_ -__ Y Y rY R po g N�U GovernmentatSludles
Term Definition
Incident B is not included in this data set. Each incident event with multiple arrestees,
victims or individuals involved is counted once.
Juvenile A person under 18 years of age. Juvenile arrestees were not included in the analyses.
The arrest data received from Dubuque contained arrestees with adult charges.
However there are few arrestees under the age of 18. For years 2006 through 2009,
the data file contained approximately 60 unique arrestees under the age of 18.
"Total" An individual who is charged (Arrestee) or reports a charge (Complainant) or is the
Arrestee, recipient of harm generated by a charge (Victim) is included in a "total" count for each
Victim or incident they are associated with in a given year. For example, an incident that took
Complainant place in 2006 involves an arrestee with four charges. This same arrestee is later
charged with three other crimes in a separate incident within that same year. This
arrestee will have two separate records in 2006's "total" arrestee count.
"Unique" An individual who is charged (Arrestee) or reports a charge (Complainant) or is the
Arrestee, recipient of harm generated by a charge (Victim) is counted as "unique" when they
Victim or are counted only once in a given report year regardless of whether or not they have
Complainant been involved in multiple incident events or have multiple charges. For example, if a
complainant reports an incident in 2007 and later reports another, non - related,
incident in the same year, this complainant will have only one record in the "unique"
complainants count.
Victim' "One that is acted on and usually adversely affected by a force or agent <the schools
are victims of the social system >: as a (1) : one that is injured, destroyed, or sacrificed
under any of various conditions <a victim of cancer> <a victim of the auto crash> <a
murder victim> (2) : one that is subjected to oppression, hardship, or mistreatment <a
frequent victim of political attacks> b : one that is tricked or duped <a con man's
victim >"
Crime Incident Mapping
The Dubuque Crime Incident Map shown below provides a spatial analysis of the location of
unique crime incidents from 2006 to 2009. This map was prepared by geo- coding the records provided
by the City's Information Services Department. The maps depict the location of unique crime incidents
for the years 2006 to 2009. The maps for the individual years can be found in the full report. The spatial
analysis shows incidents scattered throughout the City for each of the years in addition to a general
5 http: / /www.merriam- webster.com /dictionary /complainant (accessed December 9, 2010).
2010 Dubuque Crime and Poverty Study Summary Report Page 27 of 779 NIUCenlertor
q Y Y Got_Tr en laiSludles
concentration in the eastern half of the City's geography. A closer review of the mapping indicates that
for unique incidents (2006 to 2009), the concentration of incidents in the eastern part of the City tend to
fall within an area along Central Avenue and north of Dodge Street. (See map the follows ES -Map 2)
2010 Dubuque Crime and Poverty Study Report N 1I cnn terror
y Y Summary Page 28 of 779 e
� $ N1�Gov emmentalStudles
ES -Map 2
Distribution of Unique i7ii Incidents 2006-2009
r v..., , - ' ---- '''''''' -.) ),
Log ''' )7 .) .,-1" .---/ '11,...(,....: -
f
fj _ i
•
f
_ a�.
—' " .
J .. cam ▪ - x • - - * ' - � Ai
�^�, �°". i ce ; �ry 4 r` 1 ,� ,`�'
t 111Mfi ■
t �
d
u p
: 1 --
_-- -.�� - - •'- , a `t I � i� l,�
tti'�� ry
ti s
f . Y J
e tit. . 6., :
---/— r ' ' — „v._ ,,,., ,_.! ,„
—)11'±1 .r.yd r• °.� .l). ..Y.' Y � F ire
Unigt Incidents 1 y
LLI
k R y
I I Mississippi River •- ,
y
Dubuque Bounds r �`
g
Data Sour: Incident File, i__4\:_„,, ' ' /
Information Services Dept, City of Dubuque y
j Created fruit egmbined_ fiinne_'rx:dent. NOVB.dbr / f
Cermter for �, s '•., ' '1-. •'�
G Studies Miles
,..,_,•�, I \0 1 2
Created iinatiO 1, t , .—• —! ,
,°
2010 Dubuque Crime and Poverty Study Summary Report Page 29 of 779 N11 lcente
Governmental Studies
.0.hexr
Arrest Mapping
The Dubuque Arrest map that follows shows the distribution and concentration of home
addresses of arrestees from 2006 to 2009. This summary map was prepared by geo- coding the file
provided by the City's Information Services Department. The map depicts the home address of the
arrestee (as reported) for the years 2006 to 2009. The maps for each of the years 2006 through 2009
(see Appendix) indicate that the addresses of arrestees are dispersed throughout the City. The spatial
analysis does depict a more notable concentration of arrestee home addresses in the eastern part of
the City. Again, following the pattern found in the Crime Incident records, home address of all arrestees
for 2006 to 2009 portray a concentration in the eastern part of the City, along Central Avenue and north
of Dodge Street. (See Map that follows ES -Map 3)
2010 Dubuque Study Summary e Crime and Poverty Report Page 30 of 779 *' ce n teemmr ror
Y Re P g N �� Gov entalStudies
ES -Map 3
Reported Home Address of Arrestees 2006 -2009
e / 4
5 ��5� \ 1
pp I 1
1 - ,- e 11.. i �
w ry^ . _
r y s
1 4
— i� �� � � lid � ��� y 4 � � t., (� I
h ta 2 'ri ti_ .
' � ° � 7'7 � r I I. 't4:11.1''''', I v R " r,
d S S ily y II
r � Y 4 � � F
4 I I L A CT • es
a � i �
� � 1 ..' i5 ,4 „ ', �
�.�,- °- ri' am: ` ♦r - "1 • w �' ... ,_,_, -- - 1 , - i — ; 7 1 - ',7 , -,, ifi”
�t
-----)' r ,' ../1 — -,,,,,,,,,..„ L.
, ,,,i\v,
I e
Arrester Address Pti r
r . ‘",r Mississippi River � _ �
n Dubuque city Limits �.- f `` ,ma ��
,Data Sour: A.rrest File
i t Information ecvicee Dept. City of Dubuque ', , t, //
r Caeatetlioma� elee iurna adyrees rt1�8.[3rf . . fi r � .'" , 1 ,�,,r'
Cw.wr{r,llk{ R.,iw '� Miles
n ., 1
mapernatea 11/10/201° ', D 1 2
Lk 2010 Dubuque Crime and Poverty Study Summary Report _ Page 31 of 779 center for
Ni lSludlea
h ^J Qk ea�
Authorized Section 8 Housing Mapping
The Dubuque Authorized Section 8 Housing map that follows depicts unique Section 8 Housing
addresses obtained from the City's Housing & Community Development Department data base. This
map shows the distribution of authorized Section 8 housing for each of the years 2006 through 2009.
The aggregated block group maps (see Appendix) show that there is a general concentration of
authorized Section 8 housing in the eastern part of town indicating a pattern that tends to parallel
sections along Central Avenue. (Note: the 2006 -2009 map depicts all occupied Section Housing Units for
those four years, if occupied for one month or more, per the records provided). While the map depicts
Section 8 units "occupied ", participants may move in and out creating vacancies at times (also referred
to as turnover in the analysis). This turnover could result in more units available than households or
vouchers. Conversely, there are also voucher wait lists at times. Thus, there may also be scenarios
where vouchers and households exceed the number of units available. (See Map that follows ES -Map 4)
2010 Dubuque Crime and Poverty Study Summary Report Page 32 of 779 NIU l centerfor
y Governmental Studies
ES -Map 4
Section 8 Housing Units Occupied 2006 -2009
I ` -, i r ' i x,_ `
f>
_� t - - ,.<
I i I 11 l'
i c� I ' .1 ! . e .-! 4 , � 11j I-1
�3
q ctt, - 1
4
L ' � j j 1 � 4‘ . ' ,..,, r 1
i -r -. w r A
r' ." 1 .....-
6 t
l 1 ' — k = v s r �f
Ts r _,�I r f'_ ire ,ra f Ii '
♦1 0 i r E a
}
ii I'ar h �f rr - . VV 4. F
+miU � t s s '�F a , x1
6 � -A ,' ` --� G - , f r ' ; �` l
11
------- 1DC tt.St ,
i_a t. r ' a J
h q I'L''' i t -I
, r Ir ; I r :, i
, B 4 et - E 1 al siglallz ti� 1 yY�� i P i y ' �7 �-7 ` � \
Q nosa4 rmae ,darn i ` `,,
seta 3aroc aec11on a kip Fie, ' i }
H7v1n0 and Commit/ Deebpment
exit, cr.,Wa,sruz it '
i I Mules
Crested RnRS ros ee
SeceYseR ,- _,'� —0 0.5 1
NIU �7 vA-r Mr
Ge n� rld hine :*}ii - Z i U
t]eded 1111621310 -.. - ° -� a�;",�t tale.;
2010 Dubuque Crime and Poverty Study Report Pa 33 of 779 A rll ieenfert
y G ov e mm enfal S tudies
Frequency Analysis of Dubuque's Crime and Arrest Data
In order to conduct an analysis of the patterns of crime and arrest in Dubuque, the study team
requested electronic copies of key records and files kept by the City as part of their public safety records
and housing program records. The records were transferred in a secure environment between the
organizations and analytical protocols were employed to convert or protect certain information in the
files. In total, over 300,000 individual records from crime incident, arrest and housing program files
were received. It should be further noted that juvenile arrests were not included in the files sent
(except for juveniles charged as adults) which will create differences in statistics tracked internally by
the City, including the reporting of arrests within the statistical section of the City's Comprehensive
Annual Financial Report (CAFR).
For the timeframe of 2006 through 2009, Dubuque arrested about 15,900 people (exclusive of
juvenile arrests per the exception above). Refer to the chart below for an illustration of total arrest
trends compared to unique arrest trends. Unique offenders are unique for each year reported but could
have been arrested in either previous or subsequent years and could have been arrested multiple times.
In addition, it is important to recognize that where the analysis refers to Section 8 participation, the
data represents only authorized Section 8 cross - matching, (using the data analysis protocols listed in
this section) and should not be inferred to include non - authorized persons who may be associated or
implicated in Section 8 crime incidents or arrests. It is important to note that there could be overlap in
the arrestees, victims and complainants. For instance, it is possible for a complainant to also be a victim
and /or an arrestee.
Dubuque experienced an increase in incident and arrestee (unique and total) counts from years
2006 to 2008, but these trends were reversed in 2009 when the counts began to drop. Not only did the
counts drop in 2009, but they dropped below 2006's figures.
2010 Dubuque Crime and Poverty Study Summary Report Page 34 of 779 N1U Centerfor
—___ Y Y Governmental Studies
ES -Image 3: Dubuque's Arrestee and Incident Trends: Years 2006 to 2009
Dubuque's Arrestee & Incident Counts
9,000 -
8,000 -
7,000 -
# of Unique
Incidents
6,000 - # of Total Arrestees
5,000 -
# of Unique
4,000 - Arrestees
3,000
2,000
2006 2007 2008 2009
# of Unique Incidents 8,563 8,630 8,701 8,231
# of Total Arrestees 3,962 3,977 4,100 3,883
#of Unique Arrestees 2,804 2,774 2,863 2,718
Source: Dubuque Arrests & Incidents ata file.
CGS File Name: S: \common \Dubuque \Report - Final \Charts_Tables_Figures\
Final Frequency Stats\ INDIVIDUAL_ FRECI_TABLE_ TOTAL_INCIDENTS_VALIDATION.xlsx
Note 1— Figures here were obtained from Dubuque's arrestee data file and are exclusive of juvenile
arrest counts, with the exception of a few juveniles charged as adults.
Note 2— Each incident event with multiple arrestees, victims or individuals involved is counted once.
Also, Incidents presented in this chart only include events with associated arrestees, victims or victims'
businesses, complainants and other involvements (i.e., accident drivers, business owners or vehicle
passengers associated with the accidents, businesses, cited persons, deceased persons, missing persons,
owners of vehicles or property involved, pedestrians and other individuals impacted by the event) —
referred to involvement records. Individuals such as adult or juvenile suspects and witnesses are not
included in this data. Those incidents containing additional records pertaining to juveniles, witnesses or
suspects were adjusted to exclude the witnesses and /or suspects associated with the incident. For
example, incident A has one arrestee, two victims and one witness. The record of this witness is dropped
from this data set and Incident A is counted as ONE incident for this incident count table. Given another
example, Incident B has one Juvenile suspect and has no arrestees, no victims and no other individuals
involved. Incident B is not included in this data set.
2010 Dubuque Crime and Poverty Study Summary Report Page 35 of 779 NIU Center for
Governmental Studley
1.0 6,
In terms of adult arrests that took place during the period 2006 through 2009, CGS identified
more male arrestees (9,182) than females (3,528) in the data, with the majority of the males within the
age group "between 18 and 40" years old (see the following chart).
ES -Image 4: Age and Gender Composition of Dubuque's Adult Arrestees
Dubuque Total Adult Arrestee Counts By Gender &Age Group
For the Time Period 2006 - 2009
10,000 -
9,000 -
APFX
8,000 ° t
7,000 -
6,000 -
5,000 -
4,000 - . r
3,000
I Female
2,000 pz2,3 W' rgt .7
1,000 Male
4 -.
BETWEEN 18* BETWEEN 40 AND GREATER THAN
AND 40 TOTAL 65 TOTAL 65 TOTAL
Female 3,528 694 21
Male 9,182 2,373 86
Note: There was a total of 26 null and unidentified records.
Source: Dubuque Arrests Data File..
CGS File Name: S: \common \Dubuque \Report - Preliminary\Charts_Tables_Figures \Final Frequency
Stats \IN DIVIDUAL_FRECQTABLE_TOTAL_I NCIDENTS_VALIDATION.xlsx
* Figures here are exclusive of "juvenile arrest" counts. While juveniles are defined as individuals under
the age of 18, there are some arrestees younger than 18 but charged as adults (deemed "adult arrests"
which are included in these "adult arrestee" counts.
2010 Dubuque Crime and Poverty Study Summary Report Page 36 of 779 N'U centertor
q Y Y Govemmental Studies
wu a�
The chart below depicts the breakdown of the City's, State's and Nation's population by race
(based on a 3 -year average from 2006 to 2008). Per the American Community Survey's 3 -year average
(2006 to 2008), Dubuque has a higher percentage population of "White" residents than the State or the
Nation, with a lower balance of "Other" races or "Black or African American" residents compared to the
State and the Nation — though Dubuque's race compositions are somewhat comparable to the State.
ES -Image 5: Race Composition of Dubuque's Population Compared to Iowa's and the Nation's Race Compositions
(3 -Year Average)
Race Composition of Population:
3 -Yr Avg (2006 - 2008)
120.0%
100.0%
3.6% 4.8% 13.3% • Others
80.0% J. ., i
1
_ White
0
60.0 /o
AL .,
40.0%, : -1 Black or African
� � American
20.0 % — ez
0.0%
1.7% 2.4% 12.3%
Dubuque Iowa Nation
Source: American Community Survey (ACS); 3 -Yr Avg for 2006 - 2008
CGS File Name: S: \common \Dubuque \Report - Final \Charts and Tables \dbq_arrest_race_census.xlsx
2010 Dubuque Crime and Poverty Study Summary Report Page 37 of 779 Nlu center for
Govern mental Studies
a:, aka
Illustrated in the chart below is the race composition of adult arrestees (arrestees by race as a
percent to total arrestees) on a comparative basis: Dubuque compared to Iowa's rate and the Nation's
rate of arrestees. Dubuque and the State have similar arrest compositions, with Dubuque's arrest rate
of "Other" races being slightly lower than the State's (.8% compared to 2.3 %, respectively). Compared
to the Nation, both Dubuque and Iowa have higher arrest rates of "Whites" (79.3% - Dubuque; 78.5% -
Iowa; and 69.8% - Nation) with lower arrest rates of "Black or African Americans" and "Others."
ES - Image 6: Adult Arrests by Race as a Percent to Total Arrests (3 - Year Average)
Adult Arrests by Race as a % to Total Arrests:
3 -Yr Avg (2006 - 2008)
120.0% -
0.8% 2.3% 2.3% ■ Other Arrests asa %to
100.0% -
Total Arrests
80.0% �� m
WhitesArrested as % to
60.0% Total Arrests
r
40.0 % P,4 h f a Blacks or African Am as
� a % to Total Arrests
20.0% r
27.9 %'
19.9 % f 19.3%
0.0%
Dubuque Iowa Nation
Source: American Community Survey (ACS); 3 -Yr Avg for 2006 - 2008
CGS File Name: S: \common \Dubuque \Report - Final \Charts and Tables \dbq_arrest_race
Note — Figures here are exclusive of juvenile arrest counts, with the exception of a few juvenile arrests
that were included in the file forwarded by Dubuque to CGS.
Table 14 represents the figures used for the analysis in this section comparing authorized
Section 8 participants to non - Section 8 participants (arrestees, victims and complainants; unique vs.
total). The data protocol used for the analysis in this section is important to note here.
The data protocol rule that applied to determine authorized Section 8 participants is as follows:
Each person within the Merged Incident & Arrest Data File contains a unique
identification number based on a unique ID number (SSN) or other identification
2010 Dubuque Crime and Poverty Study Summary Report Page 38 of 779 center for
Y Y ry p g iUGovernmental Studles
resa_
assigned by Dubuque Police Department (i.e., Name Number). This data was then
converted, in a secure environment, to a unique identification number for each
arrestee, victim, and complainant or authorized housing program participant. This
unique identification number from the crime files (arrestee, victim, and complainant)
was compared to the unique identification number given to authorized persons
contained in the housing data file. If an identification number for a given arrestee,
victim or complainant within the Merged Incident & Arrest Data File matched an
identification number contained in the Housing "In and Out" Participation Data File AND
the incident event date Merged Incident & Arrest Data File matched the authorized
Section 8 participation time frame (e.g., arrest occurred January 2,2007 and individual
was listed as an authorized participant from December 15, 2006 through July 31, 2008)
in the in the Housing "In and Out" Participation Data File, then the arrestee, victim
and /or complainant was deemed to be an authorized Section 8 participant arrestee,
victim and /or complainant. It is understood that individuals from time to time switch
addresses within the Section 8 program, thus this rule will account for the individual
based on their identification regardless of the address given.
The CGS study team recognizes that there are weaknesses in this data and cautions the reader
of this analysis to recognize that the results and findings can be limited by a variety of factors in the
data that are not easily resolved including the following: individuals falsifying information, giving wrong
or misleading addresses and /or errors in record keeping (e.g., providing or recording incorrect SSN's or
other identifications), or lag time in data entry. Thus, there are limitations to the analyses presented
here. Furthermore, subsequent analyses of crime incident data, arrest data, victim data or Section 8
data that attempts to assess the impact or involvement of non - authorized Section 8 participants may
produce different results. It should also be noted that the analyses that include authorized Section 8
participants concurrently represents, by definition, a proportion of the population living near or below
the poverty level.
2010 Dubuque Crime and Poverty Study Summary Report Page 39 of 779 Nil t Center for
Governmental Studies
The data in the following chart illustrates the relative arrestee rate of persons arrested by
Dubuque Police for the time period of 2006 -2009. The data tabulates unique arrestees and compares
the unique arrestee ratio of non - Section 8 program arrestees to the non - Section 8 population against
the unique arrestee ratio of authorized Section 8 program participant arrestees to the total authorized
Section 8 population in Dubuque. As the data in following chart indicates, on average, unique arrestees
within the authorized Section 8 population experienced a higher ratio of arrests within their population
group (5.4 %) than did the non - Section 8 arrestees within the non - Section 8 population (3.4 %) during
the period that was examined.
ES -Image 7: Comparison of Dubuque's Unique Authorized Section 8 Arrestees Per Capita to non - Section 8 Unique Arrestees
Per Capita
Comparison of Dubuque's
Unique Authorized Section 8Arrestees Per Capita (% to TtI Section 8 Participants) to
Non- Section 8 Unique Arrestees Per Capita (% to TtI Non - Section 8 Population)
7.0% -
6.0%
6.0% - 5.5% 5 . 6 % 5.4%
48% F
5.0% : rz Unique Authorized Sect 8 Arrestees % to
4.0% - 0 3.5% 3.5% 3.4% 3.4 Sect 8 Population
��'3.4/0
3.0% i o Unique non -Sect 8 Dubuque Arrestees % I to non -Sect 8 Dubuque Population
2.0%
1.0%
2006 2007 2008 2009 4 -Year Avg
Source: Dubuque Housing and Arrest Data Files and U.S Census Bureau.
CGS File Name: S: \common \Dubuque \Report - Preliminary\Charts_Tables_Figures \Final FrequencyStats \S8 to NonS8 Comparisons.xls
Viewing the authorized Section 8 arrest data another way (see chart below), authorized Section
8 participants, who on average represent approximately 5.2% of the City's population from 2006 -2009,
account for unique arrestees in the City at only a slightly higher average rate (5.8 %) than the
percentage of the population they represent over the time period of 2006 -2009. However, it is
important to recognize that this data represents only authorized Section 8 cross - matching, (using the
data analysis protocols listed earlier in this section) and should not be inferred to include non -
authorized persons who may be associated or implicated in Section 8 crime incidents or arrests.
2010 Dubuque Crime and Poverty Study Summary Report Page 40 of 779 NJ Center for
Governmental Studies
kv e.s.®cn
ES -Image 8: Comparison of Dubuque's Authorized Section 8 Participants per Capita (Total City) to Authorized Section 8
Unique Arrestees as a Percent to Dubuque's Unique Arrestees
Comparison of Dubuque's Authorized Section 8 Participants per
Capita (Total City) to Authorized Section 8 Unique Arrestees as a
Percent of Dubuque's Unique Arrestees
8.0% -
Sect 8
Participants 7.0%
per Capita
-
(Represented 6.0%
as percent of 5,0% -
total City f 1, 1
population) 4.0% ; - R ,
■ Sect8 Share 3.0% ' u
of Dubuque's e
Unique 2.0% - 1
Arrestees 10% -
0.0% . =
2006 2007 2008 2009 AVG
Sect8 Participants per Capita
(Represented as percent of total 5.3% 5.4% 4.9% 5.4% 5.2%
City population)
Sect8 Share of Dubuque's Unique
Arrestees 5.8% 5.3% 5.4% 6.8% 5.8%
Source: Dubuque's Arrest Data, Housing Data Files and U.S. Census Bureau.
CGS Data File: C: \Documents and Settings \yl3gtkl \My Documents \Dubuque 2010 Study\
Dubuque Sec8 Participants as % of City Population Compared to Unique Arrest.xlsx
2010 Dubuque Crime and Poverty Study Summary Report Page 41 of 779 N11 lcenier...
a e nial St udies
Further analysis indicates that authorized Section 8 participants, on average, had similar
victimization rates as non - Section 8 residents for the same time period (2006 through 2009), with
Section 8 participants having a slightly higher victimization rate than non - Section 8 residents (5.9% to
5.4 %, respectively). A review of the trends below indicates the tendency for authorized Section 8
participants to be victimized was slightly higher in 2006 but leveled off to a fairly equivalent
victimization rate of non - Section 8 residents beginning in 2008.
ES -Image 9: Comparison of Dubuque's Unique Authorized Section 8 Victims Per Capita to non - Section 8 Victims Per Capita
Comparison of Dubuque's
Unique Authorized Section 8 Victims Per Capita (% to Ttl Section 8 Participants) to
Non-Section 8 Victims Per Capita (% to Non-Section 8 Population)
7.0% - 6.5% 6.3%
5.9%
6.0% _ - 5 5% t 5.6% 5.6% 55% -, 5.4%
5.0%
5.0% F
5.0% p r
r = "
Unique Authorized Sect 8 Victims %to
4.0% r I E Sect 8 Population
3.0% • Unique non -Sect 8 Dubuque Victims % to
o non -Sect 8 Dubuque Population
1.0%
l
r=
0.0% r.,.
2006 2007 2008 2009 4 -Year Avg
Source: Dubuque Housing and Arrest Data Files and U.S Census Bureau.
CGS File Name: S: \common \Dubuque \Report - Preliminary\Charts_Tables_Figures \Final FrequencyStats \S8 to NonS8Comparisons.xls
2010 201 Dubuque Crime and Poverty Study Summary Report Page 42 of 779 center for
--- ry a 4
P g --- NW Studies
. ate..,,
From years 2006 to 2009, the City's non - Section 8 residents experienced a somewhat higher
complainant rate (6.2 %), on average, than its Section 8 participants' rate (5.7 %) during the same time
period. See chart below.
ES- Imagel0: Comparison of Dubuque's Unique Authorized Section 8 Complainants Per Capita to non - Section 8 Complainants
Per Capita
Comparison of Dubuque's
Unique Authorized Section 8 Complainants Per Capita (% to Ttl Section 8 Participants) to
Non - Section 8Complainants Per Capita (% to Non - Section 8 Population)
7.0% -
5.9% 6.2% 6.0% 6.4% 6.0% 6.1% 6.2%
e
0
6.0% 5.6% 5.7 0
5 . 3 %
5.0% ;
D Unique Authorized Sect 8 Complainants %
4.0% to Sect 8 Population
3.0% r ,_ m Unique non -Sect 8 Dubuque Complainants
i - % to non -Sect 8 Dubuque Population
2.0%
1.0%
0.0%
2006 2007 2008 2009 4 -Year Avg
Source: Dubuque Housing and Arrest Data Files and U.S Census Bureau.
CGS File Name: S: \common \Dubuque \Report - Preliminary\Charts_Tables_Figures \Final Frequency Stats \S8 to NonS8 Comparisons.xls
On average, as shown in the chart below, from years 2006 to 2009, authorized Section 8
participants experience somewhat similar rates of unique arrestees, victims and complainants per
Section 8 participants, or per capita (5.4 %, 5.9% and 5.7 %, respectively). Authorized Section 8 unique
arrestees per capita remained fairly flat over the 4 year period of 2006 to 2009. In 2009, authorized
Section 8 unique arrestees represented 6.0% per capita, compared to 5.5 % per capita in 2006. At the
same time, authorized Section 8 unique victims per capita has declined (5% per capita in 2009 compared
to 6.5% per capita in 2006) and unique complainant counts decreased slightly for the same time period
(5.9 % per capita in 2006 compared to 5.6% per capita in 2009).
2010 Dubuque Crime and Poverty Study Summary Report Page 43 of 779 N1U Center for
Governmental Studies
�, awn
ES- Imagell: Dubuque's Authorized Section 8 Unique Arrestees, Unique Victims & Unique Complainants: % to Dubuque's
Authorized Section 8 Participants
Dubque's Authorized Section 8 Unique Arrestees, Unique Victims & Unique Complainants:
% to Dubuque's Authorized Section 8 Participants
:- Sect 8 Unique Arrestees %
7.0% - to Sect 8 Unique
Participants
6.0% - t,
tt 0 Sect8 Unique Victims %
; t to Sect 8 Unique
5.0/.: ° 'z Participants
� t,
4.0% - _ G Sect8 Unique
ti Complainants % to Sect 8
AZ 3.0% Unique Participants
2.0% [ - ` t
1.0% -
� r
4
2006 2007 2008 2009 Year
Avg
Sect8 Unique Arrestees% to Sect 8 Unique
5.5% 4.8% 5.6% 6.0% 5.4%
Participants
Sect 8 Unique Victims % to Sect 8 Unique
Participants 6.5% 6.3% 5.6% 5.0% 5.9%
Sect8 Unique Complainants % to Sect8 Unique
Participants 5.9% 6.0% 5.3% 5.6% 5.7%
Source: Dubuque Arrests, Incidents & Housing Data Files.
CGS File Name:S:\ common \ Dubuque \ Report - Preliminary \Charts Tables Figures\ Final Frequency Stets \ S8 to NonS8 Comparisons.xls
Per the following chart, total authorized Section 8 arrestee, victim and complainant rates are, on
average, relatively similar to each other. Also, total authorized Section 8 arrestees, victims and
complainants experienced similar trends as unique arrestees, victims and complainants; authorized total
Section 8 victims decreased at a more noticeable rate from 10% (2006) to 7.1% (2009) and authorized
Section 8 complainants remained fairly flat (8.5% in 2006 to 8.1% in 2009) during the time period under
study. Authorized Section 8 arrestee percentage (per authorized Section 8 participants) remained the
same in years 2006 and 2009 (8.9% for both time periods).
2010 Dubuque Crime and Poverty Study Summary Report Page 44 of 779 Ntucenrer for
q y w overnmenfal Studles
ES- Imagel2: Dubuque's Authorized Section 8 Total Arrestees, Total Victims & Total Complainants
Authorized % to Dubuque's
Section 8 Participants
Dubuque's Authorized Section 8 Total Arrestees, Total Victims &Total Complainants:
to Dubuque's Authorized Section 8 Participants
12.0% - +=I Sect8 Total
Arrestees % to
Sect 8 Unique
10.0% - Participants
8.0% . ; 1 uSect8Total
Victims %to
Sec t8 Unique
6.0% _ _ Participants
n
f
4.0% ®SectBTotal
r t� Complainants
_ to Sect 8
2.0% - _ Unique
h — _ - Participants
i :',a r' "s
2006 2007 2008 2009 4 -Year Avg
Sect8 Total Arrestees %to Sect8 Unique
Participants 8.9% 7.8% 8.3% 8.9% 8.5%
SectBTotal Victims %to Sect8Unique Participants 10.0% 9.7% 8.0% 7.1% 8.7%
SectBTotal Complainants % to Sect8 Unique S 5% 9 9% 7.5% 8.1%
8,5%
Participants
Source: Dubuque Arrests, Incidents & Data Files.
CGS File Name:S: \common \Dubuque \Report - Prelim (nary\Ch arts_Tables_Figures \Final Frequency Stets \S8 to NonS8 Comparisons.xls
2010 Dubuque Crime and Poverty Study Summary Report Page 45 of 779 A /II lcenterror
Governmental Studies
tiQ
Lastly, the chart below depicts the percentages of Dubuque's authorized Section 8 unique
arrestees as a percent of all unique arrestees in the data base from years 2006 -2009 as well as total
authorized Section 8 arrestees as a percent of total arrestees in the City. Total and unique arrestees
follow the same trends from 2006 to 2009 but vary in their rates possibly due to increases or decreases
in multiple offenders.
ES -Image 13: Dubuque's Authorized Section Arrestees as a % of All Arrestees: Unique Authorized Section 8 Arrestees % to All
Unique Arrestees vs. Total Authorized Section 8 Arrestees % to Total Arrestees
Dubuque's Authorized Section 8 Arrestees as a % of All Unique Arrestees:
Unique Authorized Section 8 Arrestees vs. Total Authorized Section 8 Arrestees
8.0% - 7.1%
7.0% 6.7% 6.8% 6.4%
5.8% 6.0% 5.7% ��tis . 5.8%
6,0% - 53% 5.4%
5.0%
ta Sect 8 Unique Arrestees % to Unique
4.0% Arrestees
3.0% ® Sect 8 Total Arrestees % to Total Arrestees
2.0% k
Th" CIA
1 3 " -s a
2006 2007 2008 2009 4 -Year Avg
Source: Dubuque Housing and Arrest Data Files.
CGS File Name: S: \common \Dubuque \Report - Preliminary\Charts_Tables_Figures \Final FrequencyStats \S8 to NonS8 Comparisons.xls
2010 Dubuque Crime and Poverty Study Summary Re Page 46 of 779 centertor
Governmen Studies
Section III -C: Quantitative Statistical Modeling Analysis and Spatial Analysis
Loglinear Analysis of Dubuque's Crime and Arrest Data
In order to provide a more structured assessment of the possible association of crimes, victims,
and arrestees, the study team undertook a loglinear analysis of key variables thought to be good
indicators of possible associations or interactions between phenomenon related to the prime questions
being explored in the present study. The loglinear approach is a statistical tool used to identify multi -
way associations between three or more variables. In loglinear analysis, there are no dependent or
independent variables. Instead, the test simply looks for interactions or associations that might exist
between the selected variables (for a discussion of Loglinear Analysis see Knoke and Burke's Loglinear
Models, 1980).
To examine how or if S8 housing units are implicated in Dubuque's crime, the study team
combined and coded information received from Dubuque into a Combined Crime Incident data file. This
file contains only one record for each crime incident. Through data coding, the analysts captured
information about the type of crime, number of victims, number of arrestees, and addresses of the
incidents, victims, and arrestees. To further explain, a crime incident[I] is a combination of the type of
crime, the location of the incident, the victim's, and the perpetrator's address. The associations among
these locations tells us something about the implication of S8 addresses in the distribution of crime in
Dubuque, and more generally about the associations of crime addresses in the social and economic
distributions that are reflected in S8 or non -S8 address status. Since the primary focus of this report is
on the implication of S8 addresses in Dubuque crime, each of the associations involving the addresses of
the incident, victim, and arrestee are detailed.
The analysis is looking for interactions between variables. A positive 2 -way interaction between
any of the location measures implicating S8 addresses indicates that the probability of the location
involving a S8 address increases if the other location variable involves a S8 address. The association
indicates a greater probability than one would expect from independent, random crime, victim, and
arrestee locations. If the probability of one aspect of the crimes location, say its victim, having a 58
address is increased beyond random expectation if the arrestee also has a 58 address, the pattern of
crime is circumscribed, probabilistically speaking, by S8 location of the victim and arrestee.
2010 Dubuque Crime and Poverty Study Summary Report Page 47 of 779 Center for
q Y y p - N'U GovernmentalStudies
air
Results of a loglinear analysis of the three location indicators show statistically significant,
strong, and positive two -way associations between all three S8 categories indicating an event location,
victim, and arrestee address (see Table 22 in Section III -C ). These patterns indicate that crime in
Dubuque, to the extent that S8 housing units are implicated, consists of a greater proportion of shared
S8 addresses among incident locations, victims, and arrestees than can be attributed to random
patterns. Crime is circumscribed by mutual S8 addresses.
The distributions of 12 types of crime by S8 location and non -S8 location show that violent
crimes make up nearly twice as high a percentage of crimes occurring at S8 addresses than the share
reported with non -S8 addresses. (1.91 times as high for UCR- Violent crime and 1.71 times as high for
non -UCR violent crime (see Table 16 in Section III -C ). The only other type of crime that occurs at a
much higher proportion at S8 addresses than non -S8 addresses involves local ordinance charges; the S8
rate is over five times larger than the non -S8 rate (10.56/1.95) (see Table 16 in Section III -C ). Property
crimes, DUIs, drug and alcohol related crimes, and civil disorders make up a lower proportion of the
crimes at S8 locations than they do at non -S8 locations (see Table 16 in Section III -C ). The types of
crime committed at S8 locations don't change much over the period 2006 -2009 with one notable
exception; there is a large increase in share of crimes classified as violations of local ordinances that
occurred in 2008 and 2009. The distributions of crime types at non -S8 city locations are very stable over
the same period (see
2010 Dubuque Crime and Poverty Summary Report Page 48 of 779 N1 i centerfor
q ___ -. St Y Governmental Studies
faV al
Table 17 & Table 18 in Section III -C).
Analysis of arrestee and victimization rates indicate that the biggest gaps between S8 residents
and other City of Dubuque residents are found when one adds in the S8 "persons" who gave S8
addresses in the crime incident file but were not matched to a S8 address by personal ID in the housing
file. When these S8 "persons" are added together with the official /authorized S8 residents, the rates of
victimization are overall 1.60 times greater for S8 residents than non -S8 residents living in Dubuque; and
the arrest rates for S8 residents are overall 2.49 times as high as the rates for non -S8 Dubuque City
residents (see Table 19 & Table 20 in Section III -C). If these address matches are trusted at face value,
city officials and S8 participants need to be aware that this "unseen" S8 population increases even
higher both victimization and arrest rates for this group of citizens relying on public housing who are
already at above average risk levels.
Associations exist between locations of incidents, victims, and arrestees indicating the following:
• The probability that a victim with a S8 address was victimized at a S8 location is greater
than expected on the basis of randomness;
• The probability that a victim with a S8 address was victimized by an arrestee with a S8
address is greater than expected on the basis of randomness;
• The probability that an arrestee had a S8 address when the crime occurred at an S8
address is greater than expected on the basis of randomness (see Table 23— Table 25 in
Section III -C).
The above associations involving Section 8 housing units implicate S8 housing in patterns of
crime that circumscribe the incident & victim, incident & arrestee, and victim & arrestee locations more
than random, independent processes predict. When crimes occur at S8 locations, the victims are more
likely to be S8 residents than predicted by chance. When crimes occur at S8 locations, the arrestees for
those crimes are more likely to be S8 residents than predicted by chance. And when victims live at S8
locations, the arrestees are more likely to live at S8 locations than predicted by chance. In sum, the
pattern of interactions involving S8 address locations among incidents, victims, and arrestees is
consistent with the interpretation that crime in Dubuque is probabilistically circumscribed by S8 address
involvement rather than diffused across address locations.
Spatial Autocorrelation Analysis
2010 Dubuque Crime and Poverty Study Summary Report Page 49 of 779 N 11 l cent e rto r
Governmental Studies
kv e.+n.rn
The purposes of this analysis were to depict the extent of spatial autocorrelation (henceforth,
simply referred to as spatial correlation) among crimes at the block level and, to display the locations of
areas with high levels of spatial correlation within the City of Dubuque. For comparative purposes, the
study team selected two years of data to explore for the spatial correlation analysis. The years selected
were 2006 and 2008 because of the completeness and quality of the incident data for these years. The
visual depiction of the results can be found in the accompanying maps.
Generally, the analysis indicates that the level of spatial correlation of blocks' crime rates overall
is low to moderate in Dubuque. The crimes exhibiting the higher levels of spatial dependency were UCR
Violent, Civil Disorder, and Drugs & Alcohol. UCR Violent and Non -UCR Violent crimes appeared to
diffuse, especially in downtown blocks: some blocks with low or average rates of violent crime in 2006
that neighbored blocks with high rates became blocks with higher than average rates in 2008. UCR
Property crimes, in the downtown area, exhibited the opposite tendency, to become more average
when neighboring blocks with high rates in 2006.
Spatial autocorrelation analyses are not as robust as temporal or cross - sectional correlation
analysis. Before assumed definitive, the results presented here should be compared to other results
with different specifications. Furthermore, explanation of crime patterns across geographical areas
should also take into account the social and economic characteristics of those areas along with any
spatial dependence before firm conclusions are drawn. Because of the relatively small population counts
in block areas, detailed demographic information about residents in those areas is not available. The
units of analysis in this statistical exercise began with complete data set of crime incidents. Those
incidents were aggregated up to the census block level for each of the 1085 census blocks in Dubuque.
Using the Combined Crime Incident analytical file that contains the census block identifier, the geocoded
latitude and longitude location of each unique crime incident, the type of crime (coded by CGS staff into
12 ranked categories), and the year the incident took place, the study team then analyzed the crime
incidents by census blocks to examine the spatial correlation. For each block and year, the numbers of
12 types of crimes occurring in each census block were tabulated. These crime count variables became
the focus variables of the spatial correlation analysis. Analysis was conducted on the following types of
crime counts aggregated by blocks:
• UCR Violent
• UCR Property
• Non -UCR Violent
2010 Dubuque Crime and Poverty Study Summary G
Report Page 50 of 779 NIU ceove nier for
-- -_ _ . rnm ental Studies
kv a
• Non -UCR Property
• Civil Disorder
• Drugs & Alcohol Related Crime
From the measures of the local Moran I, maps are generated that characterize every census
block as one of four types:
High Value - High Neighbors
A block with a high standardized crime count (> 2) — meaning the block's value on the
standardized crime variable is more than 2 standard deviations above zero, the average
standardized crime score for all blocks, and the block is connected to neighboring blocks with
similarly high standardized crime counts.
Low Value — Low Neighbors
A block with a low standardized crime count (< 2) — meaning the block's value on the
standardized crime variable is more than 2 standard deviations below zero, the average
standardized crime score for all blocks, and the block is connected to neighboring blocks that
have averaged standardized crime counts that are similarly 2 standard deviations below the
mean value of crime.
High Value — Low Neighbors
A block with a high standardized crime count (> 2) — meaning the block's value on the
standardized crime variable is more than 2 standard deviations above zero, the average
standardized crime score for all blocks, and the block is connected to neighboring blocks that
have averaged standardized crime counts more than 2 standard deviations below the average.
Low Value — High Neighbors
A block with a low standardized crime count (< 2) — meaning the block's value on the
standardized crime variable is more than 2 standard deviations below zero, the average
standardized crime score for all blocks, and the block is connected to neighboring blocks with
standardized crime counts that average more than 2 standard deviations above the mean crime
count.
UCR Violent Crime
The UCR violent crimes of aggravated assault, murder, rape, and robbery exhibit a relatively high
level of spatial dependence. Many if not most of the downtown area blocks are classified as High Value
— High Neighbor types of blocks (n =86 in 2006 and n =118 in 2008). The spatial pattern in 2008 shows a
slightly expanded area of blocks of this type than appears in 2006, indicating some diffusion of these
2010 Dubuque Crime and Poverty Study Summary Report Page 51 of 779 IVIU Cenler for
Governmental Studies
kv
most violent types of crime to adjacent areas. There are also a few more blocks in 2008 with higher than
average UCR violent crime rates connected to neighboring blocks with lower than average violent crime
than appeared in 2006 ( 18 v. 8). Most of these blocks are located in the center of Dubuque's
geographical area but west of downtown.
UCR Property Crime
These serious property crimes of arson, auto theft, burglary, and larceny show less spatial
dependence and less clustering than did UCR Violent crime. For these crimes, the number of blocks with
above average levels connected to neighboring blocks with above average levels numbered 42 in 2006
and 40 in 2008. The clusters and outlier blocks in the southern and northwestern portions of the city
were similar in 2006 and 2008. In the downtown and northeastern areas of the city, the spatial
distribution of UCR property crime appeared more average in 2008 than it did in 2006 when more blocks
were more likely to be classified as high -high clusters or high -low outliers.
Non -UCR Violent Crimes
These types of crimes exhibit the weakest spatial dependence of any of the crimes examined.
Again, blocks designated as belonging to the High Value — High Neighbor are the modal type for those
blocks fitting into one of the four types. We also observe 12 blocks in 2006 and 18 blocks in 2008 with
above average Non -UCR Violent crime rates with below average neighboring blocks. We also note that
the number of blocks with above average Non -UCR Violent crime increased from 2006 to 2008 of both
types, those neighboring higher than average blocks and those neighboring lower than average blocks.
As did UCR Violent crimes, the Non -UCR Violent crimes appear to involve more blocks at higher
levels in 2008 than 2006 diffused somewhat more into neighboring blocks, particularly in the south and
southwest areas of the downtown.
Non -UCR Property Crimes
These types of crime show a lower level of spatial dependence among the types of crimes
examined. The most common clustering again is of blocks into the High Value — High Neighbor clusters;
and the number of these blocks so clustered increased from 55 in 2006 to 92 in 2008. In the west areas
of the city, there is some evidence of spatial diffusion of these types of crimes from 2006 to 2008: 2006
blocks with above average crime levels neighbored by blocks with average or low levels of crime
2010 Dubuque Crime and Poverty Study Summary Report Page 52 of 779 center rot
Y Y ry P g ��UGoernm enialSludles
became, in 2008, blocks with high crime neighbors. There is some expansion of the number of high
crime blocks with high crime neighbors in the southern part of downtown.
Civil Disorder
These types of crimes, along with UCR Violent crimes, display the most pronounced spatial
dependencies of the crimes analyzed. The large numbers of blocks with above average levels of civil
disorder neighbored by other blocks with high levels of civil disorder are concentrated in the central and
southern downtown areas (100 such blocks in 2006, 106 in 2008). Interestingly there are a substantial
number of downtown blocks with neighboring blocks having high levels of civil disorder that themselves
have low levels of disorder (n =33 in 2006 and n =36 in 2008). The spatial distribution of civil disorder
seems to implicate adjoining blocks; there are few instances of blocks with high rates of civil disorder
neighboring blocks with low rates (n =10 in 2006 and 2008).
Drugs and Alcohol Crimes
These crimes, as did civil disorder and UCR Violent crime, show higher levels of spatial
correlation. Nearly all of the blocks implicated spatially are classified as blocks with high levels of crime
connected to neighboring blocks with equally as high average levels. Nearly all of these blocks are
located downtown, and the number of blocks like this expanded between 2006 (n =89) and 2008
(n =103). All but one of the newer blocks added to this type of cluster were in the downtown area and
neighboring blocks with high drug and alcohol crime levels in 2006.
Table 27 summarizes the results of the spatial correlation analysis conducted on six crimes
comparing the spatial correlations for the data sets from 2006 and 2008. Samples of the spatial results
are presented below. The full presentation of the data and maps can be found in Section III -C of the
Report.
UCR Violent Crimes Spatial Correlation Maps
The UCR violent crimes of aggravated assault, murder, rape, and robbery exhibit a relatively high
level of spatial dependence. Very similar to the Non -UCR Violent Crime clustering, the UCR violent
crimes are primarily clustered along Central Avenue, south of 25 th to 16 th , though some scattered blocks
as far south as Emmett also have High value high neighbors. (See Maps that follows ES -Map 5 and 6)
2010 Dubuque Crime and Poverty Study Summary Report Page 53 of 779 l I I Center for
Governmental Studies
ES- Map 5
Spatial Correlation of 2008 UCR V Crimes
k ,„\\
r
r k
_Y''''
-7 ''
fr
5 '*
r t 'tkh t
~�
d Y- 1 r mit_ � j
� 4f
`y J Llllralj �4 +3idliuyEyt( o•d
I lb. t r%,..,.„„i„,.,,,,,..„4„,.A. kt1t1 +., ,,,a .
:
e ...i....,„ izs . fl i a, ail
. /, 41, J f El i
Ol — \.7.— V :z '''' , -x,...
Cluste r - Outlier Type
# Ho significant clustering (rt=81)
High value ( >2 SD)1wiithhigh neighbors (n =118) 1
} Huh value (>2 SD) with tow neighbors (n =18)
Low value ( -2 SD) witihhigh neighbors (n=58)
Gala Enloe Cemtirsed_OrtnejadderdfietriS_UORantiLdattiad dot CGS
lilt of araftatr Canna aWclts 21X1 0 (n -10851
isxsrs It 0.1217-15.935
HOh vanes ale mare than Um standard dertadan abase he mean
tats times are mve Maw twost5rdant dedetera team the meal
Created 1Z1512:110 MILS
NtiJ' ,A`,1:IL.,a 0 0.5 1 2
2010 Dubuque Crime and Povert Stud Summa Re port Page 54 of 779 centertor
p Y Y Re g Governmental Studies
w.nh
ES- Map 6
Spatial Correlation of 2006 UCR Violent Crimes
\\ N :
-
a
-
%. It,
.\ 5.t -
ti W � 4 lam¥ r 11 11 tf r— i s ►� k ~ - its 4 s I
1 -�� • art 1NJ L 1 Y -r
_ .: .90.11,,,,lok,liv,,,,,...L._„,,,,i,:lkalietv:,,,r,.:;..,,,,,,;..tior,,_
, 1 l : a
is li
.4 _...._......., �� 9:0, .t.z
iii .„4„....e ...,,....1.,.........„ 4
it .4. . 1 jo -.
U r, x , ill
Cluster - Outlier Type`
1 No signiTcant clustering (n='85)
High vaue (>2SD)with high neighbors (n46) ) 1
I High value {> 2 SD) with law neighbors (n =8)
I Low value {<-2 SD) with high neighbors (9 =28) 1
a u 5ou[E C zm bred_cr re_Lsc nav2g ucranndLaratlonder, CGS
thief alxysb: Census Sexes. 2005 tn.16851 .
Aim" 1. 1.07 LZ 516Y
Hph vafue s are r ae M an two stnndad de tton, abase the mum
Lour nans a re rmare 5.0 st»dard deslffikrs beat the reon
Created 12t13P201fl Miles
NIU. 'n rI:1she. 0 0.5 1 2
2010 Dubuque Crime and Poverty Stud Summary Report - Page 55 of 779 N l leenler
q Y Y Govarnmantol Studies
UCR Property Crime Spatial Correlation Maps
These serious property crimes of arson, auto theft, burglary, and larceny show less spatial
dependence and less clustering than did UCR Violent crime. Like the Non -UCR Property Crime, the High
value high neighbors' blocks of UCR Property Crimes are clustered in three areas. The most number of
such blocks are again along the Central Avenue corridor, west of the street between Primrose and
Valeria and east of Central along 24 The second largest clustering is in the south along Dodge,
between Dodge and Brunswick. The third cluster is in the northwest part of the City, along Arterial and
Plaza. (See Maps that follows ES -Map 7 and 8)
2010 t e
and Poverty Dubuque Crime anovr Study Summary Report Page 56 o 779 NW an Y d Y rY f p g NIU Govem er m ror
ental Studies
ES- Map 7
Spatial Correlation of 2008 UCR Property roperty Crimes
i,
_ \\.
. \\\
_ ..---- __\:, A ,, a.
,.._.... .
_...-,
_,—..........., .,,„ 4,4_
nig II tilittLTI 7,, /
- -
:VP(' iZiali ' ''
....sost. .... . 1 . ,ia=_,__orealiE-..,....27rs.Weigthi.
' I lo pt i _- ---- 4,111------- 4
/
. ..„,,,i,.t.7,..,5
.....„0...- , ,., ...:
:,-,.. .._,..... ,., 1.74 4
; ,,,,
1 .4 -v. ni
,_.- ' AMPI
k • '-'-z- „ ,,,,,
T ,aiir
- 1
Cluster-Outlier Type ,1
{---] go signirreantchrstenng (n=1g31) -T-1
H with high neighbors (n=40) 1
SD) with tow neighbors 0=11)
High value (
1 Low value (<-2 SD) with high neighbors (tr
ozonsel.rem tiandSaing_Inctonnt, ralf4-. CGa
Lint of on.yob :Cennut Stocks 21:1110 in-10051
/Amara I: 0.115 [Z-a -39) ff.
fig ft VSMS nit mu r c than 11143 W tk
on't ilaanns .bom y: . e
mare lann two atalrd tla ne bet.t
st
I"' 1Z1503t0
imm==wommi Miles
0 0.5 1 2
2010 Dubuque Crime and Poverty Study Summary Report Page 57 of 779
NIUge.._!,%Lr`nt_ntal
ES- Map 8
Spatial Correlation of 2006 UCR Property Crimes
,\,,,,,,,,\.,
--\
):- - '\,,,,,,,
\' I. - )
0 1` _ a a1� -ttis y + P-- _
i S a ..tic.1 1g#,,,,..-
'.. c,-� 5,� w .
Jllilt t °tl 1ll an 4 t..5t.
-.2111111111 r. +�'� r ll° II � +� ��'+ � r1LW1'� r l ._ " 1 r , 1 it
w,Ac°s.�1,,y Pe
• 0..rag. II y }
a .01. k t
r
Gloster- QutlierType ,
' ( No signifcant ciustenng (n= 1029)
High value ( >2 SD) with high neighbors (o...2)
i ( High value ( >2 SD) with low neighbors (n =1-2)
Low value (e..-2 SD) with f u gh neighbors (n
soup Comoro d_rxxne_thcSder MCM29_UCAnndlom6xidbr, CGS
Una etarutys5: Censra Soda 2.9611 In -1x951
tkorarts I: e.tIE (2-52E)
Hphvarues am mare than bra standard deadens above the mean
Low veins are max 5SSrI. Iwo id ndam de lan:vs below the mean
a Miles
' edtVt3rtn 0 (� C �a
NIll: "M,1,..' l:Ifl,Nw 0 0.5 1 2
2010 Dubuque nd Study Summary Report a 58 of 779 *!tt center or
ue Crime Poverty Y Y rY ort Page _.. g NI�GOVern fmental Studtes
Analysis of Crime and Section 8 Hotspots in Dubuque
Hotspot analysis is a useful, summary descriptive analysis to increase understanding of the
patterns of concentrated distributions of crime in spatial areas. It is not designed nor is it capable of
explaining those distributions. Causal attributions about the role of Section 8 housing are not warranted
based on the analysis presented here and in Section III -C.
The summary of the hotspot analysis that follows depicts the extent of spatial overlap of
geographical concentrations of various types of crime and concentrations of Section 8 housing units in
the City of Dubuque. Concentrations of crime in space are termed "hotspots" and there are various
analytical tools used to find these hotspots, and all require user defined assumptions that can alter the
results.
Please note: the results reported here are conditional on the specifications the statistician
imposes on the analytical tools to identify Hotspots. The results have not been subjected to sensitivity
analysis to learn how the descriptions produced are dependent upon the algorithm choices made by the
analysts. Only after running several models using different specifications can robust findings emerge.
Readers should treat these results as exploratory only, and are conditional on how the analysis was
conducted.
The units of analysis were the unique crime incidents occurring in Dubuque between 2006 and
2009. For this analysis, 2006 and 2008 crime incidents are analyzed, the years that provided the most
complete set of addresses for analysis of total incidents.
Hotspot analysis was conducted on the following types of crime incidents and their geocoded
latitude and longitude coordinates:
• UCR Violent
• UCR Property
• Non -UCR Violent
• Non -UCR Property
• Civil Disorder
• Drugs & Alcohol Related Crime
All the results reported here are based on clustering algorithms that specified a threshold
distance of 500 feet. This is among the minimum distances used in hotspot analysis and seems
appropriate for capturing concentrated events in city center areas characterized by higher densities. Ten
2010 Dubuque Crime and Poverty Study Summary Report Page 59 of 779 Ntl Center for
Govemmental Studies
was the minimum number of events needed to create a hotspot cluster. Since data were analyzed by
year, this minimum corresponds to nearly an average of one event per month occurring at locations
within 500 feet of one another to classify as a hotspot.
The maps summarizing the results are included as the main part of this report. Each crime
analyzed produced geographically shaped hotspots (convex hulls that more closely trace the exact shape
of the hotspot than an ellipse would) that outline the polygon shape of the area enclosing all of the
crime events at that location. The hotspots based on Section 8 housing are overlaid on the same map so
one can visually determine the extent to which concentrated crime occurs at sites where there is also a
concentration of Section 8 housing.
A summary of the interpretations of the hotspot maps produced in this analysis are highlighted
below:
Section 8 Hotspots
On all of the maps, Section 8 hotspots are overlaid with the specific crime hotspots. The
distribution of Section 8 hotspots, n =63, are identical in 2006 and 2008.
From a housing policy perspective, Section 8 concentrations appear to be fairly dispersed, albeit
across the central, downtown areas of Dubuque. This is the more dense area of the city so it is not
unusual to see subsidized housing stock in these types of locations. Making use of the standing rental
stock and avoiding large scale, dense housing units is currently considered a best practice. However, if
these more scattered sites remain in poverty stricken areas, the cumulative disadvantage associated
with such areas may reduce the quality of life of residents and generate higher criminal victimization
and perpetration rates experienced by residents in these areas. Rehabilitation of housing stock and
locating Section 8 sites in mixed income neighborhoods can ameliorate these environmental conditions.
UCR Violent Crime
In 2008, only 13 UCR Violent crime hotspots were identified; only 17.5% of the Section 8
hotspots overlapped or were contingent with the UCR Violent crime hotspots. The analysis identified 10
UCR - Violent Crime hotspots in 2006 and 13 in 2008. Seven of the 10 in 2006, and 11 of the 13 in 2008
were contingent or overlapped with Section 8 hotspots. Aggravated assaults make up the majority of
these violent crimes, and poverty is correlated with this type of criminal violence, so this degree of
2010 Poverty P
and Dubuque Crime no Study Summary Report Page of N*r Center for
_ — Y d Y ry R 60 779 p g IY� Governmental Studies
6hw.ch
overlap is not unusual. Furthermore, all but two or three of the UCR Violent crime hotspots are in the
downtown area, suggesting that this geographical area with its commercial and entertainment
establishments increases the concentration of violent crime. Importantly, one must note that the vast
majority (82.5 %) of Section 8 hotspots are not implicated with the few UCR - Violent hotspots. Section 8
concentration does not 'generate' violent crime hotspots at high rates. See Table 27 in III -C. (See Map
that follows ES -Map 9)
2010 Dubuque Crime and Poverty Study Summary Report 61 of 779 t center for
q Y Std Y rY t Page P .. _ g NlU GovemmentalStudies
ES- Map 9
Hotspot Analysis of 2008 UCR Violent
grimes and Section 8 Housing
Hotspots
,
Section 8 Housing (n3) I,
1 1 - ---___-) --- .„ - .--s - - -, ,„- - .7-, - - A
UCR Violent Crimes (n=13) - 7 ; . -- ,
1
Dubuque City Boundary , s . T1, 2-_L, r
,. .._,-- ,, I '-•-' '' ,,_ 'Y- -- ',
---,
___ \ "'r ,'
, 1 -
I . ,,,..--" . s
` ' P ------' '' \ -i ' ''' -'—'-4_,,12,''?-, '-.,. ' ' r , ..' 1' ^; :O ■ 1 ' rs. -
•
- L _
I L
,,,,, 444_1,1 _u1
1 ii? .Icktt,i_ 9 'r, , I . , - .'" ---- ta .1 " -.1
Eli LJ___ -' r" ', 013 , tTi I'-v', ,, ,.;- "s,..`,:'1V
1 , —H—fT,.„,' f ff , ti ci __:. , f 1 ___\ 0 , , f '[, ' ,., ''' , c:'-iv - , ff.
- ' '-' — u 'L'''-- ' '7 1— k ri 1 Tr i — .Fr ''].ttHtL.--'4
H ,
+-
1 1 1:7) - 4 14 1 r 7 1 I rf j FL' "AA • 4---i 1- - -- , _.4y,
[-
A 1
- ' _ --
----'1
11 1
c =11' 1 47 '. :- '" ''' I \ I )
---- , ,,
,,,--
- -
L l''' ':rr } ,
1 ir
D , -- - ---rr ,
M Soinrz: Con'tnien... -z_
- '--;--,.
Crentztt 12114611113 \ 1
0 0.5
: Miles
- 1 ‘
Page 62 of 779 ar■ IS udles
2010 Dubuque Crime and Poverty Study Summary Report NIU
UCR Property Crime
Property crimes are much more frequent than violent crimes so the large numbers of UCR
Property crime hotspots are not surprising (n =46 in 2006, n =57 in 2008). Given also that the largest
category of property crimes, larceny, are more likely to occur in public places or shopping areas, it is not
surprising that the UCR Property hotspots appear in the denser, downtown area. We also see a much
greater level of overlap in the Section 8 hotspots and UCR Property crime hotspots; in 2008, two - thirds
of the Section 8 hotspots overlap or are contingent to crime hotspots (see Table 27); and all but a
quarter of the crime hotspots overlap or are contingent to Section 8 hotspots. (See Map that follows ES-
Map 10)
2010 Dubuque Crime and Poverty Study Summary Report Page 63 of 779 NIU Center for
Governmental Studley
ES- Map 10
Hotspot Analysis of 2008 UCR Property
Crimes and Section 8 Housing
Hotspots `
UCR Property Crime (n =57)
Section 8 Housing (n=63) 1
Dubuque City Boundary , L
I
, ,, , ,, _,-.,,, 1 — EI,_._
L ' ''
t'i r i� x ,
f
�^ cr , i
r1- - I I } r.v ±° 7 - X11- '71-:, '° r
' lfi l 1�, �gr i-` I ,
4 }I
t
-, ° �� 1 ` ' -`'` "'e
� ] " n 'om d
i r }z ,
J d y i :.
1
i {
a
1
Data altar!: eb_ CrF ! ° I
C_3$7Ad
created tvieilata \ t i ....:3 Miles
0. 1
2010 Dubuque Crime and Poverty Study Summary Report Page 64 of 779 All lcenfe for
wv xo-e�
- — C3avemmenfal Studies
Non -UCR Violent Crimes
Many Section 8 housing hotspots, even those in the downtown area, are not contingent with no
overlap Non -UCR Violent hotspots (in 2008, 84.1% in Table 27). However, ten of the 16 Non -UCR
Violent crime hotspots overlap or are contingent with Section 8 hotspots in 2006; and ten of 15 overlap
with Section 8 housing in 2008. Again, nearly all of the overlapping is situated in the downtown areas
where other factors likely mitigate the generation of non -UCR Violence beyond poverty per se. (See
Map that follows ES -Map 11)
2010 Dubuque Crime and Poverty Study Summary Report Page 65 of 779 NIU Centertor
Q Governmental Studies
ES- Map 11
Hotspot Analysis of 2008 Non -UCR
Violent Crimes and Section 8 Housing
Hotspots
Non -UCR Violent Crimes (n =15) C-)
a Section 8 Housing (n =63)
Dubuque City Boundary .'i.
.. 1 _ _
a S
} (,' �� _ � t
r
+ r- a, f Lam^ , 5 Y;
t,'-',/ 1 P i I (s L i i 1 4 YV ''
. mss t 1 1 1 S� �1 -, ` . f
t L',) ) A Es (t 1, i '' S IL, -f y '. r: 0" ° _,� ' y, ! 4 -
IIJ9� i,7-I r y S s ,tt �� �c�t '" 7
�' �I r Jp i a 1+1, Gil i S f
� .
fir.
� I
s
l v ri ' i _- f 5' I� l�,
� \
�l
k. t lit
1 I j
w �' — ci f� '��
/,: t
Deal source: eu- nr�ed—..Ed uw v u c�sm c2 .
created 121!.2019 1
cr
MIJcK, —,'•'' \ Miles
0 a 1
2010 Dubuque Crime and Poverty Study Summary Report Page 66 of 779 NII lcente
Govammentot Studies
by O.trn+.n
Non -UCR Property Crimes
Again the majority (81% in Table 27) of Section 8 hotspots are not implicated in Non -UCR
Property crime in 2008, especially those outside of the dense downtown area. All but three of the 21
Non -UCR Property crime hotspots overlap or are contingent with section 8 hotspots in 2006; and again
only three of the 19 crime hotspots show appreciable distance from Section 8 hotspots in 2008. Nearly
all of these hotspots are in the downtown area. (See Map that follows ES -Map 12)
2010 Dubuque Crime and Poverty Study Summary Re Page 67 of 779 *r Cen terror
q Y Y rY P 6 N�U GovernmenlalStudies
ES- Map 12
C
Hotshot Analysis 8 Housing
and Section Property Crimes a
Prop
Non-UCR
Hotspots
Non-UCR Property Crimes (n=
Section 8 Housing (n
4
Dubuque City Bo
)
-, - iii: .'
1H,I, ,, c t'i - cr
, c _i l , y , lLC -.0 -
1-4,-1 - , -1--:aigi
— ; y ) ,■,--..„ 13,J, li ._•.=, ,,,\,. '--_
' \ i ' ''''
, .„01-15.- , f 1 - ----.:)--- It
'.
--
-,kt4. ___,, ■ , t, I L.: --..— ) r 1, I- L,,,r ,Fe r„, -"1,.,A; ,,i,v,,,17j. 4 , " r. ,. , i ,- -
• ,. -_, ....,,,,.i.,,J: _Hz_Ht-,___,-----„,-,„•-,,,,,,----T,,,,,_.,,,,;„,,..0.„,c,i,„,„ -,, ..,,,— -. ,, _„.
1
: LI -.) 1 'i7-----_,::::,*--FTTE.-----jzili li ::::- , p •
1 1 kt*
, ot --•,,--, , \ L., ,, 0
2 "'• =1 . 1' ' ' '-'7=1-'7. Fri-Iii-1*-T-t5 • i I 'fFn A :"• : e,2 - (::-'; ' ,,, ,' ° ' : . r =LA - - dWill" -, X - 1--,s :1' , T -- r=li r
/ 7 -'''= - -: -
,..,, ., 1 =,4
:-1 ,' ,-.' ?a ...hil ig ...,
_71 — .-:11_ --• -, 1.-:-.,:,-,:;-,1 -- -# ill ,-
L____/ ,„, , , ,
_ — ,i,,_4...11:,_,,,
--,.-'
,,,
,
,., , f----,
F ,, t
I ,L
■
-- " ;11 II ',L____,,,',, ,/,' V ' ----\\‘'.--..
1 La__
_
1 2LticRam.......ct,
D.,. :CEmbfriedCTITP-4ndd"-N°V
C1111-611.1
I ' ---- 0 Miles 0 '' ' '
_, 7, , 0,5
creates! toterlow L '1, ,,,
[ .---,.- -4*----,-- u
1
I -1
NIU Center for
Page 68 of 779
2010 Dubuque Crime and Poverty Study Summary Report ,,,, Go , v ,a, ernmental Studies
Civil Disorder
The spatial distribution of Civil Disorder hotspots shows somewhat more of an implication with
Section 8 hotspots, in 2008, 25.4% are contingent with or overlap a Civil Disorder hotspot. All but five of
the Civil Disorder hotspots (n =20 in 2006, n =22 in 2008) were situated in the downtown area, and all but
one or two of these overlapped or were contingent with Section 8 hotspots. None of the Section 8
hotspots outside of this area were implicated with a Civil Disorder hotspot. (See Map that follows ES-
Map 13)
2010 Dubuque Crime and Poverty Study Summary Report Page 69 of 779 N1U Center tor
Governmental Studies
nom «.r„
ES- Map 13
Hotspot Analysis of 2008 Civil Disorder
Crimes and Section 8 Housing
Hotspots
I
Civil Disorder Crimes (n=22) I
Section 8 Housing (n=63)
-
i 1 1 T yL, _,-- --- -f.- -- __,,-_,
Dubuque City Boundary 7: I t L
I— .___
, ---- '
± ''`
tt
T
'
_
,
? __
—-- ,
1 ,I1 ,"
".., , „,:',.--:':"
u ' + -, -.,., , ..---__T----- ,-. ,. ;,,,,,.',!, , f-.: t =--
L ---=
j .. n --;-'' ,
= L
) '-=, r
i .
Li „„.:-. , Ji_LL_,,,,-,
_H Li— i—Trivr),'Ne,:,',;,--..,'[..,
'.y , ',
: ',,H.,'`'',1;1-'1,1:70:1T_,Ir:_
t
---:' -
0
,,,,11,:, k 417,, ,, i • 1— -1 I :-
I '-" _ --'-'-' r4 84 ,'''',
I 1 , j ,
- ` - '-- - -'-` .1- ,2,-),jn—nn'--,-----.!r.
") ' ''--,_.--n—d---,;:,-- d --
' lA ' i ' -
1 n t
—..---,,'.
■ — ' ra i ' , '
n.'T" , --
tt n. --- " -- .‘ \ i /
3'
I
_ 1
r: Carromecl_CrtrejnctcleTLNOV29_UCRenciLoc...VIanAte -'' ';''''' r ' '''''.',,_,
.
\
-J ,_., , ••■=:=3 Miles
Created 12/1461110
a 0.5 1
i I A
Page 70 of 779 .. IS udles
NtUar r ental
2010 Dubuque Crime and Poverty Study Summary Report
Drugs and Alcohol Crimes
There are few Drugs and Alcohol hotspots; one of the reasons why so few Section 8 hotspots are
implicated; in 2008 only 12.7% of the Section 8 hotspots overlapped or were contingent with these
crime hotspots. However, in 2006, all five Drug and Alcohol crime hotspots were located in the
downtown area and all were contingent with or overlapped Section 8 housing hotspots. In 2008, nine of
the eleven Drug & Alcohol crime hotspots were in the downtown area; all but one overlapped or was
contingent with a Section 8 hotspot. The analysis indicates an increase in concentrated drug and alcohol
crimes from 2006 to 2008. (See Map that follows ES -Map 14)
2010 Dubuque Crime and Poverty Study Summary Report Page 71 of 779 N�,r Center for
q Y Y fY r
p g 1V � GovernmentalStudies
tiA
ES- Map 14
Hotspot Analysis of 2008 Drugs and
Alcohol Crimes and Section 8 Housing
Hotspots ti `
4
Drugs and Alcohol Crimes (n =11) ;1
4 Section 8 Housing (n =63)
® Dubuque City Boundary , 1,-
vv L
. 7 i _, J ` 7 L 'f
— Cti 17 P t
1,—.— i9 \ o �, r- is
+ I 1- 7 —71-) a, am j .d s 1
, ' '� f T ; , f
) ,, 1 i r-2, I -11 ,,_ L''0'-r?, - " -- '.,i, , ', ... 1:":,, ° - 7 2 -- - , -, , itle; .-
� - c. 16 '1. n 4tn i t--
9 1 i i ) LI , A ' L",_4-i - V
i l l IN 0 LIi�1 , n i t 1 i � t
1 h� r �
I 19 � i l l � � , ,� - �
r 9e $t �
i t Y 1' , s a . 5�
, 1 T, y i 5 1 1 ,
— 1 , ] Z
r.
_, ., r _ 1 --jr + ,fir^
e sr f 1l r ' - 1
�s
Pf
tl u
Data Source: Car ed_Crtme_InCderLNCP/24_Lf4 .ac on.a:4
CC -S`NA!
Created t
131 he11 �1 - i
hJliJ` " " " t . r 4 Mires
I r -1 .�� o ai. 1
2010 Dubuque Crime and Poverty Study Summary Report Page 72 of 779 tal Studies
NIUCenter for
Governmen
Findings of Hotspot Analysis
The maps reveal that given the parameter settings we imposed on the hotspot analysis, there
were many more Section 8 hotspots than crime hotspots. It is also clear that many section 8 hotspots do
not overlap, nor are they contingent with crime hotspots.
The presence of a Section 8 hotspot in a location is not a necessary nor sufficient condition for
the generation of a crime hotspot. That is to say, crime hotspots emerge in other locations than areas
with concentrated Section 8 housing. More exhaustive analysis is recommended involving several
different specifications of the clustering parameters before results are deemed robust and definitive.
Summary
It is difficult to attempt to summarize an analysis as complex and multi- faceted as the one
commissioned by the City of Dubuque to study crime and poverty. However, this concluding section will
endeavor to bring forward some notable elements along with valuable context on the broader question
of crime, poverty and community experience. The readers of the analysis that has been presented here
are strongly cautioned to not take singular statements, findings, maps or graphs contained in the report
and examine or present them as a stand -alone finding. Rather, the analysis and report must be viewed
in the full context of the breadth of the examination, the totality of the findings and the broader social
factors that underlie the phenomenon under study.
For those who are committed to serving and leading our civic surroundings, it can be said that
great communities come with great expectations. The situation is no different in Dubuque where
decades of collective effort have elevated the City to notable recognition. Known as the "Masterpiece
on the Mississippi," the City of Dubuque has been recognized as one of the 100 Best Communities for
Young People, a finalist and gold- standard community at the International Awards for Livable
Communities (LivCom), the Most Livable Small City, and an All- America City. In 2009, Dubuque received
ICMA's America's Crown Community Award as well as the Excellence in Economic Development Award
for Excellence in Historic Preservation -led Strategies from the U.S. Commerce Department. Dubuquers
also take pride in their community, as 86% of residents rate the community as an "excellent" or "good"
place to live (NIU 2010 Comm. Survey). This high level of citizen satisfaction is a reflection of the quality
2010 Dubuque Crime and Poverty Study Summary Report Page 73 of 779 NIU ( Center for
Y Y Governmental Studies
of life Dubuquers associate with their community, their daily activities, and the expectation that local
leaders provide a safe, prosperous and progressive community.
Even in progressive communities, such efforts can be a challenge as local conditions are
susceptible to subtle societal forces that can alter the local landscape in terms of prosperity,
demographics, and cultural patterns. Throughout this report, levels of income and poverty are shown
to be strongly associated with various elements of crime. As local economic conditions are mostly
determined by public and private factors located counties, states, or oceans away, the preconditions for
crime within a city, including Dubuque, are often established apart from any community action. Indeed,
a recent report from the Brookings Institution demonstrates how Midwestern states have seen the
largest increases in poverty, which were further exacerbated by the 2008 economic downturn. As
times have turned tough, many cities have faced an increasingly noticeable crime problem.
Concerns regarding crime in Dubuque have been amplified by several recent high - profile
incidents, including: 1) the 2005 fatal stabbing of a pregnant teenager; 2) the 2007 fatal stabbing of a
man outside of a downtown bar, which stirred racial discussions; 3) two fatal stabbings in 2009, 4) the
2009 shootings of a police officer and suspect during an armed robbery attempt; and 5) the 2010
shooting of a man at a pawn shop. Due to the general location of the incidents, the perception that
those living in Section 8 housing units are increasing crime in Dubuque has been advanced by some
segments of the community. The City of Dubuque has responded to crime concerns by: 1) creating the
Safe Community Task Force, tasked with making recommendations to the City in order to increase public
safety; 2) establishing several programs to improve the Washington neighborhood; 3) focusing police
efforts upon troubled downtown areas; 4) creating a number of programs to assist the unemployed and
impoverished in regaining solid financial footing; and 5) tightening Section 8 housing eligibility
requirements and aggressively enforcing tenant responsibilities. These efforts represent a proactive
response that matches the progressive approach the City has displayed in a variety of community
initiatives including redevelopment, economic development, housing and public infrastructure. The
exploration and study of the relationships between crime and poverty in the present study represent a
continuation of the City's proactive approach to community issues.
The City of Dubuque has been committed to a professional approach to managing its affairs
based on analysis, expertise, review, dialogue, and objective decision - making. From this orientation,
6 Kneebone, Elizabeth, and Emily Garr. 2010. The Suburbanization of Poverty: Trends in Metropolitan America,
2000 to 2008. Washington, DC: Brookings Institution.
2010 Dubu Crime and Poverty Study Summary Report Page 74 of 779 N l cente rtor
Dubuque Y y Gov Studies
a.,Yn
and recognizing that crime remains a chief concern among residents, the City asked that a broad study
of crime and poverty be undertaken. In this context, the purpose of the analysis that has been presented
here is not to identify a "silver bullet," but to establish a foundation of understanding based on objective
data analysis and sound research principles that will be useful to frame future dialogue regarding crime
and poverty in Dubuque. There is no "one- size - fits -all" solution or strategy that can be prescribed for
the City of Dubuque or any other city. As discussed below, there is not a mysteriously elusive program
or policy decision that can be implemented to change the complex factors that contribute to the
presence of crime in a community. As the study team confirmed in the search of the literature, there is
not a singular recommendation available to direct a community on how to fight crime —every strategy
must be tailored to the needs, context, values and assets of the individual community. However, by 1)
gauging resident perceptions, 2) comparing crime attributes with similar communities, 3) examining
patterns of crime in Dubuque, 4) exploring potential connections of various demographic groups to
crime, including Sec. 8 participants, and, 5) canvassing the academic literature regarding crime and
poverty, this study provides Dubuque policymakers with an objective assessment of local crime and
poverty that can move the discussion beyond competing perceptions toward reasoned solution.
Based on the public perception survey conducted as part of this study, a majority of Dubuque
residents (57 %) believe the City of Dubuque is doing an "excellent" or "good" job addressing public
safety. Likewise, 68% believe the Dubuque Police Department is doing an "excellent" or "good" job
fighting crime. However, these high grades do not translate to satisfaction regarding the level of crime
in Dubuque. Approximately three - quarters (76 %) of those surveyed stated that crime was a "major" or
"moderate" problem in Dubuque. For those respondents who have lived in Dubuque for at least five
years, 89% say that crime has "significantly" or "somewhat" increased over the past five years. The
safety issues most often identified as major problems are: 1) drugs (33 %), 2) violent crime (23 %), 3) gang
activity (22 %), 4) unsupervised children (18 %), and 5) property crime (17 %). Crime concerns tend to be
specific in nature, as 88% of respondents rate their respective neighborhoods as "excellent" or "good"
places to live and only 12% rate crime as a "major" or "moderate" problem in their neighborhoods.
Conversely, 19% stated that downtown was unsafe during the day and 71% believed downtown was
unsafe at night. The study also found differing views among respondents' fear of crime, as a greater
fear was generally associated statistically with those who 1) had lived in Dubuque longer, 2) were prior
crime victims, 3) were less- educated, 4) were women, and 5) were older residents.
2010 Dubuque Crime and Poverty Study Summary Report Page 75 of 779 NIU Center for
Governmental Studies
While not overlooking that a portion of residents perceive a "crime problem" in Dubuque,
comparative data demonstrates that the level of crime and poverty in Dubuque is at or below the
average of the ten largest Iowa cities. Here perception and reality differ. Based on the most current
data, Dubuque is not statistically different from the other nine communities in terms of crime, poverty,
income, income inequality, law enforcement staffing, or amount of assisted housing units. The
following is a listing of Dubuque's rank among the ten cities in these areas:
• 7th in property crime
• 5 in UCR violent crime (includes simple assault, 3r without simple assaults included)'
• 6 in family poverty, 7 in household poverty, near income inequality average
• 5 in median household income
• Is staffed with 1.7 sworn officers per 1,000 residents, compared to a 1.6 average
• Is staffed with 1.8 law enforcement employees per 1,000 residents, compared to a 2.0 average
The data indicate that the crime and poverty situation in Dubuque is not significantly different than that
faced in other large Iowa communities, and in some cases it is better. Although Dubuque is a typical
large Iowa community regarding crime and poverty, resident perceptions seems to indicate that they
hold the community to a higher standard. Some selected perceptions from the RFP and survey data
with corresponding findings are highlighted below:
Perception: Crime in Dubuque is worse than in other large cities in Iowa
Finding: As noted in the comparative analysis, the inspection of comparative UCR crime rates during
2004 to 2009 for the ten comparably sized Iowa cities reveals that, overall; Dubuque's crime profile is
not uniquely different from the other cities. However, the analysis did point to some differences with
regard to offenses classified as violent crime, using the standard UCR coding approach.
Perception: there is a higher rate of crime in the center city but that most neighborhoods are safe
Finding: Hotspots are more likely to occur downtown, except for UCR property (mostly larceny). so
perception is accurate.
Perception: that Section 8 could be the cause of crime
Finding: Crime is multi - causal and poverty does play a big role. The association between poverty and
crime shows in the city level comparative analysis and in the moderate overlap of section 8 (the poorest
7 See pages 192 -194 for a discussion of this distinction
2010 Crime Dubuque rime and Poverty Study Summary Report Page 76 of 779 NIU Center 4 Y y Governmental Studies
wv ax,rn
of the poor net of the homeless) hotspots and crime hotspots. But no causal assertions can be made -
many Section 8 concentrated areas do not have concentrated crime hotspots.
Perception: crime is expanding or diffusing from Section 8 to other areas
Finding: The loglinear analysis shows that the probabilistic tendency is for crime to remain local. Section
8 victims are more likely to be victimized at section 8 locations. Crimes with section 8 victims are more
likely to also see section 8 arrestees. Crimes committed at section 8 locations are more likely to have
section 8 arrestees.
Dubuquers expect safe neighborhoods, and as the survey results indicate, a large majority
believes that most are safe. However, spatial analysis confirms that crime hotspots do exist in the city
center, especially along the Central Avenue corridor. Does the presence of Section 8 housing generate
these crime hotspots? Having the full benefit of spatial analysis, the results indicate that Section 8
hotspots have a low to moderate overlap with crime hotspots (on average, 26% of Section 8 hotspots
coincide with eight different types of crime hotspots, mostly due to property crime). As the body of
research examined indicates, other factors such as poverty, population density, security, neighborhood
characteristics and the business composition are contributors to generating crime hotspots in a
community. The analysis further shows that most crime attributed to Section 8 housing does not diffuse
into other areas; victims of crime who live in Section 8 housing are more likely to be victimized at
Section 8 addresses by those who live at Section 8 addresses, and crimes occurring at Section 8
addresses are more likely to be perpetrated by arrestees reporting a Section 8 address. Simply put, no
statistically determined causal assertions between Section 8 housing and crime can be made from the
analysis. At the same time, the analysis does indicate that there is a greater propensity for crime
victimization, location or arrests within clusters of the Section 8 community.
In particular, the data indicates that authorized Section 8 participants have nearly the same
proportional arrest rate in Dubuque as the non Section 8 segment of the population. For example, for
the period under study, authorized Section 8 participants represented 5.2% of Dubuque's total
population and 5.8% of total unique arrestees. It is important to note, however, that the analysis did
demonstrate that victimization and arrestee rates increase noticeable when those reporting a Section 8
address but who are not included in the Section 8 program database are added to the analysis. If these
addresses are taken at face value, an "unaccounted" Section 8 population may be contributing
additional disadvantage to those who are already at significant crime risk. Section 8 addresses also
witnessed a substantial increase of local ordinance charges in 2008 and 2009, perhaps reflecting more
aggressive enforcement practices by police. The analysis indicates that the pattern of crime in Dubuque
2010 Dubuque Crime and Poverty Study Summary Report Page 77 of 779 NI l centertor
Governmental Studies
1,10 Sas,
has remained fairly constant, and that crime, in general, is not diffusing to other neighborhoods. To
summarize, broader perceptions of increased crime attributable to Section 8 related locations contains
an element of validity, but the analysis indicates that the preponderance of crime in Dubuque, whether
or not it is associated with authorized or non - authorized Section 8 persons, is concentrated around the
city center and that crimes that reflect an element of Section 8 involvement (victim, incident location or
arrestee) tend to be clustered within the Section 8 community.
To further understand the multi - causal nature of crime, a review of academic literature reveals
the linkages between public housing, population density, crime, and poverty. The presence, extent and
type of crime that takes place in a community is attributable to a number of complex social factors that
extend well beyond the mere presence of Section 8 housing. Typically, assisted housing tenants have
been placed in dense residential settings (such as multi -unit or high -rise buildings) located in resource-
poor neighborhoods. Research shows that these neighborhoods suffer from cumulative disadvantage,
where the simultaneous presence of negative factors such as families below the poverty line, families
receiving public aid, the unemployed, and female- headed households with children in a dense setting
combines to create social and physical disorder. As disorder increases and crime blossoms, residents
feel that they are unable to exert formal and informal control over the neighborhood and become
fearful of crime, further weakening collective efficacy and social networks. This creates hotspots of
crime that can be difficult, if not impossible, to immediately reverse. It is the concentration of low
incomes and poverty in resource -poor neighborhoods that can be blamed for many high -crime
neighborhoods.
Nonetheless, academic research has also pointed to a number of promising strategies to address
crime- susceptible neighborhoods with concentrated public or assisted housing. Foremost among these
strategies is the following: public housing units should be located in garden -style settings, widely
geographically disbursed into neighborhoods with sufficient social resources. This helps ensure that
residents are supported socially, can create defensible space where residents can control their
immediate external surroundings, dampens the fuel that fires crime hotspots, and, has been proven not
to diffuse crime into the surrounding neighborhood. In addition, police should engage residents by
involving themselves into the problem neighborhoods. Since no single community policing strategy can
be applied successfully in every setting, the guiding principles should be to increase contact with
residents, pursue community partnerships, and be proactive in preventing crime.
2010 Dubuque Crime and Poverty Study Summary Report Page 78 of 779 AIIU center tor
Y rY p Gov
ti air
According to our review, other strategies that offer promise include methods that target
hotspots of crime with police resources. The literature indicates that this strategy has proven to reduce
reports of crime in the hotspot without diffusing crime to adjacent areas, especially when coupled with
a problem- oriented policing approach that identifies the root causes of the problem and develops
solutions that will prevent future incidents. Next, addressing physical and social disorder through
"broken windows" policing has been demonstrated to reduce fear of crime, but only after community
policing partnerships have been developed because the heightened police presence may increase fear
absent a strong relationship between the police and residents. Finally, researchers caution against
blanket strategies such as installing closed- circuit cameras and enacting juvenile curfews as these
approaches may not be sufficiently tailored to the particular needs of a distressed neighborhood.
Although the crime condition in Dubuque is similar to those faced across Iowa's largest cities, it
is evident from the survey results that residents do have concerns about crime in Dubuque. Those
concerns were found to vary depending on respondent characteristics. When dealing with an emerging
problem, it is essential to step back and assess the situation. This study and other steps the City has
taken via the Safe Community Task Force represent progressive and proactive steps in understanding
and addressing the concerns that filter through the community. The collection, presentation and
analysis of the data collected in this study provides an objective foundation to frame future dialogue
regarding crime and poverty in Dubuque. The purpose of the study is to increase understanding; it will
be the work of Dubuque policymakers and residents that will craft solutions tailored to the particular
needs of the Dubuque community. With this in mind, the research team offers the following broad
principles to assist in developing strategies to address crime and poverty:
• Utilize this study as a springboard for objective dialogue — Decision- making absent information
can create dangerous perceptions and exacerbate existing conditions. The report offers a
wealth of data and analysis to help understand crime and poverty in Dubuque. Policymakers,
staff, and residents are encouraged to adopt the analytical approaches contained within the
study now and into the future to guide public discourse on these vital community issues.
• Invest in, partner with, and empower at -risk neighborhoods — By rehabilitating housing stock,
addressing physical and social disorder, establishing relationships between police and
neighborhood residents, partnering with neighborhood businesses and nonprofits, Dubuque
policymakers can help those living in suffering areas to reclaim their neighborhoods. It is not by
accident that Section 8 units have a disproportionately higher amount of crime arrestees and
2010 Dubuque Crime and Poverty Study Summary Report Page 79 of 779 AIII I Centerfor
q Governmental Studies
victim, but a lower share of complainants. Residents in these areas must have trust in each
other and police in order to reclaim their neighborhoods.
• Address downtown crime hotspots — Focusing resources toward these areas through efforts
such as saturation patrol and minimizing disorder can have a tremendous effect upon crime;
however, it is important to note that these efforts are most effective after community policing
strategies to build relationships in troubled neighborhoods have been successfully
implemented.
• Disperse Section 8 housing unit locations — Although not densely concentrated, Section 8
housing remains largely centered in the downtown area of Dubuque. Further efforts are
necessary to disperse Section 8 housing units into neighborhoods with greater social resources
in order to avoid the cumulative disadvantage negative effects discussed above.
• Address poverty wherever it occurs — As poverty and low income are the greatest predictors of
crime, Dubuque policymakers would be wise to provide assistance to the impoverished so that
they can regain their financial footing. Such efforts could prove to be the most significant crime
prevention strategy available and could help ensure that other neighborhoods do not fall prey
to the cycles of poverty and crime.
Dubuque, like its peer cities in Iowa and the U.S., faces emerging social issues like changing
crime patterns, social change, economic strain and evolving demographics. The completion of this study
confirms that the City is exploring the question carefully and that policy alternatives will likely be
considered to meet the challenges that lay ahead.
2010 Dubuque Crime and Povert Study Summary Report Page 80 of 779 NlU center for
- -- q _ -- _-- Y Governmental Studies