Citation
Social disorganization and crime

Material Information

Title:
Social disorganization and crime
Creator:
Laslo, Jennifer Kristen
Publication Date:
Language:
English
Physical Description:
45 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Homicide -- Illinois -- Chicago ( lcsh )
Crime and race -- Illinois -- Chicago ( lcsh )
Crime and race ( fast )
Homicide ( fast )
Social conditions ( fast )
Social conditions -- Chicago (Ill.) ( lcsh )
Illinois -- Chicago ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 43-45).
General Note:
Department of Sociology
Statement of Responsibility:
by Jennifer Kristen Laslo.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
54531114 ( OCLC )
ocm54531114
Classification:
LD1190.L66 2003m L37 ( lcc )

Full Text
SOCIAL DISORGANIZATION
AND
CRIME
by
Jennifer Kristen Laslo
B.A., Anderson University, 2000
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
Sociology
2003


This thesis for the Master of Arts
degree by
Jennifer Kristen Laslo
has been approved
by
Susan Allison-Endriss


Laslo, Jennifer Kristen (M.A., Sociology)
Social Disorganization and Crime
Thesis directed by Candan Duran-Aydintug
ABSTRACT
The purpose of my study is to disaggregate homicide victimization
with a special focus on two issues. First, I look at racial differences in
homicide rates; the Black homicide rate is viewed in aggregate with the
White and Hispanic homicide rate; the Black homicide rate is viewed in
aggregate with the White and Hispanic homicide rate. Second, I seek
clarification of the effects of residential turnover, economic deprivation,
and population heterogeneity. In addressing these issues, data at the
census tract level are used to advance analyses.
The census tracts in Chicago are the unit of analysis. The present
research directs attention to the census tracts in the analysis of the Black,
White, Hispanic population. The Community Area data are aggregated
from the 1990 Population and Housing Summary Tape File 3 A (STF3A,
U.S. Bureau of the Census 1991) Census Tract Data. The homicide rates
came from the 1990 Chicago Homicide Data Set comes from the Chicago
Division of Police investigation file. My results indicate that the
characteristics of the population of an area, the characteristics of an area
itself, and the lack of economic resources all have significant effects on
overall rates of Black, White, and Hispanic Homicide.
This abstract accurately represents the content of the candidates thesis.
Signed
Candan Duran-Aydintuj


DEDICATION
I dedicate this thesis to my parents for their constant love, encouragement
and support to me.


ACKNOWLEDGEMENT
My thanks to my professors Dr. Bruce K. MacMurray and Dr. Kenneth J.
Litwin for their assistance and guidance with my studies in criminology.


CONTENTS
Tables.................................................vi
CHAPTER
1. INTRODUCTION....................................... 1
2. REVIEW OF THE LITERATURE.............................4
Theoretical Framework..................... 4
Previous Research...........................7
Present Research...........................15
3. METHODS.............................................18
Setting....................................18
Unit of Analysis...........................19
Dependent Measures.........................19
Independent Measures.......................20
4. RESULTS.............................................22
White Male Homicide Rate...................22
Black Male Homicide Rate...................23
Black Female Homicide Rate.................25
Hispanic Female Homicide Rate..............27
5. SUMMARY AND CONCLUSIONS.............................29
APPENDIX...............................................42
BIBLIOGRAPHY.......................................... 43
v


TABLES
Table
3.1 Operationalization of Residential Turnover.....................35
3.2 Operationalization of Economic Deprivation.....................36
3.3 Operationalization of Population Heterogeneity................37
4.1 Regression of White Male Homicide Rates........................38
4.2 Regression of Black Male Homicide Rates........................39
4.3 Regression of Black Female Homicide Rates......................40
4.4 Regression of Hispanic Female Homicide Rates...................41
vi


CHAPTER 1
INTRODUCTION
Three years ago I became very interested in the considerable
effects of unstable communities on individuals attitudes and criminal
behaviors. One of my roommates had been mentoring a young black girl
living in the inner city, and when this roommate moved to another state,
decided to step into this girls life to provide some sort of consistency and
guidance. I quickly developed an intimate relationship in which I was
somewhat of a parental figure to this girl. As I mentored her, I became
increasingly aware that the ever-present problem of crime has a more
direct effect on some people than on others.
At 18 years old, this girl was emancipated from her impoverished,
female-headed household. For two years she attended a liberal arts
college and I watched her succeed and improve academically, as well as
socially. Though this girl was making great accomplishments in her life,
her mind was occupied and influenced by the reality of her inner-city
environment. The reality of an unstable environment became apparent to
l


her once again as a result of the increase of contact she had with family
members and friends in the inner-city where she spent the majority of her
childhood. I noticed her attitudes and behaviors begin to change. Soon
she was on academic probation and was eventually told that her grade
point average was lower than the universitys standards and policies. The
same night that I helped her move back to her grandmothers house in the
inner-city, she committed a crime. Her involvement in criminal activities
began when she moved back to her unstable community.
This is one situation in which I have experienced the effects of an
unstable community on the attitudes and behaviors of individuals.
According to Schmalleger (2002), the structure of a society and its
comparative extent of organization are significant factors contributing to
the occurrence of criminal behavior.
This thesis probes criminal homicide victimization with a special
focus on two issues. First, this thesis is to look at racial differences in
homicide rates; the Black homicide rate is viewed in aggregate with the
White and Hispanic homicide rate. Many data reveal that Blacks in major
cities in the United States are unreasonably represented in homicide rates.
Blacks make up 13% of the population (Bureau of the Census, 1995), but
explain approximately one-half of all homicide offenders and homicide
victims (Federal Bureau of Investigation, 1995). The disproportionate
2


homicide victimization of Blacks is just as clear when viewed through the
lens of city-based data. Blacks represented the highest homicide rate in
Chicago from 1970 to 1993 (Block, 1993). Race specific examination has
been promoted as one practical option to exploring social disorganization
and homicide rates.
Second, I seek clarification of the effects of residential turnover,
economic deprivation, and population heterogeneity. In addressing these
issues, data at the census tract level are used to advance analyses. The
well-established positive association between structural conditions and
overall homicide rates is supported by the present findings. Within recent
advancements in the area of social disorganization and homicides rates,
better theory and research are needed. Little research exists on the racial
difference in homicide rates.
3


CHAPTER 2
LITERATURE REVIEW
Theoretical Framework
As discussed in the introduction, this research studys theoretical
basis is Social Disorganization Theory. Much of the research on theories
of crime refers to the failure of a community structure to recognize the
common values of its residents and maintain social control. Social control
attempts to direct behavior of individuals to norms and values of our
society (Horowitz, 1990). Controls of a social institution are essential in
determining whether or not individuals are deviant. According to Messner
and Rosenfeld (1999), when controls of an institution weaken, individuals
are more likely to engage in violence because they are less connected to
the social bonds that provide an understanding of conformity to norms.
Social structure theories consist of concepts that explain crime as a
result of the structure of society. These theories clarify characteristics of
4


society that supplement the low socioeconomic status of groups of people
as pertinent causes of crime. Siegel (1998) proposes a relationship
between a persons position in the socioeconomic structure of society and
the chance that the person will become a criminal. Social structure
theorists identify individuals of socially and economically disadvantaged
groups as being more apt to commit crime.
Social disorganization theories are a type of social structure
theories, which focus on the conditions of urban communities that affect
crime rates. This social structure approach is associated with the
ecological school of criminology viewing society as an organism and
crime as a kind of deviance. Social disorganization theories depict social
change, social conflict, and lack of social consensus as the root causes of
crime and deviance (Schmalleger, 2002).
These theories link crime rates to community ecological
characteristics. Indicators of social disorganization in a community are
high unemployment, low-income levels, and large single-parent
households (Siegel, 1998). Social disorganization theories explain crime
and violence as a product of the social forces existing when community
controls weaken providing inadequate social control. Community controls
can weaken because of residential turnover, population heterogeneity, and
economic deprivation (Bursik, 1988; Shaw& McKay, 1969).
5


Park (1915) looked at humans and the environment living in a city
of a growing organism. Parker, McCall, and Land (1999) believe that
social economic changes in a community tend to lead to the deterioration
of group solidarity and to a failure in social control factors. Social
disorganization theories support that crime and delinquency can be rooted
geographically or ecologically within particular neighborhoods or areas
within a city. Furthermore, Shaw and McKay (1942) have argued that
there are natural areas for crime, such as deteriorated inner-city
neighborhoods, where poverty rates are high. These areas are called
natural because they are unplanned and assist to arrange the operations
and needs of diverse populations within the city. Despite a persons race,
religion, or ethnicity, the everyday behavior of people living in these areas
is controlled by the social and ecological climate (Siegel, 1998).
When considering the impact of social economic conditions
within this theoretical framework, it is necessary to address the issue of
why some key structural factors, such as residential turnover, population
heterogeneity, and economic deprivation are more influential in producing
violence for one racial group than another. Homicide is a violent crime
that affects people within all socioeconomic levels. However, the chance
of being involved as offender or victim is not distributed equally among
U.S. residents (Ousey, 1999). Instead, violence is concentrated
6


unreasonably among Black Americans as both offenders and victims,
mainly living in major U.S. cities. Black homicide rates are
disproportionate because blacks currently make up only 13% of the U.S.
population (Bureau of the Census, 1995), however they account for
roughly one-half of all homicide offenders and a similar proportion of
homicide victims (Federal Bureau of Investigation, 1995).
Previous Research
#h
In the early 20 Century Robert Ezra Park, Ernest W. Burgess, Louis
Wirth, and their associates in the Sociology Department at the University
of Chicago conducted research on the social ecology of the city Chicago.
The Chicago School focused upon studying crime within the concept of
social disorganization. Many scholars of this time extensively studied
urbanization and mobility as being two social factors causing many
problems in Chicago (Shaw and McKay, 1942). The 1950s introduced the
new variable, ecology of the city, into the study of crime provided a
foundation for criminologists to concentrate on the social influences on
criminal behavior.
In the 1980s, a group of criminologists, social ecologists
attempted to develop a purer form of structural theory. To them,
structural theory emphasized community deterioration and economic
7


decline. With this idea, they were able to link crime rates with community
deterioration, lack of employment opportunities, community change,
community fear, poverty concentration, weak social controls, population
turnover, and social altruism.
Roncek (1981) examined how the physical characteristics of the
residents and of the housing units of communities affect crime rates by
looking at household composition, features of the housing environment,
and the interactions between the characteristics of the residents and
housing environments. Communities are not random collections of
individuals, instead they are very structured groups of people and
surroundings (Hunter, 1974).
Warren (1973) describes community as that aspect of the structure
of social systems which is observable and analyzable to the territorial
location of persons. The author gives more explanation of community by
examining three elements: as a social unit, the relationship as to daily life,
and collective action. The authors first research was the community as a
social unit. The community serves as a place for individuals. Social unit
is an important part of a community.
The second research concluded that the community reveals the
relationship as to the daily life. The community determines how people
8


do certain things and what they want their institutions and collective goals
to be. The attitudes and behaviors of an individual stem from the
community of which an individual has a sense of membership. Often,
individuals adopt the norms of the community, which can be revealed in
their behaviors and attitudes. The ending of the authors research explains
collective action. Collective action in a community refers to individuals
having the ability to regularly act together in the common concerns of life.
From Warrens research, communities were found to be small or
large clusters of people living in close proximity in an area that contains
industries, local stores, and/or other service facilities for the sustenance of
local people whose produce is to be distributed throughout a much wider
area. The author feels that a community model is an area in which the
majority of the population lives and makes a living.
Warren (1972) discusses how the clustering of people plays an
important role on an individuals daily life. According to Warren (1972),
community is the combination of social units and systems which perform
the major social functions having locality relevance. The five major
functions are:
1. production-distribution-consumption
2. socialization
3. social control
4. social participation
5. mutual support
9


This book reveals the research of the function of production-
distribution-consumption, which deals with the local participation in the
method of producing, distributing, and consuming goods and services that
are accessed on a daily basis. The goods and services are to be distributed
by various community institutions ranging from educational to religious
institutions. The research findings from the function of socialization deal
with how individuals develop their attitudes and behaviors. Society and
societys components affect the thoughts and reactions of individuals. The
author found from research that social control is the power that a group
has over the behavior of its members. Social control of a group influences
its members to the point of conformity to the groups norms and is
prevalent among many social units, especially the school. Another
research function is the community of social participation. Social
participation is widely seen in religious organizations.
The final research result of the community function is mutual
support. Mutual support refers to helping others. For example, mutual
support can be talking to a friend who needs somebody to listen to
him/her, the exchange of a tenant paying the landlord and a landlord
providing a house for rent. Clearly, all five of these community functions
have significance in individuals daily lives.
10


The following is East St. Louis Action Research Projects list of
trends most commonly found among inhabitants in areas with racial
problems: high unemployment rate, low family median income, low
median house values, lower high school test scores, higher percentage of
female head of household and single-parent families, higher occurrence of
substance abuse, higher crime and delinquency rate, high number of
homeless, racial isolation and exclusion from other groups of other people,
lack of necessary services (garbage, street repair, utilities, police, etc.), and
lack of profitable development. Many cities have experienced Whites
moving out of the city to avoid living in close proximity to Blacks because
of the mentioned above trends associated with all Blacks. Research
studies have examined racial attitudes of Whites and the results reveal that
Whites are more tolerable of equal treatment of Blacks when they are not
involved in their private areas of their life. Consequently, communities
become racially isolated through powerful racism techniques by
neighbors, federal agencies, and financial institutions.
Blau and Blau (1982) look at the effect of unequal economic and
social conditions crime rates. Research conducted by Blau and Blau
(1982) instituted the majority of the significant limits of macro-level
studies of urban homicides. They concluded, the relative deprivation
11


produced by much inequality rather than the absolute deprivation
produced by much inequality rather than the absolute deprivation
produced by much poverty provides the most fertile soil for criminal
violence (Blau and Blau, p. 122). Blau and Blau reveal in their study that
racial inequality leads to strong forces for individuals to commit crime and
to the weakening of social controls in a community.
A considerable amount of research examines the structural
determinants of homicide rates, but only in the last decade have
researchers begun to recognize that general structural models apply
equally well to African American and White violence (Harer &
Steffensmeier 1992; Krivo & Peterson, 1996; Laffee and Drass 1996; Laf
Free et al. 1992; Messner & Golden 1992; Parker & McCall 1999;
Peterson & Krivo 1993; Philips 1997; Sampson 1987; Shihadeh & Flynn
1996; Shihadeh & Ousey 1996,1998; Shihadeh & Steffensmeier 1994).
The results of previous macro-level studies of homicide can be
partitioned into two groups, based upon relevance to the racial invariance
issue. The first group consists of many early studies of aggregate
homicide rates, which did not employ race-specific data to measure
homicide and its structural covariates (Blau and Blau, 1982). These early
studies of aggregate homicide rates are consistent with structural theory,
assuming that the effects of important structural factors are consistent for
12


all racial groups. This limitation provides little knowledge about how the
effects of structural factors can vary by race.
The second group consists of recent studies that utilized racially
disaggregated data to examine the relationship between homicide and
structural condition (LaFree and Drass, 1996; LaFree et al., 1992; Messner
and Goden, 1992; Parker and McCall, 1999, Sampson, 1987; Shihdah and
Ousey, 1996). These studies lessen the racial invariance assumption and
represent a potential advancement in aggregate homicide research. The
results of past research has been a hindrance for this body of research not
allowing for firm conclusions about racial differences in the effects of
structural factors on homicide rates.
Attempts have been made by previous researchers to examine the
major disparity in homicide rates between black and white center-city
communities in the United States. Ousey (1999) investigated racial
differences in the relationship between homicide and structural factors that
have become known because of criminological theory. According to
Ousey (1999) poverty, unemployment, income inequality, female-headed
households, and the deprivation index are all significant predictors in
white communities. However, poverty and deprivation index are the only
two variables that significantly effect black communities homicide rates.
13


The empirical results from Ousey (1999) lend support to structural
perspectives from early researchers of the Chicago School.
Research has also outlined and supported the extent to which
structural conditions within localities produce four distinct racial patterns
of homicide measured in the following models: black intraracial, white
intraracial, black interracial, white interracial (Parker and McCall, 1999).
Further, Parker and McCall (1999) have also examined the effects of
structural conditions of race-specific and offender homicide rates.
While researchers have argued for the importance of examining
separate black and homicide models, research on race-specific victim and
offender homicide rates is limited (Parker & McCall, p.447). Messner and
Goldens (1992) study explores the effects of racial inequality on intra-and
interracial homicide rates for 1990. Still they included all incidents of
homicide involving victims and offenders of differing racial groups as
their interracial homicide measure.
Krivo and Peterson (2000) considered concentrated disadvantage,
community stability, racial residential segregation, and interracial
socioeconomic inequality as being indicators of factors for race-specific
homicide rates.
14


Present Research
This study examines racial differences in the effects of several key
structural factors (economic deprivation, residential turnover, and
population heterogeneity) which are significant in theory and previous
studies on aggregate homicide rates. The significant question of the
present research is, in the structural context of homicide, does the
association between the measures of weakening community controls and
homicide rates differ for racial groups? Based upon previous research, I
have developed a total of nine general hypotheses. I have divided them
up into three sets of hypotheses by race specific homicide rates. The first
set of hypotheses includes White Male and Female Homicide Rates with
three social disorganization variables. My first hypothesis is that the
higher residential turnover, the higher the White Male Homicide Rate and
the lower White Female Homicide Rate. My second hypothesis is that the
higher the economic deprivation the higher the White Male Homicide Rate
and the lower the White Female Homicide Rate. The third hypothesis is
higher the population heterogeneity the lower the White Male Homicide
Rate and the lower the White Female Homicide Rate.
The second set of hypotheses developed is for Black Male and
Female Homicide Rates. The first hypothesis is that the higher residential
turnover the higher Black Male and Female Homicide Rates. The second
15


hypothesis is the higher economic deprivation the higher Black Male and
Female Homicide Rates. The third hypothesis is the higher the population
heterogeneity the higher the Black Male Homicide Rate and the lower the
Black Female Homicide Rate.
The last set of hypothesis is for Hispanic Male and Female
Homicide Rates. The first hypothesis is that the higher the residential
turnover the higher Hispanic Male Homicide Rate and the lower Female
Homicide Rate. The second hypothesis is the higher the economic
deprivation the higher the Hispanic Male and Female Homicide Rate. The
last hypothesis is the higher the population heterogeneity the higher the
Hispanic Male and Female Homicide Rate. The main issue I address is
which weakening community control factor will significantly increase
homicide rates among Blacks, Whites, and Hispanics. These are
essentially the same hypotheses, but are focused specifically upon
homicide rates within a racial category. Furthermore I am examining the
same sets of hypotheses, but with race specific homicide rate
Within the theoretical framework provided by Parker, McCall, and
Land (1999) crime causation is when group solidarity in a community
tends to deteriorate because of social economic changes social economic
changes and failure in social control. Furthermore, areas in which
residents experience social disorganization nourishes individuals to
16


experience conflict and despair, which flourishes antisocial behavior.
Social disorganization theories explain crime and violence as a product of
the social forces existing when community controls weaken providing
inadequate social control.
The present research makes a contribution to the existing studies of
social disorganization on urban homicides rates. First, I examine a more
comprehensive set of key structural factors than most previous studies on
this issue by looking at residential turnover, economic deprivation, and
population heterogeneity. The conceptualization of each of these variables
leads me to examine them each separately. Second, I examine Black,
White, and Hispanic homicide victimization rates (also see Peterson and
Krivo, 1993). As a result of disaggregating homicide rates, I seek greater
precision with regard to the social forces that explain urban homicide.
Also, I am able to detect whether structural factors differ with black,
white, and hispanic homicide rates. Finally, I use 1990 data, which
provide a better indication of racial differences in the association between
current structural conditions and rates of homicide.
17


CHAPTER 3
METHODS
Setting
According to mortality statistics from the Center for Disease
Control and Prevention (CDC)s National Center for Health Statistics
(NCHS), in 1987 homicide was the 12th leading cause of death in the
United States. Chicago has experience high homicide rates in the past
couple of year. Between 1965, with 395 homicides, and 1974 with the
city's record high of 970 homicides, Chicago has experienced both the rate
and number of homicides more than double (OBrien and Benedict, 2003).
Chicago homicide rates have remained high ever since the citys record
high. The setting for the present research is Chicago, Illinois. In 1990,
there were 77 community areas in Chicago. The present research directs
attention to all 77 community areas in the analysis of the White, Black,
Hispanic, male and female populations.
18


Unit of Analysis
The census tract is the unit of analysis. The census tract data are
aggregated from the 1990 Population and Housing Summary Tape File 3 A
(STF3A, U.S. Bureau of the Census 1991) Census Tract Data.
Dependent Variables
The dependent measures came from the 1990 Chicago Homicide
Data Set comes from the Chicago Division of Police investigation file.
The Chicago homicide data set is the prevalent, most detailed data set
obtainable for a city (Zahn and McCall, 1999). Black Male Homicide Rate
is the total number of Black male homicide victims in the census tract
1990 divided by the population of the census tract then multiplied by
100,000 to create the Black male homicide rate per 100,000 residents.
Black Female Homicide Rate is the total number of Black female
homicide victims in the census tract 1990 divided by the population of the
census tract then multiplied by 100,000 to create the Black female
homicide rate per 100,000 residents. White Male Homicide Rate is the
total number of White male homicide victims in the census tract 1990
divided by the population of the census tract then multiplied by 100,000 to
19


create the White male homicide rate per 100,000 residents. White Female
Homicide Rate is the total number of White female homicide victims in
the census tract 1990 divided by the population of the census tract then
multiplied by 100,000 to create the White female homicide rate per
100.000 residents.
Hispanic Male Homicide Rate is the total number of Hispanic male
homicide victims in the census tract 1990 divided by the population of the
census tract then multiplied by 100,000 to create the Hispanic male
homicide rate per 100,000 residents. Hispanic Female Homicide Rate is
i
the total number of Hispanic female homicide victims in the census tract
1990 divided by the population of the census tract then multiplied by
100.000 to create the Hispanic female homicide rate per 100,000 residents.
These variables were created in order to know the gender and the race of
the homicide victims.
Independent Variables
The independent measures for each community area in Chicago
came from the 1990 U.S. Census of Population and Housing STF 3 A
(STF3A, U.S. Bureau of the Census 1991). According to the social
disorganization theory, residential turnover, economic deprivation, and
population heterogeneity can weaken community controls. Residential
20


turnover measures the characteristics of an area itself and is
conceptualized as percent homeowner and percent people living in a
different house.
Percent homeowner is the percentage of residents in the census
tracts that are homeowners. The percent of people living in different
house is the percentage of people who are 5 years old and older living in a
different house in 1985.
The measures of economic deprivation are conceptualized into the
following variables: median family income, percent unemployed, and
percent below poverty to explain the lack of economic resources. Median
family income is median family income of the tract population. Percent
unemployed is the percent of the tract population who are unemployed.
Lastly, the percent below poverty is the tract population who are below
poverty.
Population Heterogeneity is measured by percent foreign bom,
which is the percent of the tract population who are foreign bom. The
control variable for this study is percent of young males, conceptualized as
the percent of the tract population who are male age 16-34.
21


CHAPTER FOUR
RESULTS
White Male Homicide Rate
Table 4.1 begins the process of disaggregation by regressing the
White Male Homicide Rate on the six independent variables that represent
census tract characteristics. This table contains the stepwise results for
White Male Homicide Rate regressed on the independent variables
creating the most effective equation possible with only one independent
variable, which is the Percent Unemployed. As measured by the Percent
Unemployed, the proportion of each census tracts population in one of the
groups often involved in crime is statistically significantly linked with
White Male Homicide Rate.
The Beta is .470, which means the unit change is .470 in White
Male Homicide Rate per unit change in Percent Unemployed. White
male homicide rate has a strong positive relationship with the Percent
Unemployed. White Male Homicide Rate increases as Percent
22


Unemployed increases. In the zero-order correlation, the coefficient of
determination, r squared .470 = 22% indicates that the Percent
Unemployed associates 22% of the total variation in White Male
Homicide Rate. There is a positive relationship when White Male
Homicide Rate is high and Percent Unemployed is high.
Indicators of residential turnover and population heterogeneity do
not add anything to the equation. Indicators of economic deprivation
appear to have weaker effects on homicide rates for Blacks than for
Whites (Messner & Golden, 1992; Peterson & Krivo, 1993). Based upon
these data, the characteristics of the population of an area and the
characteristics of an area itself do not have significant effects on overall
rates of White Male Homicide. This shows that the lack of economic
resources in an area is the only significant effect on overall rates of White
Male Homicide.
Black Male Homicide Rate
Table 4.2 presents the results of the regression of Black Homicide
Rate on the independent variables that represent census tract
characteristics. This table contains the stepwise results for Black Male
Homicide Rate regressed on the independent variables creating the most
effective equation possible with only one independent variable, which is
23


Percent Below Poverty. The Beta is .572, which means the unit change is
.572 in Black Male Homicide Rate per unit change in Percent Below
Poverty. As measured by Percent Below Poverty, the proportion of each
census tracts population in one of the groups often involved in crime is
statistically significantly linked with Black Male Homicide Rate.
Black Male Homicide Rate has a strong positive relationship with
Percent Below Poverty. Black Male Homicide Rate increases as the
percent below poverty increases. In the zero-order correlation, the
coefficient of determination, r squared .572 = 33% indicates that Percent
Below Poverty associates 33% of the total variation in Black Male
Homicide Rate. There is a positive relationship when black male
homicide rate is high and percent below poverty is high.
The measures of residential turnover and population heterogeneity
do not help to explain the variance in Black Male Homicide Rate. The
characteristics of an area itself and the characteristics of the population of
an area do not have significant effects on Black Male Homicide Rates.
Similar to Table 4.1 with White Male Homicide Rate, lack of economic
resources indicates an association with Black Male Homicide Rate.
24


Black Female Homicide Rates
Table 4.3 Black Female Homicide Rate presents the results of the
regression of Black Female Homicide Rate on the six independent
variables that represent census tract characteristics. This table contains
the stepwise results for Black Female Homicide Rate regressed on the
independent variables creating the most effective equations possible with
three independent variables, which are Percent Below Poverty, Percent
Population Living in a Different Household in 1985, and Percent Foreign
Bom.
Model 1 shows the Beta Percent Below Poverty is .518 and this
variable is statistically significantly associated with Black Female
Homicide Rate. Black Female Homicide Rate has a strong positive
relationship with Percent Below Poverty. Black Female Homicide Rate
increases as Percent Below Poverty increases. In the zero-order
correlation, the coefficient of determination, r squared .610 = 37%, which
indicates that Percent Below Poverty associates 37% of the variance in
Black Female Homicide Rate. There is a positive relationship when Black
Female Homicide Rate is high and Percent Below Poverty is high.
Model 2 reveals the beta for Percent Population Living in a
Different Household in 1985 is .348. Black Female Homicide Rate has a
moderate positive relationship with Percent Population Living in a
25


Different Household in 1985. Black Female Homicide Rate increases as
Percent Population Living in a Different Household in 1985 increases. In
the zero-order correlation, the coefficient of determination, r squared .646
= 42 % indicating that Percent Population Living in a Different Household
in 1985 explains 42% of the variance in Black Female Homicide Rate.
There is a positive relationship when Black Female Homicide Rate is high
and Percent Below Poverty is high.
The final model, Model 3, indicates the beta as -.272. Black
Female Homicide Rate has a moderate negative relationship with the
Percent Foreign Bom. Black Female Homicide Rate increases as the
Percent Foreign Bom decreases. In the zero-order correlation, the
coefficient of determination, r squared .684 = 47 %, indicating that Percent
Foreign Bom associates 47% of the variance in Black Female Homicide
Rate. There is a negative relationship when Black Female Homicide rate
is high and Percent Foreign Bom is low. Areas characterized by high
levels of residential turnover and economic deprivation and low levels of
population heterogeneity are the same areas that experience higher rates of
I.
Black Female Homicide Rate.
26


Hispanic Female Homicide Rate
Table 4.4 completes the process of disaggregation by regressing
Hispanic Female Homicide Rate on the six independent variables.
This table contains the stepwise results for Hispanic Female Homicide
Rate regressed on the independent variables creating the most effective
equation possible with only one independent variable, which is Percent
Foreign Bom. The Beta is .487, which means that the unit change is .487
in Hispanic Female Homicide Rate per unit change in Percent Foreign
Bom. As measured by Percent Foreign Bom, the proportion of each
census tracts population in one of the groups often involved in crime is
statistically significantly linked with Hispanic Female Homicide Rate.
Hispanic Female Homicide Rate has a strong positive relationship
with the Percent Foreign Bom. Hispanic Female Homicide Rate increases
as the Percent Foreign Bom increases. The coefficient of determination, r
squared .487 = 24%, which indicates that the Percent Foreign Bom
associates 24% of the total variation in Hispanic Female Homicide Rate.
There is a positive relationship when Hispanic Female Homicide Rate is
high and Percent Foreign Bom is high.
Based upon these data, the characteristics of an area itself and the
lack of economic resources do not have significant effects on overall rates
of Hispanic Female Homicide. This shows that the characteristics of the
27


population of an area is the only significant effect on overall rates of
Hispanic Female Homicide.
28


CHAPTER FIVE
SUMMARY AND CONCLUSIONS
This thesis probed homicide victimization with an emphasis on
race-specific homicide rates and the comparative effects of ecological
characteristics of a tract population. Several conclusions come from the
present research, with four of primajy import. First, the present research
suggests continued attention to measures of economic deprivation and
racial inequality. In the context of the present research, for instance, the
measure of economic deprivation was important in the context of White
male homicides, Black male homicides, and Black female homicides.
Economic deprivation was the only measure that was consistently linked
to 3 of the 6 homicide rates.
These findings lend support of absolute deprivation; when
resources are lacking, rates of homicide are higher. It seems clear from
the present results, however, that the lack of resources in an area, rather
29


the characteristics of an area itself and the characteristics of the
population, is of more crucial importance to race-specific homicide rates.
Secondly, the effect of race in the United States today permeates
all aspects of life. Krivo and Peterson (2000) considered concentrated
disadvantage, community stability, racial residential segregation, and
interracial socioeconomic inequality as being indicators of factors for
race-specific homicide rates. Blacks experience residential discrimination
based upon race and as a result are trapped in areas with limited resources.
Residents also deal with higher rates of Black homicide victimization in
these geographical areas. The difficulties faced by Blacks increase
because of racial segregation and isolation of poor residents in urban areas
is greatly increasing (Siegel, 1998).
Major theories of crime lead one to expect similar results for
different racial groups (Krivo and Peterson, 2000). Yet a common finding
is that some crime-generating conditions have stronger effects on black
females than black males and white males and black males than Hispanic
males. I argue that the varying effects of predictors may result from
significantly different social positions of Whites, Blacks, and Hispanics in
U.S. cities.
30


The present research suggests that the study of urban homicide can
benefit from redeveloped attention to community areas as the unit of
analysis. Much previous research has aggregated community area
homicides in order to conduct comparisons of homicide rates between
different community areas. As a result, a large amount of previous
research has aggregated census tract data to create explanatory variables at
the community, city, or urban area levels of analysis. When researchers
aggregate homicide data, they limit studies by giving an overall
explanation of homicides. Accordingly, in places where the circumstances
of Blacks, Whites, and Hispanics are similar, crime-generating social
conditions should have comparable effects on homicide (Krivo and
Peterson, 2000). The present research therefore suggests the need for
focused attention to census tracts as valid and essential units of analysis in
urban homicide research.
Most importantly, the present research exemplifies the significance
of examining structural variables by race-specific homicide rates. If the
present research would have been limited to aggregate homicide
victimization rates, then the results would have not been specific to each
race. This limitation would have made it difficult to generalize the
findings of the study. By disaggregating the homicide data into race-
specific homicide rates, the present research reveals differing explanatory
31


importance of frequently important explanatory variables. For example,
race has been a variable that has been theoretically studied. The present
research suggests that the effects of the measure of race are measured by
structural conditions in a community.
In conclusion, by disaggregating the homicide data the present
research was able to determine which factor actually has a positive
association with weakening community control and race-specific homicide
rates. My results conclude that the impact of structural predictors on
homicide is not the same for Whites, Blacks, and Hispanics. Race, for
instance, continues to be a theoretically interesting variable, but the
present research suggests that the effects of this measure are influenced by
other structural conditions.
The results suggest the association between poverty and homicide
seems to cut across racial boundaries because White Male, Black Male,
and Black Female victimization rates increase with an elevation of
poverty. When White male, Black Male and Female residents of a
community are living in poverty, they are more likely to be a victim of
homicide. Accordingly, in places where the circumstances of White
males, Black males, and Black Females are similar, crime-producing
social conditions such as poverty should have comparable effects on
homicide.
32


Population Heterogeneity has a negative effect on Black female
homicide. This is because black females living in close proximity to
people who were not bom in the U.S. decreases their chances of being a
victim of homicide. However, Hispanic female residents are more likely
to be victims of homicide when they live in close proximity with
individuals who were not bom in the U.S.
The measure of residential turnover, however, yielded consistent
but unexpected results and was significantly related to Black Female
Homicides and Hispanic Female Homicide Rates. This suggests that
Black and Hispanic Female residents of a community have higher chances
of being a victim of homicide when they have lived in multiple households
within five years or less. The structural measures in this study (residential
turnover, economic deprivation, and population heterogeneity) have no
associations with White female and Hispanic male residents.
The limitations of this study are analyzing each structural variable
*
separately, 1990 Census data, and weak variable. First, analyzing each
variable separately limited the examination of each structural variable
making it difficult to know how the variables affect each other with race-
specific homicide rates. Secondly, using the 1990 Census data limits my
study because it is not the most recent data collected for Chicago.
However, when I began my analyses the 2000 Census Data was not
33


complete, so all of the data was not available to be used in this study.
Finally, measuring population heterogeneity by only the percent of the
population that is unemployed limited my results in this study. Therefore,
this was a weak representation of the structural variable population
heterogeneity.
In the future, studies should examine other major U.S. cities, use
more current data set, and create multiple index indicators. First, future
studies looking at other major U.S. would determine whether the results of
this study could be generalized to other cities in the U.S. with 100,000
residents. Secondly, I suggest other studies using the 2000 Census Data
when it is fully complete to better examine social disorganization and
homicide. Lastly, the creation of multiple index indicators would solve
the problem of reliability and validity. Future studies should create
composite measures using multiple index indicators to indicate residential
turnover, economic deprivation, and population heterogeneity.
34


Table 3.1: Operationalization of Residential Turnover into Independent
Variables and Expected Directions of Effects
Independent Variable Operationalization Expected Direction
Percent Homeowner percent of tract population who own homes +
Different House 1985 percent of tract population who are 5 years old & older living in Different House in 1985 +
35


Table 3.2: Operationalization of Economic Deprivation into Independent
Variables and Expected Directions of Effects
Independent Variable Operationalization Expected
Direction
Median Family median family income of tract +
Income population
Percent Unemployed percent of tract population who are +
unemployed
Percent Below percent of tract population who are +
Poverty below poverty
36


Table 3.3: Operationalization of Population Heterogeneity into an
Independent Variable and Expected Direction of Effect
Independent Variable Operationalization Expected
Direction
Percent Foreign Born
percent of tract population who are
foreign bom
37


Table 4.1: Regression Models of White Male Homicide Rates on Census
Tract Characteristics
Pearson Correlations
Residential Turnover Percent Homeowner Percent Different House in 1985 -.229 -.081
Economic Deprivation
Percent Below Poverty .416
Percent Unemployed .470
Median Family income -.321
Population Heterogeneity
Percent Foreign Bom -.279
Standardized Coefficients R R Square Adjusted R Square N
Beta
Percent Unemployed .470 .470 .221 .211 77
38


Table 4.2: Regression Models of Black Male Homicide Rates on Census
Tract Characteristics
Pearson Correlations
Residential Turnover Percent Homeowner -.481
Economic Deprivation Percent Different house in 1985 .209
Percent Below Poverty .572
Percent Unemployed .467
Median Family income -.454
Population Heterogeneity Percent Foreign Bom .017
Standardized Coefficients R R Square Adjusted R Square N
Beta
Percent Below Poverty .572 .572 .327 .317 72
39


Table 4.3: Regression Models of Black Female Homicide Rates on Census
Tract Characteristics
Pearson Correlations
Residential Turnover Percent Homeowner Percent Different House in 1985 -.588 .244
Economic Deprivation
Percent Below Poverty .610
Percent Unemployed .463
Median Family Income -.482
Population Heterogeneity
Percent foreign bom -.246
Models Standardized Coefficients R R Square Adjusted R Square N
Beta
1 Percent Below Poverty .518 .610 .372 .363 70
2 Percent Different House in 1985 .348 .646 .417 .400 70
3 Percent Foreign Bom -.272 .684 .468 .444 70
40


Table 4.4: Regression Model of Hispanic Female Homicide Rates on
Census Tract Characteristics
Pearson Correlations
Residential Turnover Percent Homeowner Percent Different house in 1985 -.123 .290
Economic Deprivation
Percent Below Poverty -.049
Percent Unemployed -.151
Median Family income -.022
Population Heterogeneity
Percent Foreign Bom .487
Standardized Coefficients R R Square Adjusted R Square N
Beta
Percent Foreign Bom .487 .487 .237 .227 77
41


APPENDIX
Excluded Variables for White Male Homicide Rate
Beta In
Percent homeowner .036
Percent Different House in 1985 -.028
Percent Below Poverty -.053
Median Family Income .162
Percent Foreign Born -.089
Predictors in the Model: (Constant), Percent Unemployed
Excluded Variables for Black Male Homicide Rate
Beta In
Percent Homeowner 130
Percent Different House in 1985 .167
Percent Unemployed -.261
Median Family Income .064
Percent Foreign Bom .197
Predictors in the Model: (Constant), Percent Below Poverty
Excluded Variables for Black Female Homicide Rate
Beta In
Percent Homeowner -.305
Percent Different house in U.S. in 1985 .213
Percent Unemployed -.443
Median Family Income .073
Percent Foreign Born -.086
Predictors in the Model: (Constant), Percent Below Poverty, Percent Different
House in 1985, and Percent Foreign Born
Excluded Variables for Hispanic Female Homicide Rate
Beta In
Percent Homeowner -.089
Percent Different House in 1985 .092
Percent Below Poverty .090
Percent Unemployed .079
Median Family Income -.056
Predictors in the Model: (Constant), Percent Foreign Born
Dependent Variables Entered/Removed
White female homicide rate and Hispanic male homicide rate
42


BIBLIOGRAPHY
Blau, J.R., & Blau, P.M. (1982). The cost of inequality: Metropolitan
structure and criminal violence. American Sociological Review, 47,
114-129.
Block, C.R. (1993). Lethal violence in the Chicago Latino community. In
A.V. Wilson (Ed.), Homicide: The victim/offender connection.
267-342.
Bursik, R.J., Jr. (1988). Social disorganization and theories of crime and
delinquency: Problems and prospects. Criminology. 26, 519-551.
Federal Bureau of Investigation
1991 Crime in the United States, 1990. Washington D.C.: U.S.
Government Printing Office.
1995 Crime in the United States, 1994. Washington D.C.: U.S.
Government Printing Office.
http://www.chipuplib.org/004chicago/disasters/homicide enidemic.html.
OBrien, Ellen & Benedict, Lyle. Chicago Public Library web
page. 18. April 2003.
http://www.hewett.norfolk.sch.uk/curric/soc/crime/chicago.htm. Hewett
School web page. 22. November 2001.
Horowitz, A. Y. (1990). The logic of social control. 8-11.
Hunter, Albert. (1974). Symbolic Communities.
Krivo, Lauren J. & Peterson, Ruth D. (1996). Extremely disadvantaged
neighborhoods and urban crime. Social Forces. 75. 619-50.
Krivo, Lauren J. & Peterson, Ruth D. (2000). The structural context of
homicide: Accounting for racial differences in process. American
Sociological Review. 65 (4), 547-559.
43


LaFree, Gary & Drass, Kriss A. (1996). The effect of changes in
intraracial income inequality and educational attainment on
changes in arrest rates for African Americans and whites, 1957 to
1990. American Sociological Review. 61, 614-34.
LaFree, Gary, Drass, Kriss A., & ODay Patrick. (1992). Race and crime
in postwar America: Determinants of African-American and white
rates, 1957-1988. Criminology. 30,157-88.
Messner, Steven F. & Golden, Reid M. (1992). Racial inequality and
racially disaggregated homicide rates: an assessment of alternative
theoretical explanations. Criminology. 30. 421-47.
Messner, Steven F. & Rosenfeld, Richard. Social Structure and
Homicide. Homicide. 27,1999.
Ousey, Graham C. (1999). Homicide, structural factors, and the racial
invariance assumption. Criminology. 37 (2), 405-426.
Park, Robert. (1915) The City: Suggestions for the Investigation of
Behavior in the City Environment. American Journal of Sociology.
(20), 579-583.
Parker, Karen F. & McCall, Patricia L. (1999). Structural conditions and
racial homicide patterns: A look at the multiple disadvantages in
urban areas. Criminology. 37 (3), 447-478.
Peterson, Ruth D. & Krivo, Lauren J. (1993). Racial segregation and black
urban homicide. Social Forces (76), 747-74.
Phillips, Julie A. (1997). Variation in African American homicide rates:
An assessment of potential explanations. Criminology. (35),
527-59.
Roncek, Dennis W. (1981). Dangerous Places: Crime and Residential
Environment. Social Forces. 60 (1), 74-96.
Sampson, Robert J. (1987). Urban black violence: the effect of male
joblessness and family disruption. American Journal of Sociology.
(93), 348-82.
44


Schmalleger, Frank. Criminology Today: an integrative introduction. 3,
2002.
Shaw, C.R. & McKay, H. (1942). Juvenile delinquency and urban areas.
Shihadeh, Edward S. & Flynn, Nicole. (1996). Segregation and crime: The
effect of Black social isolation on the rates of black urban violence.
Social Forces. (741 1325-52.
Shihadeh, Edward S. & Ousey, Graham C. (1996). Metropolitan
expansion and black social dislocation: The link between
suburbanization and center-city crime. Social Forces. (75), 649-66.
Shihadeh, Edward S. & Steffensmeier, Darrell J. (1994). Economic
inequality, family disruption, and urban black violence: Cities as
units of stratification and social control. Social Forces. (73),
729-51.
Siegel, Larry J. Criminology 6 ed. 1998.
Warren, Ronald L.
1972 The Community in America.
1973 Perspectives on the America Community. Chicago, IL
Zahn, Margaret A. & McCall, Patricia L. (1999). Trends and Patterns of
Homicide in 20th-Century United States. Homicide. 9-23, 1999.
1995 Current Population Survey: Annual Demographic File, 1994.
Washington, D.C.: Bureau of the Census.
45