determinates and consequences of crime prevention measurescommonly used crime preventive measures by...
TRANSCRIPT
Atlanta University CenterDigitalCommons@Robert W. Woodruff Library, AtlantaUniversity Center
ETD Collection for AUC Robert W. Woodruff Library
5-1-1987
Determinates and consequences of crimeprevention measuresLarry L. ThomasAtlanta University
Follow this and additional works at: http://digitalcommons.auctr.edu/dissertations
Part of the Law Commons
This Thesis is brought to you for free and open access by DigitalCommons@Robert W. Woodruff Library, Atlanta University Center. It has beenaccepted for inclusion in ETD Collection for AUC Robert W. Woodruff Library by an authorized administrator of DigitalCommons@Robert W.Woodruff Library, Atlanta University Center. For more information, please contact [email protected].
Recommended CitationThomas, Larry L., "Determinates and consequences of crime prevention measures" (1987). ETD Collection for AUC Robert W.Woodruff Library. Paper 1260.
DETERMINATES AND CONSEQUENCES OF CRIME PREVENTION MEASURES
A THESIS
SUBMITTED TO THE FACULTY OF ATLANTA UNIVERSITY
IN PARTIAL FULFULLMENT OF THE REQUIREMENTS FOR
THE DEGREE OF MASTER OF ARTS
IN CRIMINAL JUSTICE ADMINISTRATION
BY
LARRY L. THOMAS
DEPARTMENT OF CRIMINAL JUSTICE ADMINISTRATION
ATLANTA, GEORGIA
MAY, 1987
I V
*.-.. ^*r
TABLE OF CONTENTS
PAGE
Acknowledgements
Table of Contents
List of Tables v
CHAPTER 1
INTRODUCTION 1
Statement of the Problem 2
Significance of the Study 2
Sources of Data 3
Limitations of the Study 3
Organization of the Study 3
CHAPTER 2
REVIEW OF LITERATURE 4
Preventive Measures 7
CHAPTER 3
CONCEPTUAL FRAMEWORK, MEASUREMENT OF
VARIABLES AND METHODOLOGY H
Measurement of Variables 11
Measurement of Background
Characteristics 13
Measurement of Preventive Measures.... 13
Measurement of Fear of Crime 13
in
PAGE
CHAPTER 4
DATA ANALYSIS 20
Descriptive Analysis 20
Preventive Measure 24
Fear of Crime 25
Analytical Procedures 25
Fear of Crime and Preventive
Measures 27
Fear of Crime During The Day 29
Fear of Crime During The Night 29
Pearson Product Momentum
Correlation 28
Multiple Linear Regression Analysis... 16
CHAPTER 5
SUMMARY, CONCLUSION, AND IMPLICATIONS 36
BIBLIOGRAPHY 38
IV
LIST OF TABLES
Page
Table 3.1 Reliability of Prevention Item 14
Table 4.1 Frequency Distribution of Sample
Respondents 21
Table 4.2 Pearson Correlation Coefficients 28
Table 4.3 Standard Metric Regression Coefficients
for Each Equation Model to Measure Fear
of Crime During the Day 32
Table 4.4 Standard Metric Regression Coefficients
for Each Equation Model to Measure Fear
of Crime During the Night 34
ACKNOWLEDGEMENTS
This is to thank Dr. K. S. Murty, Ms. Sylvia Daniel and
Ms. Diana Moore for their assistance without whose support
this study could not have been completed. Most of all many
thanks, love, and appreciation to my Grandmother, Mrs.
Lillie M. Bolden.
ABSTRACT
CRIMINAL JUSTICE ADMINISTRATION
B.S. SAVANNAH STATE COLLEGE, 1977THOMAS, LARRY L.
DETERMINES AND CONSEQUENCES OF CRIME PREVENTION MEASURES
Advisor: Dr. K. S. Murty
Thesis Dated: May, 1987
The fear of crime is a meaningful reality in the lives
of people apart from the fluctuation in victimization rates
and is not necessarily based on either personal experience
or empirical data. Some would agree that the fear of crime
is not significant because such fear bears little if any
direct relationship to the incidence of victimization.
This thesis measures the relationship between crime
preventive measures and selected background characteristics
(socio-demographic), and between the fear of crime and
preventive measures. We found sex to be the most significant
background characteristic associated with the fear of crime;
i.e. females fear crime much more than do males. House light
and neighborhood watch were the two preventive measures that
had a significant direct association with the fear of crime.
CHAPTER I
INTRODUCTION
This study is designed to examine the fear of crime.
The fear of crime is measured in terms of the degree of
safety the respondents perceive in this study feel in
their neighborhoods during the day and during the night.
There has been an increasing interest in the fear of crime
and its relationship to the quality of life in the last
seven years. This thesis is concerned with various kinds
of people who use different kinds of crime prevention
measures in response to their fear of crime; and, whether
these measures lessen their fear of crime. Do differentials
in age, education, sex, size of household and social-economic
class influence the use of different crime preventive
measures? What degree of personal safety do these measures
provide?
Some scholars, Braugart (1980) and Skogan (1977), argue
that the increase in the fear of crime is not related to
crime prevention measures used. This study addresses the
relationship of crime preventive measures to certain back
ground characteristics and the fear of crime.
-1-
-2-
STATEMENT OF THE PROBLEM
The purpose of the study is to test two hypotheses:
(1) there is a significant relationship between crime
preventive measures and background characteristics (age,
education, sex, size of household, and social-economic
class); (2) there is a direct relationship between the
fear of crime and crime preventive measures used (police
watch, mail postponement, neighbor mail watch, neighbor
hood house check, house light).
SIGNIFICANCE OF THE STUDY
The study enables us to identify: (1) the segments
of the population that fear crime the most; (2) the most
commonly used crime preventive measures by groups who fear
crime; (3) the relationships between the fear of crime and
the preventive measures used. These findings will provide
law enforcement officials with a knowledge of the communit
ies' fears of crime and what community memberships do about
these fears. This knowledge will enable the police to
intergrate (and articulate) their crime preventive measures
with community members' fear of crime and their deterring
strategies. Consequently, police community relations could
be strengthened.
-3-
SOURCES OF DATA
The survey data for this thesis, were extracted from
a large data set on Research On Minorities: Race and Crime
In Atlanta and Washington, D.C. (ICPSR 8459). This study
contains data from 621 black interviewees from four neighbor
hoods in each of the two cities: (1) middle income-high
crime: (2) middle income-low crime; (3) low income-high
crime; and (4) low income-low crime neighborhoods. A
complete documentation of this data may be found elsewhere
(Debro et al., 1982). Debro's study examines the relation
ship of the extent of criminal occurrence to differential
community income levels.
LIMITATION OF THE STUDY
This study is a cross-sectional survey that covers
only one period in time. Therefore, it cannot measure the
lag effect or change within a given variable over a period
of time. Secondly, the study population was black and
drawn from four neighborhood types within two cities. This
limits the general ability of the findings to metropolitan
black populations.
ORGANIZATION OF THE THESIS
This thesis consists of five chapters: (1) Introduction;
(2) Review of Literature; (3) Conceptual Framework and
Methodology; (4) Data Analysis; and (5) Summary, Conclusion,
and Implication.
CHAPTER II
REVIEW OF LITERATURE
This review is restricted to the studies that utilize
the independent variables selected for this thesis.
Fear of Crime By Age - Lee (1982) examined the growing
fear of crime among the general public by using a national
public opinion survey. He used a discriminant analysis to
find those variables that best measure the fear of crime.
Age was found to be one of the most important variables
discriminating between fearful and non-fearful respondents.
Although less likely to be victims of most types of crime
the elderly and women are more likely to experience high
levels of fear than younger people and men. [Antunes
et al., 1977; Balkin, 1979; Baumer, 1978; Braungart et al.,
1980; Clemente, Kleiman, 1977; Dubau et al., 1979; Garofelo,
1981; Garafelo and Lach, 1978; Lee, 1982; Yin, 1980.]
According to Clemente and Kleiman (1976) approximately
40 percent of the people in the age group 18 to 41 years
and 50 percent of those who are older than 50 years feared
walking alone at night in their neighborhoods. Garofelo
(1979) confirmed that 63 percent of those 65 years and older
expressed similar fears. The national crime survey of 1981,
found that the 65 and older age group was the least
victimized age category.
-4-
-5-
Fear Of Crime By Education - Earlier studies find a
differentiation in the fear of crime by education. Clemente
and Kleiman (1976) found that the fear of crime is usually
correlated inversely with educational level. He reported
that 37 percent of the respondents with high school educa
tions fear crime. Forty-four percent of those attending
high school expressed a fear of crime. Forty-three percent
of those who dropped out of high school feared crime.
Hindleng (1974) observed similar patterns to Clemente and
Kleiman. He claimed that the relationship between previous
victimization is relatively weak among educated groups
compared to uneducated groups.
Fear Of Crime By Sex - Previous data indicate that the
victimization rate for women is low. However females usually
fear crime substantially more than men Debro (1982). Ennis
(1967); Conklin (1975); and Handleng (1974) showed that
females feared crime substantially more than men. Yin (1985)
and Baumer (1978) found that women fear crime more than men
because they, women, realize they are targets of sexual
assaults. Further, it has been found that women living
alone are more fearful of crime than women living in house
holds with others. Women are also more fearful of their
neighborhoods than are men (Baumer, 1978). Yin (1981)
found several reasons why women fear crime more than men:
-6-
(1) women believe more than men that there is a higher
increase in crime rates in their neighborhoods than any
other places; (2) women have a greater tendency to be
afraid of certain areas in metropolitan cities; (3) women
are more likely to perceive the crime situation as worsen
ing. (4) women are more likely to believe that other
people have limited their activities because of crime;
(5) women are more likely to worry about crime; and (6)
women see themselves as physically weaker than their male
counterparts.
Fear of Crime by Size of Household - Although there
are no extant studies on the association of fear of crime
and the size of household, there is an indirect indication
that the fear of crime may be associated with household
size. For example, Gubrum (1979) found that elderly persons
living alone, living away from friends of the same age, and
living without locally supportive relationships feared crime
more than those living with others and family members who
enjoyed supportive relationships. Hahan (1980) found
similar results among those elderly people who live alone.
Richards and Title (1981) suggested that living alone is
a significant source for fear of crime among women.
Braungart et al., (1980) reported that people living alone
are especially fearful of crime and that the most fearful
groups living alone were middle-aged and elderly females.
For men a greater fear was expressed by elderly males.
-7-
They also reported that young females and elderly males
living with others were fearful of crime.
Fear of Crime By Social-Economic Status - Reiss (1979)
found an intensive fear of crime among lower-income groups.
Ennis (1979) arrived at similar conclusions by examining the
National Victimization Survey conducted in 1966. He reported
that fear of walking alone, in..neighborhoods at night was
greater among low-income members than among high-income
groups. Louis Harris (1974) found that 31 percent of the
people with less than $3000.00 earnings per year reported
the fear of crime was a major social problem. On the other
hand, Lee (1982) suggested that fear of crime did not vary
significantly by income.
PREVENTIVE MEASURES
The literature available on strategies to prevent crime
can be grouped into two catagories: (1) how the community
protects itself from crime; and (2) how individuals protect
themselves from crime.
Crime Preventive Strategies By Communities - One popu
lar strategy adopted by several communities is neighborhood
watch; e.g., a strategy whereby neighbors watch each other's
house and property for any signs of illegal activity. The
Presidential Commission of 1967 explicitly noted the need
for citizen involvement both in improving the performance
of the Criminal Justice System and reducing the circumstances
-8-
and situations in which crimes are most likely to be commit
ted. Citizens involvement in this type of strategy is to
prevent their own victimization through various personal
protective measures; e.g., to protect property, and lessen
burguiary through preventive measures. Neighborhood Watch
limits unlawful access by creating both physical and psycho
logical barriers to offenders.
Protective escorting and block watching are two other
strategies of community crime prevention. [Leavakas, 1980;
Levin, 1980; Pennell, 1978; Percy, 1979; Rosenpraub and
Harlow, 1980.] Leavakas and Herz (1982) observed that most
citizens become involved in neighborhood anti-crime activit
ies not because of fear of crime but as a extension of their
general tendency toward community based voluntary action.
Wright and Mayer (1981) identified four crime neighborhood
control tactics: (1) neglect and copy strategy; (2) pre
ceptive strategy? (3) intergration strategy; and (4) recon
struction strategy. They argued that any crime control
system that is not some form of social adaptation will not
work.
Louis and Maxfield (1980) surveyed residents in four
Chicago neighborhoods to determine the relationship between
fear of crime and crime rate. They found that citizens •
perceptions of dangerous areas in their neighborhoods
matched the crime rates obtained in these areas.
-9-
They argued however that citizens1 perceptions of crime are
shaped not so much by the neighborhood conditions reflected
in the official crime reports but whether or not area members
had a community feeling. They feared crime less if they per
ceived themselves to belong to a close knit community.
Citizen involvement in crime prevention programs has
gained much attention, not only from the community but from
the Presidential Commission of Neighborhood Watch in 1967.
Other community anti-crime programs include the National
Sheriffs Association, Neighborhood Love Programs, Associat
ed Federation of Women Clubs, Helping Hands Programs, and
several youth preventions programs (Roebuck, 1985).
Crime Preventive Strategies By Individuals - Most of
the crime preventive strategies that individuals adopt are
primarily aimed to ensure home and personal safety. Some
individual crime prevention strategies follow: making
homes appear occupied when occupants are away; locking up
gasoline tanks, placing serial numbers on household items
and personal equipment (e.g. farm machinery, placing coded
buts of confetti into beams of hay bails for identification
purposes); security alarms, monitors, watch dogs, National
Rural Crime Prevention registers NCRC 26 and NCRC 27; Davis
and Proctor (1980); Crime Prevention Indicators System 1981;
Carter and Bealiea (1982). Still other individual preventive
measures include: dead bolt locks, secure windows, security
-10-
for sliding glass and other problem doors, auxilliary locks
for all doors, self analysis for home security, burglary
alarms system, night lighting for the home, and pad locks
for out buildings Roebuck (1985) .
Although the above studies offer a great deal of infor
mation about the relationship between the fear of crime and
people's background characteristics, none of these studies
systhesize these components into a meaningful casual model.
The fear of crime may be the covariant of preventive measures
adopted by those with differential characteristics.
This review of the literature indicates that one's
background characteristics influences one's fear of crime,
and that different kinds of people use different kinds of
preventive measures. This survey found no attempt to estab
lish a possible relationship between people's crime preven
tive measures and their fear of crime. This literature
review suggests two tasks: (1) to determine if the observed
relationship between individuals characteristics and fear of
crime hold true for the data set under study and (2) to
examine the.-relationship, if any, between preventive measures
and the fear of crime.
CHAPTER III
CONCEPTUAL FRAMEWORK
MEASUREMENT OF VARIABLES AND METHODOLOGY
The conceptual model for this thesis is presented in
Figure 3.1. The available information on fear of crime, as
revealed in Chapter II, indicates that one's background char
acteristics determine one's fear of crime. We also find
that differential groups use a wide variety of crime pre
ventive measures.
This study examines the relationship of respondents'
age, educational level, sex, size of household and social-
economic class to the degree of safety they feel during the
night and during the day. The study also examines the
relationship between preventive measures and the respondents
fear of crime. These measures are namely police watch,
mail postponement, neighbor mail watch, neighborhood house
check and house light.
MEASUREMENT OF VARIABLES
In order to maintain clarity, this section is organized
into three subsections: (1) measurement of background char
acteristics; (2) measurement of preventive measures; and
(3) measurement of fear of crime.
Measurement of background characteristics
Education is measured by the amount of schooling
reported by the respondents. Is there a correlation
-11-
-12-
between these two variables and if so what is the signific
ance of this correlation: For example do college graduates
fear crime differentially than do high school graduates?
Do these two groups use different types of preventive
measures? Do these two memberships vary in the degree of
safety they feel when using a given preventive measure?
Educational level categories are grouped as follows:
(1) enrolled in high school; (2) dropped out of grade
school; (3) grade school only; (4) dropped out of high
school; (5) high school -vocational; (6) vocational- no
high school; (7) zero to two years of college; (8) two-
three years of college; (9) college graduate; (10) graduate
or professional; (11) obtained GED; (12) never attended
school.
Age in this research is the number of respondents'
chronological years. Also it is a correlate of fear of
crime. The age groups utilized follows: (1) 15-18; (2)
19-25; (3) 26-64; (4) 65+.
Sex is the gender of respondents. Do females differ
significantially in fear of crime from their male counter
parts? For quantitative purposes this variable is coded as
(1) for male; and (2) for female.
Size of Household is the number of persons in a dwel
ling unit. Also household size correlated with fear of
crime.
-13-
MEASUREMENT OF PREVENTIVE MEASURES
Five preventive measures were selected for this analy
sis: police watch, mail postponement, neighbor mail watch,
neighborhood house check, and house light. Respondents were
required to answer "yes or "no" to each of the preventive
measures. We measured for the relationship, if any, be
tween preventive measures and background characteristics
on one hand; and, between preventive measures and fear of
crime on the other. Table 3.1 shows that the data on pre
ventive measures is fairly reliable (A=.61).
Social Economic Class is measured by a self-rating
variable from the following list of options: (1) lower;
(2) middle; and (3) upper.
MEASUREMENT OF FEAR OF CRIME
Fear of crime as defined in this research is the
degree of safety that respondents feel during the day or
during the night by age, education, sex, size of household
and social-economic class. The specific questions included
to measure the fear of crime among respondents were:
(A) how safe do you feel during the day: and (B) how safe
do you feel during the night: The rating involved a four
point scale from "very safe to "very unsafe."
ANALYTICAL PROCEDURES
In the analysis of the data two techniques are utilized:
(1) Pearson's Product Momentum Correlation; and
Variable
Niwhpr
V28
V29
V30
V31
V32
V33
Scale Mean
If Item
Deleted
7.385
7.441
8.029
7.767
7.647
8.037
Scale Varience
If Item
Deleted
1.9143
1.8684
1.5713
1.4356
1.5643
1.7884
Corrected Item
Total
Correlation
.26120
.23366
.48263
.47968
.37706
.26698
Squared
Multiple
Cdirrelation
.12007
.12587
.26916
.30210
.22161
.09595
Alpha
If Item
Deleted
.60459
.61465
.51922
.51269
.56293
.60509
-14i
Alpha - .61773Standardized Item Alpha - .61055
Item Alpha - .61055
-15-
(2) Multiple Linear Regression Analysis.
A brief description of each of these techniques follows.
Pearson's Product Momentum Correlation
Correlation is an inferential statistical measure
which measures the magnitude and direction of association
between any two variables. The magnitude of association
can be: (1) independent and dependent, (2) independent and
independent, or (3) dependent and dependent variables. The
direction of association between the variables is another
advantage of the correlation analysis, and can be positive
or inverse. Inverse is a negative association, while di
rect is a positive association. In explaining the vari
ables, the correlation coefficients help to measure the
degree of association or the strength of the relationship
between one pair of variables. The correlation coefficients
were obtained from the Pearson's Correlation Computer Pro
gram. Pearson's Product Momentum Correlation is a statis
tical measure of the amount of spread around the linear
least squared reduction. This correlation is commonly
used in studying the relationship between severity and
certainty of the coefficient it may be concluded that the
larger the magnitude the better the cohesiveness. If the
coefficient equals zero there is no correlation between the
two variables. The formula for computing Pearson's
Coefficient is:
-16-
(» - I) <y - y)
Qy - y)2J \l <*2> <y2)
Operationally, the correlation coefficient consist of
the ratio of the co-variation to the square root of the
product of the variation in x and the variation in y. The
square of the coefficient (r2) can be defined as the coef
ficient of determination or proportion of variance that is
explained.
MULTIPLE LINEAR REGRESSION ANALYSIS
A multiple regression analysis utilizes more than one
independent variable to predict the value of the dependent
variable. Regression analysis is used to measure the impact
of independent variables on dependent variables. In regres
sion analysis one can have a number, of independent variables,
but only one dependent variable in a given equation.
Regression techniques are of several types: Simple, Multi
ple, and Step-Wise Regression. A regression can be either
linear or nonlinear. A linear regression is one in which
the data inclines to fall along a straight plane. For the
present study multiple linear regression was used.
The method of multiple linear regression extends the
idea of simple regression with one independent
variable. Multiple linear regression deals with more than
one independent variable. Multiple regression allows the
FIGURE 3.1
Conceptual Model
Age
Education
Sex , .Size of Household
Social Economic Status
V-2B Past Police WatchV-29 Past Mail Postponement
V-30 Neighborhood Watchv-31 Neighborhood Mail WatchV-32 Past Neighborhood House Check
V-33 Past House Light
V-41 Safe Alone During the DayV-42 Safe Alone During the Might]
Background
Characteristics
Preventive Measures
Fear of Crime
-18-
researcher to predict the effect of two or more independent
variables xl, x2..., xk on the dependent variable y. The
fundamental regression equation is:
yl = a + bl + bk x k
Where,
yl = predicted scores of the dependent variable
xl, ..., xk = the scores of the independent variables
X X / • • • X&
a = intercept constant
bl, ..., bk = regression coefficients for the
independent variables xl, ..., xk
The above general regression model helps to develop a
series of models that should be used to analyze the concept
ual relationships among the variables in this study.
The symbolic form of these specific equations are given
below:
Fi = ao + blV28 + e (1)
Fi = ao + blV28 + b2V29 + e (2)
Fi = ao + blV28 + b2V29 + b3V30 (3)
Fi = ao + blV28 + b2V29 + b3V30 + b4V31 + e (4)
Fi = ao + blV28 + b2V29 + b3V30 + b4V31 + b5V32 + (5)
Fi = ao + blV28 + b2V29 + b3V30 + b4V31 + b5V32 +b6V33
+ e (6)
Fi = ao + blV28 + b2V29 + b3V30 + b4V31 + b5V32 + b6V33
+ b7V200 + e (7)
-19-
Fi = ao + blV28 + b2V29 + b3V30 + b4V31 + b5V32 + b6V33 +
b7V200 + b8V260 + e (8)
Fi = ao + blV28 + b2V29 + b3V30 + 4bV31 + 5bV32 + 6V33 +
b7V200 + b8V260 + b9V261 + e (9)
Fi = ao + lbV28 + 2bV29 + 3bV30 + b4V31 + b5V32 + b6V33 +
b7V200 + b8V260 + b9V261 +blOV295 (10)
Fi = ao +blV28 + b2V29 + b3V30 + b4V31 + b5V32 + b6V33 +
b7V200 + b8V260 + b9V261 + blOV295 + bllV325 (11)
Where,
Fi = Fear of crime (during the day if i = 1 and during
the night if i = 2)
V28 = Past Police Watch
V29 = Past Mail Postponement
V30 = Past Neighborhood Watch
V31 = Past Neighbor Mail Watch
V32 = Past Neighborhood House Check
V33 = Past House Light
V200 = Amount Of Schooling
V260 = Respondent Age
V261 = Respondent Sex
V295 = Total Number In Household
CHAPTER IV
DATA ANALYSIS
For the sake of clarity, this chapter is divided into
two sections: descriptive and analytical procedures. The
descriptive part of the analysis focused on the sample
distribution according to the variables selected for the
study. The analytical procedures assess the possible re
lationship between the presumed dependent and the presumed
independent variables.
DESCRIPTIVE ANALYSIS
Table 4.1 provides the frequency distribution and
corresponding percentages for the sampled respondents on
selected variables. The following patterns are observed:
Age - nearly forty-six percent of the respondents are
in the age group of 26-64 years, followed by twenty-one
percent between 19-25 years and nineteen percent between
15-18 years. Though the elderly comprise only thirteen
percent of the total sample they make up a large enough
group to compare with the other age categories. Only eight
percent of the respondents did not report their age.
Education - this variable has thirteen options varying
from never attended school to college and above. Nearly
twenty-eight percent of the respondents did not complete
high school; forty-two percent has completed high school
but did not complete college, and thirteen percent had
-20-
-21-
TABLB 4.1
FREQUENCY DISTRIBUTION OF SANPLBD RBSPONDBNTS_____—~ ——— •• —••———— — —•••- — —— —— — —— _*•_.»_ — — — —. — — — — — — —-• — — — ••••»«.,— — — .
Variable Nuaber Percent
(N=621)
Respondents Age
rV260)
15-18
19-25
26-64
65+
DK/NA
118
129
288
78
8
19.0
20.8
46.4
12.6
1.3
Respondents
Education
(V200)
Enrolled In High School 90 14.5
Drooped Out Grade 31 5.0Grade School Only 32 5.2Dropped Out High School 99 15.9High School Only 131 21.1High School-Vocation 44 7.1Vocation No High School 4 0.6Zero to Two College 56 9.0Two-Three College 33 5.3College Graduate 34 5.5Graduate or Professional 50 8.1Obtained GBD 1 °-2Never Attended School 3 0.5
DK/NA 11 1-8
41.2
57.5
1.3
Respondents Sex
(V261)
Male
Female
DK/NA
Respondents Household
Size
(V295)
1
2
3
4
5
6
7
256
357
8
78
148
107
97
70
46
30
12.6
23.8
17.2
15.6
11.3
7.4
4.8
Table 4.1 (Continued)
-22-
Variable Nuaber Percent
(N=621)
8
9
10
11
12
13
14
15
99
DK/NA
Fear of Criae
(V17)
Not A Problem
Soaewhat A Problem
Big Problem
DK/NA
Past Police Watch
(V28)
Yes
No
DK/NA
Past Mai 1 Postponement
(V29)
Yes
No
DK/NA
Neighborhood Watch
14
4
4
1
2
1
1
1
0
15
420
125
64
12
79
526
16
110
475
35
2.3
0.6
0.6
0.2
0.3
0.2
0.2
0.2
0.3
2.4
67.6
20.1
10.3
1.9
12.7
84.7
2.6
17.7
76.5
5.6
Yes
No
DK/NA
Neighbor Mail Watch
464
140
17
74.7
22.5
2.7
Yes
No
DK/NA
292
282
47
47.0
45.4
7.6
-23-
Table 4.1 (Continued)
.. * Nuaber PercentVariable »*u-ber ^^
Past
House Check
(V32)
007 38•2
Ye9 364 58.6No "ZZ 3 9
DK/NA zu
Past House Light
(V33)
v . 467 75-212S 20.3
No 27 4-3DK/NA z'
Safe Alone During
The Day
(V41)nn 9
436 *Very Safe * 21.4Reasonably Safe J« 3#1
Somewhat Unsafe 1» 2'6
Very Unsafe *)> 2.7
DK/NA 17
Safe Alone During
The Night
(V42)
«>in 33.8Very Safe ^1 26.4Reasonably Safe |" 18-0Sonewhat Unsafe ||^ 18 0
Very Unsafe *■)* 3 7DK/NA c*
-24-
graduated from college. Nearly fifteen percent of the
respondents are still in high school. Nearly two percent
did not answer the question.
Sex - the sample has a slight over representation of
females (fifty-eight percent) over males (forty-one percent),
A few respondents (8) were reluctant to provide information
on their gender.
Size of Household - some households are as large as
10-15 members. But, nearly one-third of the sample consists
of 2-3 members followed by those with 4-5 members. The no
response rate is larger for this variable (15 respondents)
than compared to age and sex.
PREVENTIVE MEASURES
Preventive measures adopted by respondents in the past
are measured in terms of six individual variables: (1)
police watch, (2) mail postponement; (3) neighborhood watch;
(4) neighborhood mail watch; (5) neighborhood house check;
and (6) house light. Of there variables house light (75
percent) and neighborhood watch (74.4 percent) were the
most common measures opted by respondents. Next in order
were neighbor mail watch, neighborhood house check, mail
postponement and then police watch. Police watch was per
ceived to be insignificant.
-25-
FEAR OF CRIME
The fear of crime is often a confusing variable to
measure at operational levels. The study collected data
on fear of crime via the following three questions:
1. "Was fear of crime a problem in the community?"
Approximately 30 percent responded from somewhat a
problem to a big problem.
2. "Were you safe alone during the day in your community?"
While 91 percent said that they were very safe to rea
sonably safe, 6 percent indicated somewhat to very
unsafe.
3. "Were you safe alone during the night in your com
munity?" As expected more people expressed feeling
unsafe during the night (61 percent) that during the
day (36 percent). From the reliability and validity
analysis it is concluded that the third question,
"were you safe alone during the night in your com
munity?" is more valid as well as reliable (Alpha=.73)
then the first (Alpha=.62) and second (Alpha=.43)
questions.
ANALYTICAL PROCEDURES
Correlation Analysis the Pearson's Product Momentum
Correlation Coefficients were computed between independent
and dependent variables and tested at the significance
level of 5 percent. The results of this analysis are
-26-
discussed under two headings: (1) Correlation between
fear of crime and background characteristics and (2) cor
relation between fear of crime and preventive measures.
Correlation Between Fear of Crime and Background
Characteristics
Age is negatively correlated with all fear of crime
variables and significantly safe alone during the night;
that is, as an individual advances in age he or she tends
to fear crime more, and this fear increases during the night.
This is expected because as age increases the perceived
vulnerability to being victimized increases. Several
studies support this relationship. (For Example, see
Clemente and Kleiman, 1976). Education is intensely as
sociated with "feeling safe alone during the day and night."
That is, as educational level increases the fear of crime
decreases. Hindeleang, (1974) provided an explanation for
less fear of crime among educated than uneducated persons.
He found that educated people do not permit victimization
to totally influence their perception of criminal activity
in the community. We suggest that educated persons tend
to be more affluent than uneducated persons, and conse
quently feel more capable than uneducated people to replace
victimization losses.
The significant positive relationship between sex
and fear of crime clearly indicates that females fear
-27-
crime at a higher rate than their male counterparts (Table
4.2) .
The relationship between the fear of crime on house
hold size is significant. Thus, the smaller the household
size the higher is the fear of crime. The elderly tend to
feel more vulnerability to victimization when they live
alone. Braungart et al., (1980) also reported that people
living alone are more fearful of crime that those living
with others.
FEAR OF CRIME AND PREVENTIVE MEASURES
The direction of the correlations between preventive
measures and fear of crime is negative. This inverse
relationship is observed consistently between fear of crime
and all preventive measures. Thus, the larger the adoption
of various preventive measures the less will be the fear of
crime. However, only two of the four preventive measures
are found to be significant: (1) police watch and (2)
neighborhood check.
Although the above analysis provides valuable infor
mation regarding the association of fear of crime with the
respondent's background characteristics and preventive mea
sures adopted (in the past), it does not distinguish bet
ween the dependent and independent variables. Such infor
mation is essential to test the conceptual model of the
study because one's choice of adopting or not adopting a
-28-
TABLB 4.2
PEARSON CORRELATION COEFFICIENTS
Preventive Measures
Respondents Age
(V260)
Respondents Education
(V200)
Respondents Sex
.0659* (V261)
Respondents Size
of Household
(V295)
Past Police Watch
(V28)
Past Mail Postponement
(V29)
Past Neighborhood Watch
(V30)
Past Neighbor Mail Watch
.0284* (V31)
Past Neighborhood
* House Check
(V32)
Past House Light
(V33)
Prevent
V17
-.0569
.1164*
.1024
-.0901
.0653*
-.0233
.0398
-.0957*
-.0957
-.0252
-.0854*
V41
.0275
-.0568
.1975*
-.0022
.0239
.0377
-.0019
-.0066
-.0066
.0024
V42
. 1189
-.1390
-.0659
.0286
.0077
.0412
. 1004
.0722*
.0739*
Note: *Significant at least .05 level.
y_17 Fear of crime as a problem in the community
V-41 Safe alone during the dayV-42 Safe alone during the nightPrevent-Preventive Measure Index
-29-
preventive measure depends on one's own background; and,
one's fear of crime is the outcome of his or her adoption
of various preventive measures. Therefore, the regression
analysis is employed.
REGRESSION ANALYSIS
The regression analysis was conducted separately for
fear of crime during the day and during the night. For
each analysis ten regression equations were tested. The
first six models tested the relative contribution of each
additional preventive measure and fear of crime. The fol
lowing three models measured the relationship of all pre
ventive measures (plus a single background characteristic
age, sex, or household size) to fear of crime. Finally,
model ten includes all variables in the model.
FEAR OF CRIME DURING THE DAY
The regression results of the chosen independent
variables of fear of crime during the day are provided in
Table 4.3 where it is observed that neighborhood watch has
as relatively larger relationship to fear of crime than the
other preventive measures. Secondly, these preventive mea
sures could explain a larger proportion of variance when
sex was included in the model than when age and education
were included. Thus, the predictability of preventive mea
sures increases more significantly when they interact with
the gender of the respondents than when they interact with
-30-
age or education. Interestingly, the predictability did
not increase even when age and education were included in
the model along with sex (equation 10). Thus, on the whole
neighborhood watch among preventive measures and sex among
background characteristics are significant variables in the
prediction of fear of crime.
FEAR OF CRIME DURING THE NIGHT
The predictability of the independent variable is
higher for fear of crime during the night than during the
day (Table 4.4). While the inclusion of education in the
model increased the ability to explain variance in fear
of crime during the night be one percent, age increased
it by two percent and sex by eight percent. Thus sex
continued to be a dominant background variable in explain
ing fear of crime during the night as well.
Summing up the regression results, this indicates
that preventive measures per se do not have a significant
relationship to fear of crime. But the predictability of
the preventive measures increases significantly when they
interact with sex rather than with education, age, or
household size. Therefore, sex plays a significant role
in measuring the relationship of preventive measures to
the fear of crime. In short, as the number of preventive
measures increase the degree of the fear of crime decrease
significantly. This is more the case for females than for
-31-
males, regardless of age, education, household size or
socio-economic class.
-32-
TABLB 4.3
STANDARD MBTRIC RBGRBSSION COBFFICIBNTS FOR BACH BQUATIONNODIl TO MBASURB FBAR OF CRIMB DURING THB DAY
V-41
Independent Model Raw
Beta
Standard
Beta
R2
V-29
V-28
Intercept
V-30
V-28
V-29Intercept
V-31
V-28
V-29
V-30
Intercept
V-32
V-29
V-30
V-28
V-31Intercept
V-33
V-28
' V-31V-29
V-32
V-30
Intercept
V-200
V-33
V-28
V-31
V-29
V-32
V-30
Intercept
.04336
.06824
1.17506*
.02948
.06637
.03880
1.15104*
-.06941
.07096
.04539
.06648
1.18962*
.00040
.04601
.06563
.06984
-.07080
1.18718*
-.01053
.06968
-.07013
.04723
-.00040
.06822
1.19392*
-.01610
-.00387
.06796
-.07134
.01912
-.00061
.05355
1.37551*
.02556
.03488
.01783
.03393
.02288
-.05143
.03627
.02676
.00029
.02713
.03970
.03570
-.05246
-.00063
-.63561
-.05196
.03784
.00029
.04127
-.07585
-.00052
.03474
-.05385
.01127
-.00044
.03239
.44192 .00250
.33011 .00281
.42987 .00489
.34338 .00490
.28740 .00493
.50848 .01015
-33-
Table 4.3 (Continued)
Independent Model Raw Standard
Variable »•*• >•*•
R2
V-260
V-28
V-30
V-200
V-33
V-32
V-29
V-31♦Intercept
V-261
V-31
V-260
V-28
V-29
V-32
V-30
Intercept
V-295
V-200
V-261
V-33
V-28
V-31
V-29
V-260
V-32
V-30
Intercept
.00098
.06730
.04907
-.01677
-.01943
.00048
-.02542
-.06681
1.34081
.22815
-.04316
.00061
.02658
.04770
.01552
.01263
1.08862*
.00097
.01873
.22856
.03609
.02645
.04317
.04813
.00057
.01487
.01267
1.09139
.02786
.03440
.02968
-.07904
-.01164
-.04950
-.04950
-.04190
.16715
-.03197
.01732
.01348
.06873
.01124
.00076
.00328
.08828
.16745
.02161
.02838
.03199
.02838
.01622
.01077
.00076
47470 .01086
1.4898 .03740
1.3369 .03741
♦Significant at least .05 level.
-34-
TABLB 4.4
STANDARD MBTRIC RBGRBSSION COBFFICIBNTS FOR BACH BQUATIONMODBL TO MBASURB FBAR OF CHIMB DURING NIGHT
V-42
IndependentVariable
V-28
Intercept
V-29
V-28
Intercept
V-30
V-28
V-29
V-31
V-28
V-29
V-30
Intercept
V-32
V-29
V-30
V-28
V-31
Intercept
V-33
V-28
V-32
V-29
V-30
V-31
Intercept
V-200
V-33
V-28
V-31
V-29
V-30
Intercept
V-260
V-28
V-30
V-200
V-33
V-32
Model
1
2
3
4
5
6
7
8
Raw
Beta
.13349
-.92773*
.18309
.20839
2.11810*
.18793
.19661
-.21200
.07665
.19158
.21923
.14707
1.92199*
.25879
-.17954
.09234
.12063
-.01202
1.76659*
.07657
.12175
.25906
-.18856
.07325
-.01698
1.71839*
-.04126*
.08138
.11717
-.01885
-.26129
.03603
2. 18365*
.00851
.11117
-.00089
-.04695
-.01219
.24429
StandardBeta
.042666
-.06745
■06659
.06706
.06282
-.07811
.03538
.06122
-.08078
.11692
-.06615
.03473
.03855
.00005
.02866
.03891
.11704
-.06947
.02755
-.00783
-. 12158
.03046
.03744
-.00869
.16153
.01355
.15077
.03553
-.00337
-.13833
-.00045
.11037
F
.68809
1.01743
1.24721
1.01995
1.5507
1.3325
1.8390
2.5754
R2
.00182
.00580
.01604
.01162
.02192
.02265
.03607
.05666
Table 4.4 (Continued)
-35-
Independent Model Raw StandardVariable Beta Beta
R2
V-29
V-31
V-261
V-31
V-260
V-28
V-200
V-33
V-29
V-32
V-30
Intercept
V-295
V-200
V-261
V-33
V-28
V-31
V-29
V-260
V-32
Intercept
9
10
-.20660
.02126
.63938
.08693
.00029
-.00029
-.05224
-.05565
-.14217
.21784
-.11224
1.17143*
-.03145
-.05063
.65140
.06380
-.00068
.08620
-.12930
.00061
-.10953
1.26279*
-.07612
.00091
.29187
.04012
.13098
.00009
-.15393
-.05238
.09844
.01325
-.04222
-.05751
-.14917
.29736
-.02388
.00021
.03978
-.04764
.10883
-.04120
6.0526 .13740
5.5539 .14006
CHAPTER V
SUMMARY, CONCLUSION AND IMPLICATIONS
This study determines the relationship between three
sets of variables: (1) background characteristics; (2)
preventive measures; and (3) fear of crime. The respondents
age, educational level, sex, household size and social-
economic class were examined in relationship to the degree
of fear of crime during the day and night. In addition to
background characteristics, several preventive measures
(past police watch, past mail postponement, past neighbor
mail watch, past neighborhood house check, and past hour
light) were examined in relationship to the fear of crime.
The research indicated that preventive measures do not
have a significant relationship to the fear of crime per se
but, a relationship is found when these measures interact
with sex, but not with age, educational level, household
size and social-economic class. Preventive measures reduce
the degree of fear of crime significantly among females
than among males regardless of their age, education, house
hold size or social-economic class.
The implications of this study suggest that additional
variables may be used in similar future research. For
example, respondents for this study were black and from
only two metropolitan areas. A study is needed including
several different kinds of population groups: non-urban
-36-
-37-
blacks, other metropolitan blacks, urban whites, non-
metropolitan whites.
This study suggests that females and elderly Americans
should be educated in some way to the fact that their fears
of crime are not related to their actual victimization.
The police may use these findings to design more effective
intergrated neighborhood crime preventive strategies-
whereby they and neighborhood members cooperate in programs:
(1) to reduce the fear of crime; (2) to implement preventive
measures, and to (3) protect the citizenry. The findings
suggest that Police-Community relations programs should
improve the perceived image of the police held by the
neighborhood members. Overall the results show that fear
of crime is a variable that affects people's quality of
life.
BIBLIOGRAPHY
Books
Conklin, John.
1975 The Impact of Crime.
New York: Macmillan Publishers.
Ennis, P.
1976 Criminal Victimization in the United States; A
Report of a National Survey
Washington, D.C.: U.S. Government Printing Office
Ennis, Richard.
1984 Endorsement of Criminal Research
New York: Richard Ennis Publications
Hahan, Paul H.
1980 Crime Against the Elderly
Santa Cruz: Davis Publishing Company, Inc.
Kahana, Ralpha J. and Broteng, Crete, L.
1969 Lectures in Psychology
New York: International Press.
Levin, Jack.
1983 Elementary Statistics in Social Research
New York: Harper and Row Publishers 1-318:
(50-60).
Louis, Harris.
1975 The Myth and Reality of Aging In America
National Council on Aging, Washington, D.C.
Reiss, Albert J.
1979 Report on Crime Preventions and Police and 1979
the Social Order
National Swedish Council for Crime Prevention,
Research and Development Division Vallingby,
Sweden.
Skogan, W.G. and W.R. Kelecka.
1977 The Fear of Crime.
Washington, D.C: The American Political Science
Association.
Yin, Peter.
1980 Victimization and the Aged
Springfield, ILL.: Charles Thomas Publisher.
-38-
-39-
Journal and Periodicals
Antunes, G.E., F.L. Cook, T.D. Cook and W.G. Skogan.
1977 "Patterns of Personal Crime Against the Elderly."
Gernotologist 17(4): 321-327.
Balkins, S.
1977 "Victimization Rates, Safety, and Fear of Crime."
Social Problem 26 (February): 343-358.
Baumer, T. L.
1978 "Research on Fear of Crime in the United States."
Victimology 3 (3-4): 254-264.
Braugart, M.M., Braugart, R.G., and Hoyer, W.J.
1980 "Age, Sex, and Social"
Sociological Focus 13(1): 55-66.
Carter, J.W., Bealiea S.
1982 "Age and Fearfulness: Personal and Situation
Factors."
Journal of Gernotology 30: 696-700.
Clark, John P. and Eugene P. Wenninger.1962 "Socio-economic class and area as correlates of
illegal behavior among juveniles."
American Sociological Review 27: 826-34.
Clemente, F., and Kleiman, M.B.
1977 "Fear of Crime in the United States: A Multivariate
Analysis."
Social Forces 56 (December): 519-531.
Davis, J. Proctor M.
1980 "The Fear of Crime in the United States."
Perspectives on Crime Victims
St. Louis: The C.V. Bosby Co.
Dubau, J., et al.
1979 "The Perception of Fear of Crime: Implications
for Neighborhood Cohesion, Social Activity, and
Social Forces 58(1) 176-193.
Garofalo, J. and Laub J.
1980 "The Fear of Crime: Broadening our Perspective."
Victimology 13 (3-4): 242-253.
Garofalo, James.
197J "Victimization and the Fear of Crime."
Journal of Research in Crime and Delinquency
16-17: 80-87.
-40-
Garofalo, J.
1977 "Fear of Crime in Large American Cities,"
Analytical Report Sd-Vad-19, The Application of
Victimization Survey Results Project (Criminal
Justice Research Center), Albany, New York.
Hindeling, M. J.
1974 "Public Opinion Regarding Crime, Criminal Justice,
and Related Topics."
Journal of Research in Crime and Delinquency
11(2): 101-116.
Levin, S.
1983 "The Perception of Fear of Crime: Implications
for Neighborhood Cohesion, Social Activity, and
Community Affect."
Social Force 58(1) : 176-193
Lee, R.
1963 Class Structure in the Social Consciousness
New York: Free Press.
Maxfield, A.
1980 "The Concept of Class and Answers Sociology."
Social Research 25: 53-59.
Nye, F. Ivan, James F. Short, Jr., and Virgil J. Olson.
"Socio-economic Status and Delinquent Behavior."
American Journal of Sociology 63: 381-9.
Pennell F.
197 8 "The Nature of Fear."
Paper, Survey Research Program, university of
Mass and The Joint Center for Urban Studies of
MIT and Harvard.
Percy, S.
1979 "American Lawlessness: An Inquiry."
American 1979 Journal of Sociology 18(1): 77-91.
Richards, C. and Title, W.
197 8 "The Myth of Social Class 1978 and Criminality:
An Emperical Assessment of Evidence."
American Sociological Review 18: 149-62.
Reynolds, P. and Blyth, D.
1976 "Occurrance, Reaction to, and Perceptions of
Victimization in an Urban Setting: Analysis of a
Survey in the Twin Cities Region."
(Unpublished Monograph), University of Minnesota.
-41-
Roebuck, J.
1985 Victimization, Crime Prevention, and Rural Crime
Surveys In the United States
Sociology Research Report Series 844 MAFES Report
5922 Mississippi State University.
Smith, Douglas, Charles R. Title, Wayne Villemez.
1978 "The Myth of Social Class and Criminality: An
Emperical Assessment of Emperical Evidence."
American Sociological Review 18: 149-62.
Wright, J.N., Sr. and Mayers, D.
1982 "Fear of Crime: Determinants and Consequences."
Unpublished Ph.D. Dissertation, The University of
Connecticut.
Yin, Peter and N.N. Anderson.
1981 "A Test of the Resignation Hypothesis: The Elderly
in Nursing Homes.: Paper presented at the Annual
Meeting of the Midwest Sociological Society,
Minneapolis.
Yin, Peter.
1985 "Fear of Crime Among the Elderly:
and Suggestions."
Social Problems 27(4): 492-504.
Some Issues