1
Chapter 1: Introduction to the Study
Introduction
Cultivation theory (e.g., Gerbner, 1958; Morgan & Shanahan, 2010; Morgan,
Shanahan, & Signorelli, 2015; Potter, 2014; Reber & Chang, 2000; Shanahan & Morgan,
1999) states that the more time that people spend in contact with mass media, the more
likely they are to equate reality with what they hear and see on those sources. In 1969,
Gerbner simplified his theory when argued that as people’s amount of media exposure
increases, so does their fear of crime. Due to the substantial focus by mass media on
violent crime, the public’s sense of safety may be significantly reduced in public spaces.
In this study, I examined this phenomenon in Los Angeles County by surveying
150 residents of Los Angeles County, who were at least 18 years old, and who indicated
that they have watched, listened to, or read about crime stories in the mass media during
the previous two weeks. This study used a series of three questionnaires to answer all
research questions.
By utilizing Rosen, Whaling, Carrier, Cheever, and Rokkum’s (2013) Media and
Technology Scale, which employs a 5-point Likert scale, I identified the amount of
exposure by participants to TV and the Internet.
By utilizing the Harmonisation Office of National Statistics’ (2015) Crime and
Fear of Crime Scale, which employs a 5-point Likert Scale, I identified participants’ level
of fear of crime upon exposure to crime stories through TV or Internet sources.
2
I used Mattick and Clarke’s (1998) Social Interaction Anxiety Scale, which
employs a 5-point Likert Scale, to identify participants’ level of social interaction anxiety
upon exposure to crime stories on TV or the Internet and the level of their fear of crime.
In this chapter, I describe why I chose the topic of media and fear of crime, then
review the problem and purpose of the study. Next, I described the significance of the
study, followed by descriptions of the theoretical foundation and nature of the study. I
then presented operational definitions of the terms used in the study. These definitions
were followed by a description of assumptions, limitations, and delimitations. After a
discussion of the scope of the study, the chapter concludes with a review of major
elements in the study.
Background of the Study
In 2014, I, an African American who grew up in Texas, moved to the city of Palos
Verdes in Los Angeles County, where I soon experienced a kind of racism that I had
never encountered in my native state. Caucasian Texans may harbor racist attitudes
privately, but are outwardly friendly. In California, my experience was that Caucasians
tended to avoid me rather than show open geniality or hostility. I pondered on the reasons
for this behavior and considered the possibility of mass media playing a role. When I
encountered cultivation theory (see Callanan, 2012; Gerbner, 1969; Kohm & Waid-
Lindberg, 2012; Shanahan & Morgan, 1999) in my graduate studies, I believed that I
discovered an explanation for this evasive behavioral trait. This study is the result of that
speculation.
3
Statement of the Problem
The mass media was recognized by researchers (e.g., Surette, 2007) as a primary
source of the public receiving information pertaining to crime. Dixon and Linz (2000)
found that approximately 30% of all news stories in mass media, both print and
broadcast, included reporting on criminal activity. Reiner (2007) contends that the mass
media distort the public’s perception of crime occurrence by disproportionately focusing
on violent crimes, thereby inhibiting people from engaging fully with others in public
spaces. Gibson (2014) argues that when the media in American cities presents stories
focused on criminality, the public’s fear of crime increases, therefore decreasing social
interaction.
Statement of the Purpose
In this study, I examined the phenomenon of increased social interaction anxiety
in public spaces in Los Angeles County through the lens of cultivation theory (Gerbner,
1969), to determine if the amount of media exposure to crime and the level of fear of
crime contributes to this behavior. The point of this study was to evaluate the
relationships between societal consumption of media messages, level of fear of crime,
and social interaction anxiety.
Significance of the Study
Up to this point, no one has studied the level of fear and presence of social
interaction anxiety of the people in Los Angeles County, and to what degree this fear can
be attributed to the mass media. Thus, this study makes a significant contribution to the
literature. Furthermore, when members of state and local governments have access to the
4
results of this study, they will be able to enact laws and codes to regulate the mass media
for the purpose of moderating the public’s level of fear regarding violent crime in public
spaces. In addition, they can use the study’s results to establish social programs to
alleviate public fear. Policymakers and public safety directors may also allocate
resources, like law enforcement, to communities in need. Also, the mass media may be
willing to self-regulate their programming for the benefit of the public.
Theoretical Framework
This study used Gerbner’s cultivation theory (see Gerbner, 1958; Morgan &
Shanahan, 2010; Morgan et al., 2015; Potter, 2014; Reber & Chang, 2000; Shanahan &
Morgan, 1999; Riddle, 2009). According to this theory, the more time people spend
watching television, listening to the radio, reading newspapers and magazines, and
participating in social media on the Internet, the more likely they are to equate reality
with what they hear and see on those mass media sources. In other words, the pictures
and messages conveyed by the mass media shape the public’s view of reality (Riddle,
2009). In Gerbner’s (1958) words, “massive attention to [the media] results in a slow,
steady, and cumulative internalization of aspects of those messages, especially the
aspects with ideological import” (p. 95). Or, as Shanahan and Morgan (1999) phrased the
issue in relation to television, “watching a great deal of television will be associated with
a tendency to hold specific and distant conceptions of reality” (p. 3).
5
Nature of the Study
Rationale of Selection of the Design
When developing the research questions for this study, I considered why it is
important and how the results can help law enforcement agencies. Therefore, a goal of
this study was to bring awareness to poor or weak social relationships due to citizens’
perceptions of property and violent crimes presented throughout the media.
Participants
As identified by counties, 150 Los Angeles County residents, who were at least
18 years old, and who indicated that they had watched, listened to, or read about crime
stories in the mass media during the previous two weeks, were surveyed.
Instruments
The consent form. Prior to taking the survey, participants read and provided their
consent by returning a completed survey. To protect their privacy, no consent signatures
were requested. This assured that their identities remain anonymous.
The survey one instrument. Rosen et al. (2013). The media and technology
usage scale. Computers and Human Behavior, 29, 2501-11.
This survey measured the study’s first independent variable, the amount of
exposure the participants have to TV and the Internet.
The survey two instrument. Harmonisation Office of National Statistics. (2015).
Crime and fear of crime scale. Titchfield, England: Author.
This survey measured the study’s second independent variable, the participants’
fear of crime.
6
The survey three instrument. Mattick and Clarke. (1998). Social interaction
anxiety scale. Behavior Research and Therapy, 36, 455-470.
This survey measured the study’s dependent variable, the participants’ level of
social interaction anxiety.
The Statistical Analysis Software
I used IBM SPSS software and SurveyMonkey to analyze statistical data—
determining, among other factors, how the Los Angeles County public’s social
interaction anxiety varies by demographic characteristics.
Procedures
The participants were located, identified, and surveyed by SurveyMonkey, which
sent me the results. I then analyzed the data to answer the research questions.
Research Questions and Hypotheses
The research question one. How does the Los Angeles County public’s amount
of media exposure and level of fear of crime impact social interaction anxiety?
The alternative hypothesis one. The public’s amount of media exposure and
level of fear of crime in Los Angeles County have a high social impact on individuals’
anxiety to interact socially.
The null hypothesis one. The public’s amount of media exposure and level of
fear of crime in Los Angeles County have no social impact on individuals’ anxiety to
interact socially.
7
The research question two. In Los Angeles County, what is the relationship
among the public’s amount of media exposure, level of fear of crime, and social
interaction anxiety after controlling for demographics (race/ethnicity, age, and gender)?
The alternative hypothesis two. There is a relationship between the public’s
amount of media exposure, level of fear of crime, and social interaction anxiety after
controlling for demographics.
The null hypothesis two. There is no relationship between the public’s amount of
media exposure, level of fear of crime, and social interaction anxiety after controlling for
demographics.
Research Variables
The first independent variable. In this study, the first independent variable was
the Los Angeles County public’s amount of media exposure to crime stories.
The second Independent Variable. In this study, the second independent
variable was the Los Angeles County public’s level of fear of crime geared toward
violent or property crimes in the county.
The dependent variable. In this study, the dependent variable was the level of
social interaction anxiety among the residents of Los Angeles County.
The mediating variables. In this study, the mediating variables included certain
demographic characteristics of the participants, including that of race/ethnicity, age, and
gender.
8
Operational Definitions
Crime: Behavior related to the commission of law-breaking (Dixon & Linz,
2000a).
Property crime: A theft-type offense that includes the taking of money or
property, but no use or threat of force is brought against the victim (Federal Bureau of
Investigation, 2010).
Demographic characteristics: Factors that identify an individual’s race or
ethnicity, educational level, income, gender, age, and etcetera (Lane & Meeker, 2003).
Fear of crime: An emotional response to criminal acts or prior victimization
(Ferraro & LaGrange, 1987).
Media: The different types of news broadcastings, either local or national (Lane
& Meeker, 2003).
Messages: Propositions, assumptions, and points of view that are understandable
only in terms of the social relationships and contexts in which they are produced
(Shanahan & Morgan, 1999).
Purposive sampling: Is a nonprobability sampling technique in which units are
selected because the investigator judges that the units somehow are representative of the
population (O’Sullivan, Rassel, & Berner, 2008).
Social interaction or engagement: The participation of individuals in desirable
activities with others (Glass, De Leon, Bassuk, & Berkman, 2006; Thomas, 2012).
9
T-test: A test of statistical significance requiring an interval dependent variable. It
is often used to test whether the difference between the arithmetic averages of two groups
is significant (O’Sullivan et al., 2008).
Two-step model: Sampling method in which the developer selects respondents
based on two criterions. For example, the selection of 142 cities or villages, then the
random selection of 20 addresses within these cities from the telephone directory (Custers
& Van Den Bulck, 2011).
Victimization: The process of being physically and illegally harmed by another
person or persons (Austin, Furr, & Spine, 2002).
Violent crime: An offense that involves the use of physical force or threat of
physical force (Federal Bureau of Investigation, 2013).
Assumptions
I assumed that the participants would answer the survey questions honestly,
including their age and their residency in Los Angeles County. I also assumed that the
methodology would answer the research questions. I had no assumptions about the role of
demographic variables in the study, especially race/ethnicity, age, and gender.
Limitations of the Study
One limitation of this study is that the results for the population of Los Angeles
Count may not be able to be generalized across populations of other cities in America.
More generally, the study’s quantitative approach to the subject limits the responses of
the participants to quantifiable characteristics, unlike the unrestricted nature open-ended
essay questions or personal interviews. Secondly, one of the survey instruments
10
considered for this study was developed in the United Kingdom and may present a
different perspective then that of the United States. In the effort to address these
limitations, I made sure that all scores aligned with each survey instrument, as well as
provided a response participants could mark if they chose to not respond to a question.
Scope of the Study
In this study, I only investigate the effect of mass media and fear of crime on
residents in Los Angeles County. Individuals who were not Los Angeles County
residents or who consumed no media were not included in the study.
Delimitations of the Study
One delimitation of this study, related to the sample, was my decision to study
one county in the nation, albeit a major one, rather than all counties in the country, which
would be beyond the means of one researcher to accomplish in a reasonable amount of
time. A second delimitation placed upon the research was that the study was completed
within one year.
Summary
This study was based on cultivation theory, which states that the more time people
spend in contact with mass media, the more likely they are to equate reality with what
they hear and see on those sources. Given the mass media’s heavy focus on violent crime,
the public’s sense of safety in public spaces may be significantly reduced. In this study, I
examined this phenomenon in Los Angeles County.
I use three questionnaires (Harmonisation Office of National Statistics, 2015;
Mattick & Clarke, 1998; Rosen et al., 2013) to survey 150 residents of Los Angeles
11
County, who were at least 18 years old, and who indicated that they had watched, listened
to, or read about crime stories on TV or the Internet during the previous two weeks.
In Chapter 2, I review 50 studies that cover the topic in various cities of the
United States, Mexico, and Europe.
12
Chapter 2: Literature Review
Introduction
Restatement of the Problem and the Purpose
The mass media distorts the public’s perception of crime rates by
disproportionately focusing on violent crimes. This focus inhibits people from engaging
fully with others in public spaces. In this study, I examined this phenomenon in Los
Angeles County through the lens of cultivation theory (Gerbner, 1969) to determine if the
residents of that county have high, medium, or low levels of social interaction anxiety
about entering public spaces, and to what degree they attribute their fear to the mass
media and their level of fear of crime.
Summary of the Content of the Literature Review
The content of the literature review is divided into eight major themes: () general
fear of crime, (2) avoidance of public spaces, (3) fear of violent crimes, (4) fear of
property crimes, (5) fear of property and violent crimes combined, (6) perceived risk and
vulnerability, (7) fear of online crimes, and (8) miscellaneous fear.
Organization of the Literature Review
For each article reviewed, I describe the purpose of the study, and then its
location, participants, and research method. This is followed by a review of the study’s
results and a summary of its recommendations for future research.
Literature Search Strategy
The library databases/search engines used. In the search for relevant literature,
I utilized Academic Search Complete, with which I was able to review articles from
13
several databases at once. Those databases included but were not limited to SAGE
Premier, Political Science Complete, and PsycINFO. With ProQuest, I was able to search
through topics such as communication, criminal justice, and health science. As an
alternative approach, I also reviewed ProQuest Dissertation and Theses Global as well as
Walden University Dissertations and Theses. In addition, I used the search engines of
Google and Google Scholar to obtain literature on topics of interest.
The search terms. In the search for relevant literature, I utilized key words such
as crime, fear, media, risk, and victimization. In addition, I used a combination of words
such as media and fear of crime, age and fear of crime, and media consumption and fear.
The scope of the literature. When reviewing literature for this study, I went back
5 years while focusing on the topics of media and fear of crime. Once I found relevant
articles, I looked at the references at the end of each one to find more literature related to
media and fear of crime. When determining seminal literature in the field of interest, I
found that the three most cited works in the articles she reviewed were Gerbner and
Gross (1976), Ferraro and LaGrange (1987), and Dixon and Linz (2000).
Theoretical Framework
Cultivation Theory
Cultivation theory was first devised by Gerbner (1958) and later elaborated upon
by other authors (see Morgan & Shanahan, 2010; Morgan et al., 2015; Potter, 2014;
Reber & Chang, 2000; Riddle, 2009; Shanahan & Morgan, 1999). The essence of the
theory is that the more time people spend watching television, listening to the radio,
reading newspapers and magazines, and participating in social media on the Internet, the
14
more likely they are to equate reality with what they hear and see on those mass media
sources. That is, the pictures and messages conveyed by the mass media shape the
public’s view of reality (Riddle, 2009). In Gerbner’s (1958) words, “massive attention to
[the media] results in a slow, steady, and cumulative internalization of aspects of those
messages, especially the aspects with ideological import” (p. 95). Or, as Shanahan and
Morgan (1999) phrased the issue in relation to television, “watching a great deal of
television will be associated with a tendency to hold specific and distant conceptions of
reality” (p. 3).
Rationale for Choosing Cultivation Theory
I chose this theory because it was directly related to the research questions.
Because I wanted to establish how and to what extent exposure to mass media’s coverage
of crime stories impacts the residents of Los Angeles County in terms of their feeling safe
enough to enter public spaces, this was the theoretical approach most relevant to the
topic.
How Cultivation Theory Has Been Used in Similar Prior Studies
The Boda and Szabo, 2011 study. In this study, conducted in Budapest,
Hungary, the participants reported that they largely ignored the news media and fictional
crime series in their assessment of crime and the criminal justice system, since they felt
that the media manipulate audiences. Thus, cultivation theory was not applicable to this
population.
The Callanan, 2012 study. This author found that as consumption of newspaper
and television news increased in southern California, so did fear of crime, just as
15
cultivation theory predicts. However, newspaper and television drama had little impact
on fear of crime.
The Callanan and Rosenberger, 2015 study. These authors used cultivation
theory to argue that fear of crime is increased by consumption of television programming.
The authors assumed that the fear levels of women would be elevated more than those of
men, which proved to be correct. However, they also assumed that the fear levels of
white women would be elevated more than those of women of color, which proved to be
incorrect.
The Custers and Van den Bulck, 2011 study. These authors, who relied on
cultivation theory, found that increased consumption of television in Flanders, Belgium,
predicted higher levels of fear of crime, just as the theory predicts.
The Custers and Van den Bulck, 2013 study. Again studying a population in
Flanders, Belgium, these authors, still relying on cultivation theory, found that media
consumption of sexually violent news stories was not a predictor for the level of that fear,
and thus cultivation theory was not confirmed in this case.
The Gibson, 2014 study. This author, relying on cultivation theory, found that
when the media in American cities featured stories about criminality, the public’s fear of
crime increased, confirming the theory.
The Jamieson and Romer, 2014 study. These authors, relying on cultivation
theory, found that as violence in TV programming rose and fell, so did the participants’
fear of crime in general. This correlation confirmed the predictions of cultivation theory.
16
The Kohm, Waid-Lindberg, Weinrath, Shelley, and Dobbs, 2012 study. These
authors, relying on cultivation theory, found that media consumption of crime stories was
indeed a predictor of fear of crime among American and Canadian undergraduate
students, thus confirming the theory.
The Nellis and Savage, 2012 study. These authors, relying on cultivation theory,
found that the amount of exposure to TV news about terrorism was positively associated
with fear of terrorism among participants in New York City and Washington, D.C., thus
confirming the theory. The findings indicated that the female participants were more
afraid of terrorism than the males, which also confirmed the theory.
Review of Other Studies That Have Used Similar Methodologies
There are many studies that used methodologies similar to this current study. The
following reviews of eight relevant studies reflect the extensive research that is being
done in this field and demonstrates the manner in which it continues to progress.
In 2011, Custers and Van den Bulck used a two-step model to select 1,394
participants for their study. A linear regression was used to analyze the participants’ fear
of crime. A multiple regression was used to analyze the correlation between the
participants’ television viewing and their fear of crime.
Callanan (2012) used probability sampling of California residents to determine
fear of crime with a varied demographic. The study was conducted statewide between
March and September of 1999. The author used a random digital dialing protocol and a
computer-generated telephone interviewing system to carry out the research. There were
4,245 completed surveys, which averaged 40 minutes to complete.
17
The dependent variables included questions related to fear of crime. The
independent variables were media outlets, including newspapers, local TV news
broadcasts, reality crime shows, and TV crime dramas. The mediating variable was the
respondents’ likelihood of becoming a victim of specific crimes in their community. The
demographic variables included the respondents’ education, age, income, and gender. An
ordinary least-square regression test, a standard error test, and a standard regression test
were used to analyze the data. These tests allowed the author to determine the
relationship between crime-related media stories and perceived risk and fear of crime.
The authors Kohm et al. (2012) examined fear of crime among college students
from four universities using a self-administered survey. Three schools were in the United
States and one was in Canada. Kohm et al. (2012) looked at three different vicarious
victimization experiences among social and personal situations (e.g., hearing about a
friend or relative being victimized). The students at the American universities were
examined in March, April, and August of 2007 and 2008. The students at the Canadian
university were examined in September of 2010. This gave Kohn et al. time to adapt their
survey to Canadian terminology and practices.
A purposive sampling technique was used to obtain data from volunteers in
different departments at each university. There were 1,466 students who participated in
the study; 397 students from the Canadian university and 1,069 from the three American
universities. A t-test was conducted to analyze the reliability of the different responses
from the Canadian and American students. The authors used an ordinary least-square
regression and regression coefficients to examine the data.
18
Nellis and Savage (2012) used a telephone survey to decide whether media
consumption influenced perceived risk of victimization and fear of terrorism. The study
was conducted in March and April of 2006, of which 532 surveys were collected. The
participants had to be at least 18 years old and reside in New York City or Washington,
D.C. Through the utilization of a random sampling tool called Survey Sampling
Incorporated, the authors obtained the information needed to accomplish their study’s
purpose. Each survey lasted approximately 15 minutes and focused on exposure to
terrorism news stories throughout media outlets and the participants’ perceived risk of
violence to themselves.
In 2013, Custers and Van den Bulck once again studied fear of crime, this time by
utilizing a standardized self-administered survey to examine the relationship between
exposure to crime-related stories in the media and fear of crime. The data was collected
in March of 2010 from 546 participants who were over the age of 18. A two-step
randomization process was used to select participants for the study. A total of 55 cities in
Flanders, Belgium, were chosen, in which 40 addresses were randomly selected from the
telephone book.
A structural equation test was conducted to determine the best fit of the
coefficients and their significance. Chi-square was used to verify the fit of the model and
the chi-square degrees of freedom ratio. The comparative fit index and the root means
square error of approximation were used in the statistical analysis.
Jamieson and Romer (2014) conducted a content analysis of the Coding of Health
and Media Project (CHAMP) when analyzing changes in the national exposure to violent
19
TV content from 1950 to the present (Annenberg Public Policy Center, 1993). The coders
were given specific definitions of violence, so that they understood what to look for in
each episode. The Gallup Poll was used to asses citizens’ fear of crime as well as their
perception of the occurrence of crime. SPSS 20.0 was used to run statistical tests on each
study question. A adjusted R-square analysis was also used as well as , a best-fitting
polynomial function, robust standard errors, and the Tucker-Lewis Index to analyze their
data.
Callanan and Rosenberger (2015) examined the relationship between various
crime-related media stories by using a computer-generated phone system. The purpose of
the study was to determine the impact that crime stories had on respondents of different
genders. The participants were 4,245 California citizens over the age of 18. The survey
consisted of 100 questions pertaining to perceived victimization, fear of criminality,
crime story consumption, and thoughts on the criminal justice system. An evaluation of
two dependent variables (perceived risk of neighborhood crime and the level of
participants’ fear), two independent variables (media consumption and prior criminal
victimization), and six moderating variables (race, gender, age, education, income, and
living status) was used to address the research question. An ordinary least-square
regression model was used to determine risk of criminal victimization and fear. A z-test
was used to identify whether the regression analysis differed across racial groups and
gender. A t-test was ran to determine if a gender difference existed.
20
Peer Reviewed Literature
Peer Reviewed Literature
Zhao, Lawton, and Longmire (2010) evaluated the relationship between property
and violent crime and fear of crime at the individual level. By utilizing a telephone
questionnaire between May 1 and June 3, 2008, the researchers collected surveys from
652 residents of Houston, Texas; 319 (48.93%) of were female and 333 (51.07%) were
male. Categorized by race, 333 (51.07%) of the participants were White, 156 (23.93%)
were Black, 104 (15.95%) were Hispanic, and 59 (9.05%) were “Others.”
The results indicated that participants’ were fearful of crime in direct relationship
to the number of crimes committed within an average of 528 feet of their home. Females
and older respondents reported higher levels of fear of crime than males and younger
respondents. Individuals with lower education were generally more fearful of crime than
individuals with higher levels of education. Therefore, the results concluded that there
was no correlation between the participants’ race and their fear of crime. The authors
recommended that future research replicate their study by examining specific types of
property and violent crimes in relation to individuals’ fear of crime.
In 2011, Boda and Szabo examined how and how much Hungarian citizens rely
on the media when interpreting issues of crime and their perception of criminal justice
system. The authors collected data from 27 participants between the ages of 20 and 24 in
Budapest, Hungary, between March and April 2010. The participants were divided into
three focus groups of nine individuals each.
21
The participants reported that they largely ignored the news media and fictional
crime series in their assessment of crime and the criminal justice system, since they felt
that the media can manipulate their audiences. Nevertheless, the participants felt that
violent crime is one of the biggest problems in Hungary. Furthermore, they felt that the
criminal justice system does not protect them or society at large from crime. The authors
recommended that future research pay more attention to participants’ personal
experiences.
In this same year, Custers and Van den Bulck determined the extent to which
media consumption correlated to fear of crime. Cultivation theory argued that individuals
with prior victimization experienced higher levels of fear when they view crime-related
stories throughout the media. The authors assumed that viewing crime-related stories on
TV would have a stronger correlation to fear of crime than experiencing similar stories
through other mediums.
The authors surveyed 142 undergraduate students from a communication course
in Flanders. There were 1,394 completed surveys obtained from citizens over 18 years of
age. The results supported the authors’ hypothesis that increased consumption of
television predicts higher levels of fear of crime. The authors recommended that future
research explore multidimensional factors associated with fear of crime.
Heber (2011) studied how fear of crime is affected by reports of crime in
newspapers. After studying the types of violent and property crimes reported in 167
articles in four Swedish national newspapers, the author learned that correlation existed
between fear of crime, gender differences, and the level of exposure readers had to stories
22
about crime. In Sweden, 81% of adults read a national newspaper daily. The results
indicated that women interviewed for newspaper articles were more fearful of becoming
the victims of sex crimes, whereas men interviewed for the articles were more fearful of
becoming the victims of crimes related to their occupations. The author suggested that
future research quantify gender differences regarding fear of crime by actually surveying
male and female victims and non-victims.
In addition, Rhineberger-Dunn (2011) assessed juvenile crime by reporting
similar accuracy by newspapers in metropolitan areas of different sizes. A total of 953
newspaper articles were related to juvenile delinquency from the Metropolitan Statistical
Area from 2002 to 2006, also included were statistics from the U.S. Census Bureau’s
2000 survey. From the data, newspapers in larger metropolitan areas reported higher rates
of juvenile crime than newspapers in small metropolitan areas. Furthermore, newspapers
cover violent crimes at a higher rate than property crimes in large metropolitan areas. The
author suggested that future research examine how juvenile crime in different sized
metropolitan areas is covered by other media, compared to coverage by newspapers.
In 2012, Alper and Chappell examined three theoretical models that explained
fear of crime. The vulnerability model argued that vulnerable people such as women,
blacks, seniors, the poor, and the physically disabled are more likely than others to fear
being criminally victimized (see Clemente & Kleiman, 1976; Hindelang, 1974; Kennedy
& Silverman, 1985; Warr, 1984). The disorder model argued that decreased social and
physical interaction in an environment leads to a greater fear of crime (Wilson &
23
Kellings, 1982). The social integration model argues that when residents respond
collectively to neighborhood problems, fear of crime can be reduced (Hale, 1996).
The authors collected data from 628 participants living in a city in the
southeastern United States by using a telephone survey. When they reviewed the three
theoretical models, the authors found that gender was not significantly related to fear of
property crime or fear of violent crime. The participants’ racial makeup was not
significantly related to fear of violent crime but was related to fear of property crime.
Individuals who had actually been criminally victimized had a greater sense than other
individuals of environmental disorder, had a greater distrust of their neighbors, and had
more fear of being victimized by violent or property crime. In general, the researchers
found that all three theoretical models were equally useful in explaining fear of property
and violent crime. The authors recommended that future research should replicate their
study by using a larger sample size and analyzing specific types of violent and property
crime.
In 2012, Callanan studied the impact that multiple forms of crime-related media
stories have on White, Latino, and African American respondents’ perception of crime
risk and fear of crime in their neighborhood. By employing a probability sampling
technique, the authors collected 3,712 surveys for their statistical analysis. The author
found that as media consumption increased, so did fear of crime. Fear of crime was
significantly higher among African Americans and Latinos than among Whites.
However, newspaper and television drama had little impact on fear of crime.
Local television news reports about crime elevated the perception of neighborhood crime
24
risk more than other variables (e.g., income, education, and age). Female perception of
neighborhood crime risk was significantly higher than that of their male counterparts.
The author recommended that future research should include more variables that measure
dimensions of media information-processing.
Cook and Fox (2012) examined whether individuals who had been physically
assaulted or sexually assaulted in the past were more fearful of violent crime than
individuals who had not been physically or sexually assaulted in the past. After obtaining
surveys from 282 undergraduate students at a southeastern university, the researchers
found that women were more fearful of violent crime than men. However, contrary to
their expectation, the researchers found that victims of physical assault and sexual assault
were no more fearful of violent crime than non-victims. Furthermore, both men and
women had high levels of fear of home invasion. The authors suggested that future
research should more carefully compare the fear of violent crime among victims of
physical assault and victims of sexual assault.
Foster, Giles-Corti, and Knuiman (2012) determined if fear of physical offense
was a deterrent from walking in public. The authors first hypothesized that individuals
who report high levels of fear of crime are less likely to walk in their neighborhood than
individuals who report low levels of fear of crime. The authors then hypothesized that
fear of walking would be greater for recreational walkers than for transport walkers. The
authors collected data from 1,044 first-time home buyers in Perth, West Australia, who
were over the age of 18 and had lived in their neighborhood for at least 12 months.
25
A self-report questionnaire asked the participants questions pertaining to
environments within a 10-to-15-minute walk from their home. The researchers found that
both of their hypotheses proved to be true. The authors recommended that future research
determine whether neighborhood efforts to deter crime will minimize the fear of crime
among recreational walkers.
In addition, Jorgensen, Ellis, and Ruddell (2012) examined individuals feelings of
safety in public parks that have other people around than in public parks in which they
were alone. The authors collected 540 surveys from volunteer participants in Salt Lake
City and the campus of the University of Utah. The participants were presented with 24
photos that illustrated various locations in a park, from deserted to crowded. The authors
found that the participants were more fearful of entering deserted spaces than crowded
ones. Furthermore, the female participants were more fearful than the male participants,
irrespective of how crowded the spaces were. The authors recommended that future
research place participants in actual parks rather than just viewing pictures.
Additionally, Kohm et al. (2012) examined three victimization perspectives
connected to college students’ social and personal environments. The indirect
victimization model (Weinrath, Clarke, & Forde, 2007) argued that secondhand
information influences a person’s fear of crime. Similarly, cultivation theory argues that
as the number of hours individuals spend consuming crime stories increases, so will their
fear of crime.
The authors argued that the relationship between media consumption of crime
stories and fear of crime was associated with traditional and situational factors. The
26
authors used a self-administered questionnaire to obtain information from 1,466
undergraduate students in three American universities and one Canadian university.
The authors discovered that media consumption of crime stories was an accurate
predictor of fear of crime. The participants also reported higher levels of fear of crime
when they used the Internet for dating and social communication. The authors
recommended that future studies examine similarities and differences between American
and Canadian television news programs about crime.
Lai et al. (2012) studied the relationship between violent crime in neighborhoods
and the residents’ specific fear of being burglarized. After surveying 737 residents of
Houston, Texas, between May and June 2008, the researchers discovered that there was
no correlation between non-burglary crimes in the neighborhoods and the residents’ fear
of burglary. However, there was a correlation between non-burglary crimes in the
neighborhoods and the residents’ fear of crime in general.
When the authors took distance into account, they found that if there were
burglary crimes within a half-mile to a mile of residents’ homes, the residents’ fear of
being burglarized increased as the distance from their homes to the burglarized homes
decreased. Nevertheless, the African American residents were less fearful of being
burglarized than the Caucasian residents, irrespective of their distance from the burglary
crime scenes.
The authors suggested replicating their study, but with an increased focus on
socioeconomic factors in the neighborhoods. They also suggested that future studies
27
distinguish residents’ fears of burglary, depending on how recently burglaries had
occurred in the neighborhoods.
Lane and Fox (2012) obtained 2,414 questionnaires from inmates in 14 jails in
Florida. Some of the inmates were current or former members of gangs, and other
inmates had never belonged to gangs. The authors concluded that the female inmates
were more fearful of violent crime than the male inmates. As for the males, the current,
former, and non-gang members all feared personal crimes more than property crimes.
However, the former gang members and non-gang members feared property crimes more
than the current gang members did. The authors suggested that future research compare
fear of sexual assault on the street as opposed to in jails or prisons.
Lee and Hilinski-Rosick (2012) determined whether lifestyle risk behaviors of
college students related to a decrease in the likelihood of becoming a victim of crime.
Routine activities theory (Cohen & Felson, 1979) argued that individuals who engage in
activities outside their home are at greater risk of being victimized. The lifestyle exposure
hypothesis (Hindelang, Gottfredson, & Garofalo, 1978) “argued that there are unusual
personal and lifestyle factors that either increase or decrease the danger of becoming a
victim” (p. 649). The authors expected college students who engaged in risky behaviors
to be less likely to fear crime on campus than students who do not participate in risky
behaviors.
The authors collected data from 3,472 undergraduate students from 12
universities across the United States. Regarding race and age, the authors found that fear
of crime was greater among younger, non-white students than among older, white
28
students. Gender was identified as being positively related to fear of crime, as the female
participants reported higher anxiety levels than the males. Prior victimization and
engagement in risky behaviors was positively correlated with fear of theft, but not with
any other crime. Thus, the authors recommended that future research further explore the
relationship between fear and prior victimization.
Lorenc et al.’s (2012) article on “Crime, fear of crime, environment, and mental
health and wellbeing: Mapping review of theories and causal pathways” examined the
linkage between fear of crime and participants’ well-being. The authors found that fear of
crime significantly affected people’s health, anxiety, social well-being, and avoidance
behaviors (e.g., in unsafe neighborhoods). On the other hand, the authors note that crime
prevention interventions that raise the public’s awareness of crime can sometimes have
adverse effects by increasing public anxiety. Since the authors did not conduct their own
empirical study, they did not make recommendations for future research.
Radar, Crossman, and Porter (2012) studied the physical and social vulnerability
levels among individuals, focusing on understanding why certain groups of people fear
crime more than others. By utilizing information from the Center for Race, Religion, and
Urban Life (CORRUL, 2006) and the U.S. Census Bureau (2000), the authors collected
2,610 surveys from respondents across the United States.
Slightly more than one-third (37%) of the respondents felt unsafe in their
neighborhood within the past year, whereas slightly less than two-thirds (63%) of the
respondents felt safe in their neighborhood within the past year. When the authors
considered racial makeup, 48% of the participants were White, 20% were Black, 20%
29
were Hispanic, 7% were Asian, and 4% classified themselves as “Other.” Three-fourths
(75%) of Blacks and slightly more than four-fifths (82%) of Hispanics were more likely
to report being fearful than Whites, whereas Asians and “Others” were not significantly
different from Whites in this respect.
Female respondents reported higher levels of insecurity than their male
counterparts. Older respondents and those in poor health were also more likely to report
higher levels of insecurity. The authors recommended that future research explore the
interconnections between physical and social vulnerability in relation to fear of crime in
greater depth.
Rengifo and Bolton (2012) studied the impact that various dimensions of fear of
crime, and behavioral adaptations to that fear, have on individuals’ perception of risk and
disorder. The authors used information from the British Crime Survey (2007-2008), to
examine the data from 11,315 respondents in England and Wales. They found that
individuals with higher levels of fear of crime and lower levels of disorder were also
those with higher voluntary participation in their places of work. Not only did higher
participation in voluntary activities correlate with fear of crime and perception of
disorder, but it also shaped the behavioral patterns of the respondents. That is to say, the
more the respondents engaged in voluntary activities in their work, the lower their level
of fear of crime. The authors recommended that future research study how individuals’
use of leisure time relates to their fear of crime.
Vilalta (2012) focused on residents’ fear of crime in Mexico City after they
installed home security systems, which included high walls, reinforced windows, and
30
watchdogs. After surveying the residents of 1,549 homes during an eight-week period in
August and September 2007, the author found that nearly half of the respondents (49.7%)
felt secure being alone in their homes, while the other half felt insecure. In fact, the
residents with the high walls felt the most insecure. Furthermore, only slightly more than
a quarter of the residents (28.4%) felt secure when they were out in their neighborhoods.
The author suggested that future research study whether or not security systems deter
criminal activity.
To conclude this year, Nellis and Savage (2012) studied participants’ fear of
terrorism. In March and April 2006, the researchers surveyed 527 adult participants, of
whom 296 (56.2%) were females and 231 (43.8%) were males, all of them living either in
New York City or Washington, D.C. Of the 527 participants, 381 (72.3%) were White
and 146 (27.7%) were non-White.
Regarding victimization by terrorists, the participants reported being more fearful
for their family members than for themselves. When considering the credibility of the
media, the participants reported that the media were only moderately accurate in their
presentation of news about terrorism (3.82 on a scale of 1 to 7). The amount of exposure
to TV news about terrorism was positively associated with fear of terrorism. Again, as
evidenced by other studies, female participants were more afraid of terrorism than males.
Furthermore, the non-White participants thought that the risk of terrorism was greater
than the White participants thought. The authors recommended that future research
replicate their study by evaluating participants’ fear of terrorism over a longer period of
time.
31
In 2013, Chadee and Ng Ying determined that a general fear level was a better
predictor of fear of crime than perceived risk of victimization. The self-interest model
used by these authors argued that the short- or long-term impact of an issue takes a toll on
a person’s well-being.
The authors used a comparative multistage sampling method in June 2009 to
gather 1,197 responses from residents living in Trinidad in the Caribbean. Regarding
gender, the female respondents reported higher stress levels than the male respondents.
The authors recommended that future research explore a broader area of general fear to
determine its role in emotional fear behaviors.
Custers and Van den Bulck (2013) article on “The cultivation of fear of sexual
violence in women: Processes and moderators of the relationship between television and
fear” identified a relationship between fear of crime and sexual violence shown on
television news programs in Flanders, Belgium. These authors relied on cultivation
theory, similarly to this study, which assumed that the consumption of crime-related
stories over time influences individuals’ perceptions of crime.
In March 2010, the authors used a standardized questionnaire to gather data from
546 female respondents over the age of 18 who lived in one of the 40 zip codes selected
at random from the Flanders telephone book. The authors concluded that fear of sexual
violence among women was related to perceived risk, but media consumption of sexually
violent news stories was not a predictor for the level of that fear. The authors suggested
that future research study males as well as females.
32
In this same year, Goodall, Slater, and Myers (2013) examined the impact of
alcohol-related crime stories in newspapers on individuals’ fear and anger. The authors
collected data from 789 adult men and women randomly chosen from across the United
States. The authors concluded that when alcohol played a role in violent crimes, the
respondents’ anger toward the accused increased. Furthermore, the respondents who were
more frightened by the crimes tended to hold society responsible, whereas those who
were less frightened tended to blame the accused. The authors suggested that future
research should differentiate the role of alcohol in different types of violent crimes.
Hanslmaier (2013) studied the impact that personal criminal victimization and
news of local crime rates have on fear of local crime and life satisfaction. The author
collected 3,245 questionnaires from respondents living in 413 German counties in
January and February 2010. The author collected 195 (6.0%) questionnaires from
participants reported having been criminally victimized in the previous two years, of
whom 162 (83.08%) experienced theft, 28 (14.36%) experienced assault, and 5 (2.56%)
experienced both crimes.
The victims of crime reported higher levels of fear of local crime and lower levels
of life satisfaction than the non-victims. Furthermore, those participants who were highly
exposed to crime news through their local newspapers were more fearful of local crime
than the participants who consumed fewer stories about local crime news in their local
newspapers. The author recommended that future research replicate this study by
collecting data over a longer period of time.
33
Hawdon, Rasanen, Oksanen, and Vuori (2013) identified how fear of violent
crime affected individuals’ and groups’ sense of well-being. The authors collected 700
surveys randomly from participants between the ages of 18 and 74, who lived in Helsinki,
Finland, in the southeastern part of the country (and its capital), and Ostrobothnia, in the
western part of the country. The authors concluded that when the crimes targeted
individuals, social solidarity declined; however, when the crimes targeted groups or
communities (as in the case of terrorism), social solidarity increased. The authors
suggested that future research evaluate how crime and fear operate among a population in
daily life after mass tragedies.
Henson, Reyns, and Fisher (2013) examined individuals’ intensity of fear of
online interpersonal victimization and its predictors. After collecting data from 838
students in the Midwest, the authors found that students were more afraid of online
interpersonal victimization by a stranger than by a friend, acquaintance, or current or
former intimate partner. Perceived risk influenced respondents’ fear of online
interpersonal victimization for all types of victim-offender relationships (i.e.,
relationships between victims and strangers, friends, acquaintances, or current or former
intimate partners).
The authors concluded that significantly positive correlation existed between
direct (in-person) victimization and fear of online interpersonal victimization by a current
or former intimate partner. The authors recommended that future research further explore
the nature and predictors related to fear of online crimes.
34
Hirtenlehner and Farrall in (2013) examined the connection between
modernization and fear of crime by comparing two theoretical approaches suggested by
Hough (2009): the generalized insecurity approach and the expanded community concern
approach. The former argued that “free-floating, amorphous anxieties about
modernization are directly projected onto crime” (p. 12); whereas the latter argued that
“abstract anxieties about social change require the prism of local conditions in order to
convert into fear of crime” (p. 18). In other words, fear of crime is either caused by social
changes on an international scale or social changes on a local scale.
After collecting 651 questionnaires from 312 males (47.93%) and 339 females
(52.07%) over the age of 20 in Linz, Austria, the authors found support for both
theoretical approaches, with a slight trend towards the generalized insecurity model. The
authors acknowledged, however, that “pathways into fear of crime may differ from
country to country, depending on the sociocultural and political-institutional makeup of a
society” (p. 5). The authors recommended that future research “take into consideration
the broader cultural and institutional makeup of a society and their interaction with
sentiments of insecurity” (p. 20).
Kappas, Greve, and Hellmers (2013) examined how older adults express greater
precautionary behaviors toward crime than their younger counterparts by, for example,
not leaving home after dark, avoiding certain streets, avoiding strangers, and so on. The
authors collected data from 528 young, middle-aged, and older adults in Lower Saxony,
Germany, in the summer of 2009 and January 2010. Of the 528 participants, 308
(58.33%) were between the ages 18 and 30, 106 (20.08%) were between the ages 50 and
35
64, and 114 (21.59%) were between the ages of 65 and 84. (It was unexplained why
individuals between the ages of 31 and 49 were not included in the study).
The authors concluded that older adults were more fearful of crime than younger
and middle-aged adults, and felt less safe than the other two groups. However, there was
no significant difference between the three age groups in regard to their evaluation of
their own neighborhood safety. The authors recommended that future research study what
specific behavioral factors account for these differences among the three age groups.
Lane and Fox (2013) article “Fear of property, violent, and gang crime:
Examining the shadow of sexual assault thesis among male and female offenders”
examined the impact of sexual and nonsexual assault on female and male inmates in 14
(70%) of the 20 jails in the state of Florida between 2008 and 2009, in regard to those
inmates’ fear of being victimized in jail by other inmates. The researchers collected data
from 2,345 inmates, 1,746 (74.46%) of whom were males, and 599 (25.54%) of whom
were females.
The authors found that a majority of the inmates were young, non-White, non-
gang members. The female inmates were more likely to be White than the male inmates;
the female inmates were also less likely than the male inmates to be involved in gangs,
either currently or previously. Furthermore, more than half (51%) of the female inmates
reported having been sexually assaulted in jail, as compared to only 7% of the male
inmates. On the other hand, the male inmates reported a higher rate than the female
inmates of having been violently victimized non-sexually, especially by gang members,
in incidents that did not involve theft of property. The female inmates were more fearful
36
than the male inmates of being victimized sexually or non-sexually, but they were less
fearful than the male inmates of being victimized by gang members. The authors
recommended that future research compare levels of fear of sexual assault among inmates
and non-inmates.
Examining a theoretical approach, Ozascilar (2013) tested the shadow of sexual
assault theory, which argued that women who fear being sexually assaulted experience
increased fear of other crimes. The author surveyed 1,051 undergraduate students of both
genders at Lund University, in southern Sweden. The author concluded female students
were indeed more afraid of crime than males. In fact, females fear of crime was twice as
high as that of the males, and applied equally to violent and nonviolent crime. The author
suggested that future research test the validity of the shadow of sexual assault theory
among the general population.
In this follow-up to the previously reviewed study, Rhineberger-Dunn (2013)
examined how 231 articles published in newspapers from five small metropolitan areas
portrayed juvenile offenders and their victims. Once again utilizing statistics from the
Metropolitan Statistical Area table of 2002 to 2006, as well as statistics from the U.S.
Census Bureau’s survey in 2000, the author found, first, that the newspapers portrayed
juvenile offenders as committing more violent crimes than property crimes; second, that
the newspapers reported that most juvenile offenders are male. The author suggested that
future research focus more on the victims of juvenile offenders than on the offenders
themselves.
37
In addition to prior research, Stodolska, Shinew, Acevedo, and Roman (2013)
evaluated the impact of crime on outdoor recreational activities among Mexican
American youth in the South Lawndale neighborhood of Chicago, which the locals refer
to as “Little Village.” The authors interviewed 25 adolescents between May and
November of 2010, found that most of the participants reported that crime was an issue in
their community, which had a greater impact on the older participants than the younger
ones. All the youths felt safest participating in leisure activities near their home, near the
home of a relative, and during school hours, when many people are around. The authors
recommended that their study be replicated with participants from other racial groups.
Vieno, Roccato, and Russo (2013) examined fear of crime as a function of one’s
environment and as a function of individual and societal characteristics. After obtaining
data from 16,306 participants in 27 European countries, the authors found that the lowest
levels of fear of crime occurred in Scandinavian countries, whereas the highest levels
occurred in Eastern European countries. In general, the level of fear of crime increased as
the researchers examined data from north to south and from west to east. Furthermore,
the authors found that living in big cities was associated with increased levels of fear of
crime. As for individual and societal characteristics, the researchers found the highest
levels of fear of crime among women, seniors, unemployed or poor individuals, and
persons with low levels of education. The authors recommended that future research
replicate their study by evaluating psychological vulnerabilities associated with fear of
crime.
38
To conclude this year, Visser, Scholte, and Scheepers (2013) studied how fear of
crime and feelings of vulnerability impact individuals at the national level rather than the
local level. Using two cross-sectional surveys conducted in 2006 and 2008, the authors
analyzed the responses of 77,674 individuals from 25 European countries. The authors
found, for example, that individuals in Eastern European countries expressed higher
levels of fear of crime and vulnerability than did individuals in Nordic countries.
Surprisingly, there was no relationship between feelings of vulnerability and the size of
the immigrant population in the countries. Also surprisingly, a higher level of crime in a
country resulted in increased trust in law enforcement. Female respondents showed
higher levels of fear of crime and feelings of vulnerability than their male counterparts.
The authors recommended that future research replicate their study by using a smaller
sample size over a longer period of time.
In 2014, Breetzke and Pearson identified the occurrence of crime in one’s
neighborhood impacts one’s level of fear of crime. The authors analyzed police records in
New Zealand between 2008 and 2010, during which 347,679 incidents of crime were
reported. The researchers, who surveyed 8,000 random participants, found that females
were significantly more fearful of crime than males. Also, individuals who had
previously been victims of crime were more fearful of crime than were non-victims.
Furthermore, individuals who lived in poverty areas were more fearful of crime than
individuals who lived in more prosperous areas. The authors recommended that a similar
study be conducted by other nations.
39
Creighton, Walker, and Anderson (2014) article “Coverage of black versus white
males in local television news lead stories” examined which male racial group, Black or
White, was represented more often in leading stories on TV, and how this representation
related to police reports of arrests for that period of time. The authors used data from the
four main TV stations in Omaha, Nebraska to review 364 news stories reported between
September and November 2012, analyzing the 188 that were related to crime.
In September 2012, there were a total of 101 lead stories, 68 (67%) of which were
associated with criminal activity. In those 68 stories, Black males were the primary
suspects in 51 (75%), and White males were the primary suspects in 17 (25%).
In October 2012, there were a total of 50 lead stories, 33 (66%) of which were
crime-related. Out of the 33 stories, 23 (70%) presented a Black male as the primary
suspect, compared to 10 (30%) that presented a White male as the primary suspect.
In November 2012, there were a total of 37 stories, 15 (41%) of which were
crime-related. Of the 15 stories, Black males were the primary suspect in 6 (40%), and
White males accounted for 9 (60%).
Of the 116 crime-related stories in the three months, 80 (69%) presented Black
males as the primary suspect and 36 (31%) presented White males as the primary suspect.
When the researchers examined the police records of actual arrests during those three
months, they found that Whites accounted for 61% of the primary suspects, and Blacks
accounted for 39%. Since only the suspect and arrest results for November were in line
with the actual statistics, the authors found that there was a clear racial bias in the media
portrayal of criminal suspects. The authors recommended that future research replicate
40
their study by evaluating television news in Omaha, Nebraska, over a longer period of
time.
Gibson (2014) meta-study found that when the media in American cities featured
stories about criminality, the public’s fear of crime increased. Furthermore, the public
regarded black neighborhoods as more unsafe than white neighborhoods. Gibson
recommends that urban communication scholars should consider how fear of crime
affects quality of life in urban settings.
Jamieson and Romer (2014) to examined the accuracy of cultivation theory’s
prediction that prolonged exposure to TV violence increases the public’s fear of crime in
general and their perception of local crime rates. By utilizing Brooks and Marsh’s (2009)
Coding of Health and Media Project, which contains data about the top 30 prime-time
dramas on network television between 1972 and 2009, the authors statistically analyzed a
total of 475.4 hours of commercial-free programs.
On a scale of 1 to 10, the authors concluded that violence in TV programming
decreased from 6.5 in 1972 to 1.4 in 1996, and then rose to 3.7 in 2009, as determined by
20 undergraduate students who were trained to master a code book of rules for the
identification of violence and other types of content. During this same period from 1972
to 2009, the public’s fear of crime declined from its highest point in the 1980s, when 42%
of the public feared crime, to 2001, when 30% of the public feared crime, and then rose
again in 2009, when 37% of the public feared crime.
Furthermore, national crime rates predicted the public’s perception of local crime
rates, but (despite cultivation theory) violence on TV did not. On the other hand,
41
prolonged exposure to TV violence did increase the public’s fear of crime. The authors
recommended that future research systematically evaluate how the accuracy of the
predictions of cultivation theory has changed over time.
In 2014, Krause studied the impact of crime-related media coverage on citizens’
attitudes toward crime control. The authors hypothesized that there would be a positive
correlation between exposure to crime and support for authoritarian crime control. After
collecting data from 503 residents of Guatemala City, Guatemala, the author concluded
that there was not a significant relationship between fear of violent crime and support for
authoritarian crime control. On the other hand, if the respondents distrusted their local
governments, they tended to support strict crime control. The author suggested that future
research compare Guatemala City with other Central American cities that experience
problems with crime control.
Malinen, Willis, and Johnston (2014) studied the impact of media reports of
sexual offenses on the attitude of the public toward recently released sexual offenders.
The authors hypothesized that individuals would have less negative attitudes toward
sexual offenders if they were exposed to informative stories rather than the fear-inducing
stories that are common in the media. The authors also expected female respondents to
have a greater negative impression of sexual offenders than male respondents.
After collecting data from 87 first-year psychology students at a New Zealand
university, the authors found that sensationalized stories did indeed increase negative
attitudes toward sexual offenders among both males and females, but female attitudes
42
were significantly more negative than those of males. The authors suggested that future
research examine the longevity of attitudes toward sexual offenders.
Stein (2014) used the broken window model (cf. Hinkle, 2015), to evaluate the
relationship between community disorder and fear of crime in three small American
cities: a city in the Midwest with a population of 26,985; a city in the Northeast with a
population of 17,967; and a city in the East with a population 5,563.
From police reports and responses to a questionnaire from 892 residents, the
author found that the majority of the residents felt safe in their neighborhoods.
Nevertheless, there was a strong relationship between community disorder and fear of
crime. The author suggested that future studies be made of trust among neighbors as a
determinant of how safe people feel in their communities.
Steinmetz and Austin (2014) article “Fear of criminal victimization on a college
campus: A visual and survey analysis of location and demographic factors” studied fear
of crime among 235 college students over the age of 18 in relation to the six most
dangerous places on the campus of the University of Louisville. After showing the
students twelve photographs of six locations on campus, the authors found that enclosed
walkway photos brought about the highest levels of fear and victimization among both
genders, although the males were less fearful than the females. The authors suggested
that future research explore the association between fear of crime and different types of
school events.
The study (Yu, 2014) examined the impact that perceived crime seriousness,
perceived risk of victimization, and actual victimization experiences have on fear of
43
cyber-crimes. The author surveyed 270 students at an urban university in the Midwest:
148 females and 122 males. Perceived risk of victimization and perceived crime
seriousness were significant predictors for fear of online scams. Students who frequently
shopped online had more fear of online crimes than those with minimal online shopping
experience. Not surprisingly, female students were more fearful of online crimes than the
male students. The author suggested that future research replicate this study by using a
different sampling population.
In 2015, Brands, Schwanen, and Van Aalst interviewed 30 students between the
ages of 18 and 25 in Utrecht, the Netherlands. Each student was interviewed three times.
The researchers found that the respondents felt safer in areas that were well lit as opposed
to non-well-lit places. Surprisingly, the presence of police in the area only made a third of
the respondents feel safer, whereas the other two-thirds reported no change in their fear
of crime from this factor. The researchers suggested that future research should use larger
samples.
Callanan and Rosenberger, 2015 studied how crime-related stories had different
impacts on female and male residents of California over the age of 18 (surveying 2,454
females and 1,791 males). Like the present author, these authors used cultivation theory
to argue that fear of crime is increased by consumption of television programming. The
authors assumed that the fear levels of women would be elevated more than those of men,
which proved to be correct. However, they also assumed that the fear levels of white
women would be elevated more than those of women of color, which proved to be
incorrect, since the fear levels of both groups were statistically the same. The authors
44
recommended that additional studies be made of gender and racial differences in relation
to fear of crime.
Dixon (2015) studied the racial representation of perpetrators of crime, victims,
and police officers in television programs broadcast in the Los Angeles area between
2008 and 2012. By utilizing the Equal Probability of Selection Method (Rosenthal &
Steen, 2012), the author statistically analyzed 117 news programs, finding that
approximately 30% of those programs featured some criminal act, approximately half of
which were violent, including murder. As for racial makeup, Whites represented 27% of
the persons in the 117 programs, Blacks represented 27%, Latinos represented 41%, and
“Others” represented 5%.
The author found that Blacks were accurately portrayed as victims, offenders, and
police officers. Latinos were overrepresented as offenders and underrepresented as
victims and police officers. Whites were overrepresented as victims and police officers,
and underrepresented as offenders. The author recommended that future research
replicate this study by utilizing a larger sample size of news programs and also by
considering nonviolent as well as violent crimes.
In this follow-up study, Dixon and Williams (2015) extended Dixon’s (2015)
prior research by evaluating whether cable news outlets differed from network news
outlets in their treatment of social categories in crime portrayals. The authors analyzed a
total of 146 news programs broadcast in the Los Angeles area, of which 90 (61.64%)
presented at least one crime story.
45
In both cable news and broadcast news programs, the authors found that Whites
were accurately represented as both perpetrators and victims, whereas Blacks were
underrepresented in those same roles. Latinos were overrepresented as both legal and
undocumented immigrants, whereas Muslims were overrepresented as terrorist suspects.
The authors recommended that future research replicate their study by analyzing the
representation of additional racial and ethnic groups, especially Asians.
Examining a theoretical approach, Hinkle (2015) used the broken window thesis,
which expects the residents of more organized communities in eastern Los Angeles (i.e.,
those with higher levels of building maintenance, policing, etc.) to have lower levels of
fear of crime than residents of more disorganized communities, which proved to be
correct. Like Callanan and Rosenberger (2015), Hinkle found higher levels of fear of
crime among females than among males. However, unlike Vieno et al. (2013), Hinkle did
not find elevated levels of fear of crime among older residents. Hinkle recommended that
future research examine all the emotional and perceptual reactions of individuals to
crime.
In connection to prior avoidance studies, Hughes, Gaines, and Pryor (2015)
examined how victimization, bullying, drug use, and media exposure were related to the
avoidance of school by 15,425 students in grades 9 to 12 in all 50 states and the District
of Columbia. The authors found that the avoidance of school was due to the examined
factors varied by racial demographic. For example, Hispanic students were more likely to
miss school than White students. The threat of being victimized by a weapon
significantly increased school avoidance among all racial groups.
46
In addition, fear of sexual assault was a significant predictor for school avoidance
for White and Black students while being hit by a partner was significant among the
Hispanic and multiracial population. For female respondents, electronic bullying, being
Hispanic, and carrying a weapon were predictors for school avoidance. For males, being
threatened with a weapon, forced sex, and property damage were predictors were school
avoidance. The authors suggested that future research compare the effects of crime in
neighborhoods to crime in schools.
Luo, Ren, and Zhao (2015) examined fear of crime by neighborhood and by home
by using a random sample of landline phones, thereby obtaining data from 2,393
participants over the age of 18 in Houston, Texas, between 2010 and 2012. The authors
found that female and senior participants felt safer in public than in their homes, whereas
the male and younger participants felt equally safe in both settings. The authors suggested
that similar research be conducted on other geographical locations.
To conclude, Ozascilar and Ziyalar (2015) identified predictors of fear of crime in
association with the impact of fear of sexual assault and perceived risk of crime among
college students in Istanbul, Turkey. A total of 723 questionnaires were administered to
undergraduate students at eight universities in Istanbul. From the data, the authors
determined that female students’ fear levels of sexual assault were significantly higher
than those of male students. Perceived risk was the strongest predictor for fear of
nonsexual crimes (burglary, robbery, theft, etc.). Also, perceived risk was higher among
women for all fear categories. The authors recommended that future research explore
these gender differences in greater depth.
47
Research by Theme
There were eight themes that emerged from the studies reviewed, which were
presented in the order of importance to my research.
Figure 1. research themes by importance.
Figure 2. research themes by importance.
General Fear of Crime
•Custers and Van den
Bulck, 2011
•Kohm, Waid-Lindberg,
Weinrath, Shelley, and
Dobbs, 2012
•Lee and Hilinski-Rosick,
2012
•Chadee and Ng Ying,
2013
•Hirtenlehner and Farrall,
2013
•Vieno, Roccato, and
Russo, 2013
•Gibson, 2014
•Callanan and
Rosenberger, 2015
•Hinkle, 2015
Avoidance of Public
Spaces
•Fosters, Giles-Corti, and
Knuiman, 2012
•Jorgensen, Ellis, and
Ruddell, 2012
•Stodolska, Shinew,
Acevedo, and Roman,
2013
•Breetzke and Pearson,
2014
•Stein, 2014
•Steinmetz and Austin,
2014
•Brands, Schwanen, and
Van Aalst, 2015
•Hughes, Gaines, and
Pryor, 2015
•Luo, Ren, and Zhao,
2015
Fear of Violent Crimes
•Cook and Fox, 2012
•Lane and Fox, 2012
•Custers and Van den
Bulck, 2013
•Goodall, Slater, and
Myers, 2013
•Hawdon, Rasanen,
Oksanen, and Vuori,
2013
•Ozascilar, 2013
•Krause, 2014
•Malinen, Willis, and
Johnston, 2014
Fear of Property Crimes
•Lane and Fox, 2011
•Lai, Zhao, and Longmire,
2012
•Vilalta, 2012
Fear of Property and
Violent Crimes
•Zhao, Lawton, and
Longmire, 2010
•Heber, 2011
•Rhineberger-Dunn,
2011
•Alper and Chappell,
2012
•Rhineberger-Dunn,
2013
•Creighton, Walker,
and Anderson, 2014
•Ozascilar and Ziyalar,
2015
Avoidance of Public
Spaces
•Callanan, 2012
•Radar, Crossman,
and Porter, 2012
•Rengifo and Bolton,
2012
•Visser, Scholte, and
Scheepers, 2013
Fear of Violent Crimes
•Henson, Reyns, and
Fisher, 2013
•Yu, 2014
Fear of Property
Crimes
•Boda and Szabo,
2011
•Lorenc, Clayton,
Neary, Whitehead,
Petticrew, Thomas,
Cummins, Sowden,
and Renton, 2012
•Nellis and Savage,
2012
•Hanslmaier, 2013
•Kappas, Greve, and
Hellmers, 2013
•Jamieson and Romer,
2014
•Dixon, 2015
•Dixon and Williams,
2015
48
Current Research Based on Research Variables
All the studies reviewed here examined the same variables as the ones that is
explored in the present study—that is, (1) amount of media exposure, and (2) level of fear
of crime. Furthermore, all the studies reviewed here examined similar demographic
variables. Nine of the studies reviewed here, like the present study, examined avoidance
of public spaces as a form of social interaction: (1) Brands et al. (2015); (2) Breetzke and
Pearson (2014); (3) Foster et al. (2012); (4) Hughes et al. (2015); (5) Jorgensen et al.
(2012); (6) Luo et al. (2015); (7) Stein (2014); (8) Steinmetz and Austin (2014); and (9)
Stodolska et al. (2013).
Rationale for Variables Selection
This study determined whether media exposure to news stories and fiction stories
about violent crime and property crime (independent variable no. 1) increased the fear of
crime (independent variable no. 2) among the residents of Los Angeles County, and
decreased their level of social interaction anxiety (dependent variable). The demographic
variables (race/ethnicity, age, and gender) were the controlling variables for this study.
Research Design
Differing Mythologies with Similar Outcomes
The Gibson, 2014 study. This was neither a quantitative nor a qualitative study,
but a meta-study—that is, a review of studies conducted by other researchers.
Nevertheless, the author confirmed the predictive value of cultivation theory, since all the
studies reviewed found that when the media in American cities featured stories about
criminality, the public’s fear of crime increased.
49
The Heber, 2011 study. This author used a qualitative method of interviewing
men and women in Sweden about the kinds of crimes of which they feared to be victims.
The author found that women were most fearful of being the victims of sex crimes,
whereas men were most fearful of being the victims of crimes related to their
occupations.
The Lorenc et al. (2012) study. In this meta-study, the authors reviewed
literature regarding the association between individuals’ fear of crime and their physical
and psychological well-being. The reviewed literature showed that fear of crime
significantly affects individuals’ health, level of anxiety, their perception of their social
well-being, and their avoidance behaviors (e.g., in unsafe neighborhoods). Surprisingly,
crime prevention interventions that raise the public’s awareness of crime can sometimes
unintentionally increase public anxiety.
Summary
With the exception of one study conducted in Hungary (Boda & Szabo, 2011),
where the population was distrustful of the media and government authorities, all the
studies reviewed here, both quantitative and qualitative, confirmed the predictive value of
cultivation theory. Specifically that increased media exposure to violence increased the
public’s fear of crime in general, and more so among females than males.
What was unknown was how the public’s fear of crime and the degree of their
exposure to media related to their level of social interaction anxiety. This gap in the
available literature was filled by the present study—at least in relation to the residents of
Los Angeles County.
50
In Chapter 3, Methodology, I described how I obtained participants for the study,
the survey instruments I used, the procedures for conducting the surveys with the
participants, the ethical issues involved with the study, and the methods used to analyze
my data.
51
Chapter 3: Methodology
Introduction
The mass media distort the public’s perception of crime rates by
disproportionately focusing on violent and property crimes. This distortion inhibits
people from fully engaging with others in public spaces, therefore creating a sense of
fear. In this chapter, I first describe how the research design derived from the problem
statement, then I review the role of the researcher. Next, I describe the population, or
participants, in this study. This description is followed by an explanation of the sampling
and setting procedures, and then by the procedures for recruitment, instrumentation, and
operationalization.
I then present operational definitions of the terms used in the study. This is
followed by a description of how I analyzed the data. After a discussion of the threats to
the validity of the study, the chapter concludes with a review of the relevant ethical
procedures.
Research Design and Rationale
How the Research Design Derived from the Research Questions
When developing the research questions for this study, I considered the question
of “why it is important” and “how such results will help law enforcement agencies.”
Therefore, a goal of this study was to bring awareness to poor or weak social
relationships due to citizens’ perceptions of property and/or violent crimes presented
throughout the Television and Internet. As a result of not addressing this concern, there
may be a decrease in the level of trust and unity among different racial groups.
52
The first research question of this study was: How does the Los Angeles County
public’s amount of media exposure and level of fear of crime impact social interaction
anxiety in public spaces?
The second research question of this study was: In Los Angeles County, what is
the relationship among the public’s amount of media exposure, level of fear of crime, and
social interaction anxiety after controlling for demographics (race/ethnicity, age, and
gender)?
This study used the following three questionnaires, which employ a 5-point Likert
scale, to answer both research questions.
The Media and Technology Scale, developed by Rosen et al. (2013), allowed me
to identify the amount of exposure by participants to TV and the Internet—thereby
collecting data in response to media exposure aspect of each research question.
The Harmonisation Office of National Statistics’ (2015) Crime and Fear of Crime
Scale enables me to identify participants’ level of fear of crime—thereby collecting data
in response to the fear of crime aspect of each research question.
Mattick and Clarke’s (1998) Social Interaction Anxiety Scale allowed me to
identify participants’ level of social interaction anxiety—thereby collecting data in
response to the social interaction anxiety aspect of each research question.
By using the Demographic Questionnaire, I was able to correlate the data
collected by the other three surveys with demographic variables—thereby collecting data
in response to the second research question.
53
Restatement of the Study’s Variables
The study’s first independent variable was the amount of the participants’
exposure to TV and the Internet.
The study’s second independent variable was the participants’ fear of crime.
The study’s dependent variable was the participants’ level of social interaction
anxiety.
The study’s moderating variables were the participants’ demographic
characteristics.
Time and Resource Constraints Consistent with the Design Choice
Since the data for the three surveys and the Demographic Questionnaire were
collected by SurveyMonkey, the research design did not face any time constraint when
reaching the target goal of sixty days or 300+ surveys. The research design did not face
any resource constraint particularly that of not having enough participants that fit the
study criterions.
How the Design Choice Was Consistent with the Research Designs Needed to
Advance Knowledge in the Discipline
The three survey instruments and the Demographic Questionnaire used in this
study obtained quantitative data about the first and second independent variables, the one
dependent variable, and the three demographic factors of gender, age, and race/ethnicity.
54
The Role of the Researcher
The Researcher as Observer
I only analyzed data collected online from anonymous participants provided by
SurveyMonkey.
The Role of the Researcher During Data Collection
I had no role during the collection of data other than to pay $23 per month to
SurveyMonkey for its data-collection services.
The Anonymity of the Relationship Between the Researcher and the Participants
I had no contact whatsoever with the participants, so there was no personal or
professional relationship between myself and the participants that could compromise the
objectivity of the study. In other words, no biases or power relationships needed to be
managed for the study to remain objective.
Other Ethical Issues
There were no ethical issues since the identities of the participants were unknown
to me. I did not, for example, conduct a study in my work environment, so there were no
questions of power differentials or conflicts of interest between myself and the
participants. The only issue that might have conceivably caused ethical questions is the
fact that SurveyMonkey attracts participants to respond to surveys by offering to make a
small contribution to their favorite charity and by giving them opportunities to win
sweepstakes. But this was unknown since there was no personal information collected
from participants. However, these rewards are strictly regulated by the laws of the state of
California, and, in any case, I had nothing to do with their distribution.
55
The Study Population
In order to be surveyed for this study, individuals needed to be 18 years of age or
older, reside in Los Angeles County, and have regularly watched the local news on
television and/or social media during the two weeks prior to their participation in the
study. So that the target population would be reached, I opened the questionnaire to all
SurveyMonkey’s panelist who met the survey criteria for 60 days or until 300+ surveys
was reached, whichever came first.
Sampling and Setting Procedures
The Sampling Strategy
SurveyMonkey used several different sampling strategies, from which I selected
voluntary response sampling. A probability sampling method was not feasible for the
study because not all participants had an equal chance of being included in the sample.
Therefore, the use of a nonprobability sampling design such as voluntary response
sampling allowed me to collect data from those who met the criteria of the study and
were willing to complete the survey. The design was also a form of case selection in
which the selection process was purposive rather than based on randomization or
probability sampling (see Jupp, 2006).
A potential negative was that nonprobability sampling leads to less trustworthy
responses when compared to the random sampling method; the usefulness of the data
collected depends on the purpose of the study, the criteria for selecting unit samples, and
how well each sample represents the population of interest. Voluntary response sampling
provided an oversampling of those with strong opinions while undersampling those who
56
care less for the survey topic. I needed to make sure participants fully understood the
purpose of the study through the informed consent form, word all questions in a manner
that did not lead participants to answer in a particular way, and verify through the use of
demographic and screening questions that participants represented the population of
interest.
How the Sample Was Drawn
I used SurveyMonkey technology as much as possible, therefore two options were
considered when reaching the target population. Option one, I could have selected
participants from SurveyMonkey’s panel based on pre-profiled targeting options such as
basic demographic questions and/or behavioral questions. Option two, I asked specific
screening questions and disqualify participants who don’t meet the criteria. Thus, I used a
combination of options one and two in which the basic demographic and behavioral
questions designed by SurveyMonkey and specific screening questions were used.
The specific screening questions for this research used only SurveyMonkey
panelists who were residents of Los Angeles County over the age of 18, lived in the
county for at least 90 days, and watched crime-related news stories on TV or the Internet
two weeks prior to participating in the study.
Inclusion and Exclusion Criteria
The participants were at least 18 years old, resided in Los Angeles County, and
watched crime stories on TV or the Internet two weeks prior to the study. Anyone who
did not meet the criteria were excluded from the study.
57
Power Analysis to Determine the Sample Size
I used the G-Power calculator developed by Faul, Erdfelder, Butcher, and
Language (2009) to compute the statistical power analysis needed for her study. For this
study, I used the F-test family with a statistical one-way ANOVA to determine the
sample size. This sample also yielded the F-distribution (degrees of freedom), non-
centrality parameters (degree to which the null hypothesis is false), and sample size. I
manually had to input the effect size, error of probability, and the power of test
significance. I used a 95% confidence level (the probability in which the sample
accurately represent the target population) and +/- 5% margin of error (the range in which
the results of the survey will fall between), therefore I was able to obtain an actual
representation of the whole population, accurately concluding that the target population
had been reached. I made sure that the total population had been reached prior to stopping
data collection so that a full sample size was acquired. This included evaluation of each
survey to make sure each question was answered as well as elimination of those with
inadequate responses.
Procedures for Recruitment, Participation, and Data Collection
Recruitment Procedures
Every month, SurveyMonkey recruits millions of participants, offering them a
choice of contributing 50 cents to a charity of their choice for every survey they answer
or allowing the participants to enter a sweepstakes for prizes. SurveyMonkey avoids
flooding participants with questionnaires, to ensure a high quality of recorded data.
SurveyMonkey also took regular self-profiling surveys to help keep its demographic
58
information updated. SurveyMonkey worked with survey panel companies to ensure that
the survey takers are willing participants who are vetted for quality.
Demographic Information to Be Collected
The demographic information collected included participants race/ethnicity, age,
and gender.
Informed Consent
Prior to taking the survey, the participants read and acknowledge their consent by
returning a completed survey, which assured them that their identity will remain
anonymous and that no physical or psychological harm came to them as a result of taking
the survey.
Collection of Data
For this study, SurveyMonkey collected responses anonymously. Therefore, no
email reminder were sent out to those who partially complete or drop-out of the survey.
Participants were selected based on identifying that they live in the state of California and
reside within Los Angeles County. After establishing residency, Los Angeles County
residents moved on to the study’s two screening questions.
Debriefing Procedures
After completing the survey, participants were automatically taken to a
SurveyMonkey web page, which thanked them for taking the survey and provided
information about SurveyMonkey.
59
Follow-up Procedures
If participants wanted to receive a copy of the study’s results, they could have
provided an e-mail address to which I would send them a copy of the results upon
publication.
Instrumentation and Operationalization of Constructs
The Developers of the Survey Instruments
The survey one instrument. Rosen et al. (2013). The media and technology
usage scale. Computers and Human Behavior, 29, 2501-11.
This survey measured the study’s first independent variable, the amount of the
participants’ exposure to TV and the Internet.
The survey two instrument. Harmonisation Office of National Statistics (2015).
Crime and fear of crime scale. Titchfield, England: Author.
This survey measured the study’s second independent variable, the participants’
fear of crime.
The survey three instrument. Mattick and Clarke (1998). Social interaction
anxiety scale. Behavior Research and Therapy, 36, 455-470.
This survey measured the study’s dependent variable, the participants’ level of
social interaction anxiety.
The Appropriateness of the Survey Instruments to the Present Study
Collectively, the three surveys measured all the components of the present study’s
research questions.
60
Permission Letters from the Developers of the Survey Instruments
I received a permission letter from the authors of Survey 1. Survey 2 contains a
privacy disclosure clause that permits researchers to use the survey for legitimate
academic purposes. For permission to use Survey 3, I obtained permission from the
original author who published in the, Behavior Research and Therapy. The two letters of
permission and a copy of the disclosure clause were included in the Appendix to this
study.
The Reliability and Validity of the Survey Instruments
The survey one instrument. Rosen et al. (2013) Media and Technology Usage
Scale. All 15 subscales showed strong reliability and validity. The strongest subscales
were Internet usage (Cronbach alpha .91) and television usage (Cronbach Alpha .61),
both of which will be used by the present researcher.
The survey two instrument. The Harmonisation Office of National Statistics’
(2015) Crime and Fear of Crime Scale. The authors used four subgroups (feeling safe;
worries about crime; crime rates in areas; problems in areas) to determine the reliability
and validity of their scale. Their data were in compliance with the Statistics and
Registration Act of 2007.
The survey three instrument. Mattick and Clarke’s (1998) Social Interaction
Anxiety Scale. The authors proved their instrument to be reliable when the results yielded
high internal consistency for all its 20 questions. As for validity, the authors stated that
the discriminant validity of almost all the items is “sufficiently high to allow clinicians to
confidently interpret individual items” (p. 467).
61
The Population the Instruments Survey to Establish Validity and Reliability
The survey one instrument. Rosen et al. (2013) Media and Technology Usage
Scale. In this study, the authors surveyed two groups: undergraduate students and
community members, both in the Los Angeles area.
The survey two instrument. The Harmonisation Office of National Statistics’
(2015) Crime and Fear of Crime Scale. In this study, the authors surveyed individuals
throughout the United Kingdom who were at least 16 years old.
The survey three instrument. Mattick and Clarke’s (1998) Social Interaction
Anxiety Scale. In this study, the authors surveyed 485 undergraduate students enrolled in
introductory psychology courses at the University of New South Wales in Sydney,
Australia, and 315 non-student friends of the students.
Operationalization for Variables
Definitions of the Variables Used
The first independent variable: In this study, the first independent variable was
the Los Angeles County public’s amount of TV and Internet exposure.
The second independent variable: In this study, the second independent variable
was the Los Angeles County public’s level of fear of crime.
The dependent variable: In this study, the dependent variable was the level of
social interaction anxiety among the residents of Los Angeles County.
The mediating variables: In this study, the mediating variables included the
demographic characteristics of race/ethnicity, age, and gender.
62
How the Variables Were Measured
All three parts of the survey contain 5-point Likert scales, so the response of each
participant to each survey were totaled for analyses.
How the Variables/Scale Scores Were Calculated, What the Scores Represent, and
an Example Item
For example, the media exposure section of Survey 1 has six items, for which the
participant responded with an answer of 0 to 4, with 0 standing for no exposure, and 4
standing for exposure all day. Thus, if a participant answers with 4 to all six items, his or
her total score for that section was 24, and his or her average score for that section was
4.0.
The Data Analysis Plan
Software Used for Analyses in This Study
I exported the data from SurveyMonkey into IBM SPSS software when I analyzed
the data.
Explanation of Appropriate Data Cleaning and Screening Procedures
I used filters provided by SurveyMonkey to eliminate participants who did not
meet the criteria for participation established by myself. Other filters were used to screen
out the responses of participants who only answer a portion of the questions, who speed
through the survey, who put the same response for every question, who provided
unrealistic answers, and who gave inconsistent responses.
63
Restatement of the Research Questions and Hypotheses
The research question one. How does the Los Angeles County public’s amount
of media exposure and level of fear of crime impact social interaction anxiety?
The alternative hypothesis one The public’s amount of media exposure and
level of fear of crime in Los Angeles County have a high social impact on individuals’
anxiety to interact socially.
The null hypothesis one. The public’s amount of media exposure and level of
fear of crime in Los Angeles County have no social impact on individuals’ anxiety to
interact socially.
The research question two. In Los Angeles County, what is the relationship
among the public’s amount of media exposure, level of fear of crime, and social
interaction anxiety after controlling for demographics (race/ethnicity, age, and gender)?
The alternative hypothesis two. There is a relationship between the public’s
amount of media exposure, level of fear of crime, and social interaction anxiety after
controlling for demographics.
The null hypothesis two. There is no relationship between the public’s amount of
media exposure, level of fear of crime, and social interaction anxiety after controlling for
demographics.
The Statistical Tests That Were Used to Test the Hypotheses.
A sequential (hierarchical) multiple regression analysis was used to test the two
hypotheses in this study. For each hypothesis, the dependent variable was the measure of
64
social interaction anxiety. For the second hypothesis, the demographic variables were
included to control for any confounding that existed.
The model 1 (for Hypothesis 1):
The model 2 (for Hypothesis 2):
Statistical Methods Used to Analyze the Data
The correlational statistic used. Preliminarily, a correlational analysis was used
to investigate the relationships among all variables in the study. Specifically, bivariate
correlations investigated the dependent variable paired with each independent variables
and each of the demographic variables. Also, bivariate correlations determined each
independent variable paired with each demographic variable.
The regression statistic used. A sequential (hierarchical) multiple regression
model was used to investigate the relationship between the two independent variables
(media usage and fear of crime) and the dependent variable (social interaction anxiety).
The analysis of variance statistic used. Analysis of variance or an alternative
nonparametric method was used to determine the significance difference in the dependent
variable across levels of the demographic variables. Post-hoc multiple comparison test
with a Bonferroni correction was not used.
12
ˆ(MediaUsage) (FearofCrime)Y a B B= + +
1 2 3 4 5
67
ˆ(MediaUsage) (FearofCrime) (Race) (Education) (Income)
(Gender) (Age)
Y a B B B B B
BB
= + + + + +
++
65
The Rationale for Inclusion of Potential Covariates and/or Confounding Variables.
Since this study obtained a representative sample of Los Angeles County
residents; media usage, fear of crime, and social interaction anxiety varied among the
demographic variables. Therefore, the inclusion of demographics as the covariate
(mediating) variable was useful in controlling for a representative sample of the whole
population.
How the Results Were Interpreted
The results of this study were interpreted based upon standard statistical
guidelines.
Threats to Validity
Threats to External Validity and How They Were Addressed
Threats to the external validity of this study would come from the researcher
applying the results of her investigation to populations outside of Los Angeles County.
Since I did not make such applications, there were no threats to the external validity of
my study.
Threats to Internal Validity and How They Were Addressed
Threats to the internal validity of this study would arise from the researcher
drawing incorrect inferences from her data. I was able to eliminate this possibility, or at
least heavily mitigate it, by avoiding any subjectivity in relation to my data and relied
exclusively on statistical analyses.
66
Threats to Construct or Statistical Conclusion Validity
Threats to the construct validity of this study did not occur, specifically to that of
not adequately defining the terms and measurements used in my study. I provided an
extensive list of operational definitions of all the major terms used in my study to address
this threat.
Threats to the statistical conclusion validity of my study did not arise, specifically
to the exporting of data from SurveyMonkey. I was able to avoid this threat by going
over my exporting techniques, repeatedly and thoroughly.
Ethical Procedures
All Agreements to Gain Access to Participants or Data
I created an account with SurveyMonkey, which included an agreement (see
Appendix F) that gave me access to all the data that SurveyMonkey collects from the
participants. In addition, SurveyMonkey required all participants to read and
acknowledge their consent, by returning a complete survey. This form detailed to
participant how all information was kept anonymous, therefore advising respondents that
no identifiable information was collected.
The Treatment of Human Participants Related to Institutional Permissions
The Institutional Review Board (IRB) at Walden University received copies of
the agreement between the researcher and SurveyMonkey, as well as the consent form
between the participants and SurveyMonkey. Since I had no direct contact with the
participants, the IRB had no ethical concerns regarding my treatment of participants.
67
The Ethical Concerns Related to Recruitment Materials and Processes
Since SurveyMonkey recruited all the participants, there were no ethical concerns
related to my recruitment materials or processes.
The Ethical Concerns Related to Data Collection
There were several participants who decided to withdraw from the survey, but
were not penalized, so there were no ethical concerns related to the collection of data.
The Treatment of Data/Anonymity and Confidentiality
The identities of all the participants were anonymous, even to myself, so there
were no issues related to confidentiality. Nevertheless, I kept the data in a file that was
password-protected, and I will destroy all the data five years after the dissertation is
concluded.
Other Ethical Issues
There were no other ethical issues related to this dissertation project.
Summary
This chapter began with a description of how the research design derived from the
problem statement. Next, I reviewed the role of myself in this study. I then described the
population of the study, I followed this explanation with the sampling and setting
procedures, and then with a description of the procedures for recruitment,
instrumentation, and operationalization.
Next, I presented operational definitions of the terms I used in the study. I then
described how I analyzed the data. After discussing potential threats to the validity of the
study, I concluded the chapter by reviewing the relevant ethical procedures.
68
Chapter 4: Results
Introduction
Statement of Purpose
The purpose of this study was to examine the occurrence of increased social
interaction anxiety in public spaces in Los Angeles through the lens of cultivation theory
(Gerbner, 1969), to determine if the amount of media exposure to crime and the level of
fear of crime contributes to this behavior. Therefore, the point of this study was to
evaluate the relationship between societal consumption of media messages, level of fear
of crime, and social interaction anxiety. To accomplish this task, the study used an online
research site (SurveyMonkey) to obtain responses from residents who are 18 years of age
or older and reside in Los Angeles county.
The research question one. How did the Los Angeles County public’s amount of
media exposure and level of fear of crime impact social interaction anxiety?
The alternative hypothesis one. The public’s amount of media exposure and
level of fear of crime in Los Angeles County had a high social impact on individuals’
anxiety to interact socially.
The null hypothesis one. The public’s amount of media exposure and level of
fear of crime in Los Angeles County had no social impact on individuals’ anxiety to
interact socially.
The research question two. In Los Angeles County, what was the relationship
among the public’s amount of media exposure, level of fear of crime, and social
interaction anxiety after controlling for demographics (race/ethnicity, age, and gender)?
69
The alternative hypothesis two. There was a relationship between the public’s
amount of media exposure, level of fear of crime, and social interaction anxiety after
controlling for demographics.
The null hypothesis two. There was no relationship between the public’s amount
of media exposure, level of fear of crime, and social interaction anxiety after controlling
for demographics.
How the Research Method Unfolded
The process for this study first involved identifying what exactly I wanted to
study. Once this was determined, I was able to identify a target environment based on an
understanding of cultivation theory (Gerbner, 1969), which suggests that as the amount of
time spent consuming crime stories throughout media sources increases, so will an
individual’s fear of crime. What had not been examined through the cultivation theory
lens was whether the media and fear of crime have any connection to social interaction
anxiety among a specific population.
After moving to Los Angeles County in 2014 and witnessing how the media
displays crime stories occurring throughout the county, I found that studying Los Angeles
County would be a suitable choice for examining the relationship between media
consumption, fear of crime, and social interaction anxiety. This study collected responses
potentially from more than 10 million residents living in Los Angeles County. These
residents had to identify as having watched crime stories on TV and/or Internet sources at
least two weeks before the study and be over the age of 18.
70
In this study, there were two independent variables, which included the Los
Angeles County public’s amount of media exposure and the public’s level of fear of
crime. The dependent variable included the level of social interaction anxiety among the
residents of Los Angeles County. This study included demographic characteristics of
race/ethnicity, age, and gender as the mediating variables. A correlational analysis was
used to investigate the relationship between all variables used in the study. Specifically,
bivariate correlations were used to investigate the dependent variable paired with each
independent variable and each of the demographic variables. Also, a bivariate correlation
was used to determine each independent variable paired with each demographic variable.
When examining each hypothesis, a sequential (hierarchical) multiple regression
was used. By using a sequential (hierarchical) multiple regression model, the research
was able to account for any statistical amount of variance among the dependent variable
after accounting for all variables used in the study. A sequential (hierarchical) multiple
regression was also used to determine the relationship between the two independent
variables and the dependent variables. An analysis of variance, or alternative
nonparametric method, was used to determine the significant difference in the dependent
variable across levels of the demographic variables. No adjustments were needed in this
study, as all statistical tests were run appropriately. I am not sure what this means?
Organization of Chapter
A summary of the data collection process. The online surveying site
SurveyMonkey was used to collect responses anonymously from participants.
Participants were selected based on their residency in the state of California. Upon
71
establishing residency, California residents must have answered “Yes” to the two
screening questions: “Do you reside in Los Angeles County?” and “Have you watched
crime stories on TV or the Internet within two weeks prior to today?” Once participants
had answered “Yes” to both screening questions, they were directed to the “Informed
Consent Form” and then proceeded to the survey. Participants were asked a series of
questions directly related to media consumption, fear of crime, and social interaction
anxiety levels. To conclude the survey, participants were asked three demographic
questions pertaining to their gender, age, and race/ethnicity.
A summary of the results. To address the first research question, the results
indicated there to be a significant relationship between the independent variables of
media consumption and fear of crime, and the dependent variable of social interaction
anxiety. Therefore, the public’s amount of media exposure and level of fear of crime in
Los Angeles County has an impact on an individual’s anxiety to interact socially.
Addressing, the second research question, a series of hierarchical multiple regression
analyses indicated a relationship between the public’s amount of media exposure, level of
fear of crime, and social interaction anxiety after controlling for demographics.
Therefore, the results of this study concluded that residents of Los Angeles County
generally had increased levels of social interaction anxiety upon exposure to crime stories
published throughout the media and fear of crime levels.
A summary of presentation of the results. There is a series of 12 tables and
figures (scatter plots) used to present the results of this study. Each figure or table
includes a description specifying how to read and interpret the data presented and the
72
relation to the research variables and/or question. Table 1 illustrates the correlational
relationship between the independent and dependent variables. Table 2 interpreted the
results for the ANOVA test in which one can view how the dependent variable differed
across all demographic variables. Table 3 illustrated the descriptive statistics for
continuous variables. Table 4 presented the findings for the frequency distributions for
categorical variables. Tables 5 and 6 displayed all results from two different multiple
regression tests. Tables 7 and 8 then presented the findings from the hypothesis
coefficients analyses. Figures 1 and 2, displayed the authors of each literature review in
the order of importance. Figure 3 then featured a scatter plot illustrating the relationship
between the independent variable of fear of crime and the dependent variable of social
interaction anxiety. Figure 4 scatter plot detailed the relationship between the
independent variable of media exposure and the dependent variable of social interaction
anxiety. Figure 5 presented the results from the normal probability analysis, which
indicates the normality assumption to be satisfied. Finally, Figure 6 revealed the result
from the testing of the standardized residuals against the standardized predicted values.
A summary of answers to the research questions. Research question one
determined the impact of the amount of media exposure and levels of fear of crime had
on Los Angeles County residents’ social interaction anxiety levels. A multiple regression
analysis was used to investigate this relationship. Results indicated, that the overall model
predicted social interaction anxiety. In general, both media exposure and levels of fear of
crime contributed to Los Angeles County residents’ social interaction anxiety levels,
although only the fear of crime variable contributed significantly to participants’ levels of
73
social interaction anxiety.
The research question two determined the relationship among Los Angeles
County residents’ amount of media exposure, levels of fear of crime, and social
interaction anxiety after controlling for demographics (race/ethnicity, age, and gender)?
A hierarchical multiple regression analysis with a two-step process investigated the
relationship among these variables. The first process examined the three demographic
variables against the dependent variable of social interaction anxiety. The results yielded
a statistically significant relationship to social interaction but only accounted for 7% of
the total variance. Thus, the second step included all demographic questions and the two
independent variables; yielding a statistically significant relationship explaining 13% of
the total variance in the dependent variable of social interaction anxiety. Therefore, it was
stated that there was a relationship between Los Angeles County residents’ amount of
media exposure, level of fear of crime, and social interaction anxiety after controlling for
all demographic variables.
Data Collection
Data Collection- Time Frame
The time frame for this study was sixty days or 300+ surveys. After launching the
survey on SurveyMonkey it took roughly thirty days to obtain 590 responses. Of these
responses, 178 respondents answered yes to both screening questions. After reviewing all
responses, there were a total of 150 respondents who completed the entire survey. Thus,
the responses from these 150 respondents were included in the data analyses. The
response rate consisted of the “percentage of survey respondents from a sample who
74
respond to a questionnaire” (see O’Sullivan, Rassel, and Berner, 2008). O’Sullivan, et al.
(2008) provided the following equation to calculate the response rate (p. 174):
𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑜𝑚𝑝𝑙𝑒𝑡𝑒𝑑 𝑠𝑢𝑟𝑣𝑒𝑦𝑠 ÷ 𝑡𝑜𝑡𝑎𝑙 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑒𝑠𝑝𝑜𝑛𝑑𝑒𝑛𝑡𝑠 × 100
= 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒 𝑟𝑎𝑡𝑒
150 ÷590 ×100 =25.42% 𝑟𝑒𝑠𝑝𝑜𝑛𝑠𝑒𝑟𝑎𝑡𝑒
The base number of 100 was based on crime rates usually being evaluated per 100
of the population (see O’Sullivan et al., 2008). Therefore, the response rate was 25.42%.
Data Collection- Discrepancies
There was one discrepancy that occurred during data collection that was not
presented in Chapter 3. When evaluating the demographic variables, this study
considered gender, race/ethnicity, and age. The age category included the following
groups: 18-24, 25-34, 35-44, 45-64, 65-74, 75+, and a decline to state option. In
SurveyMonkey, the age category was separated by the following: 18-24, 25-34, 45-54,
55-64, 65+, and a decline to state option. Secondly, this study’s race/ethnicity categories
included: African American/Black, Caucasian/White, Native American, Asian/Asian
American, Hispanic/Latino, Other, and decline to state. On the other hand,
SurveyMonkey used the following categories: White or Caucasian, Black or African
American, Hispanic or Latino, Asian or Asian American, American Indian or Alaska
Native, Native Hawaiian or other Pacific Islander, another race, and decline to state.
Although these categories differentiate slightly, no data was effected as SurveyMonkey
automatically collected the information and their categories were used in the data
75
analyses. This study did benefit from extended race/ethnicity categories as there were
several respondents who selected race options that were not originally presented.
Baseline Discrepancies and Demographic Characteristics
There were a total of 590 responses to the survey. Of the 590 responses, there
were 178 participants who provided “Yes” to both screening questions. Of these 178
qualifying respondents, a total of 150 respondents provided responses to all items of the
survey. Survey participants included 96 females (64%), 53 males (35.3%), and 1 who
declined to state their gender (0.7%). The age range for the study sample was between 18
and 65+ (99.3%), with 14% between the ages of 18 and 24, 30% between the ages of 25-
34, 19.3% between the ages of 35-44, 17.3% between the ages of 45-54, 8.7% between
the ages of 55-64, 10% over the age of 65, and 0.7% who decline to state their age. The
racial/ethnicity make-up of the study included 62 Caucasians at 41.3%, 17 African
Americans at 11.3%, 39 Hispanics at 26%, 17 Asians at 11.3%, 1 American Indian at
0.7%, 2 Native Hawaiians at 1.3%, 8 respondents identifying as another race representing
5.3%, and 4 respondents who declined to state their race equaling 2.7% of the total
sample. Therefore, the analyses for the study was based on a sample size of 150.
Representation of Sample Population
O’Sullivan et al. (2008), explained the incidence rate as the common rate measure
referring to the number of people who get a disease over a specific period of time, usually
a year (pg. 340). These authors suggested this definition could be used for other situations
such as being a victim of crime or having an accident. Considering, the interest of this
study, the incidence rate measured total population data collected by the Census Bureau
76
in 2018. Therefore, the total number of Los Angeles County residents (10.11 million)
were expected to qualify for the survey during the allotted time period was divided by to
the total number of residents in the state of California (39.56 million); as seen in the
equation:
𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝐿𝑜𝑠 𝐴𝑛𝑔𝑙𝑒𝑠 𝐶𝑜𝑢𝑛𝑡𝑦 ÷ 𝑡𝑜𝑡𝑎𝑙 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝐶𝑎𝑙𝑖𝑓𝑜𝑟𝑛𝑖𝑎
= 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑖𝑛𝑐𝑖𝑑𝑒𝑛𝑐𝑒 𝑟𝑎𝑡𝑒
10.11 ÷39.56 =25.56% 𝑒𝑠𝑡𝑖𝑚𝑎𝑡𝑒𝑑 𝑖𝑛𝑐𝑖𝑑𝑒𝑛𝑐𝑒 𝑟𝑎𝑡𝑒
Thus, the estimated incidence rate was calculated at 25.56%. Upon data
collection, results yielded a 38.49% incidence rate which is 12.93% higher than the
expected incidence rate calculated prior to the collection of data.
Results of Basic Univariate Analysis
Correlations among all the variables included in the analysis were calculated and
were provided in (Chapter 4, Table 1). I used the following codes for each categorical
variables: Gender (1 = Male, 2 = Female, 3 = Decline), Race (1 = White or Caucasian, 2
= Black or African American, 3 = Hispanic or Latino, 4 = Asian or Asian American, 5 =
American Indian or Alaska Native, 6 = Native Hawaiian or other Pacific Islander, 7 =
Another Race, 8 = Decline), and Age (1 = <18, 2 = 18-29, 3 = 30-44, 4 = 45-60, 5 =
>60). It was seen in the table that both independent variables (Media Exposure and Fear
of Crime) were statistically significantly correlated to the dependent variable (Social
Interaction Anxiety), although the Media Exposure variable was only weakly correlated
with Social Interaction Anxiety. Of the demographic variables, the only one that was
strongly correlated with the dependent variable was age. The hierarchical multiple
77
regression analysis controlled for the demographic variables to determine if the two
independent variables contribute significantly to the variation in Social Interaction
Anxiety over and above the demographic variables.
Table 1
Correlations (N = 150)
Variables
SIA
ME
FC
G
A
R
Social interaction anxiety
(SIA)
1
Media exposure (ME)
0.153*
1
Fear of crime (FC)
−0.297***
−0.161*
1
Gender (G)
0.055
−0.173*
−
1
Age (A)
−
−
0.251**
0.049
1
Race (R)
−
−
−
0.033
0.006
1
Note. Statistical significance: *p < 0.05, **p < 0.01, ***p , 0.001
A One-way Analysis of Variance (ANOVA) was performed to determine if
values of the dependent variable (Social Interaction Anxiety) differed significantly
among the levels of the demographic variables (Gender, Age, and Race/Ethnicity). The
results were provided in (Chapter 4, Table 2) and shows that there was no significant
difference in Social Interaction Anxiety among the levels of gender, age group, or race.
The F statistic used in ANOVA was the ratio of the variance between subjects to the
variance expected due to chance (error) and was used as a single value to describe the
differences between independent samples. The value of the ratio was used to determine
whether differences were large enough to be attributed to a factor effect, or if they were
due simply to chance effects. An F value near 1.00 indicated that the differences between
the groups were roughly the same as expected due to chance. An F value substantially
greater than 1.00 indicated that at least one sample was significantly different from the
78
others. The p-value for the F statistic indicated a value of F as substantially greater than
1.00 (p < 0.05) or not (p > 0.05).
Table 2
ANOVA Results
Social interaction
anxiety
Variable
Level
Mean
Standard
Deviation
F
p-
value
Gender
Male
25.06
15.52
0.771
0.464
Female
27.66
17.77
Decline to state
12.00
-
Age
18-24
33.14
15.24
1.857
0.092
25-34
30.11
15.80
35-44
25.03
15.93
45-54
25.81
20.41
55-64
18.77
12.21
65+
20.73
17.68
Decline to state
12.00
-
Race/ethnicity
White or Caucasian
26.48
18.20
1.146
0.338
Black or African American
23.35
14.67
Hispanic or Latino
28.67
17.70
Asian or Asian American
25.65
14.30
American Indian or Alaska
Native
56.00
-
Native Hawaiian or other
Pacific Islander
37.00
14.14
Another race
27.25
15.47
Decline to state
12.00
-
79
Results
Descriptive Statistics
There were 178 respondents who answered “Yes” to both screening questions, are
you a resident of Los Angeles County and did you watched crime-related news stories
two weeks prior to completing the survey. Of these eligible respondents, 150 provided
responses to all items on the survey. Thus, the following analyses was based on a sample
size of 150. This was ample respondents based on two suggestions for sample size
determinations. First, Stevens (1996) suggested having 15 participants per predictor
(independent) variable. Since this study has two predictor variables, the suggestion was to
have at least 30 participants. Second, Tabachnick and Fidell (2007) suggested a sample
size at least as large as 50 plus eight times the number of independent variables. This
suggested a minimum sample size of 66. The obtained sample size of 150 was well above
both of these suggested minimums.
It was important to note that the dependent variable (Social Interaction Anxiety)
and the two independent variables (Fear of Crime and Media Exposure) were calculated
based on responses to numerous items. The dependent variable (Social Interaction
Anxiety) was calculated by summing the responses to the 20 items on the Social
Interaction Anxiety Scale. Each item allowed a response on a scale of 0 (not at all
anxious) to 4 (extremely anxious). Three items were reverse coded so that higher scores
indicated more anxiety and lower scores indicated less anxiety for all 20 items on the
survey. Thus, the sum of the responses for all 20 items ranged from 0 (not at all anxious)
80
to 80 (extremely anxious). This sum was used as the dependent variable for each
respondent.
The two independent variables were calculated similarly. The Fear of Crime
independent variable was calculated by summing the responses to the 19 items on the
Crime and Fear of Crime Scale. Each item allowed a response on a scale of 0
(unsafe/fearful/worried/crime problems) to 4 (safe/not at all fearful or worried/no crime
problems). Three items were reverse coded so that higher scores indicated less fear of
crime and lower scores indicated more fear of crime for all 19 items on the survey. Thus,
the sum of the responses for all 19 items ranged from 0 (very fearful of crime) to 76 (no
fear of crime). This sum was used as the Fear of Crime independent variable for each
respondent.
The Media Exposure independent variable was calculated by summing the
responses to the 6 items on the Media and Technology Usage Scale. Each item allowed a
response on a scale of 0 (never) to 4 (all day). Higher scores indicated more exposure to
media and lower scores indicated less exposure to media for all 6 items on the survey.
Thus, the sum of the responses for all 6 items ranged from 0 (no exposure to media) to 24
(all day exposure to media). This sum was used as the Media Exposure independent
variable for each respondent.
The following tables provided the descriptive and demographic characteristics for
a sample of 150 participants.
81
Table 3
Descriptive Statistics for Continuous Variables
Variables
Social interaction anxiety
Fear of crime
Media exposure
Mean
26.79
34.23
14.12
Standard deviation
16.98
15.08
4.55
Range
0 – 68
4 – 71
0 – 24
Table 4
Frequency Distributions for Categorical Variables
Variables
Level
Frequency
Percent
Gender
Male
53
35.3
Female
96
64.0
Decline to state
1
0.7
Age
18-24
21
14.0
25-34
45
30.0
35-44
29
19.3
45-54
26
17.3
55-64
13
8.7
65+
15
10.0
Decline to state
1
0.7
Race/ethnicity
White or Caucasian
62
41.3
Black or African American
17
11.3
Hispanic or Latino
39
26.0
Asian or Asian American
17
11.3
American Indian or Alaska Native
1
0.7
Native Hawaiian or other Pacific Islander
2
1.3
Another race
8
5.3
Decline to state
4
2.7
82
Statistical Assumptions
The statistical assumptions necessary for hierarchical multiple regression were
linearity, normality, and homoscedasticity. Scatter plots (Chapter 4, Figures 3 and 4)
were for each independent variable against the dependent variable show an oblong oval
shape suggesting a linear trend. Thus, the linearity assumption was satisfied.
Figure 3. scatter plot of fear of crime against social interaction anxiety.
83
Figure 4. scatter plot of media exposure against social interaction anxiety.
A normal probability plot was used to investigate the normality assumption. In
(Chapter 4, Figure 5) the normal probability plot provided results in which it showed a
reasonably straight diagonal line from bottom left to top right, indicating the normality
assumption was satisfied.
84
Figure 5. normal probability plot.
A plot of the standardized residuals against the standardized predicted values was
used to investigate the assumption of homoscedasticity. This plot was given in (Chapter
4, Figure 6) a rectangular scatter showed that most of the points concentrated in the
center. This indicated that the homoscedasticity assumption was satisfied. This plot was
also used to demonstrate that the sample had no outliers in that all points were less than
three units from the center.
85
Figure 6. plot of standardized residual against the standardized predicted values.
In addition to the assumptions mentioned, the independent variables were not
highly correlated with each other. Highly correlated independent variables presents a
problem known as multicollinearity. Results from the hierarchical multiple regression
analysis showed a tolerance value for the independent variables and the control variables
ranging between 0.850 and 0.977 (none less than 0.10) and the variance inflation factors
ranging between 1.023 and 1.176 (none above 10). This indicated that multicollinearity is
likely not a problem in this sample.
Findings- Exact Statistics and Probability Values
Addressing, the first research question, a multiple regression was used to
investigate the relationship between Social Interaction Anxiety and the two independent
variables, Media Exposure and Fear of Crime. The results indicated that the overall
model was statistically significant, R2 = 0.099, F(2, 148) = 8.164, p < 0.001. A summary
86
of the regression coefficients was provided in (Chapter 4, Table 5) and but also indicated
that only the fear of crime variable significantly contributed to the model.
Table 5
Coefficients for Model Variables
B
t
p
Bivariate r
Partial r
Media
Exposure
0.396
0.107
1.352
0.179
0.153
0.110
Fear of crime
−0.316
−0.280
−3.536
0.001
−0.297
−0.279
Thus, the public’s level of fear of crime in Los Angeles County had an impact on
an individual’s anxiety to interact socially. For every one unit increase in the coded level
of fear of crime (indicating a lower fear level), the social interaction anxiety goes down
by about 0.3. However, the amount of media exposure did not contribute significantly to
an individual’s social interaction anxiety.
Addressing, the second research question, a hierarchical multiple regression was
used to investigate the relationship between Social Interaction Anxiety and the two
independent variables (Media Exposure and Fear of Crime), while controlling for the
demographic variables (gender, age group, and race/ethnicity). In the first step, the three
demographic variables alone were entered into the model. While this model was
statistically significant, F(3, 145) = 3.573, p < 0.016, it explains only about 7% of the
variance in Social Interaction Anxiety, R2 = 0.069.
In the second step, the three demographic variables were retained and the two
independent variables were included as well. This model was also statistically significant,
F(5, 143) = 4.310, p = 0.001, and explained about 13% of the total variance in social
interaction anxiety. The inclusion of the two independent variables explained an
87
additional 6% of the total variance in social interaction anxiety, after controlling for the
demographic variables, R2 Change = 0.062, F(1, 143) = 5.111, p = 0.007. Using the
following codes for each categorical variables: Gender (1 = Male, 2 = Female, 3 =
Decline), Race (1 = White or Caucasian, 2 = Black or African American, 3 = Hispanic or
Latino, 4 = Asian or Asian American, 5 = American Indian or Alaska Native, 6 = Native
Hawaiian or other Pacific Islander, 7 = Another Race, 8 = Decline), and Age (1 = <18, 2
= 18-29, 3 = 30-44, 4 = 45-60, 5 = >60); in (Chapter 4, Table 6) each table gave the
regression coefficients for the models in steps 1 and 2.
Table 6
Regression Coefficients for Model Variables in Hierarchical Multiple Regression
R
R2
R2
Change
B
SE
t
Model 1
0.262
0.069*
Gender
2.366
2.766
0.069
0.856
Age
−2.798
0.878
−0.256**
−3.187
Race/ethnicity
−0.199
0.738
−0.022
−0.270
Model 2
0.362
0.131**
0.062**
Gender
2.093
2.740
0.061
0.764
Age
−1.883
0.925
−0.172*
−2.035
Race/ethnicity
−0.481
0.725
−0.052
−0.663
Media exposure
0.239
0.313
0.064
0.763
Fear of crime
−0.283
0.094
−0.247**
−3.005
R2 = amount of variance explained by the independent variables in the model
R2 Change = additional variance in dependent variable
B = Unstandardized coefficient
= Standardized coefficient (values are converted to the same scale for comparison)
SE = Standard Error
t = estimated coefficient (B) divided by its own SE. If t < 2, the independent variable does not belong in
the model
Statistical significance: *p < 0.05, **p < 0.01, ***p , 0.001
Thus, there was a relationship between the public’s amount of media exposure,
level of fear of crime, and social interaction anxiety after controlling for demographics.
88
However, the amount of media exposure did not contribute significantly to an
individual’s social interaction anxiety. In addition, gender and race/ethnicity were not
significant indicators of an individual’s social interaction anxiety.
Findings- Confidence Intervals
Hypothesis one argued that the public’s amount of media exposure and level of
fear of crime in Los Angeles County had a social impact on an individual’s anxiety to
interact socially. Therefore, a 95% confidence interval was used for each coefficient
estimated in the model for hypothesis 1. With a 95% confidence, the coefficient for
Media Exposure (B1) was estimated to be between -0.183 and 0.974 and with 95%
confidence, the coefficient for Fear of Crime (B2) was estimated to be between -0.493 and
-0.139. Additionally, only the amount of Fear of Crime had a significant impact on social
interaction anxiety (p = 0.001), as seen in (Chapter 4, Table 7).
The model 1 (for hypothesis 1):
12
ˆ(MediaUsage) (FearofCrime)Y a B B= + +
89
Table 7
Hypothesis 1 Coefficientsa
Model
Unstandardized
coefficients
Standardized
coefficients
t
Sig.
95.0% Confidence
interval for B
B
Std.
Error
Beta
Lower
Bound
Upper
Bound
1
(Constant)
31.875
5.690
5.602
.000
20.632
43.118
Media
exposure
.396
.293
.107
1.352
.179
-.183
.974
Fear of
crime
-.316
.089
-.280
-3.536
.001
-.493
-.139
a. Dependent Variable: Social Interaction Anxiety
Hypothesis two argued that there was a relationship between the public’s amount
of media exposure, level of fear of crime, and social interaction anxiety after controlling
for demographics. Using the following codes for each categorical variables: Gender (1 =
Male, 2 = Female, 3 = Decline), Race (1 = White or Caucasian, 2 = Black or African
American, 3 = Hispanic or Latino, 4 = Asian or Asian American, 5 = American Indian or
Alaska Native, 6 = Native Hawaiian or other Pacific Islander, 7 = Another Race, 8 =
Decline), and Age (1 = <18, 2 = 18-29, 3 = 30-44, 4 = 45-60, 5 = >60). A 95%
confidence interval was calculated for each coefficient estimated in the model for
hypothesis 2. With a 95% confidence, the coefficient for Media Exposure (B1) was
estimated to be between -0.380 and 0.858 and the coefficient for Fear of Crime (B2) was
estimated to be between -0.469 and -0.097. Likewise, the coefficients for Race (B3),
Gender (B4), and Age (B5) was estimated to be between -1.915 and 0.953, -3.324 and
7.510, -3.711 and -0.054, respectively. It is important to note that only the age variable
90
and the level of fear of crime had an impact on social interaction anxiety, as seen in
(Chapter 4, Table 8).
The model 2 (for hypothesis 2):
1 2 3 4 5
ˆ(MediaUsage) (FearofCrime) (Race) (Gender) (Age)Y a B B B B B= + + + + +
Findings- Effect Size
It is important to know that when using multiple regression the effect size
calculation is transferred into the R2 value and/or the R2 Change. Thus, the standard
effect size is 0.05 (see Field, 2013, pg. 472). Model 1 had a R2 value of (0.069) which
tested the demographic variable against the dependent variables of social interaction
anxiety. In addition, the R2 Change of (0.062) was a result of the testing of media
Table 8
Hypothesis 2 Coefficientsa
Model
Unstandardized
coefficients
Standardized
coefficients
T
Sig.
95.0% Confidence
interval for B
B
Std. Error
Beta
Lower
Bound
Upper
Bound
1
(Constant)
31.859
5.648
5.640
.000
20.695
43.022
Gender
2.366
2.766
.069
.856
.394
-3.100
7.833
Age
-2.798
.878
-.256
-3.187
.002
-4.534
-1.063
Race/ethnicity
-.199
.738
-.022
-.270
.788
-1.657
1.259
2
(Constant)
36.529
9.056
4.034
.000
18.628
54.429
Gender
2.093
2.740
.061
.764
.446
-3.324
7.510
Age
-1.883
.925
-.172
-2.035
.044
-3.711
-.054
Race/ethnicity
-.481
.725
-.052
-.663
.509
-1.915
.953
Media
exposure
.239
.313
.064
.763
.447
-.380
.858
Fear of crime
-.283
.094
-.247
-3.005
.003
-.469
-.097
a. Dependent Variable: Social Interaction Anxiety
91
exposure, level of fear of crime, and all demographic variables against social interaction
anxiety. Therefore, both models resulted in a large effect.
Post-Hoc Analyses Testing
There were no post-hoc analyses performed on the data in this study.
Additional Statistical Testing of Hypotheses
There were no additional statistical testing of the hypotheses that were not already
presented in Chapter 3. This study included results from a hierarchical multiple
regression analyses (linearity, normality, and homoscedasticity), a One-way Analysis of
Variance, a correlational analysis, and a multiple regression analysis.
Summary
Research Question 1 Findings
Research question one examined the impact of residents of Los Angeles County
social interaction anxiety after accounting for the amount of media exposure and level of
fear of crime. Using a multiple regression analysis, I was able to investigate the
relationship between the amount of media exposure, levels of fear of crime, and social
interaction anxiety. The results indicated that the model as a whole was statistically
significant, (R2 = 0.099, F(2, 148) = 8.164, p < 0.001). However, media exposure was not
statistically significant over and above the level of fear of crime and could therefore be
dropped from the model. While the model only explained 10% of the variation in social
interaction anxiety, this does not negate the importance of the significant relationship
between fear of crime and social interaction anxiety. Because this study was mainly
92
interested in the relationships among the variables rather than prediction, a low R2 value
is tolerable.
Research Question 2 Findings
Research question two examined the impact in which the demographic variables
of race/ethnicity, age, and gender had on residents of Los Angeles County amount of
media exposure, level of fear of crime, and social interaction anxiety. A two-step
hierarchical multiple regression analysis was used to investigate such a relationship. In
the first step, the demographic variables alone were evaluated, of which a 7% variance
explained a statistically significant relationship. The second step evaluated the three
demographic variables along with the two independent variables of which the model
explained a 13% variance. Although this R2 value is quite low, this does not negate the
importance of the statistically significant variables in the model. Since this study is
mainly interested in understanding the relationships between the variables rather than in
prediction, a low R2 value is tolerable. There was a statistically significant relationship
between residents’ level of fear of crime, and social interaction anxiety level after
controlling for all demographic variables. The media exposure variable was not
significant over and above the level of fear of crime after controlling for all demographic
variables. Age was the only demographic variable significantly related to social
interaction anxiety.
Fear of crime and social interaction anxiety were statistically negatively related.
This seems counterintuitive. For this to make sense, one needs to look at the way fear of
crime and social interaction anxiety were coded. Low values for fear of crime meant very
93
fearful while higher values meant no fear. Low values for social interaction anxiety
indicated no anxiety while higher values indicated extreme anxiety. Because a negative
relationship was observed, as the coded value for fear of crime goes up (indicating lower
levels of fear), the values for social interaction anxiety go down (indicating lower
anxiety). This is more intuitive and agrees with previous literature. For every one unit
increase in the coded level of fear of crime (indicating a lower fear level), the social
interaction anxiety goes down by about 0.3. This same relationship holds when
controlling for the demographic variables.
Age and social interaction anxiety were also negatively related. For every one unit
increase in age, the social interaction anxiety value decreases by 2 when fear of crime is
included in the model. This seems to indicate that as one ages, the level of social
interaction anxiety decreases.
Conclusion
Summary of the Results of the Study
The purpose of this study was to examine the impact that the amount of media
exposure and level of fear of crime had on Los Angeles County residents’ social
interaction anxiety level. The basis for this examination was Gerbner’s (1969) cultivation
theory in which it is assumed that as an individual’s amount of media exposure increases,
so does their fear of crime. What was not examined through the theory was whether or
not media exposure and fear of crime also impacted one’s social anxiety levels.
Therefore, this study examined such an impact.
94
This study examined the County of Los Angeles, California which had an
estimated 10.11 million residents. There were a total of 590 responses to the survey. Of
those, 178 answered “Yes” to both screening questions. After reviewing the 178 surveys,
only 150 completed the whole survey. Therefore, a total of 150 responses were included
in the data analyses. Of the 150, 53 were males, 96 were females, and one participant
declined to state their gender. Participants ages ranged from 18-65+ with 30% (45
participants) of the total population falling in the 25-34 category. Of all race/ethnicity
variables, the White and Caucasian category had the highest participation rate at 41.3%
(62 participants). The second highest group was the Hispanic or Latin category at 26%
(39 participants). This study used the estimated incidence rate (the percentage of
residents who were estimated to qualify for the survey) to determine that the sample was
a representation of the entire population. After comparing the estimated incidence rate of
25.56% to the actual incidence rate of 38.49%, there was a 12.93% difference.
Consequently, it was stated that the sample was a representation of the whole population.
Research question one examined the impact that the amount of media exposure
and level of fear of crime may have had on Los Angeles County residents’ level of social
interaction anxiety. The results revealed that the amount of media exposure and level of
fear of crime had a significant impact on Los Angeles County residents’ social interaction
anxiety. Research question two examined the relationship the demographic variables of
race/ethnicity, age, and gender had on residents’ amount of media exposure, level of fear
of crime, and social interaction anxiety. The results revealed a relationship between the
95
public’s amount of media exposure, level of fear of crime, and social interaction anxiety
after controlling for all demographic variables.
In prior research, several authors concluded with similar results as presented in
this study. For example, Zhao, et al. (2010) evaluated the relationship between property
and violent crime and fear of crime at the individual level. The authors concluded that
participants’ fear of crime directly related to the number of crimes committed within an
average of 528 feet from their homes. In another study, authors found that when crime
stories in American cities feature criminality, the public’s fear of crime increased
(Gibson, 2014). In connection to prior research, I was surprised to find that residents’
amount of media exposure and level of fear of crime impacted their level of social
interaction anxiety. I was also surprised to find that a relationship existed between the
amount of media exposure, level of fear of crime, and social interaction anxiety after
controlling for demographics. In sum, I was surprised by all results of this study and
happy to find that the results of this study provided additional knowledge and
understanding of the cultivation theory.
Transitional Material
In Chapter 5, I explain how findings from this study confirmed, disconfirmed, or
extended knowledge in the modern discipline. I also explain how the theoretical
perspective of the cultivation theory continued through the evaluation of the findings.
The presentation of the limitations of the study allowed me to see how everything
changed from the original limitations presented in Chapter 1, to the actual limitations
96
presented during data collection. Ending Chapter 5, I provided potential social change
influences and suggestions for future researchers to examine.
97
Chapter 5: Conclusions
Introduction
Purpose and Nature of Study
The purpose of this study was to evaluate the cultivation theory (Gerbner, 1969)
by determining if the amount of media exposure to crime and the level of fear of crime
contributes to social interaction anxiety. The study was conducted to bring awareness to
poor or weak social relationships due to citizens’ perception of property and violent
crimes presented throughout the media. Therefore, this study evaluated residents of Los
Angeles County who were over the age of 18 and had viewed crime stories published
across TV and Internet sources two weeks before having completed the survey. The
survey included two screening questions, 45 questions from three separate survey
instruments, and three demographic questions. The study used the IBM SPSS software
and a series of statistical tests to analyze the data.
Why and How the Study Was Conducted
In 2000, Dixon and Linz found that approximately 30% of all news stories in
mass media, both print and broadcast, included reports on criminal activity. Thus, the
mass media was recognized by researchers as a primary source of the public’s receiving
information about crime (Surette, 2007). It was also suggested that the mass media
distorts the public’s perception of crime occurrence by disproportionately focusing on
violent crimes (Reiner, 2007). Therefore, this study examined the basis of the cultivation
theory and its relationship to the residents of Los Angeles County’s social interaction
anxiety. Research question one examined how the Los Angeles County public’s amount
98
of media exposure and the level of fear of crime impacted social interaction anxiety.
Research question two then examined the relationship among the public’s amount of
media exposure, the level of fear of crime, and social interaction anxiety after controlling
the three demographic variables.
Summary of Findings
There were 590 responses to the survey. Of this, 178 respondents answered “Yes”
to both screening questions. After reviewing these 178 responses, 150 respondents
completed the entire survey. Therefore, these 150 responses were included in the data
analyses. A multiple regression analysis was used to investigate the relationship between
social interaction anxiety and the two independent variables. The analysis revealed that
residents of Los Angeles County social interaction anxiety were impacted by the amount
of media exposure and level of fear of crime. However, the amount of media exposure
did not significantly contribute to an individual’s social interaction anxiety.
Research question two used a hierarchical multiple regression analysis to
investigate the relationship between social interaction anxiety and the two independent
variables after controlling for the three demographic variables. By using a two-step
process, the results revealed a relationship between the public’s amount of media
exposure, level of fear of crime, and social interaction anxiety after controlling for the
demographics. However, gender and race/ethnicity were not significant contributors to
individual’s social interaction anxiety. In sum, the results revealed a relationship among
all variables.
99
Interpretation of the Findings
Confirmation and Extension of Knowledge
This study examined the relationship between the amount of media exposure, fear
of crime, and social interaction anxiety levels. The basis for this examination was the use
of the cultivation theory, which argued that as the amount of media exposure increased so
would an individual’s fear of crime levels (Gerbner, 1969). To extend this theory,
researchers looked at how this understanding related to some specific outcomes. For
example, Zhao, Lawton, and Lawton (2010) examined the relationship between property
and violent crime and an individual’s fear of crime levels. The authors found that an
individual’s fear of crime levels directly related to crimes that occurred within 528 feet
from their home. Of the 652 respondents, women and older residents reported higher
levels of fear in comparison to men and younger residents.
In 2012, Foster et al. examined whether physical offenses were a deterrent from
walking in public places. Based on a self-reported questionnaire with questions about
environments within a10-to-15 minute walk from their home, the authors found that those
who reported higher levels of fear of crime were less likely to walk in their
neighborhood. Also, the authors found that fear of crime was greater for recreational
walkers (those who walked for fun) than transport walkers (those who had to walk to get
to work). In a similar study, Stodolska et al. (2013) examined the impact outdoor
recreational activities had on Mexican-American youth in the South Lawndale
neighborhood of Chicago. The authors found that the majority of the 25 participants
reported that crime was an issue in their neighborhood. Therefore, all the participants
100
reported feeling safer by participating in leisure activities near their home, near relatives’
homes, and during school hours.
Across the world in New Zealand, Breetzke and Pearson (2014) examined
whether the occurrence of crime in a neighborhood impacted one’s level of fear of crime.
The analysis included 347,679 crimes reported between 2008 and 2010, of which the
authors found that women reported higher levels of fear of crime than men. Also, those
who lived in impoverished areas were more fearful than individuals living in prosperous
areas. In a final study, Luo et al. (2015) examined fear of crime in one’s neighborhood
and near their home. The analysis included data from 2,393 participants over the age of
18 and lived in Houston, Texas between 2010 and 2012. The authors found that women
and senior participants felt safer in public settings than in their homes in comparison to
men and younger participants who felt equally safe in both settings. Therefore, to extend
prior research, the current study examined the data points of the amount of media
exposure, levels of fear of crime, and levels of social interaction anxiety.
The study examined residents of Los Angeles County who were over the age of
18 and had watched crime stories published across different media sources two weeks
before completing the survey. There were 178 respondents who answered “Yes” to both
screening questions but only 150 respondents completed the survey. Therefore, 150
responses were used in the data analyses. The research found that the public’s amount of
media exposure and level of fear of crime in Los Angeles County had an impact on an
individual’s anxiety towards social interaction. I also found a relationship between the
public’s amount of media exposure, level of fear of crime, and social interaction anxiety
101
after controlling for demographics. However, gender and race/ethnicity alone were not
significant indicators of an individual’s social interaction anxiety levels.
In addition to the research questions, I examined two different hypotheses.
Hypothesis one argued that the public’s amount of media exposure and level of fear of
crime in Los Angeles County has high or no social impact on an individual’s anxiety to
interact socially. Hypothesis two argued that there would be no relationship between the
public’s amount of media exposure, level of fear of crime, and social interaction anxiety
after controlling for demographics. I found that all categories of the demographics
impacted the amount of media exposure, level of fear of crime, and social interaction
anxiety but it must be noted that only the age category significantly impacted the
independent and dependent variables. In sum, I was able to extend prior research through
examining the relationship that the amount of media exposure and the level of fear of
crime had on residents' level of social interaction anxiety.
Findings - Theoretical Foundation
This study used Gerbner’s (1969) cultivation theory which states that the more
time people spend watching television, listening to the radio, reading newspapers and
magazines, and participating in social media on the Internet, the more likely they are to
equate reality with what they hear and see on those mass media sources. Riddle (2009)
then presented the same theory in simpler terms by arguing that the pictures and
messages conveyed by the mass media shape the public’s view of reality. Therefore, I
considered literature that used the cultivation theory as the theoretical foundation while
also examining media, fear of crime, and some specific outcome.
102
Callanan (2012) found that as the consumption of newspaper and television news
increased in southern California, so did fear of crime. In the same year, Nellis and Savage
(2012) used the cultivation theory to examine how the amount of exposure to TV news
about terrorism impacted New York City and Washington, D.C. fear of terrorism. The
authors found that as residents’ amount of media exposure to terrorism news increased,
so did their fear of terrorism. In 2014, Gibson confirmed the cultivation theory when it
was concluded that the media in American cities featured stories about criminality, the
public’s fear of crime increased. Callanan and Rosenberger (2015) used the cultivation
theory that fear of crime is increased by consumption of television programming when
considering demographic characteristics. After reviewing the literature on the cultivation
theory, I considered how findings from prior research relate to the results of the current
study.
Research question one determined Los Angeles County residents’ amount of
media exposure and their level of fear of crime impacted their social interaction anxiety
level. The outcome concluded that I would reject the null hypothesis while accepting the
alternative hypothesis which states that the amount of media exposure and the level of
fear of crime had an impact on residents' level of social interaction anxiety.
Consequently, the findings confirmed the foundation of Gerbner’s (1969) cultivation
theory.
Research question two of the present study examined the relationship among the
public’s amount of media exposure, level of fear of crime, and social interaction anxiety
after controlling for demographics (race/ethnicity, age, and gender). Based on the results,
103
I rejected the null hypothesis and accepted the alternative hypothesis, which argued there
was a relationship between the publics’ amount of media exposure, level of fear of crime,
and social interaction anxiety after controlling for demographics. In sum, findings of the
study aligned with prior research and Gerbner’s (1969) definition that argued that as the
amount of media exposure increases, so will one’s level of fear of crime.
Limitations of the Study
Generalizability, Validity, and Reliability
In 2018, Los Angeles County had a total population of 10.11 million residents'
(U.S. Census Bureau, 2018). Of the 10.11 million residents, 590 responses were collected
from Los Angeles County residents. Of these responses, 178 residents answered “Yes” to
both screening questions. After reviewing the data, only 150 surveys were completed.
Therefore, results from the 150 responses revealed that residents' social interaction
anxiety was impacted by the amount of media exposure and their level of fear of crime.
Since this study had a small sample size; future research may or may not be able to
generalize to other geographical locations. Additionally, this study did not use any open-
ended questions; therefore residents’ responses may have been skewed based on the lack
of this choice.
Also, I mentioned in Chapter 1 one of the survey instruments used was developed
in the United Kingdom but did not pose a different perspective than that of the United
States. However, results revealed very similar findings to studies done in the United
Kingdom and other parts of the world that used the same instrument. When considering
validity, I was able to draw meaningful and useful information from all data collected.
104
For example, the results revealed that a relationship existed between residents of Los
Angeles County’s amount of media exposure, level of fear of crime, and level of social
interaction anxiety. Through reliability, this study was able to confirm prior scoring
measures for all three instruments used. In sum, this study confirmed several measures,
methods, and outcomes used in prior research.
Recommendations
Future Research Recommendations
In 2012, Jorgensen et al. examined whether or not participants felt safer in a
public park that had other recreating versus a public park in which there were no other
recreating. The authors found that women participants reported increased fear in public
parks depending on the number of people in the environment. In the same year, Lee and
Hilinski-Rosick (2012) examined whether lifestyle risk behaviors of college students
decreased in their likelihood of becoming a victim. Collecting data from 3,472
undergraduate students, the authors found that fear of crime was greater among younger
non-white students than among older white students. In the following year, Kappas et al.
(2013) examined whether older adults displayed greater precautionary behaviors toward
crime versus their younger counterparts. After evaluating 528 responses, the authors
found that older adults were more fearful of crime than their younger and middle-aged
counterparts. Therefore, the following strengths and limitations assisted in my
recommendations for future research.
One strength of this study was that all data was collected within the United States;
rather than other parts of the world. By doing so, this eliminated further research needed
105
on areas outside the United States and time constraints that go along with collecting data
around the world. Another strength was the ability to collect data within my current area
of residents, Los Angeles County. This helped me gain knowledge on why it was
important to collect data on residents’ amount of media exposure, level of fear of crime,
and level of social interaction anxiety. In addition to the strengths of this study, there
were several limitations. First, this study focused only on Los Angeles County, therefore
data may or may not be able to be generalized to other populations. Second, several
survey instruments were used, of which one was based out of the United Kingdom.
To address these limitations, I presented two study questions. Research question
one examined the impact of the amount of media and level of fear of crime had on
residents of Los Angeles County’s level of social interaction anxiety. Based on 150
responses, results revealed that residents’ level of social interaction anxiety was impacted
by their amount of media exposure and level of fear of crime. Research question two then
examined the relationship between the residents of Los Angeles County’s amount of
media exposure, level of fear of crime, and social interaction anxiety after controlling for
the demographics of race/ethnicity, age, and gender. The results revealed a relationship
between the race/ethnicity, age, and gender of the respondents when considering their
amount of media exposure to crime news, their level of fear of crime, and their social
interaction anxiety levels. In conjunction with these studies strengths and limitations, the
findings of this study assisted my recommendations for future research.
First, researchers could examine how the racial representation of suspects
presented throughout crime news impacts residents’ social interaction anxiety levels after
106
accounting for their amount of media exposure and level of fear of crime. By doing so,
researchers would gain knowledge on residents’ perspectives to socially interact with
others in public settings upon exposure to crime stories that may display one racial group
more than another. In addition to evaluating suspects’ race in crime news stories, future
researchers could replicate the present study by examining different larger counties
within California. This would allow for future research to determine if residents’ amount
of media exposure, level of fear of crime, and social interaction anxiety differ among
counties in the state. Individuals within state and local governments should pay close
attention to data presented from this study, as it would help enact laws and codes that
better regulate information the mass media publishes. Also, community organizations
could use the information to create programs to help alleviate public fear. In addition,
policymakers and public safety officials could use the information to better allocate
resources to communities in need.
Implication for Social Change
Implication for Social Change - Individual, Societal, and Policies
This study examined the impact that the amount of media exposure and level of
fear of crime had on residents of Los Angeles County’s level of social interaction
anxiety. Therefore, research question one examined how the Los Angeles County
public’s amount of media exposure and level of fear of crime impacted their level of
social interaction anxiety. Research question two then examined the relationship between
the amount of media exposure, level of fear of crime, and social interaction anxiety after
controlling for the demographics of race/ethnicity, age, and gender. The results revealed
107
that the overall model significantly predicted social interaction anxiety among residents
of Los Angeles County. Also, there was a relationship between the amount of media
exposure, level of fear of crime, and social interaction anxiety after controlling for
demographics. However, it is important to note that the age category was the only
demographics characteristic that significantly indicated an individual’s social interaction
anxiety. With these results in mind, the following are implications for potential social
change in Los Angeles County.
First, state and local officials can use the data when evaluating laws that regulate
how the mass media presents crime stories to the public. It is my hope that this will
alleviate the public’s level of fear regarding violent crime in public spaces. Also, these
officials can use the data to help establish social programs alleviating the public’s fear of
violent crimes. For example, officials could work with non-profit organizations to create
crime awareness programs that can be offered to residents regularly. These programs
would include educational classes in which residents are made aware of the crime
problem in the area and provided with preventative measures they can take to protect
themselves in public.
Secondly, policymakers and public safety directors can use the data to help aid
them in the allocation of resources to communities in need. For example, public safety
directors could use the information when determining which communities within Los
Angeles County need more law enforcement officers. Finally, the mass media could use
the data to self-regulate their programming measures for the benefit of the public. By
doing so, it is the hope that the media would see the damage being done to the public
108
based on the content they have released on crime stories. To conclude, the data from this
study would help individuals at all levels better understand residents’ fear of social
interaction in public spaces and how to implement measures that could alleviate such
fear.
Social Change - Theoretical Foundation
In 1969, Gerbner’s cultivation theory argued that the more time people spend
watching television, listening to the radio, reading newspapers and magazines, and
participating in social media on the internet, the more likely they are to equate reality
with what they hear and see on those mass media sources. Therefore, this study used the
basis of cultivation theory to examine the impact that the amount of media exposure and
the level of fear of crime had on residents of Los Angeles County’s level of social
interaction anxiety. Collecting data from residents of Los Angeles County who were over
the age of 18 and had watched crime stories throughout media sources two weeks before
taking the survey, this study was able to address all research questions. By doing so, I
was able to collect information on residents' level of social interaction to media exposure
and fear of crime.
The overall results revealed that residents’ social interaction levels directly related
to the amount of time spent consuming crime news stories and their overall level of fear
of crime. Therefore, the data contributed to the understanding of residents’ social
interaction anxiety levels, in turn allowing local and state officials and policymakers to
understand the need for social programs that alleviate residents’ fear of crime in public
spaces. Also, the cultivation theory and findings assist in understanding the need for the
109
mass media to self-regulate the type of information they release and to pay closer
attention to the consequences of their actions.
Social Change- Practice
Based on the findings of this study and Gerbner’s (1969) definition of cultivation
theory which stated that the more time an individual spends consuming crime news
throughout media sources, the more likely their level of fear of crime will increase; I
recommend the following practices for social change. First, residents could practice
random acts of kindness by helping those in their community who have fallen victim to
violent crimes rather than turning a blind eye. Second, I urge local and state officials to
consider the data to create programs within local organizations that educate the public on
crime awareness and how to protect themselves. Furthermore, I urge public safety
officials to evaluate the data and determine how to provide communities in need with
more law enforcement officers.
Finally, I strongly recommended that policymakers reconsider policies that
regulate the content for which the mass media releases crime-specific stories to the
public. For example, policy changes could require that the mass media uses verbiage that
alleviates fear rather than increases fear. To conclude, I recommend that all government
officials at the state and local levels consider the findings of this study when developing
public awareness programs that alleviate fear of crime while building better social
interaction among individuals.
110
Conclusion
Take Home Message
In 2007, Surette found that the mass media represented the primary source for the
consumption of information. Before this finding, Dixon and Linz (2000) determined that
approximately 30% of all stories published throughout the mass media pertained to
criminal activity. Therefore, Reiner’s (2007) study concluded that mass media has a way
of distorting the public’s perception of crime occurrence by disproportionately focusing
on violent crimes. As a result, Gibson (2014) argued that the public’s fear of crime is
impacted by stories that focused solely on crime.
This study examined the relationship between the amount of media consumption,
residents' level of fear of crime, and residents’ level of social interaction anxiety through
the lens of the cultivation theory. In Gerbner’s (1969) cultivation theory argued that the
more time people spend watching television, listening to the radio, reading newspapers
and magazines, and participating in social media on the internet, the more likely they
were to equate reality with what they hear and see on those mass media sources.
The study used a quantitative approach to address two research questions.
Research question one examined how Los Angeles County residents’ amount of media
exposure and level of fear of crime impacted their social interaction anxiety level. The
null hypothesis argued that the amount of media exposure and level of fear of crime in
Los Angeles County had no social impact on an individual’s anxiety to interact socially.
In contrast, the alternative hypothesis argued that the amount of media exposure and level
of fear of crime in Los Angeles County had a high impact on an individual’s anxiety to
111
interact socially. Research question two examined the relationship among Los Angeles
County residents’ amount of media exposure, level of fear of crime, and social interaction
anxiety after controlling for the demographics of race/ethnicity, age, and gender. The null
hypothesis argued that there would not be a relationship between the public’s amount of
media exposure, level of fear of crime, and social interaction anxiety after controlling for
demographics. In contrast, the alternative hypothesis argued that there would be a
relationship between the public’s amount of media exposure, level of fear of crime, and
social interaction anxiety after controlling for demographics.
A total of 590 responses were obtained from residents in Los Angeles County. Of
this, only 178 respondents answered “Yes” to both screening questions: “Do you live in
Los Angeles County?” and “Have you watched crime news stories throughout media
sources within two weeks before taking this survey?” After reviewing all 178 responses,
150 respondents completed the entire survey. Thus, these responses were included in the
data analyses. Research question one was addressed by using multiple regression analysis
in which the results indicated that the amount of media exposure and level of fear of
crime contributes to residents’ social interaction anxiety levels. Therefore, I rejected the
null hypothesis and accepted the alternative hypothesis. Research question two used
several steps of hierarchical multiple regression analysis to determine that a relationship
existed between the public’s amount of media exposure, their level of fear of crime, and
their level of social interaction anxiety after controlling for demographics. Therefore, I
rejected the null hypothesis and accepted the alternative hypothesis.
112
Recommendations for future researchers suggested extending on this study by
examining whether or not the racial representation of suspects in crime stories impacts
residents’ level of social interaction, after accounting for their amount of media exposure
and level of fear of crime. Also, future researchers could replicate this study by
examining different larger counties within California, to determine if levels of social
interaction anxiety differ among county residents. Considering social change, state and
local governments should consider the data to enact laws and codes that better regulate
presentation of crime stories throughout mass media sources. Public safety officials could
use the data to work with local organizations to help create educational programs on
crime awareness and self-protection when in public spaces. Additionally, public safety
officials could consider the data when allocating resources such as law enforcement to
communities in need. To conclude, this study not only extended on prior research but also
provided significant data for which state and local officials can consider.
113
References
Alper, M., & Chappell, A. T. (2012). Untangling fear of crime: A multi-theoretical
approach to examining the causes of crime-specific fear. Sociological Spectrum,
32(4), 346-363. https://doi.org/10.1080/02732173.2012.664048
Annenberg Public Policy Center. (1993). Media health coding: Capturing changes over
time— CHAMP. Retrieved from https://www.youthmediarisk.org
Austin, D., Furr, L., & Spine, M. (2002). The effects of neighborhood conditions on
perceptions of safety. Journal of Criminal Justice, 30(5), 417-427.
https://doi.org/10.1016/s0047-2352(02)00148-4
Boda, Z., & Szabó, G. (2011). The media and attitudes towards crime and the justice
system: A qualitative approach. European Journal of Criminology, 8(4), 329-342.
https://doi.org/10.1177/1477370811411455
Beetzke, G. D., & Pearson, A. L. (2014). The fear factor: Examining the spatial
variability of recorded crime on the fear of crime. Applied Geography, 46, 45-52.
https://doi.org/10.1016/j.apgeog.2013.10.009
Brands, J., Schwanen, T., & Van Aalst, I. (2013). Fear of crime and affective ambiguities
in the night-time economy. Urban Studies, 52(3), 439-455.
https://doi.org/10.1177/0042098013505652
British Healthcare Business Intelligence Association. (2016). Guidelines for the use of
secondary data. Retrieved from
https://www.bhbia.org.uk/guidelines/secondarydataguidelines.aspx
British Office of National Statistics. (2007-2008). British crime survey. Retrieved from
114
http://doc.ukdataservice.ac.uk/doc/6066/mrdoc/pdf/6066techreport1.pdf
Brooks, T., & Marsh, E. F. (2009). Complete directory to prime time network and cable
TV shows, 1946-Present. New York, NY: Random House.
Callanan, V. J. (2012). Media consumption, perceptions of crime risk and fear of crime:
Examining race/Ethnic differences. Sociological Perspectives, 55(1), 93-115.
https://doi.org/10.1525/sop.2012.55.1.93
Callanan, V., & Rosenberger, J. S. (2015). Media, gender, and fear of crime. Criminal
Justice Review, 40(3), 322-339. https://doi.org/10.1177/0734016815573308
Center for Race, Religion, and Urban Life. (2006). Kinder Institute for Urban Research.
Retrieved from http://corrul.rice.edu
Chadee, D., & Ng Ying, N. K. (2013). Predictors of fear of crime: General fear versus
perceived risk. Journal of Applied Social Psychology, 43(9), 1896-1904.
https://doi.org/10.1111/jasp.12207
Chatterton, P., & Hollands, R. (2003). Urban nightscapes: Youth cultures, pleasure
spaces and corporate power. New York, NY: Routledge Publishing.
Clemente, F., & Kleiman, M. B. (1976). Fear of crime among the aged. The
Gerontologist, 16(3), 207-210. https://doi.org/10.1093/geront/16.3.207
Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends: A routine
activity approach. American Sociological Review, 44(4), 588.
https://doi.org/10.2307/2094589
Cook, C. L., & Fox, K. A. (2012). Testing the relative importance of contemporaneous
offenses: The impacts of fear of sexual assault versus fear of physical harm
115
among men and women. Journal of Criminal Justice, 40(2), 142-151.
https://doi.org/10.1016/j.jcrimjus.2012.02.006
Creighton, T., Walker, C. L., & Anderson, M. R. (2014). Coverage of Black versus white
males in local television news lead stories. Journal of Mass Communication &
Journalism, 04(08). https://doi.org/10.4172/2165-7912.1000216
Custers, K., & Van den Bulck, J. (2011). The relationship of dispositional and situational
fear of crime with television viewing and direct experience with crime. Mass
Communication and Society, 14(5), 600-619.
https://doi.org/10.1080/15205436.2010.530382
Custers, K., & Van den Bulck, J. (2013). The cultivation of fear of sexual violence in
women. Communication Research, 40(1), 96-124.
https://doi.org/10.1177/0093650212440444
Dixon, T. L. (2015). Good guys are still always in white? Positive change and continued
misrepresentation of race and crime on local television news. Communication
Research, 44(6), 775-792. https://doi.org/10.1177/0093650215579223
Dixon, T. L., & Linz, D. (2000). Overrepresentation and underrepresentation of African
Americans and Latinos as lawbreakers on television news. Journal of
Communication, 50(2), 131-154. https://doi.org/10.1111/j.1460-
2466.2000.tb02845.x
Dixon, T. L., & Williams, C. L. (2014). The changing misrepresentation of race and
crime on network and cable news. Journal of Communication, 65(1), 24-39.
https://doi.org/10.1111/jcom.12133
116
Doob, A. N., & Macdonald, G. E. (1979). Television viewing and fear of victimization: Is
the relationship causal? Journal of Personality and Social Psychology, 37(2), 170-
179. https://doi.org/10.1037/0022-3514.37.2.170
Elliott, M. A., & Merrill, F. E. (1934). Social disorganization. New York, NY: Harper
and Brothers.
Faul, F., Erdfelder, E., Buchner, A., & Lang, A. (2009). Statistical power analyses using
G*Power 3.1: Tests for correlation and regression analyses. Behavior Research
Methods, 41(4), 1149-1160. https://doi.org/10.3758/brm.41.4.1149
Federal Bureau of Investigation. (2010). FBI—Property crime. Retrieved from
https://www.fbi.
gov/about-us/cjis/ucr/crime-in-the-u.s/2010/crime-in-the-u.s.-2010/property-crime
Federal Bureau of Investigation. (2013). FBI—Violent crime. Retrieved from
https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-
2013/violent-crime/violent-crime-topic-page/violentcrimemain_final
Ferraro, K. F., & LaGrange, R. (1987). The measurement of fear of crime. Sociological
Inquiry, 57(1), 70-101. https://doi.org/10.4324/9781315086613-15
Field, A. (2013). Discovering statistics using IBM SPSS statistics (4th ed.). United
Kingdom: Sage Publications.
Foster, S., Giles-Corti, B., & Knuiman, M. (2012). Does fear of crime discourage
walkers? A social-ecological exploration of fear as a deterrent to walking.
Environment and Behavior, 46(6), 698-717.
https://doi.org/10.1177/0013916512465176
117
Frankfort-Nachmias, C., & Nachmias, D. (2008). Research methods in the social
sciences (7th ed.). New York, NY: Macmillan Learning.
Gabbidon, S. L. (2007). Criminological perspectives on race and crime. United
Kingdom: Routledge Publishing.
Gerbner, G. (1958). On content analysis and critical research in mass communication.
Audio-Visual Communication Review, 6, 85-108.
https://doi.org/10.1007/BF02766931
Gerbner, G. (1969). Toward cultural indicators: The analysis of mass mediated message
systems. Audio-Visual Communication Review, 17, 137-148.
https://doi.org/10.1007/BF02769102
Gerbner, G. (1970). Cultural indicators: The case of violence in television drama. The
ANNALS of the American Academy of Political and Social Science, 388(1), 69-81.
https://doi.org/10.1177/000271627038800108
Gerbner, G., & Gross, L. (1976). Living with television: The violence profile. Journal of
Communication, 26(2), 172-199. https://doi.org/10.1111/j.1460-
2466.1976.tb01397.x
Gibson, T. A. (2014). In defense of law and order: Urban space, fear of crime, and the
virtues of social control. Journal of Communication Inquiry, 38(3), 223-242.
https://doi.org/10.1177/0196859914532946
Giles-Corti, B., Timperio, A., Cutt, H., Pikora, T. J., Bull, F. C., Knuiman, M., & Shilton,
T. (2006). Development of a reliable measure of walking within and outside the
local neighborhood: RESIDE's neighborhood physical activity
118
questionnaire. Preventive Medicine, 42(6), 455-459.
https://doi.org/10.1016/j.ypmed.2006.01.019
Glass, T. A., De Leon, C. F., Bassuk, S. S., & Berkman, L. F. (2006). Social engagement
and depressive symptoms in late life. Journal of Aging and Health, 18(4), 604-
628. https://doi.org/10.1177/0898264306291017
Goodall, C. E., Slater, M. D., & Myers, T. A. (2013). Fear and anger responses to local
news coverage of alcohol-related crimes, accidents, and injuries: Explaining news
effects on policy support using a representative sample of messages and people.
Journal of Communication, 63(2), 373-392. https://doi.org/10.1111/jcom.12020
Greer, C. (2013). Crime and media: Understanding the connections. Criminology, 143-
164. https://doi.org/10.1093/he/9780199691296.003.0007
Hale, C. (1996). Fear of crime: A review of the literature. International Review of
Victimology, 4(2), 79-150. https://doi.org/10.1177/026975809600400201
Hanslmaier, M. (2013). Crime, fear and subjective well-being: How victimization and
street crime affect fear and life satisfaction. European Journal of Criminology,
10(5), 515-533. https://doi.org/10.1177/1477370812474545
Harmonisation Office of National Statistics. (2015). Crime and Fear of Crime Scale.
Titchfield, England: Author.
Hawdon, J., Rasanen, P., Oksanen, A., & Vuori, M. (2013). Social responses to collective
crime: Assessing the relationship between crime-related fears and collective
sentiments. European Journal of Criminology, 11(1), 39-56.
https://doi.org/10.1177/1477370813485516
119
Heber, A. (2011). Fear of crime in the Swedish daily press: Descriptions of an
increasingly unsafe society. Journal of Scandinavian Studies in Criminology and
Crime Prevention, 12(1), 63-79. doi:10.1080/14043858.2011.561623
Heaven, B., Brown, L. J., White, M., Errington, L., Mathers, J. C., & Moffatt, S. (2013).
Supporting well-being in retirement through meaningful social roles: Systematic
review of intervention studies. Milbank Quarterly, 91(2), 222-287.
https://doi.org/10.1111/milq.12013
Helms, G. (2008). Towards safe city centres: Remaking the spaces of an old-industrial
city. Burlington, VT: Ashgate Publishing.
Hennink, M., Hutter, I., & Bailey, A. (2011). Qualitative research methods. United
Kingdom: SAGE Publications.
Henson, B., Reyns, B. W., & Fisher, B. S. (2013). Fear of crime online: Examining the
effect of risk, previous victimization, and exposure on fear of online interpersonal
victimization. Journal of Contemporary Criminal Justice, 29(4), 475-497.
https://doi.org/10.1177/1043986213507403
Hindelang, M. J. (1974). Public opinion regarding crime, criminal justice, and related
topics. Journal of Research in Crime and Delinquency, 11(2), 101-116.
https://doi.org/10.1177/002242787401100202
Hindelang, M. J., Gottfredson, M. R., & Garofalo, J. (1978). Victims of personal crime:
An empirical Foundation for a theory of personal victimization. Cambridge,
England: Ballinger Pub Co.
Hinkle, J. C. (2015). Emotional fear of crime vs. perceived safety and risk: Implications
120
for measuring “fear” and testing the broken windows thesis. American Journal of
Criminal Justice, 40(1), 147-168. https://doi.org/10.1007/s12103-014-9243-9
Hirtenlehner, H., & Farrall, S. (2013). Anxieties about modernization, concerns about
community, and fear of crime: Testing two related models. International Criminal
Justice Review, 23(1), 5-24. doi:10.1177/1057567712475307
Holbert, R. L., Shah, D. V., & Kwak, N. (2004). Fear, authority, and justice: Crime-
related TV viewing and endorsements of capital punishment and gun ownership.
Journalism & Mass Communication Quarterly, 81(2), 343-363.
https://doi.org/10.1177/107769900408100208
Holt, L. F. (2013). Writing the wrong: Can counter-stereotypes offset negative media
messages about African Americans? Journalism & Mass Communication
Quarterly, 90(1), 108-125. https://doi.org/10.1177/1077699012468699
Hough, M. (2009, June). Risk, fear and insecurity: In praise of simplicity [Paper
presentation]. CRIMPREV Conference, Milton Keynes, UK.
House, J., Landis, K., & Umberson, D. (1988). Social relationships and health. Science,
41, 540-545. https://doi.org/10.1126/science.3399889
Hughes, M. R., Gaines, J. S., & Pryor, D. W. (2015). Staying away from school:
Adolescents who miss school due to feeling unsafe. Youth Violence and Juvenile
Justice, 13(3), 270-290. https://doi.org/10.1177/1541204014538067
Jamieson, P., & Romer, D. (2014). Violence in popular U.S. prime time TV dramas and
the cultivation of fear: A time series analysis. Media and Communication, 2(2),
31-41. https://doi.org/10.17645/mac.v2i2.8
121
Jorgensen, L. J., Ellis, G. D., & Ruddell, E. (2012). Fear perceptions in public parks:
Interactions of environmental concealment, the presence of people recreating, and
gender. Environment and Behavior, 45(7), 803-820.
https://doi.org/10.1177/0013916512446334
Jupp, V. (2006). Volunteer sampling. In the SAGE dictionary of social research methods
(1st ed.). United Kingdom: SAGE Publications.
Kappes, C., Greve, W., & Hellmers, S. (2013). Fear of crime in old age: Precautious
behaviour and its relation to situational fear. European Journal of Ageing, 10(2),
111-125. https://doi.org/10.1007/s10433-012-0255-3
Kennedy, L. W., & Silverman, R. A. (1985). Perception of social diversity and fear of
crime. Environment and Behavior, 17(3), 275-295.
https://doi.org/10.1177/0013916585173001
Kim, M., & Clarke, P. (2014). Urban social and built environments and trajectories of
decline in social engagement in vulnerable elders: Findings from Detroit’s
Medicaid home and community-based waiver population. Research on Aging,
37(4), 413-435. https://doi.org/10.1177/0164027514540687
Kohm, S. A., Waid-Lindberg, C. A., Weinrath, M., Shelley, T. O., & Dobbs, R. R.
(2012). The impact of media on fear of crime among university students: A cross-
national comparison. Canadian Journal of Criminology and Criminal Justice,
54(1), 67-100. https://doi.org/10.3138/cjccj.2011.e.01
Kort-Butler, L. A., & Hartshorn, K. J. (2011). Watching the detectives: Crime
programming, fear of crime, and attitudes about the criminal justice system. The
122
Sociological Quarterly, 52(1), 36-55. https://doi.org/10.1111/j.1533-
8525.2010.01191.x
Krause, K. (2014). Supporting the iron fist: Crime news, public opinion, and authoritarian
crime control in Guatemala. Latin American Politics and Society, 56(01), 98-119.
https://doi.org/10.1111/j.1548-2456.2014.00224.x
Lai, Y., Zhao, J. S., & Longmire, D. R. (2012). Specific crime–fear linkage: The effect of
actual burglary incidents reported to the police on residents’ fear of burglary.
Journal of Crime and Justice, 35(1), 13-34.
https://doi.org/10.1080/0735648x.2011.631408
Lane, J., & Fox, K. A. (2012). Fear of crime among gang and non-gang offenders:
Comparing the effects of perpetration, victimization, and neighborhood factors.
Justice Quarterly, 29(4), 491-523. https://doi.org/10.1080/07418825.2011.574642
Lane, J., & Fox, K. A. (2013). Fear of property, violent, and gang crime: Examining the
shadow of sexual assault thesis among male and female offenders. Criminal
Justice and Behavior, 40(5), 472-496. https://doi.org/10.1177/0093854812463564
Lane, J., & Meeker, J. W. (2003). Women's and men's fear of gang crimes: Sexual and
nonsexual assault as perceptually contemporaneous offenses. Justice Quarterly,
20(2), 337-371. https://doi.org/10.1080/07418820300095551
Lee, D. R., & Hilinski-Rosick, C. M. (2012). The role of lifestyle and personal
characteristics on fear of victimization among university students. American
Journal of Criminal Justice, 37(4), 647-668. https://doi.org/10.1007/s12103-011-
9136-0
123
Lorenc, T., Clayton, S., Neary, D., Whitehead, M., Petticrew, M., Thomson, H.,
Cummins, S., Sowden, A., & Renton, A. (2012). Crime, fear of crime,
environment, and mental health and wellbeing: Mapping review of theories and
causal pathways. Health & Place, 18(4), 757-765.
https://doi.org/10.1016/j.healthplace.2012.04.001
Luo, F., Ren, L., & Zhao, J. S. (2015). Location-based fear of crime: A case study in
Houston, Texas. Criminal Justice Review, 41(1), 75-97.
https://doi.org/10.1177/0734016815623035
Malinen, S., Willis, G. M., & Johnston, L. (2013). Might informative media reporting of
sexual offending influence community members' attitudes towards sex offenders?
Psychology, Crime & Law, 20(6), 535-552.
https://doi.org/10.1080/1068316x.2013.793770
Mattick, R. P., & Clarke, J. C. (1998). Social interaction anxiety scale. Behavior
Research and Therapy, 36, 455-470. https://doi.org/10.1037/t00532-000
Mishler, E. (1990). Validation in inquiry-guided research: The role of exemplars in
narrative studies. Harvard Educational Review, 60(4), 415-443.
https://doi.org/10.17763/haer.60.4.n4405243p6635752
Morgan, M., Shanahan, J., & Signorelli, N. (2009). Growing up with television:
Cultivation processes. In Media effects: Advances in theory and research (pp. 34-
49). Taylor and Francis.
Nellis, A. M., & Savage, J. (2012). Does watching the news affect fear of terrorism? The
importance of media exposure on terrorism fear. Crime & Delinquency, 58(5),
124
748-768. https://doi.org/10.1177/0011128712452961
O'Sullivan, E., Rassel, G. R., & Berner, M. (2008). Research methods for public
administrators (5th ed.). London, England: Pearson Publishing.
Özascilar, M. (2013). Predicting fear of crime. International Review of Victimology,
19(3), 269-284. https://doi.org/10.1177/0269758013492754
Özascilar, M., & Ziyalar, N. (2015). Unraveling the determinants of fear of crime among
men and women in Istanbul: Examining the impact of perceived risk and fear of
sexual assault. International Journal of Offender Therapy and Comparative
Criminology, 61(9), 993-1010. https://doi.org/10.1177/0306624x15613334
Potter, W. J. (2014). A critical analysis of cultivation theory. Journal of Communication,
64(6), 1015-1036. https://doi.org/10.1111/jcom.12128
Rader, N. E., Cossman, J. S., & Porter, J. R. (2012). Fear of crime and vulnerability:
Using a national sample of Americans to examine two competing paradigms.
Journal of Criminal Justice, 40(2), 134-141.
https://doi.org/10.1016/j.jcrimjus.2012.02.003
Rahm, E., & Hai Do, H. (2002). Data cleaning: Problems and current approaches.
Retrieved from http://betterevaluation.org/sites/default/files/data_cleaning.pdf
Reber, B. H., & Chang, Y. (2000). Assessing cultivation theory and public health model
for crime reporting. Newspaper Research Journal, 21(4), 99-112.
https://doi.org/10.1177/073953290002100407
Reiner, R. (2007). The Oxford handbook of criminology (4th ed.). Oxford, England:
Oxford University Press.
125
Rengifo, A. F., & Bolton, A. (2012). Routine activities and fear of crime: Specifying
individual-level mechanisms. European Journal of Criminology, 9(2), 99-119.
https://doi.org/10.1177/1477370811421648
Rhineberger-Dunn, G.M. (2011). Comparing large and small metropolitan newspaper
coverage of delinquency with arrest data: Differential coverage or more of the
same? Criminal Justice Studies, 24(3), 269-290.
https://doi.org/10.1080/1478601x.2011.593966
Rhineberger-Dunn, G. M. (2013). Myth versus reality: Comparing the depiction of
juvenile delinquency in metropolitan newspapers with arrest data*. Sociological
Inquiry, 83(3), 473-497. https://doi.org/10.1111/soin.12006
Riddle, K. (2009). Cultivation Theory Revisited: The impact of childhood television
viewing levels on social reality beliefs and construct accessibility in adulthood
[Paper presentation]. International Communication Association, Washington,
D.C..
Romer, D., Jamieson, K. H., & Aday, S. (2003). Television news and the cultivation of
fear of crime. Journal of Communication, 53(1), 88-104.
https://doi.org/10.1111/j.1460-2466.2003.tb03007.x
Rosen, L. D., Whaling, K., Carrier, L. M., Cheever, N. A., & Rokkum, J. (2013). Using
the media and technology usage scale. Computers and Human Behavior, 29,
2505-2511. https://doi.org/10.1037/t62672-000
Rosenthal, P. I., & Steen, F. (2012). UCLA communication studies news archive.
Retrieved from http://newsscape.library.ucla.edu
126
Rudestam, K. E., & Newton, R. R. (2015). Surviving your dissertation: A comprehensive
guide to content and process (4th ed.). Thousand Oaks, CA: Sage.
Sampson, R. J., & Raudenbush, S. W. (1999). Systematic social observation of public
spaces: A new look at disorder in urban neighborhoods. American Journal of
Sociology, 105(3), 603-651. https://doi.org/10.1086/210356
Shanahan, J., & Morgan, M. (1999). Television and its viewers: Cultivation theory and
research. Cambridge, England: Cambridge University Press.
Shaw, C. R., & McKay, H. D. (1969). Juvenile delinquency and urban areas: A study of
rates of delinquency in relation to differential characteristics of local
communities in American cities. University of Chicago Press.
SPSS Inc. (2009). Terms and conditions. Retrieved from
http://www.spss.com.hk/terms.htm/.
Stein, R. E. (2014). Neighborhood residents’ fear of crime: A tale of three cities.
Sociological Focus, 47(2), 121-139.
https://doi.org/10.1080/00380237.2014.883860
Steinmetz, N. M., & Austin, D. M. (2013). Fear of criminal victimization on a college
campus: A visual and survey analysis of location and demographic factors.
American Journal of Criminal Justice, 39(3), 511-537.
https://doi.org/10.1007/s12103-013-9227-1
Stevens, J. (1996). Applied multivariate statistics for the social sciences (3rd ed.). United
States: Lawrence Erlbaum Associates.
Stodolska, M., Shinew, K. J., Acevedo, J. C., & Roman, C. G. (2013). “I was born in the
127
hood”: Fear of crime, outdoor recreation and physical activity among Mexican-
American urban adolescents. Leisure Sciences, 35(1), 1-15.
https://doi.org/10.1080/01490400.2013.739867
Surette, R. (2007). Media, crime, and criminal justice: Images, realities and policies.
United States: Wadsworth Publishing Company.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics (7th ed.). London,
England: Pearson Education.
Thomas, P. A. (2012). Trajectories of social engagement and mortality in late life.
Journal of Aging and Health, 24(4), 547-568.
https://doi.org/10.1177/0898264311432310
Trochim, W. M. (2006, October 20). Research methods knowledge base. Retrieved from
http://www.socialresearchmethods.net/kb/index.php
U.S. Census Bureau. (2018, July 1). QuickFacts: Los Angeles County, California;
California. Retrieved from
https://www.census.gov/quickfacts/fact/table/losangelescountycalifornia,CA/PST
045219
Vieno, A., Roccato, M., & Russo, S. (2013). Is fear of crime mainly social and economic
insecurity in disguise? A multilevel multinational analysis. Journal of Community
& Applied Social Psychology, 23(6), 519-535. https://doi.org/10.1002/casp.2150
Vilalta, C. J. (2012). Fear of crime and home security systems. Police Practice and
Research, 13(1), 4-14. https://doi-
org.ezp.waldenulibrary.org/10.1080/15614263.2011.607651
128
Visser, M., Scholte, M., & Scheepers, P. (2013). Fear of crime and feelings of unsafety in
European countries: Macro and micro explanations in cross-national perspective.
The Sociological Quarterly, 54(2), 278-301. https://doi.org/10.1111/tsq.12020
Walden University. (2016). Research ethics and compliance: Academic guides at Walden
university. Retrieved from http://academicguides.waldenu.edu/researchcenter/orec
Wargo, W. G. (2015, August 19). Identifying assumptions and limitations for your
dissertation. Retrieved from http://www.academicinfocenter.com/identifying-
assumptions-and-limitations-for-your-dissertation.html
Warr, M. (1984). Fear of victimization: Why are women and the elderly more afraid?
Social Science Quarterly, 65, 681-702. https://search-ebscohost-
com.ezp.waldenulibrary.org/login.aspx?direct=true&db=eue&AN=17076887&sit
e=eds-live&scope=site
Weinrath, M., Clarke, K., & Forde, D. R. (2007). Trends in fear of crime in a western
Canadian city: 1984, 1994, and 2004. Canadian Journal of Criminology and
Criminal Justice, 49(5), 617-646. https://doi.org/10.3138/cjccj.49.5.617
Wilson, J. Q., & Kellings, G. (1982). The police and neighborhood safety: Broken
windows. Atlantic Monthly, 127, 29-38.
http://www.theatlantic.com/doc/198203/broken-windows.
Yu, S. (2014). Fear of cyber-crime among college students in the United States: An
exploratory study. International Journal of Cyber Criminology, 8(1), 36-46.
https://web-b-ebscohost-
com.ezp.waldenulibrary.org/ehost/pdfviewer/pdfviewer?vid=4&sid=5e70c5e3-
