The Measurement of Fear of Crime*

Kenneth F. Ferraro, Northern Illinois University

Randy LaGrange, University of North Carolina at Wilmington

The volume of research on fear of crime in the United States is substantial and
continues to regularly appear in sociology and criminology journals. Despite the
amount of research on the subject, the measurement procedures most frequently
used are suspect because of theoretical and methodological shortcomings. We present
a conceptual definition of fear of crime and then systematically review the way it has
been measured in research over the last fifteen years. T h e review indicates that
whik omnibus fear of crime and risk of crime measures are only moderately corre-
lated, a substantial number of studies have used risk measures and generalized to
fear. Suggestions for future research are onered.

Introduction
Decades of rising crime rates and extensive media coverage have made
crime salient in the public’s mind and fueled a substantial research agenda.
The results of several studies conducted in the late sixties and the report of the
President’s Commission on Law Enforcement and the Administration of
Justice (1967) made fear of crime the focus of widespread national attention
(Wilson, 1975). These findings provided the impetus for nearly two decades of
serious study of reactions to crime. During this time, several national and
local opinion polls have found that crime ranks highly as one of America’s
most enduring and serious social problems (Beardwood, 1968; Erskine, 1974;
Gallup, 1982; Skogan and Maxfield, 1981; TIME, 1985). While this body of
research has aided our understanding of reactions to crime, it lacks consistency
in how these reactions-especially fear of crime-have been conceptualized
and measured.
In order to accurately describe, explain, or predict the occurrence of any
given phenomenon, the variables under consideration must be adequately
measured. Indeed, it can be said that “measurement is the basis of all sci-
ence.” Measurement problems beset a wide variety of research issues and
hinder the process of the cumulative development of scientific knowledge.
Fear of crime is one of these areas that has suffered from measurement prob-
lems. The research is replete with methodological problems that impede our
ability to make useful generalizations. The purpose of this paper is to critically
review and assess the measurement procedures that have been employed in
empirical studies on fear of crime.
 T H E MEASUREMENT OF FEAR O F C R I M E 71

Conceptualization: The Omnibus Variable Problem

Prior to accurately measuring fear of crime one must first come to terms
with its conceptual definition. Though such a statement may seem pedantic, a
substantial body of the research on fear of crime has not manifested a rigorous
concern for conceptual issues. As DuBow et al. (1979:l) have stated, “ ‘fear of
crime’ refers to a wide variety of subjective and emotional assessments and
behavioral reports. There is a serious lack of both consistency and specificity
in these reports.” Indeed, even a casual review of the literature indicates that
the phrase “fear of crime” has acquired so many divergent meanings that its
current utility is negligible.
A major problem in conceptualizing and measuring fear of crime is the
confounding of fear of crime with risk of or vulnerability to crime (Miethe and
Lee, 1984). As Yin (1980:496) states in a review of the literature, “Though
fear of crime is almost never explicitly defined by researchers, their measure-
ments suggest that such fear is implicitly defined as the perception of the
probability of being victimized.”
Figure 1 is presented to delineate the types of crime perceptions that are
variously referred to as “fear of crime.” This figure is designed primarily to
differentiate risk from fear and is adapted from the work of DuBow et al.
(1979). The vertical axis refers to the level Of reference of the perceptions. These
range from the personal or self-oriented to the general or community-oriented.
The horizontal axis refers to the gpe of perceptions ranging from cognitive to
affective; the cognitive end of the continuum includes judgments of risk and
safety while the affective end of the continuum includes fear reactions.
The crime perceptions identified in cells C and F refer to the varying
emotional reactions generated by crime. The “values” referred to in Figure 1
represent a concern one has about crime, either for others (B) or for one’s self
(E). These generally take the form of public opinion regarding the seriousness
of the crime problem or, on the personal level, an evaluation of one’s in-
tolerance of crime. “Judgments” are estimates of the rate of victimization for
a social group (A) or the risk of victimization to the person making the judg-
ment (D). It should not be inferred that judgments are reflections of actual
realities of risk. Rather, all of these reactions are perceptions laden with sub-
jective interpretations of reality. The major benefit of this taxonomy is that i t
differentiates judgments from values from emotions at both the individual arid
community levels. The concept of fear of crime is limited to the emotional
reaction arising from crime, or symbols that a person associates with crime, to
others ( C ) or to one’s self (F). As we will point out later in this review, many
researchers have not followed such a classification and have referred to fear of
crime when actually measuring either judgments or values about crime. ’
72 KENNEI’H F. FERRARO A N D RANDY LAGRANGE

Figure 1
Classification of Crime Perceptions*

Type of Perception
Cognitive Affective

Level of
Reference Judgments Values Emotions

General A. Risk to others; B. Concern about C . Fear for

crime or safety crime to others’ vic-
assessments others timization

Personal D. Risk to self; E. Concern about F. Fear for self

safety of self crime to self; victimization
personal
intolerance

*Adapted from DuBow et al. 1979

Based on the premise that fear of crime is a negative emotional reaction to

crime or the symbols associated with crime, it may be helpful to further describe
the genesis of this emotional reaction. Physiologically speaking, fear involves a
series of complex changes in bodily functioning that alerts an individual to poten-
tial danger. These bodily changes, especially in the endocrine system, can be
either functional or dysfunctional to the individual (Silberman, 1981). On the one
hand, fear may aid the responding capability of the individual to the point that he
or she may be able to accomplish feats that were previously defined as impossible.
However, fear can also be counterproductive. The bodily changes that can give
added power can also result in physiological dysfunction and even incapacitation.
This is most likely if the “fearful” stimuli persist without being resolved, as has
been demonstrated in research on the stress process (Selye, 1956, 1974; Stagner,
1981). In comparison to the activities of everyday life, the most potent forms of
fear involve an intense emotional and physiological reaction to potential danger.
Erving Goffman (1971:4) has described two basic modes of activity which
differentiate fear reactions from everyday life:
’I‘hry go about their business p i i n g , gazing, mothering, digesting. building, resting, playing,
placidly attending to easily managed matters at hand. Or, hlly mobilized, a lury ol‘ intent,
alarmed, they gct ready to attack or to stalk or to flee.

In addition, Goffman asserts that we make use of “dissociated vigilance” to

 T H E MEASUREMENT O F FEAR OF C R I M E 73

monitor the environment while engaging in everyday activities. Humans can

sometimes anticipate danger in certain environments and estimate risk (i.e., a
judgment in Figure 1). When estimated risk increases, one is likely to increase
the monitoring of the environment. T h e added information guides the sub-
sequent activity whether it is a return to previous activity, additional monitor-
ing, or preparation for fight or flight.
Knowledge of the environment also influences the estimate of risk and
concomitant levels of fear. Specifically, familiarity with an environment in-
creases one’s assessment of its safety (DuBow et al. 1979; Silberman, 1981).
People who live in high crime areas oftentimes do not feel there is a high risk
of victimization. Instead, “strange” people or environments are more likely
to be evaluated as dangerous and provocative of fear of crime. Thus, fear of
crime, as it frequently has been measured, actually may be indicating fear of
strangers and inflating fear of crime with the amount of contact with strangers
in a neighborhood-obviously higher in urban areas (Garofalo and Laub,
1978).
Fear of crime may be part of a more general tendency to be fearful which
detracts from one’s quality of life. Erikson’s (1976:234) research on disaster
victims indicates that the typical individual experienced “ . . . a sense of
vulnerability, a feeling that one has lost a certain natural immunity to mis-
fortune, a growing conviction, even, that the world is no longer a safe place to
be.” Fear reactions may also be accompanied by feelings of violation, help-
lessness, anger, outrage, and frustration (DuBow et al. 1979).
It is this tendency to be afraid regardless of the circumstances that is most
likely to be dysfunctional to individuals. It should be expected that people will
be fearful when confronted by a dangerous object; fear may at times be a
wholly appropriate reaction (Sarnoff and Zimbardo, 1961). However, fear of
crime may also be deleterious to the individual if it is aroused by stimuli that
are fundamentally innocuous. O u r objective is not to extend this line of think-
ing into discussions of the “rationality” of fear. Rather, o u r purpose is to il-
lustrate that fear, as an emotional reaction, is both an effect and a cause in its
relationship to judgments of risk. Fear is influenced by judgments of risk, but
also affects such judgments. To assume, however, that when one measures judg-
ments of risk that one is measuring fear of crime is both invalid and obscures
the processes that generate these perceptions. Fear of crime refers to the nega-
tive emotional reaction generated by crime or symbols associated with crime
and is conceptually distinct from either judgments (risks) or concerns (values)
about crime.
Another major problem with the conceptualization of fear of crime is the
generic reference of the term “crime.” Crime refers to a wide variety of
activities including violent personal crime, property crime, organized crime,
74 KENNETH F. FERRARO AND RANDY LAGRANGE

occupational crime, public order crime, political crime, etc. While there is
some value in having an overall indicator of fear of crime, it should be obvious
that the fear of being victimized varies by the type of crime considered. Most
research indicates that violent personal crimes, such as assault, robbery, and
rape are the most frightening.
A welcomed trend in research on fear of crime has been the tendency of
researchers to provide specific types of crimes for respondents to estimate their
fears. Most studies, however, possess only implicit referents which are usually
violent personal crimes. In order to get the most valid and reliable indicators
of fear of crime, it is best to specify the type of crime to the respondent rather
than leave it up to the respondent’s own inference. A conceptual reference of
crime is needed, and if the respondent is not provided with one, he or she will
select one. Unfortunately, the selection procedure will probably not be a
random function, thereby posing threats to the measurement properties of the
instrument. If omnibus estimates of fear of crime are desired, it would be far
more desirable to transform a set of individual items about specific crimes into
an unobserved variable either by weighted or unweighted mathematical
functions.* While we do not generally advocate the use of omnibus measures,
there may be specific reasons for wanting to do so-perhaps if one is interested
in the way a particular crime is related to the complex of residual categories of
crime (e.g., Warr, 1984).
Review of the Literature
The Appendix summarizes the results of a review of the empirical re-
search on fear of crime.3 The entries are presented in alphabetical order with
the author’s name and year of publication appearing in the left-hand column.
The center column lists the purported measure(s) of fear of crime utilized in
each study. Whenever possible, each fear of crime measure has been recorded
verbatim in the Appendix in order to preserve the operational integrity of the
original work. In the few instances when the article did not provide sufficient
information concerning the measurement procedures, the implied measure(s)
was included in the table. Finally, the right-hand column of the Appendix
notes if fear of crime was measured by a single indicator-or a series of single-
item indicators-or whether a composite measure of fear of crime was con-
structed.
Focusing attention first upon the far right-hand column, it is readily
apparent that fear of crime is frequently measured and analyzed as a single-
item indicator. Indeed, more than forty percent of the forty-six studies re-
viewed rely solely upon a single-item indicator of fear of crime. (Thirteen
additional studies, or 28.3 percent, employ more than one “fear” measure yet
analyze them individually rather than as a multiple-item construct.) The
 T H E MEASUREMENT O F FEAR O F CRIME 75

methodological adequacy of this type of measurement strategy is questionable

for certain types of analyses. In light of the substantial advance in sophistica-
tion of social science measurement, the desirability of multiple-item indices
over single-item indicators is well documented (Bohrnstedt and Borgatta,
1981:10).4This is especially true when one is attempting to measure abstract
theoretical constructs such as fear of crime (Miethe and Lee, 1984). Relating
this knowledge to the problem at hand, any purported single-item measure of
the rather complex concept of fear of crime must be viewed with some degree
of caution, and study results should be assessed accordingly.
It may appear that the suggestion to use multiple indicators contradicts
the suggestion to examine specific types of crime. These are really two separate
issues. Throughout this essay we note the value of measuring fear of specific
vpes of crime. This has been approached by a few researchers with one ques-
tion per type of crime (e.g., Warr, 1984). We feel that this is an acceptable,
though not ideal, practice. However, one could also use multiple indicators for
a given type of crime. The objectives of the latter strategy are to enhance the
psychometric properties of the measuring instruments and examine some of
the range of emotional reaction typically characterized as fear. Fear may
range from relatively diffuse states such as anxiety to relatively acute states
such as trauma. Very little attention has been given to a consideration of the
range of variation in emotional reaction. Unfortunately, when researchers
have used multiple measures to form indexes, they have most frequently
blended types of crime rather than degrees of fear.
Obviously, multiple-item indices are not a panacea to social measure-
ment. Unless properly constructed and tested, one cannot be assured that
composite indices and measurement scales possess appropriate psychometric
properties. Interestingly, of the thirteen studies reviewed in the Appendix
which utilize a fear of crime iqdex, only two report reliability coefficients (Lee,
1982b; Miethe and Lee, 1984). This raises an empirical dilemma regarding
the adequacy of the measurement process and the meaningfulness of study
results. As Cohen and Cohen (1975:372) have stated, “unreliability in a
partialled variable may yield grossly inaccurate results when it is ignored.” At
the very least researchers should heed the appeal of Bohrnstedt and Carter
(1971:143):
Our plea is for sociologists engaged in substantive research to confront the unreliability of
their measurement instruments , , . we do not feel i t is either unrealistic or unreasonable to
expect sociologists to recognize explicitly the error existent in their instruments and to
take this error into account in their analyses. At the very minimum, researchers ought to
report the reliability of their measuring instruments.

Lest the reader feel that we are unduly critical of researchers who have not
76 K E N N E T H F. F E R R A R O A N D RANDY LAGKANGE

reported reliabilities because of the lack of methodological sophistication of the

times, we would like to point out that nine of the eleven studies that use a n
index without reporting a reliability have been published since 1980.‘ Clearly,
then, the vast majority of the fear of crime research falls prey to criticisms of
its measurement sophistication and design.
Purported Measures of Fear of Crime
T h e main contention of this paper is that much of the fear of crime re-
search suffers from measurement problems. Conceptual cloudiness and in-
appropriate operationalization taints the majority of this literature thereby
distorting the meaning and the utility of the fear of crime concept. A careful
scrutiny of the item content of fear of crime indicators offers testimony to their
lack of conceptual clarity and specificity. A few examples should illustrate
this point.
Consider the following question: “How safe would you feel walking
alone at night in your neighborhood?” While such a question has been used to
measure fear of crime (e.g., Baker et al. 1983), it more accurately measures
the risk to self of walking alone at night in one’s neighborhood. This is not an
emotional reaction to crime, but rather a judgment about the likelihood of
criminal victimization for the individual. A person who says he or she would
not feel very safe may not be afraid at all, but simply aware of the relative
risk. Thus, such a person may avoid walking alone in their neighborhood at
night and not really manifest any fear of crime. A related question is “ H O W
likely it is that a person walking around here at night might be held u p or
attacked?” (e.g., Block, 1971; Erskine, 1974; Mirande, 1980). Note that the
respondent is asked to make a risk assessment of the community in general; the level
of personal fear is left unmeasured.
It should be assumed that measures which do not diflerentiate emotional reactions
from judgments are invalid measures. People have perceptions of their risk of
victimization; however, the perceived risk of victimization is vastly different
from the feeling of fear of victimization. Warr and Stafford (1983) provide
evidence that risk of victimization, in and of itself, is not even a strong pre-
dictor of fear of victimization.
-4nother measure commonly employed in fear of crime research is the
National Crime Survey (NCS) question: “HOWsafe do you feel or would you
feel being out alone in your neighborhood at night?” (e.g., Baumer, 1985;
Garofalo, 1979; Liska et al. 1982; Maxfield, 1984; Riger et al. 1978). Despite
the popularity of this item, it is, nonetheless, inherently flawed. Garofalo
(1979:82) identifies four fundamental problems with this question that detract
considerably from its utility: (1) the word “crime” is not even mentioned
leaving the thrust of the question more implicit than explicit;6 (2) the geo-
 T H E MEASUREMENT OF FEAR O F CRIME 77

graphical frame of reference is the neighborhood, which means different

things to different people; (3) the respondents are asked to think about their
perceived safety when alone at night in their neighborhood-there are few instances
when this actually occurs; and (4) the part of the question that asks “do you
feel or would you feel” mixes actual with hypothetical assessments of safety
which are not necessarily equivalent. There is a fifth problem with the NCS
question similar to that noted in the above example. That is, the item fails to
differentiate relatively objective risk judgments from emotional fears of crime,
thus rendering the question conceptually vague and largely invalid.’
Finally, the question “Is there any area right around here-that is,
within a mile-where you would be afraid to walk alone at night?” has be-
come a standard in the fear of crime literature. According to our review, it is
the most frequently employed single-item indicator and has been utilized in
ten different studies listed in the Appendix (Braungart et al. 1980; Clarke and
Lewis, 1982; Clemente and Kleiman, 1976, 1977; DeFronzo 1979; Erskine,
1974; Jeffords, 1983; Lebowitz, 1975; Lee, 1982a, 1982b).
Contrary to the common problems cited thus far, .this indicator does
venture away from purely judgmental, objective assessments of one’s per-
sonal risk of victimization to touch more upon fear. This is accomplished
through a subtle yet important shift in wording from the idea of “safety” to
feelings of being “afraid.” Unfortunately, the lack of specificity in the re-
mainder of the question overrides its apparent usefulness. For example,
several of the criticisms Garofalo (1979) levels against the NCS question have
equal applicability here: (1) the question, in itself, does not explicitly specify
“fear of what” (fear of crime? fear of traffic? fear of becoming lost?); (2) the
frame of reference is vaguely defined by the term “neighborhood”; (3) the
likelihood of being alone on the neighborhood streets at night is such an un-
likely event for most of us, especially for the elderly to whom much of this
research has been directed, that it fails to touch base with everyday life ex-
periences. In short, the continued use of this question as an indicator of fear
of crime is difficult to justify.
Figure 2 seeks to further clarify some of these measurement problems in
the context of our conceptual framework presented earlier. Applying the
general model presented in Figure 1, the selected examples in Figure 2 illu-
strate which crime perceptions are most fully tapped by different “fear of
crime” indicators. For instance, in the upper left-hand corner (cell A), the
question “DO you think that people in this neighborhood are safe inside their
homes at night?” (Clarke and Lewis, 1982) is a combination ofjudgments, or
risk assessments, and a general referent (i.e., “people in this neighborhood”).
Clearly the measure is conceptually distinct from the meaning which many
researchers attach to it and probably is equally inappropriate as a “surrogate”
78 KENNETH F. FERRARO AND RANDY LACRANCE

indicator of fear. By comparison, the lower right-hand corner of Figure 2

(cell F) comes closest to the conceptual domain commonly assumed to repre-
sent fear of crime. The personal level of reference is crossed with stated feel-
ings of fear. The series of indicators employed by Warr and Stafford (1983) to
measure the amount of fear for sixteen different types of victimization are
good examples and provide useful measures of fear of crime.

Figure 2
Examples of Crime Perceptions

Type of Perceptions
Cognitive Affective

Level of
Reference Judgments Values Emotions

General A. Do you think B. Choose the C . I worry a

that people in single most great deal
this neighbor- serious do- about the
hood are safe mestic prob- safety of my
inside their lem (from a loved ones
homes at list of ten) from crime
night? (Clarke that you and crimi-
and Lewis, would like to nals. (Lee,
1982) see govern- 1982a)
ment do
something
about. (Fur-
stenberg,
1971)

Personal D. How safe do E. Are you per- F. How afraid

you feel or sonally con- are you of be-
would you cerned about coming the
feel being out becoming a victim of (six-
alone in your victim of teen separate
neighborhood crime? (Jaeh- oJenses) in
at night? nig et al. your everyday
(Liska et al. 1981) life? (Warr
1982) and Stafford,
1983)
 THE MEASUREMENT OF FEAR OF CRIME 79

Discussion
Assuming we have adequately documented the weak conceptual develop-
ment and the pervasive measurement problems of the research on fear of
crime, some important implications arise. T o the extent that this general
body of literature is subject to questions of its sophistication and appropriate-
ness, what we “know” about the phenomenon-its causes, consequences,
geographic and demographic distribution-is also open to question. Another
way of approaching this issue is to ask: What are the substantive consequences
of using different measures of “fear of crime”? A few studies give some hints
of the potential problems when one generalizes to fear from an item that
actually measures a judgment or value.
To begin, research by Lee (1982a) reveals that the correlation between a
fear of crime measure (i.e., emotion)* and risk assessments of the community
(i.e., judgment) range in absolute value between .32 and .48. Even less en-
couraging is the correlation between the most frequently used measure in the
Appendix and Lee’s fear of crime measure: r = .28. Assuredly, these correla-
tions are statistically significant; but more importantly, they reveal thatfear of
crime andperceived risk ofcrime are only moderately related. Based on Lee’s research,
one could reasonably conclude that perceived risk of crime is a poor surrogate
of fear since it can, at best, explain only about twenty to twenty-five percent of
the variance in fear.
The works of Warr and Stafford (1983) and Warr (1984) offer additional
insight into this issue. These articles are based on data which assess both risk
of victimization and fear of victimization for each of sixteen different offenses.
(Thus, while most of Lee’s measures are omnibus, W a r ’ s are offense specific.)
Because the data are offense specific, correlations between fear and risk for
each type of crime were computed by Warr (1984:690). These correlations
(between fear and risk) range between .9 and - .6 for the sixteen offenses.
These results show more dramatically the problems with ( 1 ) lumping types of
crime into one measure and (2) using risk measures to show fear.
The results also indicate that subject characteristics affect the relationship
between fear and risk. As Warr (1984:694) notes, “ . . . perceived risk of
victimization does not have a uniform impact on fear among all age/sex
groups. ” Thus anyone willing to tolerate the measurement error inherent in
interpreting risk measures as fear should also realize that the error is not ran-
dom, but related to subject characteristics and probably community type and
test conditions as well.
In short, it appears that different measures of fear, or purported fear, of
crime yield inconsistent empirical relationships. Not only are the distinctions
outlined in Figure 1 helpful for conceptual purposes but meaningful in an
80 KENNETH F. FERRARO A N D R A N D S LAGRANGE

observable way as well. The practical implications of the measurement prob-

lems are also important. Theory development and testing are hindered. More
directly, recommendations for social policy may be hampered and specific
efforts to reduce fear of crime may be misguided.
For example, consider the quite common and altogether unstartling
observation that many people report feelings of apprehension when walking
the streets of their neighborhood alone at night. Measures of this caliber
partially ensure discovering elevated levels of fear simply by the dangerous
scenario presented. Images of bad-guys lurking in the streets are created, and
only the most courageous of souls would dare travel at night. Perhaps a more
realistic and more revealing question would be “Who wouldn’t be afraid to
walk the streets alone at night?” It is worthy of note that several studies have
found that even among “high fear” groups (e.g., the elderly), the oft-stated
feelings of fear are significantly reduced when the home is the locational refer-
ence rather than the neighborhood (Davis and Brody, 1979). In other words,
fear of crime may not be as problematic in the everyday lives of people as public
opinion polls typically suggest (LaGrange and Ferraro, 1986). The standard
recommendation for reducing fear-that is, increasing patrol strength on the
streets-may be less effective than many believe. Other security measures
such as replacing old locks, installing burglar alarms, or establishing com-
munity watch programs where residents keep an eye on their neighbors would
prove less expensive and perhaps more effective in allaying feelings of fear.
Our earlier admonition that global fear of crime measures mask important
differences in the degree of fear of specific crimes should not be ignored. Not
all criminal offenses pose similar levels of threat to the public, partly because
of differences in the likelihood of victimization (e.g., mugging versus kid-
napping), and partly because of variation in the seriousness of offenses (e.g.,
property destruction versus sexual assault). It is generally agreed that the
category of offenses called street crime, such as robbery, rape and aggravated
assault, evokes the greatest fear among the public. James Q. Wilson (1975:23)
writes that “predatory crime does not merely victimize individuals, it impedes
and, in the extreme case, even prevents the formation and maintenance of
community.” McIntyre (1967) warns us that street crime reduces social inter-
action and the level of mutual trust, while Garofalo and Laub (1978) speak of
its erosive effects on the “quality of life” of the community. Therefore, it is
not surprising that assessments of the public’s fear of crime generally have
been assessments of their fear of street crime.
However, not everyone shares equally high levels of fear of predatory
crime, nor is fear of predatory crime always the highest among the list of
victimization fears. For example, Sundeen and Mathieu (1976) discovered
 T H E MEASUREMENT O F FEAR O F CRIME 81

that residents in retirement communities (i.e., an age-segregated environ-

ment) tend to fear consumer fraud more than predatory crime. In contrast,
Wolf (1977) found that residents of the community at large (i.e., an age-
integrated environment) have a greater fear of burglary and vandalism.
Brodyaga et al. (1975) found that only murder ranks higher among the fears
of women than rape, providing an additional crime in the overall fear of
crime equation that rarely applies to men (Riger et al. 1978:278). Although
global single-item indicators of fear of crime may serve as a useful barometer
of the public’s concern in general, they do not speak to the complexities of
fear of crime as a social problem, nor do they immediately lend themselves to
specific policy implementation. Since vague measures often raise more ques-
tions than they answer, researchers are advised to be more vigilant in their
pursuit of appropriate fear of crime indicators.
Suggestions for Future Research
Recognizing these problems should issue a call for a new wave of research
that improves upon this weakness in research design. Though the type of re-
search we would consider valuable should be somewhat obvious, we would
like to, nevertheless, offer some specific suggestions for future research on this
topic.
First, measures of fear of crime should tap the emotional state of fear
rather than judgments or concerns about crime. The phrase “how afraid” is a
helpful way to examine this emotional reaction. Second, questions that at-
tempt to measure fear of crime should make explicit reference to crime. Many
researchers have based their studies upon implied meanings which are prob-
ably not valid indicators. Third, as mentioned earlier, general referents about
crimes are often vague; if there is no clear crime referent, one could certainly
not expect respondents’ fear reactions to be reliable or valid. Thus, we recom-
mend that specific victimizations or categories of victimizations be used to
assess an individual’s fear reactions.’ This procedure should enhance object
consistency upon which the fear reactions are predicated. Fourth, questions
intended to measure fear of crime should be stated in a nonhypothetical for-
mat. More than ten different questions cited in the Appendix ask the respon-
dent to estimate how they would feel under certain circumstances. We suggest
that researchers should avoid the use of the word “would” in questions at-
tempting to measure fear of crime. Rather, it is better to obtain specific re-
ports about how individuals feel in everyday situations. Fifth, the qualifying
phrase, “in your everyday life” brings a touch of reality to the questions re-
garding fear of crime. Respondents can better relate to this type of question
than more abstract, hypothetical, or perhaps unlikely situations such as “when
82 KENNETH F. FERRARO A N D RANDY LAGRANGE

walking alone at night in your neighborhood.” If we are truly interested in

measuring fear of crime, our efforts will best be targeted toward examining
the world of everyday life, not hypothetical situations.
With these suggestions in mind, the careful reader will already note that a
few research studies meet all, or at least most, of these criteria. We would
like to offer such studies as exemplars for the measurement of fear of crime.
Two studies conducted during the mid- 1970s demonstrate the value of assess-
ing fear of specific victimizations rather than fear of crime in general. Sundeen
and Mathieu (1976;) asked their respondents to indicate how fearful they were
of being victimized for each of four specific offenses, and Lalli and Savitz (1976)
used 13 offenses to measure fear. These investigations saw the importance of a
crime referent for assessing fear. Lalli and Savitz (1976:405) made use of a
“fear ladder” for measuring fear of victimization for each of the offenses.
During the personal interview a card displaying an eleven-step ladder ranging
from a bottom rung indicating “no fear” (0) to a top rung indicating “extreme
fear” (10) was used.
The works of Warr and Stafford (1983) and Warr (1984) are more recent
examples of appropriate measurement coupled with sound scientific analyses.
In these studies, respondents were asked to describe how afraid they were of
becoming a victim to each of sixteen different criminal offenses. The questions
tapped the emotional reaction of fear in a direct way and were based on specific
victimizations. Even a casual glance at their research findings will show the
crudeness of attempting to only apply omnibus measures for fear of crime.
Fear reactions vary substantially by the perceived seriousness of the crime and
the individual’s judgment of the risk of victimization. These studies provide
better measures of fear of victimization than most of the other studies and are
good baselines for further analyses. As others replicate and extend their work,
we will be in a much better position to understand the etiology and reduction
of fear of crime.
Theoretical Epilogue
It should be readily apparent that the methodological problems in re-
search on fear of crime have obscured theoretical development in this subject.
Researchers’ proclivity to term various types of perceptions about crime (i.e.,
cells A, B, D, and E in Figure 1) as the fear of crime has all too frequently re-
sulted in misspecification of models a n d o r a confounded variable problem.
We exhort researchers to use the categorical descriptions in Figure 1, or
similar terms, rather than the omnibus concept of fear of crime. This will
hopefully lead to more fruitful discussions of the relationships between these
concepts as well as other theoretically relevant concepts (e.g., perceived
seriousness of a crime). A few scholars (e.g., Furstenberg, 1971; Warr and
 T H E MEASUREMENT OF FEAR OF C R I M E 83

Stafford, 1983; Warr, 1984) have engaged in this endeavor and their work
deserves serious consideration.

APPENDIX
Summary Table of Fear of Crime Studies

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

Baker et al. 1. How safe would you feel Additive index

( 1983) walking alone at night in (no reported
your neighborhood? reliability)
2. Think of the worst area
within a mile of your
house. How safe would
you feel walking alone at
night in this area?
Balkin 1. How safe do you feel Single item
(1979) being out alone in your
neighborhood during the
day?
Baumer 1. How safe do you feel or Single item
(1985) would you feel being out
alone in your neighbor-
hood at night?
Block 1. How likely is it that a Single item
(1971) person walking around
here at night might be
held up or attacked?
Braungart et al. 1. Is there any area right Single item
( 1980) around here-that is,
within a mile-where
you would be afraid to
walk alone at night?
Clarke and Lewis 1. Is there any area right Single item
(1982) around here-that is,
within a mile-where you
84 KENNETH F. FERRARO A N D RANDY LACKANGE

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

would be afraid to walk

alone at night?
2. Do you think that people
in this neighborhood are
safe inside their homes at
night?
3. Respondents chose their
three most serious prob-
lems from a list of ten
items with o n e b e i n g
“ f e a r of b e i n g vic-
timized. ”
Clemente and 1. Is there any area right Single item
Kleiman a r o u n d here-that is,
(1976) w i t h i n a mile-where
you would be afraid to
walk alone at night?
Clemente and 1. Is there any area right Single item
Kleiman around here-that is, within
(1977) a mile-where you would be
afraid to walk alone at night?
Cutler 1. Is there any area right Single item
( 1980) around here-that is, within
a mile-where you would be
afraid to walk alone at night?
DeFronzo 1. Is there any area right Single item
(1 979) around here-that is, within
a mile-where you would be
afraid to walk alone at night?
Erskine 1. In the past year, do you feel Single item
(1974) the crime rate in your
neighborhood has been in-
creasing, decreasing, or has
 T H E MEASUREMENT OF FEAR OF CRIME 85

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

it remained about the same

as it was before?
2. Would you say that there is
more crime in this com-
munity now than there was
five years ago, or less?
3. Is there more crime in this
area than there was a year
ago, or less?
4. Would you say there is more
crime or less crime in this
area than there was a year
ago?
5. Is there any area around
here-that is, within a mile
-where you would be afraid
to walk alone at night?
6. Compared to a year ago, do
you personally feel more
worried, less worried, or not
much different about your
personal safety on the streets?
7. Compared to a year ago, are
you personally more worried
about violence and safety on
the streets, less worried, or
do you feel about the same
as you did then?
8. How likely is it that a person
walking around here at night
might be held up or attacked?
9. What about walking alone
86 KENNETH F. FERRARO AND RANDY LAGRANGE

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

(in your neighborhood)

when it is dark-how safe do
(would) you feel?
10. Have there been any times
recently when you might
have wanted to go some-
where in town but stayed
home instead because you
thought it would be unsafe
to go there?
11. Compared to a year ago, do
you feel more afraid and un-
easy on the streets today,
less uneasy, or not much
different from the way you
felt a few years ago?
12. Do you feel it is safe to walk
in the streets alone in your
neighborhood?
Furstenberg 1. Respondents selected single Additive index of
(1971) most serious domestic prob- eight different
lem from a list of ten they offenses (no re-
would like to see the govern- ported reliability)
ment do something about.
2. Respondents rated the pos-
sible risk of their being
victimized on eight different
offenses.
Garofalo 1. How safe do you feel or Single item
(1979) would you feel being out
alone in your neighborhood
at night?
Hartnagel 1. The degree of personal Single item
 THE MEASUREMENT OF FEAR OF CRIME 87

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

(1979) safety the respondents felt in

their own neighborhoods.
2. The safety of the city as a
whole as respondents rated
it.

Hepburn and 1. Have you been afraid that Single item

Monti someone will hurt you or
(1979) bother you at school?
Hunter and 1. Respondents were asked to Additive index of
Baumer estimate the risk of being the five questions (no
(1982) victim of robbery, assault, reliability reported)
and theft (street crime).
2. Respondents were asked
how worried they were of
street crime.
Jaehnig et al. 1. Are you personally con- Single item
(1981) cerned about becoming a
victim of crime?
Janson and Ryder 1. A question asking respon- Additive index of
( 1983) dents what are their three the three items (no
greatest problems. reliability reported)
2. Is living in a high crime
neighborhood a serious
problem to you?
3. Does crime in the streets
cause you any special diffi-
culties in getting around?
Jeffords 1. Is there any area within one Single item
( 1983) mile of your home where
you would be afraid to walk
alone at night?
88 K E N N E l H F. FERRARO A N D R A N D Y LAGRANGE

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

2. Would you be afraid to walk

alone within one block of
your home at night?
3. Are you afraid to be alone at
night?
Kennedy and 1. How safe do you feel walk- Single item
Krahn ing alone in your neighbor-
( 1984) hood at night?
Kennedy and 1. How safe do you feel walk- Single item
Silverman ing alone in your neighbor-
( 1985) hood at night?
Lalli and Savitz 1. Adults were asked to rate the Single item
(1976) intensity of their “fear,”
“
concern,” or “worry” of
being victimized to each of
twelve different offenses and
o n e specific d a n g e r o u s
locale.
2. Youth were asked a similar
question involving eight
different offenses.
Lawton and Yaffe 1 . A scale derived from ten Additive index (no
( 1980) closed-ended and sixteen reliability reported)
open-ended questions con-
cerning personal anxiety
over crime (individual items
not reported).
Lebow it z 1. Is there any area right Single item
(1975) around here-that is, within
a mile-where you would be
afraid to walk alone at night?
 T H E MEASUREMENT OF FEAR O F C R I M E 89

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

Lee 1. Is there any area near your Single item

(1982a) home-that is, within a
mile-where you would be
afraid to walk alone at night?
2. How about during the day?
Is there any area near your
home where you would be
afraid to walk alone in the
daytime?
3. There are times during the
night when I’m afraid to go
outside.
4. If someone assaulted me, I
could protect myself.
5. In terms of crime, do you
think that your neighbor-
hood is a very safe place in
which to live?
6. When I am away from home,
I worry about the safety of
my property.
7. I worry a great deal about
my personal safety from
crime and criminals.
8. I worry a great deal about
the safety of my loved ones
from crime and criminals.
9. I worry a great deal about
the safety of my property
from crime and criminals.
10. Even in my home, I’m not
90 KENNETH F. FERRARO AND RANDY LAGRANGE

APPENDIX (continued)
~~ ~ ~

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

safe from people who want

to take what I have.
11. There is a reason to be
afraid of becoming a victim
of crime in my community.
12. My neighborhood is a very
safe place in which to live.
13. Please tell us whether crime
or fear of crime has been a
serious problem for you in
the past year.
Lee 1. Is there any area near your Single item and
(1982b) home-that is, within a additive index of
mile or so-where you would seven items (alpha
be afraid to walk alone at of .842)
night?
2. a) When I am away from
home, I worry about the
safety of my property.
b) I worry a great deal about
my personal safety from
crime and criminals.
c) I worry a great deal about
the safety of my loved
ones from crime and
criminals.
d) I worry a great deal about
the safety of my property
from crime and criminals.
e) Even in my own home,
I’m not safe from people
who want to take what I
have.
 T H E MEASUREMENT O F FEAR OF C R I M E 91

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

r) There is reason to be afraid

of becoming a victim of
crime in my community.
g) Crime or fear of crime has
been (no problemla prob-
lem/a serious problem) for
me in the past year.
Lewis and 1. How much of a problem Single item
Max field does (burglary, robbery,
( 1980) assault, and sexual assault)
represent?
2. Estimate on a scale of one to
ten the likelihood of being
victimized for each of the
above four offenses.
3. How safe do you feel in your
neighborhood at night?
Lindquist and 1. Respondents were asked if Single item
Duke fear of crime was a personal
(1982) problem.
Liska et al. 1. How safe do you feel or Single item
( 1982) would you feel being out
alone in your neighborhood
at night?
Maxfield 1. How safe do you feel, or Single item
( 1984) would you feel, being out
alone in your neighborhood
at night?
Miethe and Lee 1. a) I worry a great deal about Additive index of
( 1984) my personal safety from violent crime
crime and criminals. (alpha of .753)
b) I worry a great deal about
92 KENNETH F. FERRARO A N D RANDY LAGRANGE

APPENDIX (continued)
~

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

the safety of my loved ones

from crime and criminals.
2. a) I worry a great deal about Additive index of
the safety of my property property crime
from crime and criminals. (alpha of .745)
b) When I am away from
home, I worry about the
safety of my property.
Mirande 1. How likely is it that a person Single item
( 1980) walking around at night
might be held up or attacked?
Norton and 1. Fearfulness of vandalism. Additive index of
Courlander 2. Safety in neighborhood eight items derived
( 1982) during the day. from factor analy-
3. Safety in neighborhood at sis (no reliability
night. reported)

4. Concern over home being

broken into while away.
5. Security in home at night.
6. Fear of crime preventing
you from action.
7. Worry about assault
8. Worry about robbery.
Ollenburger 1. Do you secure your home Additive index
(1981) and other structures on the (item-level corre-
premises from burglary? lations reported)
2. Are you or any members of
your household afraid to go
out in your neighborhood
after dark?
 THE MEASUREMENT OF FEAR OF CRIME 93

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

3. All things considered, would

you say that (your town) is
safer from crime, about as
safe, or not as safe, as it was
a few years ago?
Pollack and 1. A “fear of property loss” Additive indices
Patterson scale was constructed from (no reliabilities
(1980) eleven items (only two of the reported)
items were reported as
examples).
2. A “fear of personal assault”
scale was constructed from
sixteen items (only two items
were reported).
Riger et al. 1. How safe do you feel being Single item
(1978) out alone in your neighbor-
hood at night?
2. How often do you think of
your own safety?
3. Do you ever feel afraid that
someone might deliberately
harm you?
4. T h e last time you worried
about rape, how afraid or
scared did you feel?
5. Do you fear for the safety of
others who live in your
home?
6. Indicate the extent of your
worry to the following ac-
tivities: (twelve different
activities such as (1) being
94 KENNETH F. FERRARO A N D RANDY LAGRANGE

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

home alone after dark and

(2) riding with male strangers
alone after dark).
Shotland et al. 1. Two hypothetical situations Single item
(1979) (one concerning rape, the
other burglary) were pre-
sented to college females
to measure their fear.
Smith and 1. Respondents were asked A latent variable
Patterson how likely they thought it was created with
(1984) was that they would be the three indicators
vicims of a) robbery, b) (LISREL)
burglary, and c) vandalism
during the next year.
Stafford and 1. Are you afraid to go out into Single item
Galle (1 984) your neighborhood after
dark by yourself?
Sundeen and 1. Respondents were asked to Single item
Mathieu indicate how fearful they
(1976) were of being victimized of
each of four specific offenses:
burglary, robbery, car theft,
and consumer fraud.
Taylor et al. 1. How safe would you feel Additive index
( 1984) being out alone in your (no reliability
neighborhood during the reported)
day?
2. How safe would you feel
being out alone in your
neighborhood at night?
Thomas and 1. This city’s downtown sec- Additive index
Hyman tion just isn’t safe at night (no reliability
(1977) anymore. reported)
 T H E MEASUREMENT O F FEAR OF C R I M E 95

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

-

2. I don’t really feel that the

threat of criminal behavior
is any greater today than in
the past.
3. The danger of becoming the
victim of a criminal offense
seems to be lower in this city
than in many parts of the
country.
4. I avoid shopping in the
downtown section of this city
because of the crime prob-
lem.
5. During recent years I have
become more afraid of being
victimized by criminals than
I ever was before.
My family and I feel reason-
ably safe and secure in the
community.
Crime is such a problem that
this city is not a safe place
to raise children.
The threat of crime has be-
come so great that nobody
can feel safe in his own home
anymore.
9. Crime has become such a
problem in my neighborhood
that I’m afraid to go out at
night.
Warr 1. Respondents were asked to Single item
(1984) describe how afraid they
96 KENNETH F. FEKRARO AND RANDY LAGRANGE

APPENDIX (continued)

Study Purported Measure (s) of Fear of Crime Use of Measure(s)

were of becoming a victim

to each of sixteen different
offenses.
Warr and 1. Respondents were asked to Single item
Stafford describe how afraid they
( 1983) were of becoming a victim
to each of sixteen different
offenses.
Wiltz 1. What is the likelihood of Single item
( 1982) your being victimized in the
future?
2. What is the possibility of
being burglarized?
3. What is the possibility of
having your purse snatched?
Yin 1. How safe do you feel it is to Additive index
(1982) be out alone in your neigh- (no reliability
borhood during the clay? reported)
2. How safe do you feel it is to
be out alone in your neigh-
borhood at night?

ENDNOTES

*This paper was originally prepared for presentation at the annual meeting of the Southern
Sociological Society, Charlotte, N . C . , April, 1985. We extend our appreciation to Margaret
Lyons and Ranie Hufiinan for library research assistance and to Kenneth Bechtel, Linda Ferraro,
Joseph Harry, William Minor, Mark Warr, and two anonymous reviewers for helpful comments
on earlier drafts of this paper. We thank Amber Oldham for manuscript preparation. Please
direct all correspondence to Kennt-th F. Ferraro, Department of Sociology and Center lor Gov-
ernmental Studies, Northern Illinois University, DeKalb, IL 601 15.
'Most of the articles reviewed include the phrase "fear of crime" in the title. In some cases
thr conceptual inconsistency is more subtle. For example, Mullen and Donnermeyer (1985)
 T H E MEASUREMENT O F FEAR O F C R I M E 97

attempt to focus on perceived safety from crime in their paper entitled “Age, trust, and perceived
safety from crime in rural areas.” However, in their attempt to compare their findings to pre-
vious studies on fear of crime, the authors imply that the fear of crime and perceived safety from
crime concepts are simply opposite of each other. While it might be argued that fear of crime is
the opposite of afceling of safety, the authors mistakenly assume that fear of crime is the opposite
of a cognitive asscsrmtnt of safety as measured in their research.
’The work of Smith and Patterson (1984), applying a MIMIC model to the study of vic-
timization, is a good work to consider on this topic. It does not deal with fear of crime as pre-
viously defined, but rather personal judgments about crime (cell D in Figure One). It is not an
exemplar for the measurement of fear of crime, but the method of analysis deserves considera-
tion. The authors use three indicators to create a latent variable with LISREL, called perceived
risk, and then model the relationship between the indicators, the latent variable, and the exo-
genous variables. Clearly, confirmatory factor analytic models are advantageous for a number
of reasons, especially when the variables of interest are theoretical constructs
’The studies reviewed here were derived from a complete search of the Social Science Index
as well as a number of pieces located during the process of researching the subject. Although this
review is by no means exhaustive, it is extensive and includes the main empirical works that
occupy the nucleus of the fear of crime literature.
‘Smith (1981:296) has examined the interplay between the operational reliability of a
variable and the number of items used, contending that “not only is reliability increased by using
more than one item, but also validity tends to increase with larger numbers of items used.” Thus,
for any given variable (y), there is an underlying true score component (t) and an error compo-
nent (e). The influence of item-specific errors generally decreases as items are added to the scale
since error terms have a tendency to cancel each other out (Lord and Novick, 1968).
’At least Norton and Courlander (1982) used factor analysis to select the items to be used
in their index.
‘In all fairness we should note, as Garofalo (1979) does, that the introduction to the section
in which the NCS question appears contains the following statement: “Now I’d like to get your
opinion about crime in general.” Hence, the location of the “fear” question in relation to the
introductory statement is important and will largely influence whether the respondents are likely
to be thinking specifically about crime and criminalization at the time that they answer it. For
additional information on the NCS, consult the critique by Block and Block (1984).
’Despite the fact that the item has been repeatedly criticized, researchers continue to use
it (e.g., Baumer, 1985; Maxfield, 1984).
‘The fear of crime variable selected on the basis of face validity is number 7 in the Appen-
dix under Lee (1982a). Risk assessments are best measured with items 5 and 11 in the same
section of the Appendix.
‘If measures of fear of specific victimizations are not available or desirable, at least cate-
gories of victimization should be used as referents. For instance, Miethe and Lee (1984) have
argued that it is essential to differentiate violent crime from property crime when studying fear
of crime.

REFERENCES

Baker, Mary H . , Barbara C. Nienstedt, Ronald S. Everett, and Richard McCleary

1983 “Impact of a crime wave: Perceptions, fear and confidence in the police.” Law and
Society Review 17:319-335.
Balkin, Steve
1979 “Victimization rates, safety, and fear of crime.” Social Problems 26:343-358.
98 KENNETH F. FERRARO AND RANDY LAGRANGE

Raumer, Terry 1,.

1985 “Testing a general model of fear of crime: Data from a national sample.” Journal of
Research in Crime and Delinquency 22:239-255.
Beardwood, Roger
1968 “The new Negro mood.” Fortune 77 (January):146-151.
Block, Carolyn Rebecca, and Richard L. Block
1984 “Crime definition, crime measurement, and victim surveys.” Journal of Social Issues
40:137-160.
Block, Richard L.
1971 “Fear of crime and fear of police.” Social Problems 19:91-101.
Bohrnstedt, George W., and Edgar F. Borgatta
1981 Social Measurement: Current Issues. Beverly Hills, Calif.: Sage Publications.
Bohrnstedt, George W., and T. M . Carter
1971 “Robustness in regression analysis.” Pp. 118-146 in Herbert L. Costner (ed.),
Sociolo+al Methodology 1971. San Francisco, Calif.: Jossey-Bass.
Braungart, Margaret M . , Richard G. Braungart, and William J. Hoyer
1980 “Age, sex, and social factors in fear of crime.” Sociological Focus 13:55-56.
Brodyaga, Lisa, Margaret Gates, Susan Singer, Marna Tucker, and Richardson White
1975 Rape and Its Victims: A Report for Cities, Health Facilities, and Criminal Justice
Agencies. Washington, D.C.: U.S. Government Printing Oflice.
Clarke, Alan H., and Margaret Lewis
1982 “Fear of crime among the elderly.” British Journal of Criminology 22:49-62.
Clement, Frank, and Michael B. Kleiman
1976 “Fear of crime among the aged.” Gerontologist 16:207-210.
1977 “Fear ofcrime in the United States: A multivariate analysis.” Social Forces 56519-531.
Cohen, Jacob, and Patricia Cohen
1975 Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. New
York: Wiley.
Cutler, Stephen J.
1980 “Safety on the streets: Cohort changes in fear.” International Journal of Aging and
Human Development 10:373-384.
Davis, Linda J., and Elaine M . Brody
1979 Rape and Older Women. U.S. Department of Health, Education and Welfare.
Washington, D.C.: U.S. Government Printing Ofice.
DeFronzo, James
1979 “Fear of crime and handgun ownership.” Criminology 17:331-339.
DuBow, Frederic, Edward McCabe, and Gail Kaplan
1979 Reactions to Crime: A Critical Review of the Literature. Washington, D.C.: National
Institute of Law Enforcement and Criminal Justice, U . S . Government Printing
Ofice.
Erikson, Kai T .
1976 Everything in Its Path. New York: Simon and Schuster.
Erskine, Hazel
1974 “The polls: Fear of violence and crime.” Public Opinion Quarterly 38:131-145.
Furstenberg, Frank F., Jr.
1971 “Public reaction to crime in the street.” American Scholar 40:601-610.
Gallup, George H .
1982 The Gallup Report, Report Number 198. Princeton, N.J.: The Gallup Poll.
 T H E MEASUREMENT O F FEAR O F C R I M E 99

Garofalo, James
1979 “Victimization and the fear of crime.” Journal of Research on Crimc and Dclin-
quency 16:OO-97.
Garofalo, James, and John Laub
1978 “The fear of crime: Broadening our perspective.” Victimology 3:242-253.
Goffman, Erving
1971 Relations In Public. New York: Harper Colophon Books.
Hartnagel, Timothy F.
1979 “The perception of fear ol‘ crime: Implications for neighborhood cohesion, stxial
activity, and community affect.” Social Forces 58:176-193.
Hepburn, John R., and D. J. Monti
1979 “Victimization, Fear of Crime, and Adaptive Responses among High School Sru-
dents.” Pp. 121-132 in William H. Parsonage (ed.), Perspectives on Victimoloyy.
Beverly Hills, Calif.: Sage Publications.
Hunter, Albert, and Terry L. Baumer
1982 “Street traffic, social integration, and fear of crime.” Sociological Inquiry 52:
122- 131.
Jaehnig, Walter B., David H. Weaver, and Frederick Fico
1981 “Reporting crime and fear of crime in three communities.” Journal of Comrnunica-
tion 31:88-96.
Janson, Philip, and Louise K. Ryder
1983 “Crime and the elderly: The relationship between risk and fear.” Gerontologist
23:207-212.
Jeffords, Clifford R.
1983 “The situational relationship between age and the fear of crime.” International
Journal of Aging and Human Development 17:103-111.
Kennedy, Leslie W., and Harvey Krahn
1984 “Rural-urban origin and fear ofcrime: The case for ‘rural baggage’.” Rural Sociology
49: 247-260.
Kennedy, Leslie W., and Robert A. Silverman
1985 “Significant others and fear of crime among the elderly.” International Journal of
Aging and Human Development 20:241-256.
LaGrange, Randy L., and Kenneth F. Ferraro
1986 “The elderly’s fear of crime: How serious is it?” Paper presented at the Annual
Meeting of the Southern Sociological Society, New Orleans, La.
Lalli. Michael, and Leonard Savitz
1976 “Fear of crime in the school enterprise and its consequences.” Education and Urban
Society 8:401-416.
Lawton, M. Powell, and Sylvia Yaffe
1980 “Victimization and fear of crime in elderly public housing tenants.” Journal of
Gerontology 35: 768-779.
Lebowitz, Barry
1975 “Age and fearfulness: Personal and situational factors.” Journal 0 1 Gerontology
30:696-700.
Lee, Gary K.
1982a “Sex ditkrences in fear of crime among older people.” Research and Aging 4:284-298.
1982b “Residential location and fear of crime among the elderly.” Kural Sociology 47:
655-669.
100 K E N N E T H F. F E R R A R O AND RANDY LAGRANGE

Lewis, D. A , , and Michael G. Maxfield

1980 “Fear in the neighborhoods: An investigation of the impact of crime.” Journal of
Research on Crime and Delinquency 17: 160-189.
Lindquist, John H . , and Janice M . Duke
1982 “ T h e elderly victim at risk: Explaining the fear-victimization paradox.” Criminology
20: 1 15- 126.
Liska, Allen E., Joseph Lawrence, and Andrew Sanchirico
1982 “Fear of crime as a social fact.” Social Forces 60:760-770.
Lord, Frederic M . , and Melvin R. Novick
1968 Statistical Theories of Mental Test Scores. Reading, Mass.: Addison-Wesley.
Maxfield, Michael G .
1984 “ T h e limits of vulnerability in explaining fear of crime: A comparative neighbor-
hood analysis.” Journal of Research in Crime and Delinquency 21 :233-‘250.
McIntyre, Jennie J .
1967 “Public attitudes toward crime and law enforcement.” Annals 374:34-46.
Miethe, Terance, and Gary R. Lee
1984 “Fear of crime among older people: A reassessment of the predictive power ofcrime-
related factors.” Sociological Quarterly 25:397-415.
Mirande, Allredo
1980 “Fear of crime and fear o f the police in a Chicano community.” Sociology and
Social Research 64:528-541.
Mullen Robert E., and Jest-ph F. Donnermeyer
1985 “Age, trust, a n d perceived safety from crime in rural areas.” Gerontologist 25:
237-242.
Norton, Lee, and Michael Courlander
1982 “Fear o f crime among the elderly: T h e role 0 1 crime prevention programs.” Geron-
tologist 22:388-993.
Ollenburger, Jane C .
1981 “Criminal victimization and fear o f crime.” Research on Aging 3: 101-1 18.
Pollack, Lance, and Arthur H . Patterson
1980 “Territoriality and fear of crime i n elderly and nonelderly homeowners.” Journal 0 1
Social l ’ s y c h o l o ~ 111:119-129.
~
Riser. Stephanie, Margaret Gordon, and Robert Le Baily
1978 “Women’s lear of crime: From blaming to restricting the victim.” Victimology
3:274-284.
Sarnolf, Irving, and Philip C . Zimbardo
1961 “Anxiety, tear, and social affiliation.” Journal o f Abnormal Social Psychology
62: 356-363.
Selyc, Hans
1956 The Stress o f Life. New York: McCraw-Hill.
1974 Stress Without Distress. Philadelphia, Penn.: J . B. Lippincott.
Shotland, R . , Scott Hayward, Carlotta Young, Margaret Signorella, Kenneth Mindingall,
Jolin Kennedy, Michael Rovine, a n d Edward Danowitz
1979 “Fear o f crinii, i n residential communities.” Criminology 17:34-45.
Silberman, Charles
1981 “Fear.” Pp. 5-21 in Robert C. Culbertson and Mark R. Tezak (rds.), Order Under
Law: Readings in Criminal Justice. Prospect Heights, 111.: Waveland Press, Inc.
Skogan, Wesley G . , and Michael G . Maxfield
1981 Coping- with Crime: Individual and Neighborhood Diffcrcnces. Beverly Hills, Calif.:
Sage.
 T H E MEASUREMENT O F FEAR O F C R I M E 101

Smith, Douglas A,, and E. Britt Patterson

1984 “Applications and a generalization of MIMIC models to criminological research.”
Journal of Research in Crime and Delinquency 21:333-352.
Smith, Herman W .
1981 Strategies of Social Research. Second edition. Englewood Cliffs, N.J.: Prentice-Hall,
Inc.
Statlord, Mark C., and Omer P. Galle
1984 “Victimization rates, exposure to risk, and fear of crime.” Criminology 22: 173-185.
Stagner, Ross
1981 “Stress, strain, coping, and defense.” Research on Aging 3:3-32.
Sundeen. Richard A,, and James T. Mathieu
1976 “The urban elderly: Environments of fear.’’ Pp. 51-66 in Jack Goldsmith and
Sharon S. Goldsmith (eds.), Crime and T h e Elderly: Challenge and Response.
Lexington, Mass.: Lexington Books
Taylor, Ralph B., Steven D. Gottfredson, and Sidney Brower
1984 “Block crime and fear: Defensible space, local social ties, and territorial functioning.”
Journal of Research in Crime and Delinquency 21:303-331.
Thomas, C . N., and J . M. Hyman
1977 “Perceptions of crime, fear of victimization, and public perceptions of police per-
formance.” Journal of Police Science and Administration. 5:305-317.
TIME
1985 “ U p in arms over crime.” T I M E 125: 28-34.
U.S. Government
1967 The President’s Commission on Law Enforcement and Administration of Justice:
The Challenge of Crime in a Free Society. Washington, D.C.: U.S.Government
Printing Oflice.
Warr, Mark
1984 “Fear of victimization: Why are women and the elderly more afraid?” Social Science
Quarterly 65:68 1-702.
Warr, Mark, and Mark Stafliird
1983 “Fear of victimization: A look at the proximate causes.” Social Forces 61: 1033-1043.
Wilson, .James Q.
1975 Thinking About Crime. New York: Vintage Books.
Wiltz, C . .J.
1982 “Fear of crime, criminal victimization and clderly blacks.” Phylon 43:283-294.
Wolf, Robert
1977 “An aid to designing prevention programs.” Police Chief 44:27-29.
Yin, Peter
1982 “Fear of crime as a problem for the elderly.” Social Problems 30:240-245.
1980 “Fear of crime among the elderly: Some issues and suggestions.” Social Problems
27:492-504.