Directory UMM :Data Elmu:jurnal:A:American Journal of Police:Vol15.Issue1.1996:
American Journal of Police, Vol. XV, No. 1 1996
3
RACE, COMMUNITY CONTEXT AND
CONFIDENCE IN THE POLICE
Liqun Cao
Eastern Michigan University
James Frank
Francis T. Cullen
University of Cincinnati
Although American citizens have deep suspicions about state
power (Bayley, 1976; Lipset and Schneider, 1983), studies show that the
public generally holds positive attitudes toward the police (Apple and
O’Brien, 1983; Dean, 1980; Erez, 1984; Furstenberg and Wellford, 1973;
Hindelang, 1974; White and Menke, 1978; Zamble and Annesley, 1987).
Even so, these sentiments are not displayed equally across all sectors of
the social order. In particular, minorities appear to evaluate law
enforcement less favorably (Albrecht and Green, 1977; Bordua and Tifft,
1971; Carter, 1985; Decker and Wagner, 1981; Hindelang, 1974; Jacob,
1971; Jefferson and Walker, 1993; Parks, 1984; Peek et al., 1981; Percy,
1980; Reasons and Wirth, 1975; Scaglion and Condon, 1980; Skogan,
1978; but see Brandl et al., 1994). In response to this uneasy relationship,
many of the current community policing ventures have been undertaken
in minority neighborhoods in hopes of building residents’ confidence in
the police and willingness to be coproducers in the fight against crime
(Goldstein, 1987; Skolnick and Bayley, 1988:26; Wycoff, 1988).
Although research has consistently found that African-Americans
(and other minorities) view the police less favorably than whites, the
existing literature is limited in two ways. First, many studies report the
relationship between single variables of interest and confidence in the
police with no controls introduced (Albrecht and Green, 1977; Carlson
and Sutton, 1979; Flanagan and Maguire, 1992; Hindelang, 1974;
Jefferson and Walker, 1992, 1993; Klyman and Kruckenberg, 1974;
Komanduri et al., 1990; Mirande, 1980). Second, multivariate studies
4
American Journal of Police, Vol. XV, No. 1 1996
generally include only a limited range of variables in the analysis. Some
studies, for example, include controls only for other socio-demographic
variables (e.g. age, gender, income) (Apple and O’Brien, 1983). Even
more elaborate studies consider only a limited number of
nondemographic factors, such as contact with the police (Dean, 1980;
Mastrofski, 1981; Scaglion and Condon, 1980) and having been a crime
victim (Brandl and Horvath, 1991; Koenig, 1980; Percy, 1986). It is
possible, therefore, that these models are misspecified – that the effects of
race on police attitudes might be accounted for by factors not
incorporated into the analysis.
In this regard, the current study considers the impact of a range of
variables on confidence in the police. Some of these variables have been
examined infrequently, if at all, in prior research. The analysis includes,
for example, measures of crime experiences and of conservative political
orientation. Building on the research tradition that suggests that crimerelated reactions are shaped by the social context that surrounds
individuals (see, for example, Skogan, 1990), we also assess the effects of
neighborhood disorder and informal collective security on confidence in
law enforcement. In particular, we explore whether these contextual
conditions help to explain the relationship between race and confidence
in the police.
Two aspects of “crime experiences” are investigated: fear of
crime and having been a crime victim. The assumption is that when crime
is more salient in a person’s life, confidence in the police is undermined.
Previous research has found that victimization experiences tend to
increase negative attitudes toward the police (Apple and O’Brien, 1983;
Homant et al., 1984; Koenig, 1980; Parks, 1984; Smith and Hawkins,
1973).
Second, we investigate whether crime ideology influences
assessments of the police. We hypothesize that, since support for the
police is often a plank in “law and order” conservative political
platforms, a conservative crime ideology would enhance confidence in
the police. Some evidence exists to support this hypothesized relationship
(Larsen, 1968; Zamble and Annesley, 1987).
Third, we explore whether perceptions of community social and
physical disorder – sometimes called “incivility” – lessen citizens’
confidence in the police (see Covington and Taylor, 1991; Lewis and
Salem, 1986; Skogan, 1987). It is plausible that social disorder (e.g. noisy
neighbors, loitering by rowdy teenagers) and physical disorder (e.g.
graffiti, deteriorating property) send a message that law enforcement has
lost control over or consciously abandoned the community. In short, the
“broken windows” are not going to be fixed by the police (Wilson and
American Journal of Police, Vol. XV, No. 1 1996
5
Kelling, 1982). If this is the case, then such incivility may serve as a daily
cue that the police cannot be trusted to provide protection.
Fourth, we examine how confidence in the police is affected by
“informal collective security” – whether neighbors can be relied on to
provide protection (see Luxenburg et al., 1994; McDowall and Loftin,
1983; also Smith and Uchida, 1988). The nature of this relationship is far
from clear, but two competing hypotheses can be proposed. On the one
hand, it is possible that confidence in one’s neighbors breeds confidence
in the police; or, stated in the alternative fashion, those who feel
abandoned by their neighbors may also feel abandoned by the police. On
the other hand, informal collective security may mean that the police are
seen as superfluous, a second line of defense that is too distant to make a
difference; or, alternatively stated, if confidence in neighbors is lacking,
it may be that citizens see the police as more critical to their safety and
thus invest more positive affect in law enforcement. If so, the relationship
between informal collective security and confidence in the police would
be nonexistent or even inverse.
We should note, however, that the data set did not contain a
measure of “contact with the police”. Previous research has indicated that
negative evaluations of contacts with the police generally foster negative
global attitudes toward law enforcement, while positive evaluations of
police contacts tend to have a less substantial positive impact on global
attitudes (Brandl et al., 1994; Brown and Coulter, 1983; Dean, 1980).
This omission, however, is unlikely to account fully for the results
reported below. Existing studies reveal that the strength of the
relationship between the contact variables and attitudes is moderate and
accounts for only a modest amount of the explained variation. 1 In
contrast, in the current study two contextual variables have strong
relationships with the police confidence measure, even with a range of
factors controlled. Of course, future research should examine whether,
and if so to what extent, including contact with the police in an analysis
might cause the findings we report to be revised.
METHOD
Sample
The data for the present research were drawn from a larger study
of crime-prevention issues in an urban environment. Questionnaires were
mailed to 1,000 randomly selected residents in the city of Cincinnati in
6
American Journal of Police, Vol. XV, No. 1 1996
the fourth quarter of 1992. Adjusting for residents who had moved, had
undeliverable addresses or were deceased, the final sample size was 934.
Guided by Dillman’s (1978) Total Design Method, the respondents were
contacted five times: an initial mailing of the questionnaire, a reminder
letter, two additional mailings of the questionnaire and a final telephone
call to those who had not responded. Half of the households contacted
received instructions that an adult female over the age of 21 should
complete the survey; the other half were instructed to have the survey
completed by a male over the age of 21. In the end, 539 usable surveys
were returned, a response rate of 57.7 percent.
The sample had a mean age of 47.2, was 56.7 percent female, was
79.3 percent white (17.5% African-American, 3.4% “other” nonwhites),
and had a mean family income of $35,259. The sample was closely
representative of the Cincinnati population for age and gender, but it
underrepresented minorities and lower-income people. Since previous
research suggests that minorities and the poor are less supportive of law
enforcement, the results reported here may overestimate levels of
confidence in the police (Erez, 1984; Furstenberg and Wellford, 1973;
Hindelang, 1974; Parks, 1984; Percy, 1980; Scaglion and Condon, 1980).
Measures
The coding and items comprising each of the measures used in the
study are presented in Appendix A. Below we present an overview of the
dependent variable and independent variables in the analysis.
Dependent Variable
Confidence in the police was measured by a five-item measure,
which assessed whether respondents believed that police were
responsive, cared about the neighborhood’s safety, maintained order and
were able to protect residents against crime. The respondents were asked
to use a six-point Likert scale ranging from “very strongly disagree” to
“very strongly agree” to rate the five items. A high score indicates high
confidence in the police. The Cronbach alpha for the scale is 0.77.
Independent Variables
Information was collected on five demographic variables: age,
gender, education, household income and race (see Appendix A). The
race variable is coded as 1 = white and 0 = nonwhite. For purposes of
American Journal of Police, Vol. XV, No. 1 1996
7
analysis, minorities in the “other category” (3.4%) were grouped with
African-Americans. This was done because an examination of crosstabular relationship revealed that confidence toward the police among
these minorities was more similar to African-Americans than to whites.
The analysis also included three crime-related variables. First,
crime victimization was measured by seven items to which the
respondents answered either yes (for a victimization) or no (if no
victimization had taken place). Six items listed specific common property
or violent crimes (e.g. “someone broke into your house”), while the
seventh encompassed victimizations not specifically named (i.e. “some
other crime not listed here happened to you”). This measure assessed
victimizations that had occurred in the “last two years”. The scale was
scored as the number of offenses for which the respondent had been
victimized.
Second, fear of crime was assessed by asking the respondents,
“Think back to those times when you might have felt afraid or worried
that you might be a crime victim. How many times have you felt afraid in
the last month?” This phrasing was chosen because, unlike measures of
fear typically used in the research, it assesses whether people actually
experienced worry over the potential of crime victimization (see Ferraro
and LaGrange, 1987).
Third, based on previous research (Browning and Cao, 1992;
Cullen et al., 1985; Dunaway and Cullen, 1991; McGarrell and Flanagan,
1987), we employed a five-item conservative crime ideology scale, with
a Cronbach alpha of 0.80. In general, the scale measures support for “get
tough” policies, a belief that offenders are afforded too many rights, and
a belief that crime is due to permissiveness in American society. The
respondents rated these items using a six-point, agree-disagree Likert
scale.
Finally, two community context variables were included in the
analysis. First, the concept of community disorder – “incivility” – was
measured with eight items drawn from Skogan (1987, 1990). These items
encompassed physical disorder (e.g. “garbage or litter on the street”) and
social disorder (e.g. “groups of teenagers hanging out on the corners or in
the street”). The respondents were asked to state whether each of the eight
disorderly conditions of their community were “1 = not a problem”, “2 =
some problem”, or “3 = a big problem”. The reliability coefficient for this
scale was 0.88.
The concept of informal collective security was measured by
three items regarding the willingness of neighbors to assist in providing
8
American Journal of Police, Vol. XV, No. 1 1996
protection from crime. The items were rated using a six-point, agreedisagree Likert scale. The reliability was 0.69.
RESULTS
To assess public confidence in the police, we first examine the
descriptive data on the public’s overall level of confidence in the police
items. Second, to explore sources of confidence in the police, we present
the results of multivariate analyses.
Overall Level of Confidence in the Police
Five survey items were used to measure public confidence in the
police (see Appendix A). As shown in Table 1, the distribution of
responses to these items reveals that the citizens sampled expressed
confidence in the police.
In four of the five items, a substantial majority of the respondents
agreed (“agree a little” through “strongly agree”) that the police are
responsive (79.2%), do a good job maintaining order (74.7%), care about
the safety of residents (84.3%), and do a good job protecting people from
crime (81.9%). For each item, the respondents avoided selecting the most
favorable response (“strongly agree”); nonetheless, the most commonly
selected response, the “agree” category (39.7% to 47.5%), was the second
most supportive response option. Finally, the mean level of agreement for
four of the five confidence measures is above four on the six-point
response scale that ranges from “strongly disagree” to “strongly agree”
(4.15 to 4.46).
Less confidence was evident, however, in the responses to the
fifth confidence item: “There are not enough police in my neighborhood
to deal with crime”. The respondents were almost evenly divided in their
judgment of whether police staffing levels were sufficient to control their
neighborhood’s crime. It is difficult to assess why this item elicited less
supportive responses than items 1 to 4, but one possible reason can be
noted. Even though the respondents may have felt that police officers
come when called, care, and do a good job (items 1 to 4), some citizens
still may have preferred that police have more of a physical presence in
their neighborhoods so as “to deal with crime”. In this regard, previous
research suggests that one of the most common concerns of citizens is the
lack of street-level personnel in their neighborhoods (Maguire and
Table 1
PUBLIC CONFIDENCE IN THE POLICE (PERCENTAGES REPORTEDa)
Strongly
Disagree
(1)
Disagree
(2)
Disagree
a Little
(3)
Agree
a Little
(4)
Agree
(5)
Strongly
Agree
(6)
2.4
9.1
9.3
21.7
47.5
10.0
b. The police do a good job in my
neighborhood in making sure that
no one disturbs the peace
(mean = 4.15; std = 1.22)
3.2
9.8
12.2
26.1
41.2
7.4
c. The police care a lot about the safety
in my neighborhood (mean = 4.46;
std = 1.14)
2.2
5.2
8.2
28.7
40.4
15.2
d. The police do a good job in
protecting me against crime
(mean = 4.25; std = 1.09)
2.0
7.4
8.5
35.0
39.7
7.2
e. There are not enough police in
my neighborhood to deal with
crime (mean = 3.393; std = 1.348)
4.5
29.5
17.8
24.5
17.6
6.1
a
American Journal of Police, Vol. XV, No. 1 1996
a. When people in my neighborhood
call the police, they come right
away (mean = 4.33; std = 1.2)
Due to rounding errors, some percentages do not equal 100%
9
10
American Journal of Police, Vol. XV, No. 1 1996
Pastore, 1994:192; US Department of Justice, 1977:39-40). Further, a
major “selling point” of community policing is that officers are more
visible to residents.
Precise comparisons of these findings with the results of prior
studies are not possible. Studies of public assessments of the police have
used a diversity of measures that tap a rather wide variety of attitudes.
Some surveys enquire about respondents’ global satisfaction with the
police, while others ask respondents to evaluate a variety of more specific
police tasks and officer characteristics (see Brandl et al., 1994).
Nonetheless, the general pattern of results reported here is consistent with
previous research: Overall, citizens in our sample, as in other samples,
express favorable sentiments toward the police.
One final note is in order concerning the public’s confidence. For
each confidence measure, the mean confidence level of white
respondents exceeded the mean of nonwhites, although in only two
situations do the differences achieve a conventional level of significance
(p < 0.05). Although nonwhites did express lower levels of confidence,
their responses still indicated overall confidence in the police (a scale
mean of 4.229). This finding is consistent with existing research, which
has found that while minorities express less favorable attitudes toward the
police than do whites, they still possess favorable attitudes.
Sources of Confidence in the Police
To assess whether various demographic, crime-related and
contextual variables influence confidence in the police, we undertook
three multivariate analyses (see Table 2). In each equation, the confidence
in the police scale was the dependent variable. A correlation matrix was
also computed; no correlations among the independent variables were
high enough to suggest the presence of multicollinearity (see Appendix
B).
Equation 1 examines only the impact of respondents’
demographic variables on confidence in the police. The equation reveals
that respondents’ race is significantly related to the confidence measure,
with whites expressing more confidence in the police than nonwhites.
In addition, the respondents’ gender, age and income were
statistically significant determinants of confidence in the police.
Specifically, females and older individuals were more likely to express
higher levels of confidence than were males and younger respondents.
American Journal of Police, Vol. XV, No. 1 1996
11
Table 2
THE DETERMINANTS OF CONFIDENCE IN POLICE
Independent Variables
Race (White = 1)
Gender (Male = 1)
Age
Income
Education
Equation 1
Beta
Equation 2
Beta
Equation 3
Beta
0.117**
–0.130**
0.204***
0.139**
0.040
0.098*
–0.133*
0.131**
0.099*
0.042
0.043
–0.071*
0.048
–0.004
–0.004
–0.191***
–0.149***
0.012
–0.065
–0.041
0.023
Fear of Crime
Victimization
Conservative Crime
Ideology
Community Disorder
Informal Collective
Security
R-square =
F=
Significance =
–0.368***
0.331***
0.079
9.205
0.000
0.149
11.318
0.000
0.386
33.214
0.000
* p < 0.05; ** p < 0.01; *** p < 0.001; N = 539
Income level was also positively related to confidence in the police. The
respondents’ educational level, however, was not significant.
Equation 2 reveals that the influence of race (and of the other
demographic variables) on confidence remains statistically significant
even after controlling for the effects of three crime-related variables. Age,
gender and income also retained statistical significance. Equation 2 also
reveals that two of the three crime-related variables are significant: being
afraid of crime and having been a crime victim in the last two years both
reduce confidence in the police. Holding a conservative crime control
ideology, however, was not statistically significant.
Two additional findings relative to Equation 2 are worthy of
mention. First, the proportion of variance explained by Equation 2
(R2 = 0.149) is almost twice that explained by Equation 1 (R2 = 0.079).
This finding suggests that the crime-related variables in this equation
contribute substantially to our ability to predict public confidence in the
12
American Journal of Police, Vol. XV, No. 1 1996
police. Second, the standardized regression coefficients for the variables
indicate that fear of crime and recent victimization experiences exert a
larger effect on confidence in the police than do any of the demographic
variables.
In Equation 3, the final and most complete model we present, two
contextual variables are introduced: citizen perceptions of neighborhood
disorder and informal collective security. When these variables are
included, the effect of race on confidence in the police is no longer
significant. In fact, of the eight variables brought forward from Equation
2, only the respondents’ gender remains statistically significant.
The two community context variables, first included in Equation
3, are both significant determinants of confidence in the police.
Perceptions of community disorder exerted a significant negative effect
on confidence: As citizens’ perceptions of incivility increased, confidence
in the police decreased. Additionally, the standardized regression
coefficient for this variable reveals that perceptions of disorder had the
largest effect of all the variables included in this equation.
The second contextual variable, perceptions of the willingness of
neighbors to assist the police, had a significant positive effect on public
confidence: As perceptions of informal collective security increase,
confidence in the police also increases. The effect of this variable on
confidence in the police was the second largest of the included variables.
Finally, the variance explained by variables in Equation 3 (R2 =
0.389) substantially exceeds that explained by Equation 2 (R2 = 0.149).
The community context variables, therefore, appear to have both
statistical and substantive significance for the explanation of public
confidence in the police.
DISCUSSION
This study represents an attempt to understand more clearly the
relationship between race and public confidence in the police. Although
the initial analysis (reported in Equation 1) revealed a significant race
effect, the introduction of contextual factors eliminated this relationship.
For our sample, then, respondents’ race is not a significant determinant of
confidence in the police.
Furthermore, results for Equation 3 support the proposition that
the community context is the most important determinant of public
confidence in the police. Specifically, citizens’ perceptions of disorder
American Journal of Police, Vol. XV, No. 1 1996
13
exerted the strongest impact on confidence. Thus it appears that citizens
hold the police at least partially responsible for the disorder – the “broken
windows” – in their neighborhoods. Whether the police are capable of
containing the types of incivilities that appear to undermine public
confidence is a matter that remains open to debate. In any case, our
respondents appear to perceive the police as the government’s first-line
representative, responsible for controlling neighborhood disorder.
In addition to perceptions of disorder, the effects on confidence of
the other community context variable – informal collective security, or
the willingness of neighbors to provide protection from crime – were
significant and positive. As proposed in the introduction, this finding
suggests that confidence in neighbors breeds confidence that the police
will provide protection – that is, informal collective security encourages
feelings of formal collective security (see McDowall and Loftin, 1983).2
More broadly, this finding suggests that neighborhood social integration
may provide a supportive context in which residents are reminded that
they are not isolated individuals cut off from the larger social order. In
short, social bonds may encourage their identification with and positive
evaluation of formal institutional arrangements.
When viewed together, the findings on race and community
context yield an important conclusion: Attitudes toward the police may
not be regulated by a person’s race per se, but by the social context in
which the person is situated. Accordingly, past research, which typically
has reported that minorities appraise law enforcement less favorably, may
be in need of revision. Given that race became nonsignificant when
contextual variables were controlled, it is possible that much of the
existing research, which did not include contextual variables, has based
findings on misspecified models.
A further comment about our conclusions on community context
is in order: our measures of neighborhood context were perceptual rather
than objective indicators. This qualification, however, does not mean that
the results reported are unimportant. Indeed, research using objective
indicators, such as crime rates, has generally found that these measures
explain very little of the variation in citizens’ attitudes toward the police
(Stipak, 1979). In contrast, perceptions of neighborhood conditions may
well be more salient to respondents than less proximate, objective
conditions. If so, research indicates that these beliefs about local
community conditions are more likely to be “cognitively accessed”
(Fazio and Williams, 1986), which in turn may cause them to influence
attitudes toward the police. Furthermore, the impact of these perceptions
14
American Journal of Police, Vol. XV, No. 1 1996
may be especially important where the problem is one that the public
considers within the function and control of the police (Percy, 1986;
Stipak, 1979). For similar reasons, the public’s beliefs about the level of
collective informal control within the community are likely to exert a
much greater effect than actual measures of citizen participation.
An additional and consistent finding was the significant influence
of respondents’ gender on confidence in the police: even with a range of
factors controlled, females were more likely than males to express
confidence in the police. This finding is consistent with several prior
studies (Apple and O’Brien, 1983; Thomas and Hyman, 1977).3 It has
been suggested that women hold more positive sentiments because they
generally have contact with the police that is less antagonistic than the
contact that typically occurs between males and the police. Compared
with males, females are more likely to make service requests of the police
and are less likely to be involved in situations in which the police are
acting against them. Because (as noted) our data set did not contain
measures of citizen contacts with the police, we were unable to explore
whether differential police contact can account for the gender gap in
confidence in law enforcement.
In connection with this point, we should state again that existing
research suggests that citizens’ experiences with the police influence their
evaluations of the overall quality of police performance (Brown and
Coulter, 1983; Dean, 1980; Mastrofski, 1981; Parks, 1984; Scaglion and
Condon, 1980). Even so, the model presented in Equation 3 in Table 2
performs well: The explained variance by the model’s variables
(R2 = 0.386) exceeds the variance explained in prior studies that have
included police contact variables (see, for example, Brandl et al., 1994;
Dean, 1980). It remains an empirical question, of course, whether policecitizen contact measures will significantly increase the explained
variance or specify the relations reported here. In the absence of this or
other research, however, our study suggests that an accurate
understanding of the public’s attitudes toward the police must include an
assessment of the social context in which citizens reside.
Finally, the impact of neighborhood problems on confidence in
the police should be of special concern to law enforcement
administrators. Even though many neighborhood problems are not
susceptible to control or even regulation by the police, citizens’
perceptions of these problems provide salient information that appears to
influence their confidence in the police. Thus, improving citizens’
American Journal of Police, Vol. XV, No. 1 1996
15
attitudes is likely to be a challenging task for police agencies to undertake
successfully.
NOTES
A previous version of this paper was presented at the 1994
meeting of the American Society of Criminology, Miami. Address
correspondence to: Liqun Cao, Department of Sociology, Anthropology,
and Criminology, Eastern Michigan University, Ypsilanti, MI 48197.
1.
Dean (1980) in her study estimated a series of regression
equations using a variety of contact variables. When the number
of “bad” and “good” contacts are included as two variables the R2
increased from 0.091 to 0.155. When citizen evaluations and the
nature of the contact are included (request for information,
response to a victimization, stopped by the police) the R 2
increased to 0.165.
2.
We suggest that perceptions of informal collective security
influence confidence in the police because information about
informal collective security is likely to be received on a daily
basis. However, because our data are not longitudinal, we cannot
rule out the possibility of reverse (or reciprocal) causal ordering
between informal and formal collective security.
3.
It should be noted that there are also research findings suggesting
that gender is not a significant predictor of attitudes toward the
police (Boggs and Galliher, 1965; Brandl et al., 1994; Brown and
Coulter, 1983).
REFERENCES AND FURTHER READING
Albrecht, S.L. and Green, M. (1977), “Attitudes toward the Police and the Larger
Attitude Complex: Implications for Police-Community Relationships”,
Criminology, 15:67-86.
Apple, N. and O’Brien, D. (1983), “Neighborhood Racial Composition and
Residents’ Evaluation of Police Performance”, Journal of Police Science and
Administration, 11:76-84.
16
American Journal of Police, Vol. XV, No. 1 1996
Bayley, D. (1976), Forces of Order: Police Behavior in Japan and the United
States, Berkeley, CA: University of California Press.
Boggs, S.L. and Galliher, S.F. (1965), “Evaluating the Police: A Comparison of
Black Street and Household Respondents”, Social Problems, 13:393-406.
Bordua, D.J. and Tifft, L.L. (1971), “Citizen Interview, Organizational Feedback,
and Police-Community Relations Decisions”, Law and Society Review,
6:155-182.
Brandl, S., Frank, J., Worden, R.E. and Bynum, T.S. (1994), “Global and Specific
Attitudes toward the Police: Disentangling the Relationship”, Justice Quarterly,
11:119-134.
Brandl, S. and Horvath, F. (1991), “Crime Victim Evaluation of Police
Investigative Performance”, Journal of Criminal Justice, 19:293-305.
Brown, K. and Coulter, P.B. (1983), “Subjective and Objective Measures of
Police Performance”, Public Administration Review, 43:50-58.
Browning, S.L. and Cao, L. (1992), “The Impact of Race on Criminal Justice
Ideology”, Justice Quarterly, 9:685-701.
Carter, D.L. (1985), “Hispanic Perception of Police Performance: An Empirical
Assessment”, Journal of Criminal Justice, 13:487-500.
Carlson, H.M. and Sutton, M.S. (1979), “Some Factors in Community Evaluation
of Police Street Performance”, American Journal of Community Psychology,
7:583-591.
Covington, J. and Taylor, R.B. (1991), “Fear of Crime in Urban Residential
Neighborhoods: Implications of between- and within- Neighborhood Sources for
Current Models”, Sociological Quarterly, 32:231-249.
Cullen, F.T., Bynum, T.S., Garrett, K.M. and Greene, J.R. (1985), “Legislator
Ideology and Criminal Justice Policy: Implications from Illinois”, in Fairchild,
E.S. and Webb, V.J. (Eds), The Politics of Crime and Criminal Justice, Beverly
Hills, CA: Sage, 57-76.
Dean, D. (1980), “Citizen Ratings of the Police: The Difference Police Contact
Makes”, Law and Policy Quarterly, 2:445-471.
Decker, S. and Wagner, A. (1981), “Race and Citizen Complaints against the
Police: An Analysis of Their Interaction”, in Baker, R. and Meyer, F. (Eds),
Evaluating Alternative Law Enforcement Policies, Lexington, MA: Lexington.
American Journal of Police, Vol. XV, No. 1 1996
17
Dillman, D.A. (1978), Mail and Telephone Surveys: The Total Design Method,
New York, NY: John Wiley & Sons.
Dunaway, R.G. and Cullen, F.T. (1991), “Explaining Crime Ideology: An
Exploration of the Parental Socialization Perspective”, Crime and Delinquency,
37:536-554.
Erez, E. (1984), “Self-defined Desert and Citizen’s Assessment of the Police”,
Journal of Criminal Law and Criminology, 75:1276-1299.
Fazio, R.H. and Williams, C.J. (1986), “Attitude Accessibility as a Moderator of
the Attitude Perception and Attitude-Behavior Relations: An Investigation of the
1984 Presidential Election”, Journal of Personality and Social Psychology,
51:505-514.
Ferraro, K.F. and LaGrange, R.L. (1987), “The Measurement of Fear of Crime”,
Sociological Inquiry, 57:70-101.
Flanagan, T.J. and Maguire, K. (1992), Sourcebook of Criminal Justice Statistics,
Washington, DC: US Department of Justice.
Furstenberg, F. and Wellford, C.F. (1973), “Calling the Police: The Evaluation of
Police Service”, Law and Society Review, 8:393-406.
Goldstein, H. (1987), “Toward Community-oriented Policing: Potential, Basic
Requirements, and Threshold Questions”, Crime and Delinquency, 33:6-30.
Hindelang, M.J. (1974), “Public Opinion Regarding Crime, Criminal Justice and
Related Topics”, Journal of Research in Crime and Delinquency, 11:101-116.
Homant, R.J., Kennedy, D.B. and Fleming, R.M. (1984), “The Effect of
Victimization and the Police Response on Citizen’s Attitudes toward Police”,
Journal of Police Science and Administration, 12:323-32.
Jacob, H. (1971), “Black and White Perceptions of Justice in the City”, Law and
Society Review, 5:69-89.
Jefferson, T. and Walker, M.A. (1993), “Attitudes to the Police of Ethnic
Minorities in a Provincial City”, British Journal of Criminology, 33:251-266.
Jefferson, T. and Walker, M.A. (1992), “Ethnic Minorities in the Criminal Justice
System”, Criminal Law Review, 28:83-95.
Klyman, F.I. and Kruckenberg, J. (1974), “A Methodology for Assessing Citizen
Perceptions of the Police”, Journal of Criminal Justice, 2:219-233.
18
American Journal of Police, Vol. XV, No. 1 1996
Koenig, D.J. (1980), “The Effect of Crime Victimization and Judicial or Police
Contacts on Public Attitudes toward the Local Police”, Journal of Police Science
and Administration, 8:243-249.
Komanduri, S.M., Roebuck, J.B. and Smith, J.D. (1990), “The Image of the
Police in Black Atlanta Communities”, Journal of Police Science and
Administration, 17:250-257.
Larsen, K.S. (1968), “Authoritarianism and Attitudes toward Police”,
Psychological Reports, 3:349-350.
Lewis, D. and Salem, G. (1986), Fear of Crime: Incivility and the Production of
a Social Problem, New Brunswick, NJ: Transaction Books.
Lipset, S.M. and Schneider, W. (1983), The Confidence Gap: Business, Labor,
and Government in the Public Mind, New York, NY: The Free Press.
Luxenburg, J., Cullen, F.T., Langworthy, R.H. and Kopache, R. (1994), “Firearms
and Fido: Ownership of Injurious Means of Protection”, Journal of Criminal
Justice, 22:159-170.
McDowall, D. and Loftin, C. (1983), “Collective Security and the Demand for
Legal Handguns”, American Journal of Sociology, 88:1146-1161.
McGarrell, E.F. and Flanagan, T.J. (1987), “Measuring and Explaining Legislator
Crime Control Ideology”, Journal of Research in Crime and Delinquency,
24:102-118.
Maguire, K. and Pastore, A.L. (Eds) (1994), Sourcebook of Criminal Justice
Statistics 1993, Washington, DC: US Government Printing Office.
Mastrofski, S. (1981), “Surveying Clients to Assess Police Performance”,
Evaluation Review, 5:397-408.
Mirande, A. (1980), “Fear of Crime and Fear of the Police in a Chicano
Community”, Sociology and Social Research, 64:528-541.
Parks, R.B. (1981), “Police Response to Victimization Effects on Citizen
Attitudes and Perceptions”, in Skogan, W.G. (Ed.), Sample Surveys of the Victims
of Crime, Cambridge, MA: Ballinger, 89-104.
Parks, R.B. (1984), “Linking Objective and Subjective Measures of
Performance”, Public Administration Review, 44:118-127.
Peek, C.W., Lowe, G.D. and Alston, J.P. (1981), “Race and Attitudes toward
Local Police”, Journal of Black Studies, 11:361-374.
American Journal of Police, Vol. XV, No. 1 1996
19
Percy, S.L. (1980), “Response Time and Citizen Evaluation of Police”, Journal of
Police Science and Administration, 8:75-86.
Percy, S.L. (1986), “In Defense of Citizen Evaluations as Performance
Measures”, Urban Affairs Quarterly, 22:66-83
Reasons, C.E. and Wirth, B.A. (1975), “Police Community Relations Units: A
National Survey”, Journal of Social Issues, 31:27-33.
Scaglion, R. and Condon, R.G. (1980), “Determinants of Attitudes toward City
Police”, Criminology, 17:485-494.
Skogan, W.G. (1978), “Citizen Satisfaction with Police Services: Individual and
Contextual Effects”, Police Studies Journal, (special issue):469-479.
Skogan, W.G. (1987), Disorder and Community Decline, Evanston, IL: Center for
Urban Affairs and Policy Research, Northwestern University.
Skogan, W.G. (1990), Disorder and Decline: Crime and the Spiral of Decay in
American Neighborhoods, Berkeley, CA: University of California Press.
Skolnick, J.H. and Bayley, D.H. (1988), “Theme and Variation in Community
Policing”, in Tonry, M. and Morris, N. (Eds), Crime and Justice, Volume 10,
Chicago: University of Chicago Press, 1988, 1-37.
Smith, D. and Uchida, C. (1988), “The Social Organization of Self-help: A Study
of Defensive Weapon Ownership”, American Sociological Review, 53:94-102.
Smith, P.E. and Hawkins, R.O. (1973), “Victimization, Types of Citizen-Police
Contacts and Attitudes toward the Police”, Law and Society Review, 8:135-152.
Stipak, B. (1979), “Citizen Satisfaction with Urban Services: Potential Misuse as
a Performance Indicator”, Public Administration Review, 39:46-52.
Thomas, C.W. and Hyman, J. (1977), “Perceptions of Crime, Fear of
Victimization, and Public Perceptions of Police Performance”, Journal of Police
Science and Administration, 5:305-317.
US Department of Justice (1977), The Police and Public Opinion, Washington,
DC: Department of Justice.
White, M. and Menke, B.A. (1978), “A Critical Analysis of Surveys on Public
Opinions toward Police Agencies”, Journal of Police Science and Administration,
6:204-218.
Wilson, J.Q. and Kelling, G.L. (1982), “The Police and Neighborhood Safety:
Broken Windows”, Atlantic Monthly, (March):29-38.
20
American Journal of Police, Vol. XV, No. 1 1996
Wycoff, M.A. (1988), “The Benefits of Community Policing: Evidence and
Conjecture”, in Greene, J.R. and Mastrofski, S. (Eds), Community Policing:
Rhetoric or Reality, Westport, CT: Praeger: 103-120.
Zamble, E. and Annesley, P. (1987), “Some Determinants of Public Attitudes
toward the Police”, Journal of Police Science and Administration, 15:285-290.
Appendix A:
SUMMARY OF MEASURES USED IN THE ANALYSIS
Independent Variables
1. Race
1 = whites; 0 = nonwhites
2. Age
Natural age in years at the time of the survey
3. Gender
1 = male, 0 = female
4. House Income
1 = $2,000 and below; 2 = 2,001 to 6,000; 3 = 6,001 to
10,000; 6 = 20,001 to 25,000; 7 = 25,001 to 30,000; 8
= 30,001 to 35,000; 9 = 35,001 to 40,000; 10 = 40,001
to 45,000; 11 = 45,001 to 50,000; 12 = 50,001 to
55,000; 13 = 55,001 to 60,000; 14 = 60,001 to 70,000;
15 = 70,001 to 80,000; 16 = 80,001 to 90,000; 17 =
90,001 to 120,000; 18 = $120,001 and above
5. Education
1 = did not graduate from high school; 2 = graduated
from high school; 3 = some years of college; 4 =
graduated from college; 5 = finished one or more years
of graduate school
6. Informal Collective
Security
A combination of three items: (a) The people in my
neighborhood watch one another’s home when no one
is home. (b) In my neighborhood, people will call the
police right away if they think a crime is being
committed. (c) If some criminal was trying to rob me,
my neighbors wouldn’t just stand there and watch, but
would come help me out. 1 = very strongly disagree;
2 = strongly disagree; 3 = disagree; 4 = agree; 5 =
strongly agree; and 6 = very strongly agree
American Journal of Police, Vol. XV, No. 1 1996
21
7. Community
Disorder
A combination of eight items: How much of a problem
is each of the following: (a) noisy neighbors – people
who play loud music, have late parties, or have noisy
fights; (b) garbage or litter on the streets or sidewalks;
(c) people not keeping up their houses or apartments
– like not painting or fixing windows that break; (d)
groups of teenagers hanging out on the corners or in
the streets; (e) people selling or using illegal drugs in
the neighborhood; (f) vandalism – like kids breaking
windows or writing on walls or things like that; (g)
teenage gangs who commit crime; (h) people who say
insulting things or bother people as they walk down
the street. 1 = not a problem; 2 = some problems;
3 = big problem
8. Fear of Crime
Think back to those times when you might have felt
afraid that you might be a crime victim. How many
times have you felt afraid of crime in the last month?
1 = once; 2 = two to three; 3 = three to five; 4 = more
than five
9. Victimization
A combination of seven items. In the last two years,
have any of the following crimes been committed
against you: (a) someone broke into your house; (b) had
property stolen from your house or yard; (c) someone
stole, broke into, or vandalized your car; (d) someone
held you up on the street and robbed you; (e) someone
threatened to beat you up or threatened you with a
knife, gun, or other weapon; (f) someone actually beat
you up; (g) some other crime not listed here happened
to you. 1 = yes; 0 = no
10. Conservative Crime A combination of five items; (a) stiffer jail sentences
Ideology
are needed to show criminals that crime does not pay;
(b) juvenile criminals are treated too leniently by the
courts; (c) crime has increased in recent times because
society has become too permissive; (d) these days,
criminals have too many legal rights; and (e) if we
really cared about crime victims, we would make sure
that criminals were caught and given harsh
punishment. 1 = very strongly disagree; 2 = strongly
disagree; 3 = disagree; 4 = agree; 5 = strongly agree;
and 6 = very strongly agree
22
American Journal of Police, Vol. XV, No. 1 1996
Dependent Variable
Confidence in
the Police
A combination of five items: (a) when people in my
neighborhood call the police, they come right away;
(b) the police do a good job in my neighborhood in
making sure that no one disturbs the peace; (c) the
police care a lot about the safety of the people in my
neighborhood; (d) the police do a good job in
protecting me against crime; and (e) there are not
enough police in my neighborhood to deal with crime.
1 = very strongly disagree; 2 = strongly disagree;
3 = disagree; 4 = agree; 5 = strongly agree; and
6 = very strongly agree
Appendix B:
CORRELATION MATRIX OF VARIABLES IN ANALYSIS (N = 539)
Variables
1
Confidence in
Police
1.00
2. Race
0.12
3. Age
0.17
4. Gender
–0.10
5. Household
Income
0.12
6. Education
0.04
7. Informal
Collective
Security
0.46
8. Community
Disorders
–0.51
9. Fear of
Crime
–0.25
10. Victimization –0.25
11. Conservative
Ideology
0.05
2
3
4
5
6
7
8
9
10
11
1.
Means
Standard
Deviation
1.00
–0.04 1.00
0.14 –0.08 1.00
0.16 –0.18 0.19 1.00
0.13 –0.33 0.13 0.48 1.00
0.08 0.12 –0.09 0.15 0.04 1.00
–0.15 –0.15 0.01 –0.21 –0.16 –0.27 1.00
–0.10 –0.14 –0.13 –0.13 –0.02 –0.13 0.36 1.00
–0.02 –0.24 0.12 –0.04 0.02 –0.15 0.32 0.24 1.00
0.05 0.21 –0.08 –0.09 –0.25 0.06 0.04 0.01 –0.10 1.00
16.8
0.79 47.1
0.43 9.25 3.25 11.4
4.66 2.01 0.67 18.6
4.4
0.41 18.0
0.50 4.25 1.26 2.7
3.69 1.19 0.93 5.0
3
RACE, COMMUNITY CONTEXT AND
CONFIDENCE IN THE POLICE
Liqun Cao
Eastern Michigan University
James Frank
Francis T. Cullen
University of Cincinnati
Although American citizens have deep suspicions about state
power (Bayley, 1976; Lipset and Schneider, 1983), studies show that the
public generally holds positive attitudes toward the police (Apple and
O’Brien, 1983; Dean, 1980; Erez, 1984; Furstenberg and Wellford, 1973;
Hindelang, 1974; White and Menke, 1978; Zamble and Annesley, 1987).
Even so, these sentiments are not displayed equally across all sectors of
the social order. In particular, minorities appear to evaluate law
enforcement less favorably (Albrecht and Green, 1977; Bordua and Tifft,
1971; Carter, 1985; Decker and Wagner, 1981; Hindelang, 1974; Jacob,
1971; Jefferson and Walker, 1993; Parks, 1984; Peek et al., 1981; Percy,
1980; Reasons and Wirth, 1975; Scaglion and Condon, 1980; Skogan,
1978; but see Brandl et al., 1994). In response to this uneasy relationship,
many of the current community policing ventures have been undertaken
in minority neighborhoods in hopes of building residents’ confidence in
the police and willingness to be coproducers in the fight against crime
(Goldstein, 1987; Skolnick and Bayley, 1988:26; Wycoff, 1988).
Although research has consistently found that African-Americans
(and other minorities) view the police less favorably than whites, the
existing literature is limited in two ways. First, many studies report the
relationship between single variables of interest and confidence in the
police with no controls introduced (Albrecht and Green, 1977; Carlson
and Sutton, 1979; Flanagan and Maguire, 1992; Hindelang, 1974;
Jefferson and Walker, 1992, 1993; Klyman and Kruckenberg, 1974;
Komanduri et al., 1990; Mirande, 1980). Second, multivariate studies
4
American Journal of Police, Vol. XV, No. 1 1996
generally include only a limited range of variables in the analysis. Some
studies, for example, include controls only for other socio-demographic
variables (e.g. age, gender, income) (Apple and O’Brien, 1983). Even
more elaborate studies consider only a limited number of
nondemographic factors, such as contact with the police (Dean, 1980;
Mastrofski, 1981; Scaglion and Condon, 1980) and having been a crime
victim (Brandl and Horvath, 1991; Koenig, 1980; Percy, 1986). It is
possible, therefore, that these models are misspecified – that the effects of
race on police attitudes might be accounted for by factors not
incorporated into the analysis.
In this regard, the current study considers the impact of a range of
variables on confidence in the police. Some of these variables have been
examined infrequently, if at all, in prior research. The analysis includes,
for example, measures of crime experiences and of conservative political
orientation. Building on the research tradition that suggests that crimerelated reactions are shaped by the social context that surrounds
individuals (see, for example, Skogan, 1990), we also assess the effects of
neighborhood disorder and informal collective security on confidence in
law enforcement. In particular, we explore whether these contextual
conditions help to explain the relationship between race and confidence
in the police.
Two aspects of “crime experiences” are investigated: fear of
crime and having been a crime victim. The assumption is that when crime
is more salient in a person’s life, confidence in the police is undermined.
Previous research has found that victimization experiences tend to
increase negative attitudes toward the police (Apple and O’Brien, 1983;
Homant et al., 1984; Koenig, 1980; Parks, 1984; Smith and Hawkins,
1973).
Second, we investigate whether crime ideology influences
assessments of the police. We hypothesize that, since support for the
police is often a plank in “law and order” conservative political
platforms, a conservative crime ideology would enhance confidence in
the police. Some evidence exists to support this hypothesized relationship
(Larsen, 1968; Zamble and Annesley, 1987).
Third, we explore whether perceptions of community social and
physical disorder – sometimes called “incivility” – lessen citizens’
confidence in the police (see Covington and Taylor, 1991; Lewis and
Salem, 1986; Skogan, 1987). It is plausible that social disorder (e.g. noisy
neighbors, loitering by rowdy teenagers) and physical disorder (e.g.
graffiti, deteriorating property) send a message that law enforcement has
lost control over or consciously abandoned the community. In short, the
“broken windows” are not going to be fixed by the police (Wilson and
American Journal of Police, Vol. XV, No. 1 1996
5
Kelling, 1982). If this is the case, then such incivility may serve as a daily
cue that the police cannot be trusted to provide protection.
Fourth, we examine how confidence in the police is affected by
“informal collective security” – whether neighbors can be relied on to
provide protection (see Luxenburg et al., 1994; McDowall and Loftin,
1983; also Smith and Uchida, 1988). The nature of this relationship is far
from clear, but two competing hypotheses can be proposed. On the one
hand, it is possible that confidence in one’s neighbors breeds confidence
in the police; or, stated in the alternative fashion, those who feel
abandoned by their neighbors may also feel abandoned by the police. On
the other hand, informal collective security may mean that the police are
seen as superfluous, a second line of defense that is too distant to make a
difference; or, alternatively stated, if confidence in neighbors is lacking,
it may be that citizens see the police as more critical to their safety and
thus invest more positive affect in law enforcement. If so, the relationship
between informal collective security and confidence in the police would
be nonexistent or even inverse.
We should note, however, that the data set did not contain a
measure of “contact with the police”. Previous research has indicated that
negative evaluations of contacts with the police generally foster negative
global attitudes toward law enforcement, while positive evaluations of
police contacts tend to have a less substantial positive impact on global
attitudes (Brandl et al., 1994; Brown and Coulter, 1983; Dean, 1980).
This omission, however, is unlikely to account fully for the results
reported below. Existing studies reveal that the strength of the
relationship between the contact variables and attitudes is moderate and
accounts for only a modest amount of the explained variation. 1 In
contrast, in the current study two contextual variables have strong
relationships with the police confidence measure, even with a range of
factors controlled. Of course, future research should examine whether,
and if so to what extent, including contact with the police in an analysis
might cause the findings we report to be revised.
METHOD
Sample
The data for the present research were drawn from a larger study
of crime-prevention issues in an urban environment. Questionnaires were
mailed to 1,000 randomly selected residents in the city of Cincinnati in
6
American Journal of Police, Vol. XV, No. 1 1996
the fourth quarter of 1992. Adjusting for residents who had moved, had
undeliverable addresses or were deceased, the final sample size was 934.
Guided by Dillman’s (1978) Total Design Method, the respondents were
contacted five times: an initial mailing of the questionnaire, a reminder
letter, two additional mailings of the questionnaire and a final telephone
call to those who had not responded. Half of the households contacted
received instructions that an adult female over the age of 21 should
complete the survey; the other half were instructed to have the survey
completed by a male over the age of 21. In the end, 539 usable surveys
were returned, a response rate of 57.7 percent.
The sample had a mean age of 47.2, was 56.7 percent female, was
79.3 percent white (17.5% African-American, 3.4% “other” nonwhites),
and had a mean family income of $35,259. The sample was closely
representative of the Cincinnati population for age and gender, but it
underrepresented minorities and lower-income people. Since previous
research suggests that minorities and the poor are less supportive of law
enforcement, the results reported here may overestimate levels of
confidence in the police (Erez, 1984; Furstenberg and Wellford, 1973;
Hindelang, 1974; Parks, 1984; Percy, 1980; Scaglion and Condon, 1980).
Measures
The coding and items comprising each of the measures used in the
study are presented in Appendix A. Below we present an overview of the
dependent variable and independent variables in the analysis.
Dependent Variable
Confidence in the police was measured by a five-item measure,
which assessed whether respondents believed that police were
responsive, cared about the neighborhood’s safety, maintained order and
were able to protect residents against crime. The respondents were asked
to use a six-point Likert scale ranging from “very strongly disagree” to
“very strongly agree” to rate the five items. A high score indicates high
confidence in the police. The Cronbach alpha for the scale is 0.77.
Independent Variables
Information was collected on five demographic variables: age,
gender, education, household income and race (see Appendix A). The
race variable is coded as 1 = white and 0 = nonwhite. For purposes of
American Journal of Police, Vol. XV, No. 1 1996
7
analysis, minorities in the “other category” (3.4%) were grouped with
African-Americans. This was done because an examination of crosstabular relationship revealed that confidence toward the police among
these minorities was more similar to African-Americans than to whites.
The analysis also included three crime-related variables. First,
crime victimization was measured by seven items to which the
respondents answered either yes (for a victimization) or no (if no
victimization had taken place). Six items listed specific common property
or violent crimes (e.g. “someone broke into your house”), while the
seventh encompassed victimizations not specifically named (i.e. “some
other crime not listed here happened to you”). This measure assessed
victimizations that had occurred in the “last two years”. The scale was
scored as the number of offenses for which the respondent had been
victimized.
Second, fear of crime was assessed by asking the respondents,
“Think back to those times when you might have felt afraid or worried
that you might be a crime victim. How many times have you felt afraid in
the last month?” This phrasing was chosen because, unlike measures of
fear typically used in the research, it assesses whether people actually
experienced worry over the potential of crime victimization (see Ferraro
and LaGrange, 1987).
Third, based on previous research (Browning and Cao, 1992;
Cullen et al., 1985; Dunaway and Cullen, 1991; McGarrell and Flanagan,
1987), we employed a five-item conservative crime ideology scale, with
a Cronbach alpha of 0.80. In general, the scale measures support for “get
tough” policies, a belief that offenders are afforded too many rights, and
a belief that crime is due to permissiveness in American society. The
respondents rated these items using a six-point, agree-disagree Likert
scale.
Finally, two community context variables were included in the
analysis. First, the concept of community disorder – “incivility” – was
measured with eight items drawn from Skogan (1987, 1990). These items
encompassed physical disorder (e.g. “garbage or litter on the street”) and
social disorder (e.g. “groups of teenagers hanging out on the corners or in
the street”). The respondents were asked to state whether each of the eight
disorderly conditions of their community were “1 = not a problem”, “2 =
some problem”, or “3 = a big problem”. The reliability coefficient for this
scale was 0.88.
The concept of informal collective security was measured by
three items regarding the willingness of neighbors to assist in providing
8
American Journal of Police, Vol. XV, No. 1 1996
protection from crime. The items were rated using a six-point, agreedisagree Likert scale. The reliability was 0.69.
RESULTS
To assess public confidence in the police, we first examine the
descriptive data on the public’s overall level of confidence in the police
items. Second, to explore sources of confidence in the police, we present
the results of multivariate analyses.
Overall Level of Confidence in the Police
Five survey items were used to measure public confidence in the
police (see Appendix A). As shown in Table 1, the distribution of
responses to these items reveals that the citizens sampled expressed
confidence in the police.
In four of the five items, a substantial majority of the respondents
agreed (“agree a little” through “strongly agree”) that the police are
responsive (79.2%), do a good job maintaining order (74.7%), care about
the safety of residents (84.3%), and do a good job protecting people from
crime (81.9%). For each item, the respondents avoided selecting the most
favorable response (“strongly agree”); nonetheless, the most commonly
selected response, the “agree” category (39.7% to 47.5%), was the second
most supportive response option. Finally, the mean level of agreement for
four of the five confidence measures is above four on the six-point
response scale that ranges from “strongly disagree” to “strongly agree”
(4.15 to 4.46).
Less confidence was evident, however, in the responses to the
fifth confidence item: “There are not enough police in my neighborhood
to deal with crime”. The respondents were almost evenly divided in their
judgment of whether police staffing levels were sufficient to control their
neighborhood’s crime. It is difficult to assess why this item elicited less
supportive responses than items 1 to 4, but one possible reason can be
noted. Even though the respondents may have felt that police officers
come when called, care, and do a good job (items 1 to 4), some citizens
still may have preferred that police have more of a physical presence in
their neighborhoods so as “to deal with crime”. In this regard, previous
research suggests that one of the most common concerns of citizens is the
lack of street-level personnel in their neighborhoods (Maguire and
Table 1
PUBLIC CONFIDENCE IN THE POLICE (PERCENTAGES REPORTEDa)
Strongly
Disagree
(1)
Disagree
(2)
Disagree
a Little
(3)
Agree
a Little
(4)
Agree
(5)
Strongly
Agree
(6)
2.4
9.1
9.3
21.7
47.5
10.0
b. The police do a good job in my
neighborhood in making sure that
no one disturbs the peace
(mean = 4.15; std = 1.22)
3.2
9.8
12.2
26.1
41.2
7.4
c. The police care a lot about the safety
in my neighborhood (mean = 4.46;
std = 1.14)
2.2
5.2
8.2
28.7
40.4
15.2
d. The police do a good job in
protecting me against crime
(mean = 4.25; std = 1.09)
2.0
7.4
8.5
35.0
39.7
7.2
e. There are not enough police in
my neighborhood to deal with
crime (mean = 3.393; std = 1.348)
4.5
29.5
17.8
24.5
17.6
6.1
a
American Journal of Police, Vol. XV, No. 1 1996
a. When people in my neighborhood
call the police, they come right
away (mean = 4.33; std = 1.2)
Due to rounding errors, some percentages do not equal 100%
9
10
American Journal of Police, Vol. XV, No. 1 1996
Pastore, 1994:192; US Department of Justice, 1977:39-40). Further, a
major “selling point” of community policing is that officers are more
visible to residents.
Precise comparisons of these findings with the results of prior
studies are not possible. Studies of public assessments of the police have
used a diversity of measures that tap a rather wide variety of attitudes.
Some surveys enquire about respondents’ global satisfaction with the
police, while others ask respondents to evaluate a variety of more specific
police tasks and officer characteristics (see Brandl et al., 1994).
Nonetheless, the general pattern of results reported here is consistent with
previous research: Overall, citizens in our sample, as in other samples,
express favorable sentiments toward the police.
One final note is in order concerning the public’s confidence. For
each confidence measure, the mean confidence level of white
respondents exceeded the mean of nonwhites, although in only two
situations do the differences achieve a conventional level of significance
(p < 0.05). Although nonwhites did express lower levels of confidence,
their responses still indicated overall confidence in the police (a scale
mean of 4.229). This finding is consistent with existing research, which
has found that while minorities express less favorable attitudes toward the
police than do whites, they still possess favorable attitudes.
Sources of Confidence in the Police
To assess whether various demographic, crime-related and
contextual variables influence confidence in the police, we undertook
three multivariate analyses (see Table 2). In each equation, the confidence
in the police scale was the dependent variable. A correlation matrix was
also computed; no correlations among the independent variables were
high enough to suggest the presence of multicollinearity (see Appendix
B).
Equation 1 examines only the impact of respondents’
demographic variables on confidence in the police. The equation reveals
that respondents’ race is significantly related to the confidence measure,
with whites expressing more confidence in the police than nonwhites.
In addition, the respondents’ gender, age and income were
statistically significant determinants of confidence in the police.
Specifically, females and older individuals were more likely to express
higher levels of confidence than were males and younger respondents.
American Journal of Police, Vol. XV, No. 1 1996
11
Table 2
THE DETERMINANTS OF CONFIDENCE IN POLICE
Independent Variables
Race (White = 1)
Gender (Male = 1)
Age
Income
Education
Equation 1
Beta
Equation 2
Beta
Equation 3
Beta
0.117**
–0.130**
0.204***
0.139**
0.040
0.098*
–0.133*
0.131**
0.099*
0.042
0.043
–0.071*
0.048
–0.004
–0.004
–0.191***
–0.149***
0.012
–0.065
–0.041
0.023
Fear of Crime
Victimization
Conservative Crime
Ideology
Community Disorder
Informal Collective
Security
R-square =
F=
Significance =
–0.368***
0.331***
0.079
9.205
0.000
0.149
11.318
0.000
0.386
33.214
0.000
* p < 0.05; ** p < 0.01; *** p < 0.001; N = 539
Income level was also positively related to confidence in the police. The
respondents’ educational level, however, was not significant.
Equation 2 reveals that the influence of race (and of the other
demographic variables) on confidence remains statistically significant
even after controlling for the effects of three crime-related variables. Age,
gender and income also retained statistical significance. Equation 2 also
reveals that two of the three crime-related variables are significant: being
afraid of crime and having been a crime victim in the last two years both
reduce confidence in the police. Holding a conservative crime control
ideology, however, was not statistically significant.
Two additional findings relative to Equation 2 are worthy of
mention. First, the proportion of variance explained by Equation 2
(R2 = 0.149) is almost twice that explained by Equation 1 (R2 = 0.079).
This finding suggests that the crime-related variables in this equation
contribute substantially to our ability to predict public confidence in the
12
American Journal of Police, Vol. XV, No. 1 1996
police. Second, the standardized regression coefficients for the variables
indicate that fear of crime and recent victimization experiences exert a
larger effect on confidence in the police than do any of the demographic
variables.
In Equation 3, the final and most complete model we present, two
contextual variables are introduced: citizen perceptions of neighborhood
disorder and informal collective security. When these variables are
included, the effect of race on confidence in the police is no longer
significant. In fact, of the eight variables brought forward from Equation
2, only the respondents’ gender remains statistically significant.
The two community context variables, first included in Equation
3, are both significant determinants of confidence in the police.
Perceptions of community disorder exerted a significant negative effect
on confidence: As citizens’ perceptions of incivility increased, confidence
in the police decreased. Additionally, the standardized regression
coefficient for this variable reveals that perceptions of disorder had the
largest effect of all the variables included in this equation.
The second contextual variable, perceptions of the willingness of
neighbors to assist the police, had a significant positive effect on public
confidence: As perceptions of informal collective security increase,
confidence in the police also increases. The effect of this variable on
confidence in the police was the second largest of the included variables.
Finally, the variance explained by variables in Equation 3 (R2 =
0.389) substantially exceeds that explained by Equation 2 (R2 = 0.149).
The community context variables, therefore, appear to have both
statistical and substantive significance for the explanation of public
confidence in the police.
DISCUSSION
This study represents an attempt to understand more clearly the
relationship between race and public confidence in the police. Although
the initial analysis (reported in Equation 1) revealed a significant race
effect, the introduction of contextual factors eliminated this relationship.
For our sample, then, respondents’ race is not a significant determinant of
confidence in the police.
Furthermore, results for Equation 3 support the proposition that
the community context is the most important determinant of public
confidence in the police. Specifically, citizens’ perceptions of disorder
American Journal of Police, Vol. XV, No. 1 1996
13
exerted the strongest impact on confidence. Thus it appears that citizens
hold the police at least partially responsible for the disorder – the “broken
windows” – in their neighborhoods. Whether the police are capable of
containing the types of incivilities that appear to undermine public
confidence is a matter that remains open to debate. In any case, our
respondents appear to perceive the police as the government’s first-line
representative, responsible for controlling neighborhood disorder.
In addition to perceptions of disorder, the effects on confidence of
the other community context variable – informal collective security, or
the willingness of neighbors to provide protection from crime – were
significant and positive. As proposed in the introduction, this finding
suggests that confidence in neighbors breeds confidence that the police
will provide protection – that is, informal collective security encourages
feelings of formal collective security (see McDowall and Loftin, 1983).2
More broadly, this finding suggests that neighborhood social integration
may provide a supportive context in which residents are reminded that
they are not isolated individuals cut off from the larger social order. In
short, social bonds may encourage their identification with and positive
evaluation of formal institutional arrangements.
When viewed together, the findings on race and community
context yield an important conclusion: Attitudes toward the police may
not be regulated by a person’s race per se, but by the social context in
which the person is situated. Accordingly, past research, which typically
has reported that minorities appraise law enforcement less favorably, may
be in need of revision. Given that race became nonsignificant when
contextual variables were controlled, it is possible that much of the
existing research, which did not include contextual variables, has based
findings on misspecified models.
A further comment about our conclusions on community context
is in order: our measures of neighborhood context were perceptual rather
than objective indicators. This qualification, however, does not mean that
the results reported are unimportant. Indeed, research using objective
indicators, such as crime rates, has generally found that these measures
explain very little of the variation in citizens’ attitudes toward the police
(Stipak, 1979). In contrast, perceptions of neighborhood conditions may
well be more salient to respondents than less proximate, objective
conditions. If so, research indicates that these beliefs about local
community conditions are more likely to be “cognitively accessed”
(Fazio and Williams, 1986), which in turn may cause them to influence
attitudes toward the police. Furthermore, the impact of these perceptions
14
American Journal of Police, Vol. XV, No. 1 1996
may be especially important where the problem is one that the public
considers within the function and control of the police (Percy, 1986;
Stipak, 1979). For similar reasons, the public’s beliefs about the level of
collective informal control within the community are likely to exert a
much greater effect than actual measures of citizen participation.
An additional and consistent finding was the significant influence
of respondents’ gender on confidence in the police: even with a range of
factors controlled, females were more likely than males to express
confidence in the police. This finding is consistent with several prior
studies (Apple and O’Brien, 1983; Thomas and Hyman, 1977).3 It has
been suggested that women hold more positive sentiments because they
generally have contact with the police that is less antagonistic than the
contact that typically occurs between males and the police. Compared
with males, females are more likely to make service requests of the police
and are less likely to be involved in situations in which the police are
acting against them. Because (as noted) our data set did not contain
measures of citizen contacts with the police, we were unable to explore
whether differential police contact can account for the gender gap in
confidence in law enforcement.
In connection with this point, we should state again that existing
research suggests that citizens’ experiences with the police influence their
evaluations of the overall quality of police performance (Brown and
Coulter, 1983; Dean, 1980; Mastrofski, 1981; Parks, 1984; Scaglion and
Condon, 1980). Even so, the model presented in Equation 3 in Table 2
performs well: The explained variance by the model’s variables
(R2 = 0.386) exceeds the variance explained in prior studies that have
included police contact variables (see, for example, Brandl et al., 1994;
Dean, 1980). It remains an empirical question, of course, whether policecitizen contact measures will significantly increase the explained
variance or specify the relations reported here. In the absence of this or
other research, however, our study suggests that an accurate
understanding of the public’s attitudes toward the police must include an
assessment of the social context in which citizens reside.
Finally, the impact of neighborhood problems on confidence in
the police should be of special concern to law enforcement
administrators. Even though many neighborhood problems are not
susceptible to control or even regulation by the police, citizens’
perceptions of these problems provide salient information that appears to
influence their confidence in the police. Thus, improving citizens’
American Journal of Police, Vol. XV, No. 1 1996
15
attitudes is likely to be a challenging task for police agencies to undertake
successfully.
NOTES
A previous version of this paper was presented at the 1994
meeting of the American Society of Criminology, Miami. Address
correspondence to: Liqun Cao, Department of Sociology, Anthropology,
and Criminology, Eastern Michigan University, Ypsilanti, MI 48197.
1.
Dean (1980) in her study estimated a series of regression
equations using a variety of contact variables. When the number
of “bad” and “good” contacts are included as two variables the R2
increased from 0.091 to 0.155. When citizen evaluations and the
nature of the contact are included (request for information,
response to a victimization, stopped by the police) the R 2
increased to 0.165.
2.
We suggest that perceptions of informal collective security
influence confidence in the police because information about
informal collective security is likely to be received on a daily
basis. However, because our data are not longitudinal, we cannot
rule out the possibility of reverse (or reciprocal) causal ordering
between informal and formal collective security.
3.
It should be noted that there are also research findings suggesting
that gender is not a significant predictor of attitudes toward the
police (Boggs and Galliher, 1965; Brandl et al., 1994; Brown and
Coulter, 1983).
REFERENCES AND FURTHER READING
Albrecht, S.L. and Green, M. (1977), “Attitudes toward the Police and the Larger
Attitude Complex: Implications for Police-Community Relationships”,
Criminology, 15:67-86.
Apple, N. and O’Brien, D. (1983), “Neighborhood Racial Composition and
Residents’ Evaluation of Police Performance”, Journal of Police Science and
Administration, 11:76-84.
16
American Journal of Police, Vol. XV, No. 1 1996
Bayley, D. (1976), Forces of Order: Police Behavior in Japan and the United
States, Berkeley, CA: University of California Press.
Boggs, S.L. and Galliher, S.F. (1965), “Evaluating the Police: A Comparison of
Black Street and Household Respondents”, Social Problems, 13:393-406.
Bordua, D.J. and Tifft, L.L. (1971), “Citizen Interview, Organizational Feedback,
and Police-Community Relations Decisions”, Law and Society Review,
6:155-182.
Brandl, S., Frank, J., Worden, R.E. and Bynum, T.S. (1994), “Global and Specific
Attitudes toward the Police: Disentangling the Relationship”, Justice Quarterly,
11:119-134.
Brandl, S. and Horvath, F. (1991), “Crime Victim Evaluation of Police
Investigative Performance”, Journal of Criminal Justice, 19:293-305.
Brown, K. and Coulter, P.B. (1983), “Subjective and Objective Measures of
Police Performance”, Public Administration Review, 43:50-58.
Browning, S.L. and Cao, L. (1992), “The Impact of Race on Criminal Justice
Ideology”, Justice Quarterly, 9:685-701.
Carter, D.L. (1985), “Hispanic Perception of Police Performance: An Empirical
Assessment”, Journal of Criminal Justice, 13:487-500.
Carlson, H.M. and Sutton, M.S. (1979), “Some Factors in Community Evaluation
of Police Street Performance”, American Journal of Community Psychology,
7:583-591.
Covington, J. and Taylor, R.B. (1991), “Fear of Crime in Urban Residential
Neighborhoods: Implications of between- and within- Neighborhood Sources for
Current Models”, Sociological Quarterly, 32:231-249.
Cullen, F.T., Bynum, T.S., Garrett, K.M. and Greene, J.R. (1985), “Legislator
Ideology and Criminal Justice Policy: Implications from Illinois”, in Fairchild,
E.S. and Webb, V.J. (Eds), The Politics of Crime and Criminal Justice, Beverly
Hills, CA: Sage, 57-76.
Dean, D. (1980), “Citizen Ratings of the Police: The Difference Police Contact
Makes”, Law and Policy Quarterly, 2:445-471.
Decker, S. and Wagner, A. (1981), “Race and Citizen Complaints against the
Police: An Analysis of Their Interaction”, in Baker, R. and Meyer, F. (Eds),
Evaluating Alternative Law Enforcement Policies, Lexington, MA: Lexington.
American Journal of Police, Vol. XV, No. 1 1996
17
Dillman, D.A. (1978), Mail and Telephone Surveys: The Total Design Method,
New York, NY: John Wiley & Sons.
Dunaway, R.G. and Cullen, F.T. (1991), “Explaining Crime Ideology: An
Exploration of the Parental Socialization Perspective”, Crime and Delinquency,
37:536-554.
Erez, E. (1984), “Self-defined Desert and Citizen’s Assessment of the Police”,
Journal of Criminal Law and Criminology, 75:1276-1299.
Fazio, R.H. and Williams, C.J. (1986), “Attitude Accessibility as a Moderator of
the Attitude Perception and Attitude-Behavior Relations: An Investigation of the
1984 Presidential Election”, Journal of Personality and Social Psychology,
51:505-514.
Ferraro, K.F. and LaGrange, R.L. (1987), “The Measurement of Fear of Crime”,
Sociological Inquiry, 57:70-101.
Flanagan, T.J. and Maguire, K. (1992), Sourcebook of Criminal Justice Statistics,
Washington, DC: US Department of Justice.
Furstenberg, F. and Wellford, C.F. (1973), “Calling the Police: The Evaluation of
Police Service”, Law and Society Review, 8:393-406.
Goldstein, H. (1987), “Toward Community-oriented Policing: Potential, Basic
Requirements, and Threshold Questions”, Crime and Delinquency, 33:6-30.
Hindelang, M.J. (1974), “Public Opinion Regarding Crime, Criminal Justice and
Related Topics”, Journal of Research in Crime and Delinquency, 11:101-116.
Homant, R.J., Kennedy, D.B. and Fleming, R.M. (1984), “The Effect of
Victimization and the Police Response on Citizen’s Attitudes toward Police”,
Journal of Police Science and Administration, 12:323-32.
Jacob, H. (1971), “Black and White Perceptions of Justice in the City”, Law and
Society Review, 5:69-89.
Jefferson, T. and Walker, M.A. (1993), “Attitudes to the Police of Ethnic
Minorities in a Provincial City”, British Journal of Criminology, 33:251-266.
Jefferson, T. and Walker, M.A. (1992), “Ethnic Minorities in the Criminal Justice
System”, Criminal Law Review, 28:83-95.
Klyman, F.I. and Kruckenberg, J. (1974), “A Methodology for Assessing Citizen
Perceptions of the Police”, Journal of Criminal Justice, 2:219-233.
18
American Journal of Police, Vol. XV, No. 1 1996
Koenig, D.J. (1980), “The Effect of Crime Victimization and Judicial or Police
Contacts on Public Attitudes toward the Local Police”, Journal of Police Science
and Administration, 8:243-249.
Komanduri, S.M., Roebuck, J.B. and Smith, J.D. (1990), “The Image of the
Police in Black Atlanta Communities”, Journal of Police Science and
Administration, 17:250-257.
Larsen, K.S. (1968), “Authoritarianism and Attitudes toward Police”,
Psychological Reports, 3:349-350.
Lewis, D. and Salem, G. (1986), Fear of Crime: Incivility and the Production of
a Social Problem, New Brunswick, NJ: Transaction Books.
Lipset, S.M. and Schneider, W. (1983), The Confidence Gap: Business, Labor,
and Government in the Public Mind, New York, NY: The Free Press.
Luxenburg, J., Cullen, F.T., Langworthy, R.H. and Kopache, R. (1994), “Firearms
and Fido: Ownership of Injurious Means of Protection”, Journal of Criminal
Justice, 22:159-170.
McDowall, D. and Loftin, C. (1983), “Collective Security and the Demand for
Legal Handguns”, American Journal of Sociology, 88:1146-1161.
McGarrell, E.F. and Flanagan, T.J. (1987), “Measuring and Explaining Legislator
Crime Control Ideology”, Journal of Research in Crime and Delinquency,
24:102-118.
Maguire, K. and Pastore, A.L. (Eds) (1994), Sourcebook of Criminal Justice
Statistics 1993, Washington, DC: US Government Printing Office.
Mastrofski, S. (1981), “Surveying Clients to Assess Police Performance”,
Evaluation Review, 5:397-408.
Mirande, A. (1980), “Fear of Crime and Fear of the Police in a Chicano
Community”, Sociology and Social Research, 64:528-541.
Parks, R.B. (1981), “Police Response to Victimization Effects on Citizen
Attitudes and Perceptions”, in Skogan, W.G. (Ed.), Sample Surveys of the Victims
of Crime, Cambridge, MA: Ballinger, 89-104.
Parks, R.B. (1984), “Linking Objective and Subjective Measures of
Performance”, Public Administration Review, 44:118-127.
Peek, C.W., Lowe, G.D. and Alston, J.P. (1981), “Race and Attitudes toward
Local Police”, Journal of Black Studies, 11:361-374.
American Journal of Police, Vol. XV, No. 1 1996
19
Percy, S.L. (1980), “Response Time and Citizen Evaluation of Police”, Journal of
Police Science and Administration, 8:75-86.
Percy, S.L. (1986), “In Defense of Citizen Evaluations as Performance
Measures”, Urban Affairs Quarterly, 22:66-83
Reasons, C.E. and Wirth, B.A. (1975), “Police Community Relations Units: A
National Survey”, Journal of Social Issues, 31:27-33.
Scaglion, R. and Condon, R.G. (1980), “Determinants of Attitudes toward City
Police”, Criminology, 17:485-494.
Skogan, W.G. (1978), “Citizen Satisfaction with Police Services: Individual and
Contextual Effects”, Police Studies Journal, (special issue):469-479.
Skogan, W.G. (1987), Disorder and Community Decline, Evanston, IL: Center for
Urban Affairs and Policy Research, Northwestern University.
Skogan, W.G. (1990), Disorder and Decline: Crime and the Spiral of Decay in
American Neighborhoods, Berkeley, CA: University of California Press.
Skolnick, J.H. and Bayley, D.H. (1988), “Theme and Variation in Community
Policing”, in Tonry, M. and Morris, N. (Eds), Crime and Justice, Volume 10,
Chicago: University of Chicago Press, 1988, 1-37.
Smith, D. and Uchida, C. (1988), “The Social Organization of Self-help: A Study
of Defensive Weapon Ownership”, American Sociological Review, 53:94-102.
Smith, P.E. and Hawkins, R.O. (1973), “Victimization, Types of Citizen-Police
Contacts and Attitudes toward the Police”, Law and Society Review, 8:135-152.
Stipak, B. (1979), “Citizen Satisfaction with Urban Services: Potential Misuse as
a Performance Indicator”, Public Administration Review, 39:46-52.
Thomas, C.W. and Hyman, J. (1977), “Perceptions of Crime, Fear of
Victimization, and Public Perceptions of Police Performance”, Journal of Police
Science and Administration, 5:305-317.
US Department of Justice (1977), The Police and Public Opinion, Washington,
DC: Department of Justice.
White, M. and Menke, B.A. (1978), “A Critical Analysis of Surveys on Public
Opinions toward Police Agencies”, Journal of Police Science and Administration,
6:204-218.
Wilson, J.Q. and Kelling, G.L. (1982), “The Police and Neighborhood Safety:
Broken Windows”, Atlantic Monthly, (March):29-38.
20
American Journal of Police, Vol. XV, No. 1 1996
Wycoff, M.A. (1988), “The Benefits of Community Policing: Evidence and
Conjecture”, in Greene, J.R. and Mastrofski, S. (Eds), Community Policing:
Rhetoric or Reality, Westport, CT: Praeger: 103-120.
Zamble, E. and Annesley, P. (1987), “Some Determinants of Public Attitudes
toward the Police”, Journal of Police Science and Administration, 15:285-290.
Appendix A:
SUMMARY OF MEASURES USED IN THE ANALYSIS
Independent Variables
1. Race
1 = whites; 0 = nonwhites
2. Age
Natural age in years at the time of the survey
3. Gender
1 = male, 0 = female
4. House Income
1 = $2,000 and below; 2 = 2,001 to 6,000; 3 = 6,001 to
10,000; 6 = 20,001 to 25,000; 7 = 25,001 to 30,000; 8
= 30,001 to 35,000; 9 = 35,001 to 40,000; 10 = 40,001
to 45,000; 11 = 45,001 to 50,000; 12 = 50,001 to
55,000; 13 = 55,001 to 60,000; 14 = 60,001 to 70,000;
15 = 70,001 to 80,000; 16 = 80,001 to 90,000; 17 =
90,001 to 120,000; 18 = $120,001 and above
5. Education
1 = did not graduate from high school; 2 = graduated
from high school; 3 = some years of college; 4 =
graduated from college; 5 = finished one or more years
of graduate school
6. Informal Collective
Security
A combination of three items: (a) The people in my
neighborhood watch one another’s home when no one
is home. (b) In my neighborhood, people will call the
police right away if they think a crime is being
committed. (c) If some criminal was trying to rob me,
my neighbors wouldn’t just stand there and watch, but
would come help me out. 1 = very strongly disagree;
2 = strongly disagree; 3 = disagree; 4 = agree; 5 =
strongly agree; and 6 = very strongly agree
American Journal of Police, Vol. XV, No. 1 1996
21
7. Community
Disorder
A combination of eight items: How much of a problem
is each of the following: (a) noisy neighbors – people
who play loud music, have late parties, or have noisy
fights; (b) garbage or litter on the streets or sidewalks;
(c) people not keeping up their houses or apartments
– like not painting or fixing windows that break; (d)
groups of teenagers hanging out on the corners or in
the streets; (e) people selling or using illegal drugs in
the neighborhood; (f) vandalism – like kids breaking
windows or writing on walls or things like that; (g)
teenage gangs who commit crime; (h) people who say
insulting things or bother people as they walk down
the street. 1 = not a problem; 2 = some problems;
3 = big problem
8. Fear of Crime
Think back to those times when you might have felt
afraid that you might be a crime victim. How many
times have you felt afraid of crime in the last month?
1 = once; 2 = two to three; 3 = three to five; 4 = more
than five
9. Victimization
A combination of seven items. In the last two years,
have any of the following crimes been committed
against you: (a) someone broke into your house; (b) had
property stolen from your house or yard; (c) someone
stole, broke into, or vandalized your car; (d) someone
held you up on the street and robbed you; (e) someone
threatened to beat you up or threatened you with a
knife, gun, or other weapon; (f) someone actually beat
you up; (g) some other crime not listed here happened
to you. 1 = yes; 0 = no
10. Conservative Crime A combination of five items; (a) stiffer jail sentences
Ideology
are needed to show criminals that crime does not pay;
(b) juvenile criminals are treated too leniently by the
courts; (c) crime has increased in recent times because
society has become too permissive; (d) these days,
criminals have too many legal rights; and (e) if we
really cared about crime victims, we would make sure
that criminals were caught and given harsh
punishment. 1 = very strongly disagree; 2 = strongly
disagree; 3 = disagree; 4 = agree; 5 = strongly agree;
and 6 = very strongly agree
22
American Journal of Police, Vol. XV, No. 1 1996
Dependent Variable
Confidence in
the Police
A combination of five items: (a) when people in my
neighborhood call the police, they come right away;
(b) the police do a good job in my neighborhood in
making sure that no one disturbs the peace; (c) the
police care a lot about the safety of the people in my
neighborhood; (d) the police do a good job in
protecting me against crime; and (e) there are not
enough police in my neighborhood to deal with crime.
1 = very strongly disagree; 2 = strongly disagree;
3 = disagree; 4 = agree; 5 = strongly agree; and
6 = very strongly agree
Appendix B:
CORRELATION MATRIX OF VARIABLES IN ANALYSIS (N = 539)
Variables
1
Confidence in
Police
1.00
2. Race
0.12
3. Age
0.17
4. Gender
–0.10
5. Household
Income
0.12
6. Education
0.04
7. Informal
Collective
Security
0.46
8. Community
Disorders
–0.51
9. Fear of
Crime
–0.25
10. Victimization –0.25
11. Conservative
Ideology
0.05
2
3
4
5
6
7
8
9
10
11
1.
Means
Standard
Deviation
1.00
–0.04 1.00
0.14 –0.08 1.00
0.16 –0.18 0.19 1.00
0.13 –0.33 0.13 0.48 1.00
0.08 0.12 –0.09 0.15 0.04 1.00
–0.15 –0.15 0.01 –0.21 –0.16 –0.27 1.00
–0.10 –0.14 –0.13 –0.13 –0.02 –0.13 0.36 1.00
–0.02 –0.24 0.12 –0.04 0.02 –0.15 0.32 0.24 1.00
0.05 0.21 –0.08 –0.09 –0.25 0.06 0.04 0.01 –0.10 1.00
16.8
0.79 47.1
0.43 9.25 3.25 11.4
4.66 2.01 0.67 18.6
4.4
0.41 18.0
0.50 4.25 1.26 2.7
3.69 1.19 0.93 5.0