due in 5 hours

lee_gibbs_2015.pdf

Home >Biology homework help >due in 5 hours

Race and attitudes toward police: the mediating effect

of social distance Joongyeup Lee and Jennifer Gibbs

School of Public Affairs, Pennsylvania State University, Middletown, Pennsylvania, USA

Abstract Purpose – Given the consistent finding in the literature that members of minority groups hold less favorable views of the police than white citizens, social distance may be an important, yet untested, mediator. The purpose of this paper is to examine the mediating effect of social distance net of other established correlates. Design/methodology/approach – A sample of students attending a university in the northeastern USA completed an online survey in 2013. The survey was about their contact with the police, attitudes toward the police, and lifestyles, among others. Findings – Race, along with other predictors, significantly influenced confidence in police. However, race is the only factor that turns nonsignificant when social distance is included in the model. Mediation tests confirmed that social distance mediates the relationship between race and confidence in the police. Research limitations/implications – To maximize confidence in the police, administrators should focus on closing the social distance between the public and the police through initiatives like community policing. Originality/value – While there is extensive research on public attitudes toward the police, social distance has been neglected as a determinant, despite movements like community policing that promote citizens’ relational closeness to the police – that is, to decrease the social distance between police and the public. The current study would be an exploratory study and reference for future studies. Keywords Community-oriented policing, Community relations, Public perceptions, Citizen satisfaction Paper type Research paper

In late 2014, court failure to prosecute police officers perceived as using excessive force against minorities sparked anti-police protests across the USA and especially in areas like Ferguson, MO, and New York City. A number of protesters and commentators held that the civil unrest would continue until police-minority relations change and police recognize that “black lives matter” – a movement slogan. These events resonate with an important social paradigm that has lingered on in the USA well into twenty-first century: minority citizens have little confidence in their police (e.g. Lee and Gibbs, 2014).

The police, as a government institution in a democratic society, exist to serve the public. Public attitude is a barometer of the legitimacy in the institution’s function and ultimately determines its longevity (Barbalet, 2009; Hatry, 1999). Legitimacy is essential, particularly for policing, considering that the police regulate individual rights and freedoms and their contact with the public is singularly high in frequency. If thePolicing: An International Journal

of Police Strategies & Management Vol. 38 No. 2, 2015 pp. 314-332 ©EmeraldGroup Publishing Limited 1363-951X DOI 10.1108/PIJPSM-03-2015-0034

Received 15 March 2015 Accepted 15 March 2015

The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/1363-951X.htm

The authors thank Hoon Lee, PhD at the Chosun University in South Korea, for sharing the survey instrument and conceptual ideas for this study.

314

PIJPSM 38,2

public regards their police service favorably, they are likely to abide by the law and police directives (Hinton and Newborn, 2009; Tyler, 1990; Tyler and Huo, 2002); on the other side of the coin, officers who believe citizens view them favorably have higher job satisfaction, which may affect their job performance (Yim and Schafer, 2009). Ultimately, police administration would benefit immensely from research on factors affecting citizens’ attitudes toward the police because “widespread confidence in the police makes law enforcement officers’ work easier and more effective” (Skogan, 2006, p. 118; see also Warren, 2011, for a discussion about how attitudes toward police shape individuals’ perceptions of their encounters with police).

Several indicators of public confidence in the police have been introduced in the literature, with race emerging as a prominent factor. In particular, the general finding is that members of minority groups have lower confidence in the police than whites (Dowler and Sparks, 2008; Schuck and Rosenbaum, 2005; Weitzer and Tuch, 2004), even in cities where the police are predominantly minorities (Weitzer et al., 2008). One factor missing from these analyses, though, is social distance. Social distance is “a subjective perception or experience of distance from another person or other persons” (Magee and Smith, 2013, p. 159). This distance can be: a difference between the “self and other,” a difference between “one’s own and others’ group identities,” “unfamiliarity with others,” and “degree of felt closeness to another person” (Magee and Smith, 2013, p. 159). It is a by-product of differential social experiences, rather than an invariant disposition. That is, a person with more informal and personal interaction with the police is likely to feel less social distance to the police. Therefore, a person’s social distance to the police can improve or deteriorate based on the magnitude of informal interaction. Given that the majority of the police in the USA are white and members of minority groups are disproportionately more likely to feel targeted by police, the principle of homophily suggests that they are less likely to have such positive interactions with police than whites. This suggests that citizens, regardless of race, who have more relationalities to members of police agencies may hold more positive attitudes toward them. In other words, differential social distance may mediate the racial disparity in public attitude toward the police, but this hypothesis has yet to be tested.

The purpose of this research is to address this gap in the literature. We begin by surveying the literature on attitudes toward police, with a particular focus on the importance of race. Next, we describe social distance and outline the reasons it likely mediates the relationship between race and confidence in the police. Before presenting the results, we describe the methods used to test this hypothesis. Finally, we discuss the findings, draw conclusions and offer policy recommendations to police administrators and suggestions for future research.

Attitudes toward police There is an extensive research on attitudes of US citizens toward the police, establishing that demographic factors like race and ethnicity, age, gender, and other variables like criminal victimization, direct contact with police, indirect experience through family and friends, and exposure to media coverage of police misconduct all affect citizen satisfaction with law enforcement officers. Selected studies from this literature are summarized in Table I. Indeed, a substantial body of literature finds that members of minority groups such as African-Americans have lower confidence in the police than their white counterparts (Brunson, 2007; Callanan and Rosenberger, 2011;

315

Race and attitudes

toward police

So ur ce

U ni t/p

er io d

M et ho d

D em

og ra ph

ic va ri ab le s

O th er

va ri ab le s

C on fid

en ce

in po lic e

M ill er

an d D av is

(2 00 8)

N ew

Y or k Ci ty ,2 00 2

A N CO

V A

(n ¼ 57 6)

B la ck

H is pa ni cb

O th er

ns Fe m al e ns

A ge

B or n ou ts id e U SA

O w ns

ho m e ns

E du

ca tio

n ns

B ad

po lic e- in iti at ed

co nt ac tb

B ad

re sp on de nt -in

iti at ed

co nt ac tb

B ad

vi ca ri ou s po lic e- in iti at ed

co nt ac tb

B ad

vi ca ri ou s re sp on de nt -in

iti at ed

po lic e co nt ac tb

E xp

er ie nc ed

ba d ne w s ns

N ew

s ad di ct s ns

T ab lo id

ne w s co ns um

er sb

H ig hb

ro w

ne w s co ns um

er s ns

La ia

nd Zh

ao (2 01 0)

H ou st on ,2 00 8

O LS

(n ¼ 75 6)

B la ck

H is pa ni c ns

O th er

ra ce s ns

A ge

ns G en de r ns

E du

ca tio

n ns

N ei gh

bo rh oo d fa ct or s

V io le nt

cr im

e in ci de nt

ns V ic tim

iz at io n ex pe ri en ce

ns Fe ar

of cr im

e ns

Co nc en tr at ed

di sa dv

an ta ge

Sa tis fa ct io n w ith

po lic e w or ka

Co nt ac t ex pe ri en ce

w ith

po lic e ns

Ca lla na n an d

R os en be rg er

(2 01 1)

Ca lif or ni a, 19 99

O LS

(n ¼ 4, 24 5)

M al eb

A ge

B la ck

La tin

o ns

E du

ca tio

n ns

In co m e ns

D em

oc ra t ns

R ep ub

lic an

V ic tim

Fe ar

A rr es te db

N ei gh

bo rh oo d cr im

Fr eq ue nc y of

w at ch in g tv

cr im

e dr am

a ns

Fr eq ue nc y of

w at ch in g cr im

e re al ity

tv sh ow

Fr eq ue nc y of

re ad in g ne w sp ap er

ns Fr eq ue nc y of

vi ew

in g tv

ne w sa

Fr eq ue nc y of

w at ch in g al lt v ns

(c on

tin ue d )

Table I. Selected multivariate studies of attitudes toward police in the USA

316

PIJPSM 38,2

So ur ce

U ni t/p

er io d

M et ho d

D em

og ra ph

ic va ri ab le s

O th er

va ri ab le s

T ay lo r an d La

w to n

(2 01 2)

Pe nn

sy lv an ia ,2 00 3

M ul til ev el m od el in g

(n ¼ 1, 28 9)

Fe m al ea

W hi te b

A ge

W or k fu ll tim

e ns

M ar ri ed

ns H om

eo w ne rb

Ch ild

re n in

ho us e ns

H ou se ho ld

in co m e ns

E du

ca tio

Lo ca lo

rg an iz at io na l

in vo lv em

en tb

Co un

ty ty pe

R ur al b

Su bu

rb an

M un

ic ip al ity

St ab ili ty

ns SE

S ns

% no nw

hi te

ns V io le nt

cr im

e ra te

ns Pe rc ei ve d lo ca ls oc ia lc oh es io n ns

Pe rc ei ve d lo ca lp

ro ce du

ra lj us tic ea

Pe rc ei ve d in ci vi lit ie sb

W u (2 01 4) c

Se at tle ,2 00 2- 20 03

O LS

cr os s- se ct io n

re gr es si on

(n ¼ 3, 83 0)

B la ck

A si an

ns H is pa ni c ns

O th er

Fe m al e ns

A ge

Co lle ge

ns E m pl oy ed

ns H ou se ho ld

in co m e ns

U S- bo rn

Po lic e vi si bi lit ya

Pr op er ty

cr im

e vi ct im

ns V io le nt

cr im

e vi ct im

Fe ar

of cr im

N ei gh

bo rh oo d di so rd er

N ei gh

bo rh oo d co de

of vi ol en ce

Sa tis fa ct io n w ith

po lic e

G ar ci a an d Ca

o (2 00 5)

Sm al lc ity

in no rt he as te rn

U SA

, 19 99

O LS

(n ¼ 22 7)

A fr ic an -A m er ic an

H is pa ni cb

A ge

M al ea

In co m e ns

V ic tim

iz at io n ns

Cr im

e su sp ec t ns

Po lic e fr ie nd

ns Sa fe ty

D is tr ic tb

(c on

tin ue d )

Table I.

317

Race and attitudes

toward police

So ur ce

U ni t/p

er io d

M et ho d

D em

og ra ph

ic va ri ab le s

O th er

va ri ab le s

Sm ith

(2 00 5)

W as hi ng

to n,

D C,

19 98

O LS

(n ¼ 80 0)

A fr ic an -A m er ic an

G en de r ns

A ge

ns E du

ca tio

Co nt ac t w ith

po lic e in

la st

ye ar

Pr op er ty

vi ct im

iz at io nb

Pe rs on al

vi ct im

iz at io n ns

Q ua lit y of

lif e in

ne ig hb

or ho od

N ei gh

bo rh oo d di so rd er

W ei tz er

an d T uc h

(2 00 5)

U SA

,2 00 2

O LS

(n ¼ 1, 47 0)

R ac e (w hi te ,b

la ck ,H

is pa ni c) d

E du

ca tio

In co m e ns

G en de r ns

A ge

Ci ty

si ze

ns U rb an

vs su bu

rb an

re si de nc e

ns R eg io n of

U SA

Pe rc ep tio

ns of

ne ig hb

or ho od

sa fe ty

D ur in g th e da ya

A t ni gh

Cr im

E ff ec tiv

en es s of

po lic ea

Pr es en ce

of co m m un

ity po lic in ga

M ed ia

ex po su re

ns Pe rc ei ve d po lic e m is co nd

uc tb

Pe rs on al

ex pe ri en ce

w ith

po lic e ns

V ic ar io us

ex pe ri en ce

w ith

po lic eb

D ow

le r an d Sp

ar ks

(2 00 8)

12 U S ci tie s, 19 98

O LS

(n ¼ 11 ,2 63 )

A fr ic an -A m er ic an

H is pa ni cb

M al eb

A ge

E du

ca tio

n ns

In co m e ns

Ci ty

ch ar ac te ri st ic s

% be lo w

po ve rt y lin

e ns

U ne m pl oy m en t ra te b

Po pu

la tio

n in cr ea se

19 90 -2 00 0 ns

V io le nt

cr im

e ra te a

Pr op er ty

cr im

e ra te b

Po lic e pe r po pu

la tio

R ac ia ld

is pa ri ty

in po lic e fo rc ea

Po pu

la tio

n of

ci ty

ns Se ri ou s cr im

e in

ne ig hb

or ho od

– Q ua lit y of

lif e in

ne ig hb

or ho od

Fe ar

of cr im

V ic tim

of th ef tb

V ic tim

of as sa ul tb

Po lic e co nt ac ta

R ep or te d in ci de nt

to po lic eb

(c on

tin ue d )

Table I.

318

PIJPSM 38,2

So ur ce

U ni t/p

er io d

M et ho d

D em

og ra ph

ic va ri ab le s

O th er

va ri ab le s

D ai

an d Jo hn

so n

(2 00 9

Ci nc in na ti,

O H ,1 99 7

H LM

(n ¼ 61 4

in di vi du

al s in

29 ne ig hb

or ho od s)

A fr ic an -A m er ic an

M al e ns

A ge

H om

eo w ne r ns

Po si tiv

e co nt ac t w ith

po lic e ns

N eg at iv e co nt ac t w ith

po lic eb

Q ua lit y of

lif e

H as

be co m e be tt er

W ill

be co m e be tt er

Pe rc ei ve d in ci vi lit yb

Pe rc ei ve d sa fe ty

N ei gh

bo rh oo d le ve l

Co nc en tr at ed

di sa dv

an ta ge

− Cr im

e ra te

ns R hi ne be rg er -D un

n an d Ca

rl so n (2 00 9)

Ch ic ag o (P H D CN

), 19 95

H LM

(n ¼ 8, 56 2

in di vi du

al s in

34 3

ne ig hb

or ho od s)

B la ck

La tin

M al e ns

Y ea rs

in ne ig hb

or ho od

ns M ov es

pa st

5 ye ar s ns

Fa m ily

in co m ea

A ge

Y ea rs

ed uc at io na

N ev er

m ar ri ed

ns Se pa ra te d/ di vo rc ed

ns W id ow

ed ns

H om

eo w ne r ns

N ei gh

bo rh oo d- le ve l

E co no m ic di sa dv

an ta ge

R ac ia lh

et er og en ei ty

ns E th ni c he te ro ge ne ity

R es id en tia

ls ta bi lit y ns

Pe rc ei ve d po lic e m is co nd

uc t

Sc hu

ck et al .( 20 08 )f

Ch ic ag o, 20 02

H LM

(n ¼ 42 8

in di vi du

al s in

69 ne ig hb

or ho od s)

A fr ic an -A m er ic an

H is pa ni ca

M al e ns

A ge

ns H om

eo w ne r ns

E du

ca tio

n ns

N eg at iv e po lic e in iti at ed

ns N eg at iv e re si de nt

in iti at ed

ns N eg at iv e vi ca ri ou s ex pe ri en ce

Po si tiv

e vi ca ri ou s ex pe ri en ce

ns Le ss

pe rc ei ve d di so rd er

Co nc en tr at ed

di sa dv

an ta ge

(c on

tin ue d )

Table I.

319

Race and attitudes

toward police

So ur ce

U ni t/p

er io d

M et ho d

D em

og ra ph

ic va ri ab le s

O th er

va ri ab le s

W ei tz er

et al .( 20 08 )

W as hi ng

to n,

D C,

19 98 -2 00 0

O LS

cr os s- se ct io n

an al ys is (n ¼ 2, 73 1)

B la ck

E du

ca tio

M ar ri ed

Ch ild

re n at

ho m ea

A ge

ns G en de r ns

Y ou ng

× B la ck

Y ou ng

× M al ea

Pe rc ei ve d cr im

Pe rc ei ve d di so rd er

Pe rc ei ve d co m m un

ity po lic in gb

Cr im

e- re la te d ci tiz en

in iti at ed

po lic e co nt ac ts b

N on cr im

e ci tiz en

in iti at ed

po lic e co nt ac ts a

Po lic e in iti at ed

po lic e- ci tiz en

co nt ac ts

(v eh ic le )a

Po lic e in iti at ed

po lic e- ci tiz en

co nt ac ts a

N ot es

: ns , no t si gn

ifi ca nt ; a o r

b d ir ec tio

n of

si gn

ifi ca nt

re la tio

ns hi p;

c W u (2 01 4)

us ed

fo ur

ou tc om

e m ea su re s as se ss in g ge ne ra l at tit ud

es to w ar d po lic e

(i. e. pr ob le m -s ol vi ng

,h as sl in g, ra ci al pr of ili ng

,p ol ic e bi as ). H er e, on ly

th e re su lts

of th e m od el us in g pr ob le m

so lv in g ar e re po rt ed ;d W ei tz er

an d T uc h (2 00 5)

fo un

d di ff er in g fa ct or s af fe ct in g ci tiz en

sa tis fa ct io n w ith

po lic e ac ro ss

th re e ra ci al /e th ni c gr ou ps ,a s di d Ca

lla na n an d R os en be rg er

(2 01 1) ;e A t th e in di vi du

al le ve l, D ai an d Jo hn

so n (2 00 9) fo un

d th at

A fr ic an -A m er ic an

re sp on de nt s ha d si gn

ifi ca nt ly

lo w er

sa tis fa ct io n w ith

po lic e th an

w hi te re sp on de nt s; ho w ev er ,t hi s

re la tio

ns hi p be co m es

no ns ig ni fic an t w he n ne ig hb

or ho od

le ve l ch ar ac te ri st ic s ar e in cl ud

ed ;f Sc hu

ck et

al .( 20 08 ) re po rt ed

th e re su lts

of m od el s w ith

fo ur

ou tc om

e m ea su re s; he re ,w

e in cl ud

e th e m os t re le va nt :f ea r of

po lic e m is be ha vi or

Table I.

320

PIJPSM 38,2

Dowler and Sparks, 2008; Garcia and Cao, 2005; Lai and Zhao, 2010; Miller and Davis, 2008; Rhineberger-Dunn and Carlson, 2009; Schuck and Rosenbaum, 2005; Schuck et al., 2008; Smith, 2005; Weitzer and Tuch, 2004, 2005; Wu, 2014; but see McCluskey et al., 2008; Taylor and Lawton, 2012). For example, Schuck et al. (2008) examined Chicago residents’ attitudes toward the police in terms of race; controlling for negative contact with the police, their sample reaffirmed this racial disparity (see also Schuck and Rosenbaum, 2005). Notably, not only direct but also indirect experiences through family members or friends appear to diminish public confidence.

However, one factor missing from the literature on attitudes toward police is social distance, which has the potential to mediate the relationship between race and public confidence in the police. One exception is Garcia and Cao’s (2005) study on satisfaction of police in a small city located in northeastern USA In their models, they included a variable measuring whether the respondent had a police friend, although this dichotomous variable was included as a “contact” variable rather than a measure of social distance. Having a police friend failed to reach significance in any of the models, but this nonsignificant finding could be due to using only one questionnaire item to measure this concept and failing to account for some variables known to be related to the outcome (e.g. media exposure; Garcia and Cao, 2005). The current study corrects for these limitations. Before discussing the methods, we first conceptualize social distance and elaborate on the importance of including social distance in studies of confidence in police.

Social distance The varying degrees of public attitudes toward the police among citizens may be more a function of differential social distance rather than race. Social distance is theorized to have an effect on interest in others and socially engaging and disengaging emotions (Magee and Smith, 2013). That is, the greater the social distance between two parties, the less likely they will experience guilt, embarrassment, gratitude, or compassion and the more likely they will feel pride, disgust, contempt, and anger (Anagnostopoulosm and Hantz, 2011; Magee and Smith, 2013), which could explain public attitudes toward police, regardless of race.

The literature acknowledges that lowering social distance between citizens and the police is the primary approach in community policing, mainly because the average law- abiding citizen would expect to have no contact with the police, rendering citizen-police encounter more ominous than amicable (Manning, 1992; Reisig, 2010). This inevitable unfamiliarity and even animosity generates social distance between the police and the public. Acknowledging this intrinsic problem in law enforcement, the police have accommodated the philosophy of community-oriented policing as the core context of modern policing in the USA since the 1980s (Dowler and Sparks, 2008). Although a variety of forms may exist for community policing, the central idea is to increase citizens’ support for the police by reducing social distance between them and the police (Manning, 1998; Reisig, 2010). Some research indicates that people’s perceptions of effective community policing in their neighborhoods influence their attitudes toward the police (Gill et al., 2014). In particular, Weitzer and colleagues (2008) found that residents – especially blacks – viewed police misconduct as less of a problem when they also believed police were attentive to community concerns and worked with citizens to solve problems. Accordingly, to the extent that community policing is designed to minimize social distance between the police and the public, citizens’ perceived closeness to and familiarity with the police would directly measure the procedural aspect of

321

Race and attitudes

toward police

community policing. Despite the high stake public satisfaction holds for the central theme of modern policing, little attention has been given to social distance in the research of public attitude toward the police[1].

Criminologists have been discussing the concept of social distance at least since Merton (1968). Recently, Hipp (2010) tied social distance to social disorganization theory, pointing out that those with greater social distance, which he defined using social categories, interact with one another much less, reducing social control, than those with less social distance who, thus, interact more. Studying perceptions of disorder at the neighborhood level, Hipp (2010) found that residents with more social distance perceived more disorder than residents with less social distance[2]. While Hipp focussed on the relationship between social distance and crime and disorder, the literature on the connection between social distance and attitudes toward police is scant. The goal of this study is to contribute to filling this gap by examining the mediating effect of social distance on the relationship between race and confidence in the police.

Social distance has been conceptualized in several ways (e.g. Anagnostopoulosm and Hantz, 2011; Hipp, 2010; Magee and Smith, 2013). Drawing from the definition of social distance presented in the introduction (see Magee and Smith, 2013), we define social distance as the subjective perception of distance one has with the police. That perception of distance decreases with informal interactions. For example, when a person has a friend or family member in the police forces, s/he is expected to have little social distance with police. Additionally, this distance can include the difference between one’s own identity and his or her perceived identity of the police, familiarity with the police and or the perception of closeness with the police force.

It is also important to note that familiarity, or low social distance, does not necessarily warrant favoritism. A person who is close to and familiar with a group may dislike or even abominate the group for either personal or professional reasons. For example, one may find a local police officer less professional than s/he expected before being personally acquainted. Therefore, the current study proceeds with a nondirectional hypothesis for the effect of social distance on confidence in the police.

Methods Sample The sample was drawn from over 90,000 students registered at a large public state university in the northeastern USA as of March of 2014. With approval from the Institutional Review Board, an invitation for the online survey containing a description of the study and the web link was delivered to a random sample of 1,500 e-mail addresses. After four weeks of solicitation, 65 percent did not complete the survey and were excluded from our study, leaving 532 cases for the current study[3]. The response rate was about 35 percent of those invited, substantially higher than the literature, which suggests around a 20 percent response rate to surveys using an e-mail invitation (Kaplowitz et al., 2004).

Measurement Dependent variable. Confidence in the police was conceptualized using three five-point Likert indicators of personal attitude and perception: how much confidence do you have in the police, I am satisfied with the police, and the police are responsive to the community concerns in my neighborhood. Principal axis factor analysis identified a

322

PIJPSM 38,2

single latent structure, which accounted for more than 32 percent of the variance for any of the indicators. Cronbach’s α for reliability was 0.86. Factor loadings ranged from 0.772 (police are responsive to community concerns) to 0.927 (satisfied with the police). The factor was retrieved using regression method for further analysis and ranged from −2 to 1 with a mean of −0.463 (cf. Table II).

Independent variables. Magee and Smith (2013) recommend several social psychological scales to measure social distance, which they claim resembles the concept of psychological distance. These scales include: affiliation motivation indices like the Sociability Scale and the Need to Belong Scale; feeling close vs distant indices like the Relationship Closeness Inventory Scale, the Perceived Interpersonal Closeness Scale and the Loneliness Scale; and behaving close vs distant indices like the Relationship Closeness Inventory or Perceived Partner Responsiveness scale (Magee and Smith, 2013, p. 175). While these scales generally gauge how people feel toward others, they are likely to overlap with favoritism. In other words, these concepts may share the same latent structure with favoritism toward the object such that any discussion of their relationship would be tautological. Therefore, we used indicators of relational social distance in terms of family, friends, and neighborhood. In addition, these concepts better fit the definition of social distance adopted here.

Specifically, social distance was measured with three dichotomized indicators of positive interest and relational distance: I have at least one police officer among my family members, I have at least one police officer among my close friends, and I personally know at least one police officer in my neighborhood who is neither my family member nor my close friend. A sum of the 0 (yes) and 1 (no) across the three variables produced a scale ranging from 0, indicating closeness and familiarity, to 3, where the respondent is considered relatively unfamiliar with the police (cf. Table II). Social distance shows a mean scale of 2.10.

Although a variety of races was observed in this sample, whites accounted for a substantial portion (81.3 percent). Owing to small caseloads for other races and a concern for inflated degrees of freedom, a decision was made to dichotomize race into white (0) and minority (1).

Negative contact with police was comprised of three dichotomous indicators asking about individual experience of unreasonable stops, insulting language, or physical force by police officers in the past two years. While a little more than one in four reported an unreasonable stop by police, only few (3.5 percent) experienced physical

Min. Max. Mean SD

Confidence in the police −3 2 −0.06 1.07 Social distance 0 3 2.10 0.90 Minority 0 1 0.19 0.39 Negative contact 0 3 0.38 0.73 Negative Vicarious contact 0 6 0.82 1.39 Exposure to media 0 3 1.53 0.76 Victimization 0 1 0.20 0.40 Male 0 1 0.44 0.50 Age group 0 3 0.98 0.77 Note: n¼ 532

Table II. Descriptive statistics

323

Race and attitudes

toward police

force upon them. The prevalence of the same three incidents experienced by family and friends, respectively, was scaled as vicarious experience of negative contact with the police[4]. Next, exposure to media coverage of police misconduct was measured with three five-point Likert indicators asking how often the respondent heard or read about police use of excessive force, racial profiling, and corruption. For all of the three types of media reports, more than half of the respondents reported they ever heard or read the coverage. As was the case with social distance, neither negative contact nor exposure to media coverage was a scale in the psychometric sense of multiple questions measuring a single concept, but was instead an additive scale of counting the number of similar experience or incidents. Both negative contact with the police and criminal victimization scales show a small mean, reflecting relatively scarce prevalence. Third, criminal victimization was a single indicator asking whether the respondent was victimized in a criminal incident in the past two years. Most students (88 percent) had no experience of criminal victimization (cf. Table II).

Last, gender and age were also included. The sample has more males (56 percent) than females, so male is used as the reference category. While the average age in the sample was about 21, a close examination confirmed a severe positive skewness due to extreme cases (e.g. 51 years old). Because the current study is about college student population, and considering scarcity of college students over the age of 40, age was grouped into tens, ranging from teens (0) to 40s or older (3). The mean of age group is 0.98, indicating the average respondent was in his or her 20s.

Analytic strategy Figure 1 shows a hypothesized full path model for confidence in the police. Age, gender, and race are independently associated with varying experiences of negative contact with police, exposure to media coverage of police misconduct, criminal victimization, and social distance. In turn, each of these variables individually affects confidence in the police. Not only does this accord with logical inference of the causal paths but also it abides by the temporal order between variables. That is, experiences are past events, whereas perception and attitude are relatively present states of mind. For the main purpose of this study, two separate models, one as a base model without social distance and another as a full model, were developed using the LISREL program and compared for the mediating effect of social distance.

Demographics

Negative Contacts

Negative Media Reports

Victimization

Social Distance

Confidence in the Police

Negative Vicarious Experiences

Figure 1. Hypothesized full model for confidence in the police

324

PIJPSM 38,2

Results The base model includes all of the predictor variables except for social distance (cf. Table III). Race appeared to have a significant effect. That is, minority respondents have lower confidence in the police than white respondents (B¼−0.659, po0.05), confirming the literature. Additionally, minority students showed more negative contacts with the police (B¼ 0.345, po0.05), more negative vicarious experiences (B¼ 1.974, po0.01), and more exposure to media coverage of police misconduct (B¼ 0.758, po0.01). Male students had more negative contacts with the police (B¼ 0.409, po0.01) than their female counterparts. Criminal victimization was not regressed by any of the demographic variables. In turn, negative contact with police (B¼−0.503, po0.01), negative vicarious experience (B¼−0.096, po0.01), and exposure to media (B¼−0.370, po0.01) appeared to curtail confidence in the police. The base model explained 26.6 percent of variance in the confidence in the police.

The full model includes all of the variables of the base model and adds social distance (cf. Table IV). First, social distance was regressed only by race; minority students reported larger social distance (i.e. less familiarity) with police than their white

Negative contact

Negative vicarious contact

Media exposure Victimization

Confidence in the police

Minority 0.345* 1.974** 0.758** 0.010 −0.659* Male 0.409** 0.526 0.206 0.012 −0.179 Age group 0.010 −0.222 0.049 0.001 0.062 Negative contacts −0.503** Neg’ vicarious contact −0.096** Media exposure −0.370** Victimization −0.232 Notes: R2¼ 0.266; Root mean square error of approximation ¼ 0; Normed Fit Index ¼ 1.00; Root mean square residual¼ 0; Goodness of Fit Index¼ 1.00. *,**Significance level at po0.05; po0.01, respectively

Table III. Unstandardized

regression coefficients in the

base model

Social distance

Negative contact

Negative vicarious contact

Media exposure Victimization

Confidence in the police

Minority 0.628* 0.345* 1.974** 0.758** 0.010 −0.546 Male −0.109 0.409** 0.526 0.206 0.012 −0.206 Age group 0.008 0.010 −0.222 0.049 0.001 0.061 Negative contacts −0.467** Neg’ vicarious contact −0.107** Media exposure −0.368** Victimization −0.256 Social distance −0.168** Notes: R2¼ 0.281; Root mean square error of approximation ¼ 0.033; Normed Fit Index ¼ 0.99; Root mean square residual¼ 0.013; Goodness of Fit Index ¼ 0.97. *,**Significance level at po0.05; po0.01, respectively

Table IV. Unstandardized

regression coefficients in the

full model

325

Race and attitudes

toward police

counterparts (B¼ 0.628, po0.05). In turn, social distance had a significant diminishing effect on confidence in the police (B¼−0.168, po0.01), net of all other variables. Adding social distance increased the substantive significance as the full model explained 28.1 percent of variance in the confidence in the police.

Also notably, race is no longer significant, while all of the other findings from the base model remain the same. Specific to the change in the regression coefficient from the base model to the full model, a 17 percent reduction in the coefficient for race was noticeable. Further, a mediation test confirmed that social distance does indeed mediate the relationship between race and confidence in police. Sobel’s test statistic was 1.97, with a standard error of 0.033, which corresponds to a p-value of 0.048[5].

Discussion and conclusion While public confidence in the police in the USA remains high, research consistently indicates that minority groups have much less confidence in the police than Whites do. However, one factor missing from most of these studies is social distance, which is included here. In the path model without social distance, we find that race, along with negative contacts and exposure to media coverage of police misconduct, is an important predictor of confidence in the police. In fact, it is the only significant demographic predictor. However, race becomes nonsignificant once social distance is introduced in the model, while other predictors remain unchanged. Further analysis suggests that social distance substantially mediates the influence of race on confidence in police.

These findings, though, should be viewed with caution, considering some limitations inherent in this study. First, in light of external validity, the sample was drawn from university students at a single public university. Whether these results generalize to other college-age students or to the US population as a whole is an avenue for future research. That said, a recent poll of randomly selected residents from the state in which the university is located shows similar attitudes toward police as the student sample in the present study. For example, according to unpublished data, 78.9 percent of the state residents polled said they had at least some confidence that the police could protect them from violent crime, while 85.9 percent of the student sample believed the police are doing a good job controlling violent crime in their neighborhoods. Similarly, 67.4 percent of the state residents agreed or strongly agreed that police can be trusted, compared with 68.7 percent of the student sample who had quite a lot or a great deal of confidence in the police. These statistics suggest there may be some concordance between attitudes of university students and of resident of the state where the university is located, but, again, future research should determine the generalizability of these results.

Concerning face validity of our measurement, negative contact with police reflects police intervention as is perceived by the involved citizen to be unjust and unfair, rather than actual misconduct. It is sensible to doubt how accurate this would be as an indicator of police misconduct. A college student may underestimate his or her wrongdoing and blame the officer for simply enforcing law. While the police should still try to minimize the odds of this happening, future study could validate this measurement or develop a better indicator for the actual unjust enforcement by the police. Next, social distance has a variety of conceptualizations, but the current measures pertain only to the interactive and relational dimensions. For example, the Bogardus social distance scale is used widely in the field of sociology (Bogardus, 1925, 1933). Magee and Smith (2013) recommend several social psychological indices relevant

326

PIJPSM 38,2

to, but not directly measuring, social distance. Others use sociodemographic characteristics, including race, to create social distance indices (Hipp, 2010; Takagi et al., 2013). Although the measures used here are considered the most appropriate for the current study of public attitudes toward police, future studies may benefit from exploring other dimensions or developing a single scale with multiple dimensions of social distance. For example, McCormick and colleagues (2013) suggest using aggregated relational data to measure social distance in data sets incorporating social networking. Indeed, our measure of social distance asks only whether the respondent knows an officer, not how many. Future research employing social network analysis may uncover additional details about the mediating nature of social distance on the relationship between race and public attitudes toward the police. Further, scholars attempting to clarify the conceptualization of social distance as applied to police may consider additional elements, like citizen involvement in community policing activities with their local police department. Another concern with measurement is a possible difference between people who actively seek media reports of police misconduct and others who are involuntarily and randomly exposed to media coverage. The current study used the measures based on the assumption that most students are the latter. By contrast, the former would be indicative of a unique characteristic that bears a predetermined negative attitude toward the police.

Our last caution is in reference to internal validity in the modeling. Confidence in the police may be influenced by a myriad factors absent in the current study, including frequency of local events hosted or invested in by the police; the person’s amount of time outside on a daily basis, which may determine the odds of his or her contact with the police; and neighborhood crime rate. For example, it is plausible to hypothesize that a person in a crime-ridden neighborhood has relatively low confidence in the police. Controlling for this would illuminate the net effect of social distance and other variables more clearly. On the other hand, it would require cross-neighborhood comparison as part of multilevel analysis with an appropriate data set. Conservatively, more holistic model for confidence in the police may or may not report the same findings as the current findings.

Conclusion Despite these limitations, our policy recommendation for police agencies is to decrease social distance. Among other things, militarization of police has created a growing social gap between the police and citizens, encouraging citizens to ignore police directives and disrespect the police. Social distance, or attempting to close this gap, may help decrease crime and increase citizen obedience to police. Additionally, when police view a greater social distance between themselves and the public, they perceive the public as more dangerous (Broussard et al., 2011). This can affect their interactions with the public and further undermine this relationship.

One way to reduce social distance between the police and the public they serve is to focus efforts on improving police-public relations through community policing and away from racial profiling. While these may seem like two distinct and unrelated policies, they have an underlying commonality in their relationship to social distance. According to social distance theory, those who hold high levels of power (i.e. the police) are more likely to stereotype (Magee and Smith, 2013), which is seen in racial profiling. Racial profiling highlights differences and ultimately increases distance between the police and the public by facilitating mistrust, rendering a detrimental effect on public attitudes toward police. However, the results of this study suggest reducing that

327

Race and attitudes

toward police

distance – even if a citizen simply feels that he or she “knows” a police officer – can improve confidence in the police even net of the alleged negative contacts and negative media coverage. The community policing movement, bringing the public and the police closer together, is one way to bridge that distance. Similar federal, state and local programs, such as Teen and Police Service Academy and Citizens’ Police Academy, are concrete ways to apply the principles of community policing relevant to promoting closeness. The informative and structured association with police officers would lessen residents’ social distance to the police and ultimately improve residents’ satisfaction with the police (Phillips and Trone, 2002).

Another policy for consideration is additional training on procedural justice, ensuring fair process to all citizens who interact with police. When citizens think they are being treated fairly and are receiving respect from police, they may believe, by extension, they are on a similar social status as police to deserve such treatment, thereby increasing closeness between the public and the police. Further, a simple change like scheduling modifications may allow police to develop social relationships outside of work. Police are known for working long hours on varying shifts, which makes adhering to social plans difficult at best. For this reason, police officers tend to form social groups with one another, as shift work discourages social interactions with others outside of the police force. Perhaps adjusting these shifts to encourage social interactions with citizens would aid in decreasing social distance between the police and the public.

Aside from the issue of social distance, other findings shed light on more conventional wisdom. Local police departments should endeavor to minimize the odds of police misconduct or mistreatment in the line of duty. Unjust law enforcement on the streets is heavily influential on public confidence in the police, which is ultimately vital to the legitimacy of the police. Also, such events affect not only the involved citizen but also the general public when broadcast. In a sense, the two factors originate from a single issue of police misconduct such that the police agencies should acknowledge the gravity and pay enough attention to the education for line officers.

Notes 1. Social distance can be reciprocal, too, as police who perceive greater social distance between

themselves and the public tend to have a higher suspicion of dangerousness (Broussard et al., 2011). Further, social distance has been mentioned in the policing literature, applying social distance to cooperation between police departments (McDavid, 1979), policing styles (Bayley and Mendelsohn, 1969), police attitudes toward people with handicaps like mental illness (Broussard et al., 2011) and mental retardation (Eadens, 2008), and Latino and immigration enforcement (Theodore, 2013). This, though, is beyond the scope of the present study.

2. For Hipp, social distance consisted of at least four categories, including racial heterogeneity.

3. Randomly selected respondents were compensated for their participation. Upon the completion of the survey, the participants were given an opportunity to enter a gift raffle. Each of 20 randomly selected participants won a $20 gift certificate.

4. Negative vicarious contact is a sum of two independent measures – one for family and another for friends – and ranged from 0 to 3.

5. bBMRACE�bFMSOCIAL DISTANCEffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi bFMSOCIAL DISTANCEð Þ2 � Se BMRACEð Þ2

� � þ bBMRACEð Þ2� Se FMSOCIAL DISTANCEð Þ2 � �q , where BM

indicates base model and FM indicates full model.

328

PIJPSM 38,2

References

Anagnostopoulosm, F. and Hantz, A. (2011), “Familiarity with and social distance from people with mental illness: testing the mediating effects of prejudiced attitudes”, Journal of Community & Applied Social Psychology, Vol. 21 No. 5, pp. 451-460.

Barbalet, J. (2009), “A characterization of trust, and its consequences”, Theory and Society, Vol. 38 No. 4, pp. 367-382.

Bayley, D. and Mendelsohn, H. (1969), Minorities and the Police, Free Press, New York, NY.

Bogardus, E.S. (1925), “Measuring social distances”, Journal of Applied Sociology, Vol. 9, pp. 299-308.

Bogardus, E.S. (1933), “A social distance scale”, Sociology and Social Research, Vol. 17 No. 1, pp. 265-271.

Broussard, B., Krishan, S., Hankerson-Dyson, D., Husbands, L., D’Orio, B., Thompson, N.J., Watson, A.C. and Compton, M.T. (2011), “Psychiatric crisis from the perspective of police officers: stigma, perceptions of dangerousness, and social distance”, Corrections & Mental Health: An Update of the National Institute of Corrections, US Department of Justice, National Institute of Corrections, Washington, DC, available at: http://community.nicic. gov/blogs/mentalhealth/archive/2011/08/04/psychiatric-crisis-from-the-perspective-of- police-officers-stigma-perceptions-of-dangerousness-and-social-distance.aspx (accessed August 13, 2014).

Brunson, R.K. (2007), “‘Police don’t like black people’: African-American young men’s accumulated police experiences”, Criminology & Public Policy, Vol. 6 No. 1, pp. 71-102.

Callanan, V.J. and Rosenberger, J.S. (2011), “Media and public perceptions of the police: examining the impact of race and personal experience”, Policing and Society: An International Journal of Research and Policy, Vol. 21 No. 2, pp. 167-189.

Dai, M. and Johnson, R.R. (2009), “Is neighborhood context a confounder? Exploring the effects of citizen race and neighborhood context on satisfaction with the police”, Policing: An International Journal of Police Strategies & Management, Vol. 32 No. 4, pp. 595-612.

Dowler, K. and Sparks, R. (2008), “Victimization, contact with police, and neighborhood conditions: reconsidering African American and Hispanic attitudes toward the police”, Police Practice & Research, Vol. 9 No. 5, pp. 395-415.

Eadens, D.M. (2008), “Police officers’ perceptions regarding persons with mental retardation”, (unpublished doctoral dissertation), University of South Florida, Tampa, FL.

Garcia, V. and Cao, L. (2005), “Race and satisfaction with the police in a small city”, Journal of Criminal Justice, Vol. 33 No. 2, pp. 191-199.

Gill, C., Weisburd, D., Telep, C.W., Vitter, Z. and Bennett, T. (2014), “Community-oriented policing to reduce crime, disorder and fear and increase satisfaction and legitimacy among citizens: a systematic review”, Journal of Experimental Criminology, Vol. 10 No. 4, pp. 399-428.

Hatry, H.P. (1999), Performance Measurement: Getting Results, Urban Institute Press, Washington, DC.

Hinton, M.S. and Newborn, T. (2009), “Introduction”, in Hinton, M.S. and Newborn, T. (Eds), Policing Developing Democracies, Routledge, New York, NY, pp. 1-28.

Hipp, J.R. (2010), “Micro-structure in micro-neighborhoods: a new social distance measure, and its effect on individual and aggregated perceptions of crime and disorder”, Social Networks, Vol. 32 No. 3, pp. 148-159.

Kaplowitz, M.D., Hadlock, T.D. and Levine, R. (2004), “A comparison of web and mail survey response rates”, Public Opinion Quarterly, Vol. 68 No. 1, pp. 94-101.

329

Race and attitudes

toward police

Lai, Y.-L. and Zhao, J.S. (2010), “The impact of race/ethnicity, neighborhood context, and police/ citizen interaction on residents’ attitudes toward police”, Journal of Criminal Justice, Vol. 38 No. 4, pp. 685-692.

Lee, J. and Gibbs, J. (2014), “There’s a stark racial disparity in how Pennsylvanians feel about police”. Penn Live, available at: www.pennlive.com/opinion/2014/12/public_confidence_ in_the_polic.html (accessed January 5, 2015).

McCluskey, J.D., Perez McCluskey, C. and Enriquz, R. (2008), “A comparison of Latino and white citizen satisfaction with police”, Journal of Criminal Justice, Vol. 36 No. 6, pp. 471-477.

McCormick, T.H., Moussa, A., Ruf, J., DiPrete, T.A., Gelman, A., Teitler, J. and Zheng, T. (2013), “A practical guide to measuring social structure using indirectly observed network data”, Journal of Statistical Theory and Practice, Vol. 7 No. 1, pp. 120-132.

McDavid, J.C. (1979), “The effects of social and economic distance on interlocal police cooperation”, International Journal of Public Administration, Vol. 1 No. 1, pp. 65-79.

Magee, J.C. and Smith, P.K. (2013), “The social distance theory of power”, Personality and Social Psychology Review, Vol. 17 No. 2, pp. 158-186.

Manning, P.K. (1992), Organizational Communication, Aldine de Gruyter, New York, NY.

Merton, R.K. (1968), Social Theory and Social Structure, The Free Press, New York, NY.

Miller, J. and Davis, R.C. (2008), “Unpacking public attitudes to the police: contrasting perceptions of misconduct with traditional measures of satisfaction”, International Journal of Police Science & Management, Vol. 10 No. 1, pp. 9-22.

Phillips, E. and Trone, J. (2002), Building Public Confidence in Police Through Civilian Oversight, Vera Institute of Justice, New York, NY.

Reisig, M.D. (2010), “Community and problem-oriented policing”, Crime & Justice, Vol. 39 No. 1, pp. 1-53.

Rhineberger-Dunn, G.M. and Carlson, S.M. (2009), “Confirmatory factor analyses of collective efficacy and police satisfaction”, Journal of Crime and Justice, Vol. 32 No. 1, pp. 125-154.

Schuck, A.M. and Rosenbaum, D.P. (2005), “Global and neighborhood attitudes toward the police: differentiation by race, ethnicity and type of contact”, Journal of Quantitative Criminology, Vol. 21 No. 4, pp. 391-418.

Schuck, A.M., Rosenbaum, D.P. and Hawkins, D.F. (2008), “The influence of race/ethnicity, social class, and neighborhood context on residents’ attitudes toward the police”, Police Quarterly, Vol. 11 No. 4, pp. 496-519.

Skogan, W.G. (2006), “Asymmetry in the impact of encounters with police”, Policing & Society, Vol. 16 No. 2, pp. 99-126.

Smith, B.W. (2005), “Ethno-racial political transition and citizen satisfaction with police”, Policing: An International Journal of Police Strategies and Management, Vol. 28 No. 2, pp. 242-254.

Takagi, D., Kondo, K., Kondo, N., Cable, N., Ikeda, K. and Kawachi, I. (2013), “Social disorganization/social fragmentation and risk of depression among older people in Japan: multilevel investigation of indices of social distance”, Social Science & Medicine, Vol. 83 No. 10, pp. 81-89.

Taylor, R.B. and Lawton, B.A. (2012), “An integrated contextual model of confidence in local police”, Police Quarterly, Vol. 15 No. 4, pp. 414-445.

Theodore, N. (2013), Insecure Communities: Latino Perceptions of Police Involvement in Immigration Enforcement, University of Illinois at Chicago, Department of Urban Planning and Policy, Chicago, IL.

330

PIJPSM 38,2

Tyler, T.R. (1990),Why People Obey the Law: Procedural Justice, Legitimacy, and Compliance, Yale University Press, New Haven, CT.

Tyler, T.R. and Huo, Y.J. (2002), Trust in the Law: Encouraging Public Cooperation with the Police and Courts, Russell-Sage Foundation, New York, NY.

Warren, P.Y. (2011), “Perceptions of police disrespect during vehicle stops: a race-based analysis”, Crime & Delinquency, Vol. 57 No. 3, pp. 356-376.

Weitzer, R. and Tuch, S. (2004), “Race and perceptions of police misconduct”, Social Problems, Vol. 51 No. 3, pp. 305-325.

Weitzer, R. and Tuch, S.A. (2005), “Determinants of public satisfaction with police”, Police Quarterly, Vol. 8 No. 3, pp. 279-297.

Weitzer, R., Tuch, S.A. and Skogan, W.G. (2008), “Police-community relations in a majority-black city”, Journal of Research in Crime and Delinquency, Vol. 45 No. 4, pp. 398-428.

Wu, Y. (2014), “Race/ethnicity and perceptions of the police: a comparison of white, black, Asian and Hispanic Americans”, Policing and Society: An International Journal of Research and Policy, Vol. 24 No. 2, pp. 135-157.

Yim, Y. and Schafer, B.D. (2009), “Police and their perceived image: how community influence officers’ job satisfaction”, Police Practice and Research, Vol. 10 No. 1, pp. 17-29.