evaluating the police

8/22/2019 Evaluating the Police

1/25

Black Cities/White Cities: Evaluating the PoliceAuthor(s): Susan E. Howell, Huey L. Perry and Matthew VileSource: Political Behavior, Vol. 26, No. 1 (Mar., 2004), pp. 45-68Published by: SpringerStable URL: http://www.jstor.org/stable/4151358 .

Accessed: 09/04/2013 02:43

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

Springeris collaborating with JSTOR to digitize, preserve and extend access to Political Behavior.

http://www.jstor.org

This content downloaded from 189.220.27.251 on Tue, 9 Apr 2013 02:43:21 AMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=springerhttp://www.jstor.org/stable/4151358?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/4151358?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=springer


2/25

Political Behavior, Vol. 26, No. 1, March 2004 (@ 2004)

BLACK ITIES/WHITEITIES:Evaluatingthe PoliceSusan E.Howell,Huey L.Perry,and MatthewVile

It is well known that African Americans and whites hold different views of the police,but nearly all of the previous research has been conducted in majority white settings.This research examines the relationship between race and evaluations of the police inmajority black versus majority white contexts. Social dominance theory and the re-search on racial threat predict that when the racial majority changes, the relationshipbetween race and attitudes toward police will change. We find that, in majority blackcontexts, the traditional relationship between being black and having negative evalua-tions of the police disappears, and it disappears because whites' evaluations of thepolice become more negative. Black evaluations of the police are relatively consistentacross racial contexts. Also, white racial attitudes affect police evaluations in majorityblack contexts, but not in white contexts, while African American racial attitudes areinconsequential in both contexts. Furthermore, if a white citizen is victimized by crimein a black city, it has greater ramifications for evaluations of the police than if thevictimization had occurred in a white city. All of this suggests that whites' views of thepolice may be more racialized than the views of African Americans.Key words: attitudes toward police; race and police; racial context.

It is well known that African Americans and whites hold different views ofthe police. The underpinnings of these differences are both historical andcurrent, having their basis in both the role of the police in maintainingsocialorder and the position of African Americansas a minority group. Historically,the relationshipbetween blacksand the police hasbeen problematic, luctuatingfrombeing mildlystrained o openlyconfrontational.For more thanhalf of thetwentieth century, the police enforced Jim Crow laws in the South, and theydid so in ways that rangedfrom being simply disrespectfulof blacks to being

Susan E. Howell, Professor of Political Science, Universityof New Orleans,New Orleans, LA70148 ([email protected]).Huey L. Perry,Chancellor'sProfessorof Political Science, SouthernUniversity, Baton Rouge, LA. ([email protected]).Matthew Vile, Senior Research Analyst,Goodwill International,Inc. ([email protected]).45

0190-9320/04/0300-0045/0 ? 2004 Plenum PublishingCorporation

http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/25

46 BLACK ITIES/WHITEITIESbrutallyoppressive.Currentissues involvepolice brutality,racialprofiling,andgenerallymore aggressivepolicingof blacks and their neighborhoods.Further-more, the crimeproblemin the United Stateshas become raciallychargeddueto highercrime ratesin minorityneighborhoods, exaggeratedmediaportrayals(Entman, 1992; Peffley, Shields, and Williams, 1996), and the politicalrheto-ric of politicianswho talk about crime in raciallycoded language (Edsall andEdsall, 1991; Jamieson, 1992; Mendelberg,2001). Given all of this, racial dif-ferences in attitudes toward the police are quite predictable.However, most of the researchon racialdifferences in attitudestowardthepolice has been conducted in majoritywhite contexts. There is reason to be-lieve that in majorityblackcontexts, such as majorityblackcities, the attitudesof blackstowardthe police are more positive, and possibly more positive thanwhites' attitudes. This research examines the relationshipbetween race andattitudes toward the police in majorityblack versus majoritywhite cities andaddresses the following questions. (a) Does racial context change the "typical"relationship between race and evaluations of the police? (b) What are theeffects of race compared to perceptions of crime and actualvictimizationonevaluations of police? Do these effects varywith racial context? (c) What roledo racism(amongwhites) and blackconsciousness(amongAfricanAmericans)play in evaluatingthe police? Does the effect of racial attitudeschange fromwhite to black contexts?These questionsaddressa fundamental ssue of legitimacy.The police makeit possible for Americancitizens to enjoya peaceful andorderly ife. Neverthe-less, a second, and more controversial, nterpretationof their role is that thepolice maintainracialoppressionin the interest of the majority white) society.Does the racial composition of one's context affect attitudes towardlaw en-forcement and thus the legitimacyof local government?As the United Statesbecomes more ethnically diverse, white citizens will increasinglylive in cir-cumstances where they are the racial or ethnic minority,and blacksand His-panicswill increasingly ive in circumstanceswhere they arethe majority.Howwill these changes affect citizen evaluationsof the local authorities,especiallythe most visible arm of local government,the police?

AFRICAN MERICANS'ELATIONSITH HE OLICEResearch has demonstratedthat AfricanAmericans and whites have differ-ent experiences with the police. Blacks are more likely than whites to reporthavingexperienced involuntary,uncivil,or adversarial ontactswith the police;to be stopped, questioned, and/or searchedwithout cause or due process;andto experience verbal or physical abuse personally (Browning, Cullen, Cao,Kopache,and Stevenson, 1994;Flanaganand Vaughn,1996;Harris, 1997). Inthe largest survey ever conducted in the United States regardingcontacts



4/25

HOWELL,ERRY,NDVILE 47with police, the Bureauof Justice Statisticsfound that AfricanAmericansandHispanicswere more likely to experience a threat of force, or the actual useof force, by the police as a result of contact with the police (Langan,Green-field, Smith, Durose, and Levin, 2001, p. 7). Also, AfricanAmerican driversstopped by the police were more likely than white drivers to be ticketed,arrested,or have their vehicle searched (Langanet al., 2001, p. 22).The relationshipbetween AfricanAmericans and the police is complicatedby the pattern of greater criminalactivityin black neighborhoods,which cre-ates a dilemma for the police in terms of what should be the proper balancebetween maintainingpublic order and civilityand aggressivelyfightingcrime.Vigil (1989) studied police relationsin a high crime Los Angeles housingproj-ect and described the duality of residents' relations with the police. Eitherthere is no police presence whatsoever and crime is uncontrolled,or the policeengage in violent and aggressive attempts to control crime. Naturally,resi-dents are upset by this dualityof underpolicingvs. overpolicing(Vigil, 1989,pp. 45-49). MarkMoore (1991) has also studied police/communityrelationsin areas with high levels of youth violence:

The policecould maketwo responseshatwould be racistandperceivedas such.One is to ignorethe problembecauseneithervictimsnor offendersarejudgedworthsaving.The other s to use broad ears nthewidercommunitysanoccasionforcracking own.(p. 19)Thus, the central dilemma in AfricanAmericans'relationshipwith the policeis fear of the police because of historicabuse, while at the same time desiringpolice protection from criminal elements that are disproportionatelypresentin their communities.

RACIAL IFFERENCESN VIEWSOF POLICEIn addition to the literature on police/minorityinteractions,there is a well-establishedbody of researchdemonstratingsubstantialdifferences in the waythat African Americans and whites view the police. Race has long been a

strong predictor of the attitudes toward the police, with AfricanAmericansmore likelythanwhites to expressunfavorableattitudestowardvariousaspectsof policing (Campbelland Schuman, 1968; Decker, 1981; Smith, Graham,andAdams, 1991). A report producedby the Bureauof JusticeStatisticsfinds that,while overall satisfactionwith law enforcement is high, black Americansaretwice as likelyas whites to be dissatisfiedwith the police in majormetropolitanareas (Greenfield, Langan,and Smith, 1997). More recently, another Bureauof Justice Statisticsnationalsurveyconfirms the results of earlierstudies andreports that, while the police enjoy a generally favorable image among the



5/25

48 BLACK ITIES/WHITEITIESpublic, racial differences continue to exist. AfricanAmericansare less likelythan whites to expressa favorableopinion of the police, to have confidence inthe police, or to think the police should use force in any situation (U.S. De-partment of Justice, 2000). Finally, in studies of both the Los Angeles areaand the Detroit area, wide interracialgulfs were found in perceptions of theextent of racism and discriminationamong the city's police (Sigelman,Welch,Bledsoe, and Combs, 1997; Tuch and Weitzer, 1997; Welch, Sigelman,Bled-soe, and Combs, 2001, pp. 144-149).Even on matters regarding the efficiency with which the police providetheir services to communities, African Americans hold less favorableviewsthan whites. On the issue of response time, which is universallyregardedtobe an importantstandardmeasure of the efficiency of police service, AfricanAmericans express a higher level of dissatisfactionwith police than whites(Bloch, 1974, pp. 22-25; Fogelson, 1968, pp. 217-247; Furstenbergand Well-ford, 1973, pp. 393-406).This evidence of racialpolarizationin police evaluationsis suggestive of arelationshipbetween racialattitudes and police evaluations.White racialatti-tudes have been shown to predict a wide variety of attitudes and behaviors(Bobo, 2000; Bobo and Hutchings, 1996; Sears,Van Laar,Carrillo,and Kos-terman, 1997; Sears, Henry, and Kosterman,2000). Furthermore,the effectsof racial attitudes are demonstrably strongerthan any personaldisadvantageawhite person may have experienced (Bobo, 2000; Sears et al., 1997). There-fore, we expect whites with conservativeracialattitudesto have more positiveevaluations of the police in white cities, but more negative evaluationsof thepolice in black cities. Black racialattitudesalso have empiricalsignificanceinexisting research, particularlythe concept of black consciousness (Dawson,1994; Gurin,Hatchett, andJackson,1989;Verba and Nie, 1972). We hypothe-size that blackswith greaterlevels of blackconsciousnesswill have more posi-tive evaluationsof the police in blackcities, but more negative evaluationsofthe police in white cities.CRIME-BASEDACTORS

In addition to racial factors, perceptions of crimeand experiences withcrime are criteriaby which citizens mayevaluatethe police. Since the primarypurpose of the police is to prevent and solve crime, it is reasonableto assumethat these crime-based factorswill play an importantrole in the formationofpolice evaluations. Research from Atlantaand WashingtonD.C. indicate that,as the actual rates of crime in respondents'neighborhoods ncrease,their eval-uationsof the police become more negative(Parker,Onyekwuluje,and Murty,1995). Also, fear of crime relates negativelyto evaluationsof the police in fourcommunities in Los Angeles (Hennigan, Maxson,Sloane, and Ranney,2002).



6/25

HOWELL,ERRY,NDVILE 49Nevertheless, the question of the independent effects of the crime-based

factors while controllingfor other variablesremains largelyunanswered.Thepreviousresearch relies on bivariate correlationsto demonstraterelationships,and none of this research has considered the effects of racial context as aconditioningfactor.Thus, our model of police evaluationsmust include crime-based factorsas explanatoryvariables.

THEROLEOFRACIAL ONTEXTThe relationshipbetween race and attitudestowardthe police maywell bedifferent in majorityblackcities, defined in this researchas havinga majorityblack population, an African American mayor, and a majorityblack policeforce. One of the outstandingresults of black political development over thelast 30 years has been the election of numerous AfricanAmericanmayorsinmajorityblackcities. New Orleans, Detroit, Atlanta,Birmingham,and Wash-ington D.C. have had continuous AfricanAmerican mayoral leadership for

more than 20 years. What is lacking is a rigorous comparison of the racialpolarization n these black"empowerment" ities to majoritywhite cities. Per-ry's (1997, 2000, 2002) research on Birminghamand New Orleans suggeststhat leadershipprovided by AfricanAmericanmayorscan play a critically m-portant role in improving the relationshipbetween the city's police and itsblack citizens. But the analysislacks the crucialcomparisonto whites in NewOrleansand Birmingham,and is not based on individual evel attitudinaldata.Other researchcited earliercomes from either majoritynonblacksettings, forexample, Los Angeles, or from nationaldata.What is it about a majorityblack setting that potentiallychanges the rela-tionship between race and police evaluations?Social dominance theory tellsus that societies are organizedaround the notion of dominantand subordinategroups with the dominant groups possessing a disproportionateamount ofwealth, power, status,and so forth. The dominantgroupwill naturallyseek tomaintain its dominancethroughinstrumentaland psychologicalmeans (Sidan-ius, 1993, Sidanius and Pratto, 1999, Sidanius, Singh, Hetts, and Federico,2000). Thus, the "tradingplaces"phenomenon representedby a majorityblackcity with a black mayoris a violation of expected group positions. As such ithas great significance to both the previouslysubordinategroup and the pre-viously dominant group. For the previouslysubordinategroup it means thatthey have some hope of access to the privileges and materialgoods of thepreviouslydominantgroup. For the previouslydominantgroupit means anxi-ety about losing their privilegedstatus and its benefits.Research supports the impact of racial context on both whites and blacks.Among whites, the percentage of blacks in the local population has been



7/25

50 BLACK ITIES/WHITEITIESshown to affect animosity toward blacks and racial policy attitudes (Glaser,1994; Taylor, 1998, 2000), as well as voting and party registration(Giles andBuckner, 1993; Giles and Hertz, 1994). Among blacks, the contextual factorutilized in most research is the presence of a black mayor, that is, actualpolitical empowerment, not just the potential for empowerment. Having ablack mayor is associatedwith a number of positive civic traits, such as trustin local government, efficacy, political information,and politicalparticipation(Bobo and Gilliam, 1990; Gilliam and Kaufman,1998). Empowerment clearlygives a psychic boost to blacks,which is exactlywhat social dominancetheorywould predict.Evaluationsof police are alreadyracialized,so a change in the racialbalanceof power should impact these evaluations.Specifically,we predict that a ma-jorityblackcity with a blackmayorand a blackpolice force will improveblackattitudes towardthe police. At the same time, we predict that white attitudesin majorityblack cities with blackmayorsand blackpolice forces will be morenegative than in majoritywhite cities with white mayors and white policeforces, based on the notion of white anxiety about maintainingtheir socialdominance in the face of the blackmajority.The result is that the racialpolar-ization we normallyobserve in evaluations of police should disappearor re-verse in majorityblack settings.In addition, racial context is expected to influence the impact of crime-based factors on police evaluations. While negative perceptions of crime ide-ally should be related to negative evaluations of police in all contexts andamong all racialgroups,we doubt if this is the case given the racializationofcrime and law enforcement. Whites' minoritystatus, coupled with the stereo-type of black criminality,is hypothesized to magnify the effect of crime onevaluations of the police in the black cities. We expect a parallel patternamong blacks, a magnification in the strength of the relationshipbetweencrime-based factors and evaluations of the police in white cities. This expecta-tion is based on their minoritystatus and their fear of overzealousand indis-criminanttreatment from the white police force.Finally, racial context is expected to prime racial attitudes in both racialgroups. That is, white racial attitudes are primed in a black city, and blackracial attitudes are primed in a white city. Thus, our expectation is that theracialattitudes of both blacks and whites have more impact on police evalua-tions when they are in the minority.DATAANDMETHODOLOGY

The data for this research are from surveysof African-Americanand non-black (hereafter called "white"for simplicity)'registered voters in four largecities: Chicago, New Orleans, Detroit, and Charlotte, NC. While these data



8/25

HOWELL,ERRY,NDVILE 51were not originallycollected to address the topic of police evaluations, thecities selected are quite appropriatefor a study of race and police. The fourcities represent two very different racialcontexts.Two of these cities, Detroitand New Orleans,have majorityblackpopulations,blackmayors,and majorityblack police forces. The other two, Chicago and Charlotte,are majoritynon-black, with white mayors and white police forces. In addition, the minorityrace comprises at least 20% of each city's population, and all four cities havea historyof black/whiteracial tension. We avoided western and southwesterncities with large numbers of Hispanics in order to focus on the black/whiterelationship.2With four cities the question of generalizability nevitablyarises.We makeno claim that these cities represent some larger population.We are, however,moving beyond the typicalsingle-cityapproachused in urban research to con-duct identical studies in four cities representingtwo very different racial con-texts.

There is also the question of other differences between the black andwhitecities that could account for the findings,such as levels of crime, actualpolicebehavior, or the racialcomposition of respondents' neighborhoods. Some ofthese explanationswill be examined empirically and eliminated. However,other differencesbetween the black cities and the white cities, such as level ofpoverty,are impossible to eliminate. Ourcentralargument,given the previousresearch supporting the impact of a black racialcontext and the role of raceas the primarysocial and political cleavage in the United States (Kinder andSanders, 1996), is that the "color"of one's environment has an impact overand above any related specific factors.Within each city a nearly equal number of African Americans and whiteswere interviewed. Thus, instead of a cross-sectionalrepresentation of eachcity, we have maximizedvariation on the key variableof race (King,Keohane,and Verba, 1994)."Cross-sectional randomdigit dialing samples of each city'sblacks and whites were utilized with a screen for registeredvoters. In orderto reach an equal number of minority race respondents, whether white orAfricanAmerican,oversamplesof 40% or more minorityrace concentrationwere used to obtain the needed number. More detail on the sampling pro-cedure is provided in the appendix. The final data contain 253 blacks and251 whites from New Orleans, 282 blacks and 250 whites from Detroit, 273whites and 255 blacks from Charlotte, and 250 whites and 262 blacks fromChicago.The dependent variable,evaluationsof police, is an additivescale composedof respondents' ratingsof five items: the overall qualityof police protection,police response time, police effectiveness in apprehending suspects, policecourtesy,and how well the police avoidexcessiveforce. The scale was recodedto range from 0, the lowest evaluation of the police, to 1, the highest evalua-



9/25

52 BLACK ITIES/WHITEITIEStion. The reliabilityof this scale is quite high as indicated by a Cronbach'salphaof .81 or .82 in all four cities.Respondentswere also asked whether they thought crimewas increasingordecreasing in their city, whether someone in their household had been a vic-tim of crime in the past 3 years, their racialattitudes,politicalparty,ideology,and some standarddemographics.Among white respondents, symbolic racism is measured by a four-itemscale composed of whether respondentsthink blacksare getting less thantheydeserve, whether they think blacks should just try harder, whether blacksshould "worktheir way up" like other minorities, and whether past slaveryand discrimination have limited blacks' opportunities (Cronbach'sa = .55).These four items were selected due to their inclusionin the 1986-1994 Amer-ican National Election Studies.Racial group conflict is measured by four items developed by Bobo andHutchings (1996) that measures whites views of racial competition for re-sources. The items askwhether respondents agree or disagreewith statementsabout whether good jobs for blacks mean less good jobs for other groups,whether more black influence in local politics means less influence for othergroups, whether more neighborhoodsgoing to blacks means fewer neighbor-hoods for other groups,and whether blackstry to get ahead at the expense ofwhites (Cronbach'sa = .80).Our measures of African American racial attitudesare Dawson's (1994) in-dicator of linked fate (p. 77) and four items measuringblack solidarity,all ofwhich have been utilized in one or more of the National Black Election Stud-ies. Linkedfate is measuredby a single indicatorthat assesseswhether AfricanAmericanrespondents perceive a connection between what happens to blacksin the countryas a whole and their own situationas individualAfricanAmeri-cans. The blacksolidarity items ask the respondents about the importanceofvoting for blackcandidates,participatingn black-only organizations,and hav-ing control over the economy and government in mostly black communities(Cronbach'sa for black solidarity= .78).RACEANDEVALUATIONSFTHE OLICE

As expected from the previousliterature,there is racialpolarizationon thepolice evaluation scale, with white respondents generally giving the policemore positive evaluations than African-Americanrespondents (Table 1). Themore interesting findings are the patterns across cities. First, racialpolariza-tion is clearly greater in the white cities than in the black cities, as measuredby the correlationsbetween race and the police evaluationsscale. In Detroitthere are no significant differences between AfricanAmericansand whites,and in New Orleans,the correlationbetween being blackandgivingthe police



10/25

HOWELL,ERRY,NDVILE 53TABLE 1. Police Evaluation Scale by Race and City (scale ranges from 0 to 1)

NewOrleans Detroit Charlotte ChicagoPoliceEvaluations Blacks Whites Blacks Whites Blacks Whites Blacks WhitesLow,0-0.25 11.1% 11.2% 15.7% 17.7% 7.9% 1.4% 14.8% 8.7%0.30-0.50 47.4 27.3 47.5 37.7 25.9 8.8 45.1 25.70.55-0.75 38.0 52.4 33.4 37.7 53.4 63.2 37.0 48.0High,0.80-1.0 3.6 9.2 3.6 7.0 12.7 26.6 3.4 17.7

100% 100% 100% 100% 100% 100% 100% 100%Mean .49 .54 .47 .48 .57 .69 .46 .58Std.Dev. .17 .21 .19 .21 .19 .14 .18 .20N 253 252 282 250 256 273 262 250Correlation etweenRace(Black) nd -.136** -.030 -.325*** -.300***Police Scale

*p


11/25

54 BLACK ITIES/WHITEITIES2500 ---

DetroitC 2000

2c --0 .Chicago. 1500-

S1000-.r

NewOrleans500

01997 1998 1999 2000

FIG. 1. Violentcrimerates.Source:Federal Bureau of Investigation,UniformCrime Reports.

lice as agents of social control, and the historic hostility between blacks andthe police is unlikely to be overcome by simply having a black mayorand ablack police force.5White respondents, however, evaluate the police more positively in thewhite cities than in the black cities. Combiningthis with the findingthat racialcontext makes no difference to blacks'police evaluations,we conclude that isit the white citizens who are producingthe greater racialpolarization n thewhite cities.RACEANDCRIME-BASEDACTORS SPREDICTORSOFEVALUATIONSFTHE OLICE

Next we turn to a comparisonof raceversus crime-based factorsas predic-tors of evaluations of the police. The fundamentalpurpose of the police is tocontrol crime, whether it is throughapprehendingsuspects or crime preven-tion programssuch as communitypolicing. Therefore, citizens' evaluationsofthe police should be based primarilyon how effective they believe their policeare in controllingcrime. The crime-based factorsin this researchare whetherthe respondent thinkscrime is increasing,decreasing,or remainingthe samein his or her city, and whether the respondent or a household member hasbeen a victim of crime in the past 3 years.The crime-based measures and race are entered into multivariatemodels of



12/25

HOWELL,ERRY,NDVILE 55police evaluationsalongwith controlsfor partisanship,deology, neighborhoodracial composition,6and standard demographics (Table 2). The dependentvariablein these models is the scale of police evaluationsfrom Table 1. Thefirst conclusion to be drawn from the models is that race mattersmore in thewhite cities than in the black cities; in fact, in the two black cities race is noteven statistically ignificant.The directionof the coefficients in the white citiesis in the expected direction, indicatingless positive evaluationsof the policeamong African Americans.We estimate the magnitude of this difference inblack versus white cities by pooling the data and adding an interaction be-tween race of respondent and racialcomposition of the city (Table 2B). Theresults indicate that being African Americanin a white city is associatedwithan 8.4% negative movement on the police evaluation scale. However, beingAfricanAmericanin a black city (the interaction)eliminates all of this effect.Furthermore,the racial effect in the two white cities is robust. The bivariateregressioncoefficients associatedwith race in Table 2A retain 70% and 86%of their originalvalue when 10 control variables, including the crime-basedfactors and neighborhoodracialcomposition, are added to the models.

Second, the crime-based factorsof perceptions of increasingcrime and vic-timizationare not clearlymore influential in either black or white cities. Thecoefficients forvictimizationare quite similaracrossall cities andcan be inter-preted to mean that victimizationresults in a 6%-8% movement down thepolice evaluation scale regardless of the race of the victim. The effects ofperceptions of increasingcrime are very similarin New Orleans,Detroit, andChicago, indicating that those who think crime is increasing are about 8%-10% lower on the police evaluation scale. In contrast, the Charlotte police,who are consistentlythe most highly evaluatedof the four police forces stud-ied, are barelyaffected by perceptions of increasingor decreasingcrime rates.Third, and perhaps most significant,is that the racialpolarizationin policeevaluations in the white cities is as strong or stronger than the effects of thecrime-basedfactors in those cities. This is .a strong statement regardingthestrengthof racialdifferences in the white cities. It means that, no matterhowwell the police do, they are still disliked more by AfricanAmericansthan bywhites in the majoritywhite cities. On the other hand, in the black cities, thecrime-based factors are clearlymore influentialthan race. Thus, the standardfinding that African Americans are more negative toward the police thanwhites may only apply to majoritywhite contexts.

RACIAL TTITUDESND EVALUATIONSFPOLICEWhat is the impact of racial attitudeson evaluationsof the police? Answer-

ing this question requires estimating separate models for AfricanAmericansand whites that incorporate the appropriateracialattitudes. Among whites,



13/25

56 BLACK ITIES/WHITEITIESTABLE 2. Race and Approval of the PoliceA. New Orleans Detroit Charlotte ChicagoBivariate b -.052** -.012 -.117*** -.121***

(.017) (.017) (.015) (.017)OLS RegressionModelRace -.007 -.010 -.082*** -.104***(black) (.020) (.022) (.021) (.024)Age .017** .026*** .013* .016**

(.006) (.006) (.006) (.006)Income .012a -.000 -.006 -.005(.006) (.007) (.006) (.006)Education .005 .012 .018* .001(.008) (.008) (.008) (.007)Gender .029 -.004 .003 .018

(female) (.017) (.019) (.016) (.018)PartyIdentification .019 -.013 .006 -.010(.014) (.016) (.012) (.015)Ideology .009 .016 -.006 .028a

(.013) (.014) (.012) (.013)Black Neighbrhd -.057 -.012 -.061a -.053(.037) (.043) (.036) (.037)Crime Increasing -.101*** -.096*** -.036a -.080**(.022) (.024) (.022) (.023)Victimization -.080*** -.070*** -.075*** -.064**(.019) (.020) (.018) (.018)

Adj.R2 .14 .10 .17 .19N 421 434 442 427B. Pooled ModelRace -.084***

(.015)Black City -.128***(.013)Race x BlackCity .082***(.018)Adj. R2 .19N 1724

Note:Cellentriesareunstandardizedregressionoefficients,tandardrrorsaregiven n pa-rentheses.ap< .10,*p< .05,**p< .01,***p< .001;all modelscontaina control or raceof interviewer.



14/25

HOWELL,ERRY,NDVILE 57the expectation is that the more conservative the racial attitudes, the morepositive the attitudestowardthe police. Among AfricanAmericansthe expec-tation is that the higher the black consciousness, the lower the evaluationsofthe police. However, the racialcontext is expected to alter, or even reverse,these expectations.Table 3 presents the models for white respondents in all four cities. Theresults in the white cities are different from the results in the black cities inways that confirm the racializationof whites' attitudes toward police. First,whites who perceive the most group conflict between whites and AfricanAmericansare more negative towardthe police than those who perceive lessconflict, but only in the black cities. This makes sense given the historicalposition of whites as the privilegedrace in America.When whites live in blackcities, they may see the police as less protective of their privileged position.The magnitudeof these effects is quite strong;movement fromperceivingtheleast conflict with blacks to perceivingthe most conflict with blacks is associ-ated with a change of 16%-20% down the scale of police evaluations. Incontrast, in the white cities, the effects of white racial attitudes on policeevaluationsare insignificant.The difference between the impactof white rac-ism in the white cities compared to the impact of white racismin the blackcities can be see in the second column of Table 3B. The interactionbetweengroup conflict and black city approaches significance and means that the ef-fects of white racismin the white cities (-.073) is more than doubled in theblack cities (-.073 + (-.090) = -.163).Second, the effects of perceptions of crime are much strongerfor whites inthe black cities than in the white cities. That is, if a white person perceivescrime as increasing in a black city, it results in a more negative evaluationofthe police in that city. However, there are no consequences for police evalua-tions in white cities if a white person perceives crime as increasing.The inter-action effects in the third column of Table 3B indicate that the impact ofcrime perceptions in white cities is insignificant,but in black cities, percep-tions of increasingcrime results in a 13% movement down the police evalua-tions scale. So, the police in blackcities are held accountableby white citizensfor their beliefs about the effectiveness, or lack thereof, of the police in com-bating crime in a way that the police in white cities are not held accountable.In a similarvein, the effects of victimizationare strongerfor whites in theblackcities than for whites in the white cities. In the case of victimization,theracial context effects are not as strong,but they are in the same directionandhave the same substantiveinterpretation.That is, if a white person is victim-ized in a black city, it probablyresults in a more negative evaluationof thepolice in that city. However, there are no consequences for police evaluationsin white cities if a white person is victimized.



15/25

TABLE 3. Racial Attitudes and Approval of the Police Among WhitesA. New Orleans Detroit Charlotte ChicagoSymbolic Racism .061 -.061 .006 .112"(.071) (.071) (.047) (.061)Group Conflict -.200** -.164** -.051 -.053(.056) (.060) (.048) (.054)Age .009 .035** .001 .025*(.009) (.010) (.008) (.010)Income .020* -.017 -.003 -.006

(.009) (.012) (.008) (.009)Education -.014 .016 .018a .011(.012) (.013) (.010) (.012)Gender .000 -.037 -.004 -.015(female) (.028) (.031) (.019) (.027)

PartyIdentification .010 .007 .004 -.003(.020) (.024) (.013) (.020)Ideology .026 .031 .006 .036.(.022) (.023) (.015) (.019)Black Neighbrhd -.077 -.000 -.091" -.103a(.057) (.061) (.050) (.055)Crime Increasing -.124** -.110** -.044 -.044(.038) (.039) (.028) (.038)Victimization -.079** -.077* -.026 -.039(.029) (.030) (.021) (.028)Adj. R2 .19 .17 .04 .07N 198 189 214 198B. Pooled ModelSymbolic Racism .045 .046 .043(.031) (.032) (.031)Group Conflict -.123*** -.073' -.071l(.027) (.040) (.039)Crime Increasing -.057** -.054** .011(.017) (.017) (.024)Victimization -.060*** -.059*** -.033"

(.014) (.014) (.020)BlackCity -.117*** -.090*** -.147***(.013) (.021) (.030)Group Conflictx Black City -.0901 -.074(.052) (.052)Victimizationx Black City -.049"(.027)Crime x Black City -.130***(.033)Adj. R .23 .23 .24N 799 799 799

Note:Cellentriesareunstandardizedregressionoefficients,tandardrrorsaregiven n pa-rentheses.p < .10,*p< .05,**p< .01,***p< .001;allmodelscontain control orraceof interviewer.



16/25

HOWELL,ERRY,NDVILE 59Finally, simply living in a black city is associatedwith lower evaluationsof

the police among white residents, regardless of their racial attitudes, theirperceptions of crime, or whether they have been victims.When the fourcitiesare pooled, as in Table 3B column one, whites in blackcities are 11.7%loweron the police evaluationscale, net of other factors.Additionally,the explana-torypower is quite low in the white cities (R2= .04 and .07), while the explana-tory power is respectable in the black cities (R2= .19 and .17), indicatingthatwhites' attitudes toward the police in the white cities are less dependent onthe factors in the models. All of these findingsare even more convincingwhenwe consider that the models control for both whites' income and the racialcomposition of their neighborhoods.Turningto the impact of AfricanAmericanracial attitudeson police evalua-tions, it appearsthat both types of black consciousnesshave generally nsignifi-cant effects on blacks' evaluations of the police (Table 4). In contrast, thecrime-based factors of victimization and perceptions of increasingcrime aresignificant almost across the board. Among AfricanAmericans, victimizationresults in a 6%-10% decrease in police evaluations,and perceiving crime asincreasingis associated with a 5%-10% decrease in police evaluations.Poolingthe cities and entering Black City as a variable indicates that blacks haveslightly lower evaluations of the police in black cities, but this finding is afunction of the unusuallyhigh police evaluationsin Charlotteonly. Thus, weare reluctant to draw a conclusion on the basis of this weak coefficient.7A final approachto the effects of racial context is to comparethe impact ofcrime-basedfactorson evaluationsof police amongwhites versus blacksin thesame city, that is, comparingcoefficients in Table 3 to coefficients in Table 4.A traditionalstatisticalcomparisonof coefficients in these two tables cannotbe computed because the racial attitudes are asked of only one race, andlistwise deletion would eliminate all of the cases. However, a visual compari-son across Tables 3 and 4 indicates that, within the black cities, perceptionsof increasing crime affect whites more than blacks, and, within white cities,victimizationaffects blacks more thanwhites. When whites find themselves inthe minority,they are more likely than blacks to react to collectivejudgmentsabout the social environment. Blacks,on the other hand, when in their "nor-mal"minority position are more likelythan whites to blame the (white)policefor their personalvictimization.Overall,whites' evaluationsof the police are more responsiveto racialfac-tors. Whites' opinions of the police are affected by the racialcomposition ofthe city and by their own racial attitudes if they live in a majorityblack city.Also, they are more affected by both perceptionsof crime andby being victim-ized in a black city compared to the same events/perceptionsin a white city.Some observers might wonder why the whites who stay in black cities are soquick to blame the police. Why have they not become comfortablewith black



17/25

60 BLACK ITIES/WHITEITIESTABLE 4. Racial Attitudes and Approval of the Police Among BlacksA. New Orleans Detroit Charlotte ChicagoBlack Solidarity -.120* .006 -.002 -.089(.048) (.053) (.059) (.054)Linked Fate -.044 .008 -.047 -.038(.028) (.030) (.032) (.030)Age .019* .018a .021* .013(.008) (.010) (.010) (.008)Income .017a .005 -.008 -.001

(.009) (.010) (.011) (.009)Education .011 .002 .021a -.008(.010) (.011) (.012) (.010)Gender .033 .013 .014 .033(female) (.023) (.026) (.028) (.024)PartyIdentification .012 -.040 -.003 -.034(.022) (.024) (.024) (.024)Ideology -.006 .001 -.019 .003(.018) (.020) (.020) (.017)Black Neighbrhd .022 -.054 -.063 .010

(.048) (.062) (.057) (.050)Crime Increasing -.056* -.079* -.012 -.102"**(.027) (.033) (.037) (.029)Victimization -.085** -.069* -.106*** -.088***(.024) (.028) (.030) (.025)Adj.R2 .14 .06 .09 .14N 217 242 216 222B. Pooled ModelBlack Solidarity -.047a(.026)Linked Fate -.032*(.015)Crime Increasing -.050**(.015)Victimization -.090**

(.013)BlackCity -.046***(.012)Adj.R2N

Note:Cellentriesareunstandardizedegressionoefficients;tandard rrorsaregiven n pa-rentheses."p


18/25

HOWELL,ERRY,NDVILE 61authorityor simply fled to the suburbs?Exploringthe reasons whites stay inmajorityblack cities is beyond the scope of this research, but their lack ofcomfort with black authority s clearlypredicted by social dominance theory.The theory stipulates that groups are acutely aware of their position in thesocial order, and, for whites, the social order has been reversed in majorityblack cities, producinganxietyabout change and sensitivityto their new mi-nority status. Just because whites in majorityblack contexts have "accepted"black leadership as inevitable, does not mean that they are psychologicallycomfortablewith it.8

How can we reconcile these findingswith the usual focus on negativeevalu-ations of the police by AfricanAmericans,which implies that their opinionsof the police are most affected by racialfactors? We offer three points in theway of explanation.First, given the historyof African Americanmistreatmentby police, it is certainlythe case that black wariness of police is warranted.Research has emphasized this strained black/police relationship, giving theimpression that blacks respond to racial factors more than whites, when, intruth, such inferences cannot be made. Second, African Americansare some-times openly critical of the police when public attentionis broughtto repeatedand/or highly publicized examples of excessive use of force by the policeagainstAfricanAmericans. This focuses attentionon the negative attitudes ofAfricanAmericans. Less attention is given to the subsequent reconciliationifthe mayor and/or the police chief give genuine assurancesthat these inci-dences of excessive use of force by the police are being investigatedand thatappropriateremedies will be put in place. AfricanAmericansmay then returnto their preincidence level of support of the police. This is precisely whathappened in Los Angeles following the Rodney King beating by the LosAngeles police (Tuch and Weitzer, 1997).Finally,the neglect of racialcontextvariancein the urban literaturecontrib-utes to the absence of a proper understandingof African Americans'relation-ship with the police. Most prior research has been conducted in contextswhere AfricanAmericansare the minorityrace.Examiningmajorityblackcon-texts reveals that whites' reaction to racial context is greater than AfricanAmericans' reaction to racial context. Furthermore, whites' racial attitudescome into play in majorityminoritycontexts, whereas AfricanAmericans'ra-cial attitudes are almost irrelevant in both contexts.CONCLUSION

Since the mid-1960s a growing number of citizens, scholars, and policyanalystshave expressed concerns about police-communityrelations and citi-zens' attitudes toward the police and police behavior. These concerns havealwayshad a strong undercurrent of race as the lynchpin of the tensions be-



19/25

62 BLACKITIES/WHITEITIEStween the police and citizens, and this undercurrenthas become more mani-fest over the last two decades. Previousresearchclearlyindicates that, due toa historyof differentialtreatment,AfricanAmericans are more negative thanwhites towardthe police, controllingfor many other explanatory actors.This study highlights a key deficiency in previous research on evaluationsof the police. Racial context, that is, whether the city is majoritywhite ormajorityblack, alters the effects that race and racialattitudeshave on evalua-tions of the police. In the two majorityblack cities studied, the relationshipbetween race and police evaluations disappears, and it disappears becausewhites become more negative in a majorityblack context.Our findings are indicative of the reactions of whites when the "normal"social order is reversed. In a majorityblack context, whites, who are accus-tomed to holding the dominantposition in society, show a tendency to penal-ize the police, a social institution that they typicallyembrace.White racialatti-tudes affect police evaluationsin majorityblack contexts, but not in whitecontexts, while African Americanracial attitudes are inconsequentialin bothcontexts.In addition,if a white citizen is victimizedby crime in a blackcity, orbelieves thatcrime is increasing, t has greaterramificationsorhis or her evalu-ations of the police than if the victimizationor the beliefs about crime hadoccurred in a white city. None of this can be attributedto higher crime rates,more citizen complaintsaboutthe police, or differentneighborhoodracialpat-terns in the black cities. What this does suggest is that whites' views of thepolice maybe more racialized han AfricanAmericans'views of the police.Social dominance theory predicts that when groups change their positionsin the social order, our models must also change. If current demographictrends continue, there will be more majorityAfrican American and majorityHispaniccontexts in the future,andwhites will be more likelyto face minoritystatus. Our findings imply that, when race or ethnicity is involved, models ofattitudes must take into account the sociopsychologicaleffects produced bythe contexts people experience. At a minimum doing so will allow urban re-searchers and scholars to rethink some of the conventional understandingabout the racialdynamicsof urbanlife and at the maximummay allow urbanresearchers to capturethe effects of racial transformations.Our researchsug-gests that one consequence of African American urban empowerment is thatwhites' attitudes toward the police in these empowerment cities will becomemore racialized.If we are on to somethinghere, this "tradingplaces"phenom-enon represents an importantdevelopment in urban politics and should bestudied in areasbeyond police evaluations.

Acknowledgements.e aregratefulo the National cienceFoundationorfundingthisproject,and to FrankP. ScioliJr.andMarianneC. Stewart ortheircounsel npreparinghe proposal.We would also like to thankGaryKing,SaraBenesh,and



20/25

HOWELL,ERRY,NDVILE 63MatthewVile for methodological advice along the way, and the anonymousreviewersfor comments on an earlierdraft. This project could not have been completed withoutthe efforts of the graduateassistantswith the UNO SurveyResearchCenter, MatthewVile, Heidi Unter, ManabuSaeki, and Monica Farris.

APPENDIXTheSurveysThe MayoralApprovalSurveyswere conducted between May and August 2000 infour cities-New Orleans,Detroit, Chicago,and Charlotte-by the Universityof NewOrleans Survey Research Center. The National Science Foundation funded the re-search (Award# 9986165) on which this study is based. All surveys included onlyregisteredvoters. Nearlyequal numbers of blacks andwhites were interviewedin eachcity in order to maximize variance on race, a key independent variable. Thus, thesamples do not represent cross sections of each city's registered voters, but they dorepresent a cross section of whites and blacks in each city. Samples, drawnby Survey

samplingof Fairfield, CT, used a random digit dialingprocedure to which the surveycenter applied a screen for registeredvoters.Response Rate #3 in AAPOR'S Standard Definitions: Final Disposition of CaseCodes and OutcomeRatesand RDD Telephone Surveysand In-PersonHouseholdSur-veys was utilized. The response rates in the four cities were as follows: New Orleans63%, Chicago 55%, Charlotte58%, and Detroit 53%. These response rates are quiterespectable when you consider that half of the sample is urban minorities, a largelylow-income group that is more difficult to include in telephone surveys.Interviewersfor all four surveyswere universitystudents. The interviewswere con-ducted from a central phone bank, with constant supervisionby a graduate student.The surveycenter used a CATI system to manage the sample, and as many as fifteenattempts were made to reach respondents.

QuestionWordingPolice Evaluations:I would like to askyou about governmentand governmentservicesin the city. Is each of the following aspects of governmentvery good, good, fair,poor,or very poor? Police Protection.How would you rate the XX police on the following?The abilityof the XX police torespond quicklyto calls for help and assistance?The abilityof the police to find thecriminals after a crime has been committed? Being courteous, helpful and friendly?Avoidingthe use of unnecessaryor excessive force?SymbolicRacism:(1) Over the past few yearsblacks have gotten less thanthey deserve(agree stronglyto disagree strongly).(2) Irish, Italians,Jews and other minoritiesover-came prejudice and worked their way up. Blacks should do the same without anyspecial favors. (3) In past studies, we have asked people why they think white people



21/25

64 BLACK ITIES/WHITEITIESseem to get more of the good things in life in America-suchas better jobs and moremoney-thanblack people do. These are some of the reasonsgiven by both blacksandwhites. It's really just a matter of some people not tryinghard enough. If blackswouldonly try harder they could be just as well off as whites. 4) Generationsof slaveryanddiscrimination have created conditions that make it difficult for blacks to work theirway out of the lower class.Group Conflict:(1) More good jobs for blacks means fewer good jobs for members ofother groups (agree stronglyto disagree strongly).(2) The more influence blackshavein local politics the less influence members of other groups will have in politics. (3)As more good housing and neighborhoods go to blacks, the fewer good houses andneighborhoodsthere will be for members of other groups. (4) Manyblackshave beentryingto get ahead at the expense of other groups.Black LinkedFate: Do you thinkwhat happens generallyto Blackpeople in this coun-trywill have something to do with what happens in your life? (if Yes)Will it affect youa lot, some, or not very much?

Black Solidarity:How important s it for blacks to vote for blackcandidateswhen theyrun for office: extremely important,very importantor only somewhat important?(notvery important:volunteered). To participatein black-only organizationswhenever pos-sible? To have control over the government in mostly black communities? To havecontrol over the economy in mostly black communities?Crime Increasing:Turning to crime and safety... would you say that the amount ofcrime in (city) has increased, decreased, or remained about the same over the lastseveralyears?Victimization:Have you are anyone in your family been a victim of crime in the cityin the past three years?

NOTES1. Self-identified hites dominate he nonblackregistered otersamples romeachcity:NewOrleans 95%,Detroit= 83%,Charlotte 91%,Chicago 80%.2. Theoriginal urposeof these datawasa studyof mayoral pproval,whichrequiredhat themayorhavebeen in office at leastthreeyearsprior o datacollection.We alsostipulatedhatthe blackmayorsmustnot be the firstblackmayorn order o avoid henoveltyactor.Thesecriteria arrowedhe choiceof cities.3. As a result,we havenearlyanequalnumberof AfricanAmericansndwhites iving n a citywithawhitemajoritys AfricanAmericans ndwhiteslivingn a citywithan African-Americanmajority. urthermore,t is not thelocation-Chicagoor Detroit-that matters onceptually.It is the racialcompositionf thecityanditsleadershiphat s of concern o thisstudy.Thus,the samplesweredesigned o equally epresent ellsof a 2 x 2 tableof respondent'saceandraceof the cityandthe mayor.



22/25

HOWELL,ERRY,NDVILE 654. Informationn citizencomplaintswasobtaineddirectly rom the four cities'police depart-ments.5. Alongthese lines,evidence s mixedas to whetherhavinga blackmayorand blackpolicereducesexcessiveuse of forceagainst lacks Franklin, 989;Perry, 003;PerryandDelmas,1992).6. Neighborhood racialcomposition is an importantcontrol because it is possible that neighbor-hood conditions are a determinant of evaluationsof police. This was not borne out. Further-more, an interaction of Race*BlackNeighborhoodwas insignificant n all four cities, indicatingthat the impact of neighborhoodracialcomposition did not influence one racial group morethanthe other.7. To test forthe impactof unusually igh policeevaluationsn Charlotte etweenwhitesand

blacks, the models for both blacks and whites were reestimated without Charlotte. The coeffi-cient for Black City among whites was still significant,albeit smaller, at -.069, p < .001, butthe coefficient for Black City among blackswas insignificant, indicatingthat high evaluationsin Charlottewere producingthe significanteffects amongblacks.Also, none of the interactionsbetweenblackracialattitudes ndblackcity,or betweencrime-basedactorsand blackcity,weresignificant,hustheyarenotincluded n Table4B.8. Analternativeo the socialdominanceexplanations the possibilityhat n blackcities, owerclasswhitessuffer rompoorpolicingbecausethey ive in raciallymixedneighborhoods,hilehigherSESwhitesreceivebetterpolicing.Aninteractionf blackcity*income as added othe firstmodel n Table3B,and ts insignificanceeadsus to reject hisexplanation.urther-more, amongwhites, an interactionof blackcity*neighborhoodracialmakeupis also insignifi-cant.

REFERENCESBloch, Peter B. (1974). Equality of Distribution of Police Services:A Case Study ofWashington, D.C. Washington,D.C.: Urban Institute.Bobo, Lawrence D. (2000). Race and beliefs about affirmative action: assessing the

effects of interests, group threat, ideology and racism. In David O. Sears, JamesSidanius,and Lawrence D. Bobo (eds.), Racialized Politics:The Debate About Rac-ism in America,pp. 137-164. Chicago: Universityof Chicago Press.Bobo, Lawrence D., and Gilliam, Frank D., Jr. (1990). Race, sociopolitical partici-pation, and black empowerment. American Political Science Revie 84 (June):377-393.Bobo, Lawrence D., and Hutchings, Vincent L. (1996). Perceptions of racial groupcompetition: extending Blumer'stheory of group position to a multiracial ocial con-text. American SociologicalReview 61: 951-972.Browning, Sandra, Cullen, Francis, Cao, Liqun, Kopache, Renee, and Stevenson,

Thomas J. (1994). Race and getting hassled by the police. Police Studies 17: 1-11.Campbell, Angus, and Schuman, Howard (1968). Racial Attitudesin FifteenAmericanCities. Ann Arbor:SurveyResearch Center, Institute for Social Research.Dawson, Michael C. (1994). Behind the Mule: Race and Class in African-AmericanPolitics. Princeton, NJ: Princeton UniversityPress.Decker, Scott (1981). Citizen attitudes toward the police: a review of past findingsand suggestions for future policy.Journal of Police Science and Administration9(1):80-87.Dvorak, Petula (1996). Police probe may reopen; U.S. official has New Orleans onnotice. The TimesPicayune (July3): Al.



23/25

66 BLACK ITIES/WHITEITIESEdsall, Thomas, and Edsall, M. (1991). Chain Reaction. New York:W.W. Norton.Entman, Robert (1992). Representingorder:crime, law, andjustice in the news media(book reviews).AmericanJournal of Sociology 97(May): 1776-1778.Flanagan, Timothy J., and Vaughn, Michael S. (1996). Public opinion about policeabuse of force. In William Geller and Hans Toch (eds.), Police Violence,pp. 113-128. New Haven, CT: Yale UniversityPress.Fogelson, Robert (1968). From resentment to confrontation:the police, the Negroes,and the outbreak of the Nineteen-Sixties riots. PoliticalScience Quarterly83(June):217-247.Franklin, Jimmy Lewis (1989). Back to Birmingham:Richard ArringtonJr. and HisTimes.Tuscaloosa:Universityof AlabamaPress.Furstenberg,FrankF., andWellford,Charles(1973). Callingthe police: the evaluationof police services. Law and Society Review 7: 393-406.Giles, MichaelW., and Buckner, Melanie A. (1993). David Duke and blackthreat:anold hypothesis revisited.Journal of Politics 55: 702-713.Giles, MichaelW., and Hertz, Kaenan(1994). Racial threat and partisan dentification.American Political Science Review 88: 317-326.Gilliam, FranklinD., Jr., and Kaufman,Karen M. (1998). Is there an empowermentlife cycle? UrbanAffairsReview 33: 741-766.Glazer, James M. (1994). Back to the black belt: racial environment and white racialattitudes in the south.Journal of Politics 56: 21-41.Greenfield, LawrenceA., Langan,PatrickA., and Smith, Steven K. (1997). Police Useof Force:Collectionof National Data. NCJ-165040.Washington,D.C.: U.S. Depart-ment of Justice, Office of Justice Programs,Bureau of Justice Statistics.Gurin, Patricia, Hatchett, Shirley, and Jackson,James S. (1989). Hope and Indepen-dence: Blacks' Response to Electoral and Party Politics. New York:Russell SageFoundation.Harris, David A. (1997). "Drivingwhile black"and all other traffic offenses: the Su-preme Court and pretextualtraffic stops. Journal of CriminalLaw and Criminology87: 544-582.Hennigan, Karen M., Maxson, Cheryl L., Sloane, David, and Ranney, Molly (2002).Communityviews on crime and policing:surveymode effects on bias in communitysurveys.Justice Quarterly 19: 565-588.Howell, Susan E., and McLean, William P. (2001). Performancemodels and minoritymayors.Public Opinion Quarterly65: 321-343.Jamieson, Kathleen Hall (1992). Dirty Politics: Deception, Distraction, and Democ-racy. New York:OxfordUniversityPress.Kinder, Donald R., and Sanders, Lynn M. (1996). Divided by Color. Chicago, IL:Universityof Chicago Press.King, Gary,Keohane, Robert O., and Verba, Sidney. (1994). Designing Social Inquiry:

Scientific Inference in Qualitative Research. Princeton, NJ: Princeton UniversityPress.Langan,PatrickA., Greenfield, Lawrence A., Smith, Steven K., Durose, Matthew R.,and Levin, David J. (2001). Contacts between police and the public: findings fromthe 1999 National Survey.NCJ-184957. Washington,D.C.: U.S. Department of Jus-tice, Office of Justice Programs,Bureau of Justice Statistics.Mendelberg, Tali (2001). The Race Card. Princeton, NJ: Princeton UniversityPress.Moore, Mark (1991). Violence, race, and the police. Unpublished paper. PrograminCriminalJustice Policy and Management, John F. Kennedy School of Government,HarvardUniversity.



24/25

HOWELL,ERRY,ND VILE 67Parker,Keith D., Onyekwuluje,Anne B., and Murty, KomanduriS. (1995). African

Americanattitudes toward the local police: a multivariateanalysis.Journal of BlackStudies 25(January):396-409.Peffley, Mark,Shields,Todd, and Williams,Bruce (1996). The intersectionof race andcrime in television news stories: an experimental study. Political Communication,13: 309-327.Perry,Huey L. (1997). The evolution and impact of biracialcoalitionsand black may-ors in Birminghamand New Orleans.In Rufus P. Browning,Dale Rogers Marshall,and David H. Tabb (eds.), Racial Politics in AmericanCities, 2nd edition, pp. 179-200. New York:Longman.Perry, Huey L. (2000). Mayor RichardArrington,Jr. and police-communityrelations

in Birmingham,Alabama.In JamesR. Bowers andWilburC. Rich (eds.), GoverningMiddle-Sized Cities: Studies in Mayoral Leadership, pp. 103-118, Boulder, CO:Lynne Rienner.Perry,Huey L. (2002). The evolution and impact of biracialcoalitionsand black may-ors in Birminghamand New Orleans. In Rufus P. Browning,Dale Rogers Marshall,and David H. Tabb (eds.), Racial Politics in American Cities, 3rd edition, pp. 227-254. New York:Longman.Perry, Huey L. (2003). The evolution and impact of biracialcoalitionsand black may-ors in Birminghamand New Orleans.In Rufus P. Browning,Dale Rogers Marshall,and David H. Tabb (eds.), Racial Politics in American Cities, pp. 227-254. WhitePlains, NY:Longman.Perry, Huey L., and Delmas, Judith(1992) A Reporton PoliceAbuse in New Orleans,Louisiana. Preparedfor the New Orleans,LouisianaHuman RelationsCommission.Sears, David O., Henry, P. J., and Kosterman, Rick (2000). Egalitarianvalues andcontemporaryracialpolitics. In David O. Sears, James Sidanius, and Lawrence D.Bobo (eds.), RacializedPolitics: The Debate About Racismin America,pp. 75-117.Chicago:Universityof Chicago Press.Sears, David O., Van Laar, C., Carrillo, M., and Kosterman, R. (1997). Is it reallyracism? The origins of white americans'opposition to race-targetedpolicies. PublicOpinion Quarterly61: 16-53.Sidanius,Jim (1993). The psychologyof group conflict and the dynamicsof oppression:a social dominance perspective. In Shanto Iyengar and William J. McGuire (eds.),Explorationsin Political Psychology, pp. 183-219. Durham, NC: Duke UniversityPress.

Sidanius,Jim, and Pratto, Felicia (1999). Social Dominance:An IntergroupTheoryofSocial Hierarchyand Oppression.New York:CambridgeUniversityPress.Sidanius,Jim, Singh, Pam, Hetts, John J., and Federico, Chris (2000). It's not affirma-tive action, it's the blacks:the continuingrelevance of race in Americanpolitics. InDavid O. Sears, James Sidanius,and Lawrence D. Bobo (eds.), RacializedPolitics:The Debate About Racismin America,pp. 191-235. Chicago:Universityof ChicagoPress.

Sigelman, Lee, Welch, Susan, Bledsoe, Timothy, and Combs, Michael (1997). Policebrutalityand public perceptions of racialdiscrimination:a tale of two beatings.Polit-ical ResearchQuarterly50: 777-791.Smith, Douglas, Graham, Nannette, and Adams, Bonnie (1991). Minorities and thepolice: attitudinaland behavioralquestions. In Michael Lynchand E. BrittPatterson(eds.), Race and CriminalJustice, pp. 22-35. New York:Harrowand Heston.Taylor, Marylee (1998). How white attitudesvarywith the racialcomposition of localpopulations;numbers count. American SociologicalReview 63: 512-35.



25/25

68 BLACK ITIES/WHITEITIESTaylor, Marylee (2000). The significance of racialcontext. In David O. Sears, James

Sidanius, and Lawrence D. Bobo (eds.), RacializedPolitics: The Debate About Rac-ism in America, pp. 118-136. Chicago:Universityof Chicago Press.Tuch, Steven A., and Weitzer, Ronald (1997). Trends: racial differences in attitudestowardpolice. Public Opinion Quarterly61: 642-663.U.S. Department of Justice (2000). Sourcebookof Criminal Justice Statistics 1999.NCJ-183727. Washington,D.C.: Bureauof Justice Statistics.Verba, Sidney, and Nie, Norman H. (1972). Participation n America:PoliticalDemoc-racy and Social Equality. New York:Harper and Row.Vigil, James (1989). Barrio Gangs: Street Life and Identity in Southern California.Austin:Universityof Texas Press.Welch, Susan, Sigelman, Lee, Bledsoe, Timothy, and Coombs, Michael (2001). Raceand Place: Race Relations in an American City. New York:Cambridge UniversityPress.

evaluating the police

Documents