Factors That Influence Public Opinion of the Police U.S

U.S. Department of Justice Office of Justice Programs JUNE 03 National Institute of Justice

Research for Practice

Factors That Influence Public Opinion of the Police U.S. Department of Justice Office of Justice Programs 810 Seventh Street N.W. Washington, DC 20531

John Ashcroft Attorney General

Deborah J. Daniels Assistant Attorney General

Sarah V. Hart Director, National Institute of Justice

This and other publications and products of the U.S. Department of Justice, Office of Justice Programs, National Institute of Justice can be found on the World Wide Web at the following site:

Office of Justice Programs National Institute of Justice http://www.ojp.usdoj.gov/nij JUNE 03 Factors That Influence Public Opinion of the Police

Acknowledgments This research was supported by National Institute of Justice grant number 96–IJ–CX–0069. The authors would like to acknowledge the aid of Lois Mock, NIJ grant monitor; Jack Greene, who served as consultant to the project; and Gar- rett Zimmon, Dan Koenig, and many other members of the Los Angeles Police Department.

Findings and conclusions of the research reported here are those of the authors and do not reflect the official position or policies of the U.S. Department of Justice. NCJ 197925 RESEARCH FOR PRACTICE / JUNE 03

ABOUT THIS STUDY

Police can improve public opin- have little influence on public ion by increasing their informal opinion of the police. contacts with citizens. Accord- ing to a survey of Los Angeles residents’ opinions of police What did the survey job performance and officers’ reveal? demeanor, police can increase ■ residents’ approval of their job Residents’ perception of performance by participating the level of crime and dis- in community meetings, order in their neighborhood increasing officers’ visibility in was a significant factor neighborhoods, and talking shaping their opinion of with citizens. Such informal the police. contacts had a positive impact ■ Residents with informal on job approval ratings even police contacts had more when other factors associated positive perceptions than with lower approval ratings— residents with formal such as residents’ perceptions contacts. that their neighborhoods are crime ridden, dangerous, and ■ Residents’ opinion of po- disorderly—were present. lice performance did not vary by race or ethnicity in Informal contacts with police disorderly neighborhoods. also lessened the negative impact of residents’ formal ■ Media did not affect resi- contacts with police (such as dents’ approval of police being arrested or questioned job performance or their by police). Residents with perception of officers’ both types of contact report- demeanor. ed higher approval ratings than residents with only formal contact. Race and ethnici- What were the ty were not found to be as survey’s limitations? important as neighborhood characteristics or personal Although care was taken to contacts in determining the survey residents from diverse public’s satisfaction with communities in Los Angeles, police, although race and eth- the four areas covered by the nicity did seem to play a role survey are not, nor were they in residents’ assessment of intended to be, representative officers’ demeanor. The of the entire city. ii media were also found to F ACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE

Cheryl Maxson, Karen Hennigan, and David C. Sloane

Factors That Influence Public Opinion of the Police

A new study has found that “Study Background and neighborhood characteristics Methodology.”) Using a and interactions with police series of questions, they are the factors that most influ- measured two aspects of ence public opinion of the police performance: police. The study, conducted in Los Angeles, found that ■ Job approval. The meas- residents from neighborhoods ure of job approval was perceived to be crime ridden, based on a six-question dangerous, and disorderly scale modified from Sko- were less likely to approve of gan’s work in Chicago, the police.1 In contrast, resi- which asked residents dents who had informal per- how well police in their sonal contact with police neighborhood prevent were more likely to express crime, help victims, and approval. Race and ethnicity, solve problems.2 About the Authors factors cited as influential in ■ Cheryl Maxson, Ph.D., other studies, were not found Officer demeanor. The is an assistant professor to be as important as commu- measure of officer in the Department of nity disorder in determining demeanor was formed Criminology, Law, and from five questions Society at the University the public’s satisfaction with police. Race and ethnicity that asked respondents of California at Irvine. whether police were re- Karen Hennigan, Ph.D., did affect assessment of spectful, trustworthy, fair, is a research assistant police demeanor. The media professor in psychology were found to have little or helpful and how con- at the University of influence on public opinion cerned they acted. Southern California of the police. (USC). Both authors are The study focused on four research associates in the factors that were thought to Social Science Research Measuring opinions in influence public opinion of Institute at USC. David the police: C. Sloane, Ph.D., is an Los Angeles ■ associate professor in Researchers mailed a survey Residents’ perceptions of the School of Policy, to residents in four diverse the levels of disorder and Planning, and Devel- areas of Los Angeles. (See opment at USC.

1 RESEARCH FOR PRACTICE / JUNE 03

crime in their neighbor- Previous research suggests hood and the neighbors’ that an individual’s opinion sense of mutual trust and of the police is filtered responsibility. through these factors, but relatively little is known about ■ Residents’ formal and how they work together to informal contacts with influence perceptions police and prior experience of police performance. as victims of violent and property crime. Perceptions of the ■ Residents’ demographic characteristics. neighborhood As might be expected, ■ The role of the media. residents expressed less

STUDY BACKGROUND AND METHODOLOGY Area Description fueled by the largest gang population in the The study sample was taken from four of the city, with more than 7,000 gang members in Los Angeles Police Department’s (LAPD’s) 18 approximately 40 gangs. Area B, however, geographic areas. The areas were chosen to has the lowest property crime rate among reflect contrasting rates of reported property the four areas sampled. Area B is primarily and violent crimes and demographic profiles, Hispanic (84 percent) and low income, with especially income and race/ethnicity. One many residents living in multigenerational representative from each of the LAPD’s four households. In the 1990 census, 80 percent administrative bureaus was included. However, of residents reported household incomes the four areas are not, nor were they intended below $45,000 and 32 percent below $15,000. to be, representative of the entire city. ■ Area C is on the city’s west side. It has a rel- ■ Area A is in South Central. It has one of the atively low violent crime rate but the highest highest violent crime rates in the city, more property crime rate among the surveyed than three times higher than any of the other areas. Area C has approximately 200,000 res- areas surveyed. In the 1990 census, area A idents and is ethnically diverse (61 percent had approximately 150,000 residents, who white, 22 percent Latino, 10 percent Asian, were 55 percent black and 40 percent Latino. and 6 percent black). In the 1990 census, Latinos have moved into this traditionally 59 percent of residents reported household black area in increasing numbers. Eighty-six incomes below $45,000 and 17 percent be- percent of residents reported household low $15,000. incomes below $45,000 and 43 percent ■ Area D is in the San Fernando Valley. It has a below $15,000. very low violent crime rate and the second ■ Area B is just north and east of downtown. It lowest reported property crime rate (higher has the second highest violent crime rate, only than area B). Area D is primarily white

2 F ACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE

approval of officers and the trust and responsibility.3 To way they do their job when determine the level of neigh- residents perceived problems borhood cohesion and con- with disorder or violent crime trol, residents in this survey in their neighborhood or were asked, among other reported being fearful (see questions, whether people in exhibit 1). their neighborhood got along with each other, shared the The level of social cohesion same values, could be trust- and informal social control ed, were willing to help their present in a neighborhood neighbors, and could be also influenced residents’ counted on to intervene in assessments of the police. neighborhood problems, This characteristic describes such as children skipping residents’ sense of mutual

(73 percent) with growing Latino (17 percent) Mailings to 14 percent of the selected address- and Asian (8 percent) populations. Residents es were returned by the post office, primarily of area D have higher incomes than those in because of inaccurate address information. other areas surveyed: 50 percent reported Excluding these nondeliverable surveys, the household incomes above $45,000 and 26 response rate varied from 44 percent to 65 per- percent above $75,000 in the 1990 census. cent across the four areas, with a combined response rate of 57 percent. In all, 714 complet- Survey Methods ed surveys were available for analysis. Howev- Residential addresses were obtained from city- er, respondents sometimes skipped individual wide utility service lists that were coded to per- questions, so the number of responses varies mit selection of addresses within the four in each analysis.b geographic areas. A random sample of 375 addresses for each area composed the original Notes sample. Survey procedures were adopted from a. Dillman, D.A., Mail and Telephone Survey: The Total Dillman’s Total Design Approach, including up Design Method, New York: Wiley and Sons, 1978; “The to six contacts with potential respondents: a Design and Administration of Mail Surveys,” Annual Review prenotification postcard, four survey mailings of Sociology 17 (1991): 225–249. (the last two with a $1 incentive), and a thank b. For more information about the study methods, see Henni- you/reminder following the first survey mailing.a gan, K., C. Maxson, D. Sloane, and M. Ranney, “Community All mailed materials included Spanish transla- Views on Crime and Policing: Survey Mode Effects on Bias in tions. Other languages were offered, but they Community Surveys,” Justice Quarterly 19 (3), 2002, 565–587. were requested only rarely.

3 RESEARCH FOR PRACTICE / JUNE 03

Exhibit 1. Percentage of residents who approve of police, by neighborhood characteristics

Job approval Officer demeanor

Percentage

100

0 High Low High Low High Low High Low (n = 252) (n = 372) (n = 301) (n = 338) (n = 237) (n = 369) (n = 327) (n = 304) Level of violence Level of fear Level of disorder Level of cohesion and control Neighborhood characteristic

school or the potential clos- Public opinion was associ- ing of a fire station because ated with neighborhood of budget cuts. As indicated cohesion and control for two in exhibit 1, residents who reasons. First, residents who responded positively to these reported living in neighbor- and related questions were hoods where neighbors got much likelier to approve of along, shared similar values, police performance and and could rely on each other demeanor. were likelier to have informal

4 F ACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE

contacts with police officers community meetings, police- than those who reported liv- sponsored youth activities, ing in neighborhoods where and community safety fairs. these traits were less com- Although almost half the mon. Second, these respon- respondents reported infor- dents are likelier to believe mal contacts with police, less that the community shares than one in five residents said responsibility with the police they knew or recognized for a safe and orderly neigh- police officers who worked borhood. They are therefore in their community. less likely to judge police officers harshly when crime and Exhibit 2 shows how resi- social disorder occur. dents rated police based on the types of contact they had (formal only, informal only, Contacts with the both, neither). Those with police only informal contacts hold the highest opinions of police The survey captured two performance and officer kinds of contact with officers: demeanor. Those with only formal and informal. Forty- formal contacts hold the least eight percent of the respon- positive attitudes toward local dents reported some type of police on these two meas- formal contact with local ures. Individuals with no con- police. These formal contacts tacts with police have high included residents’ calls to opinions of job performance police stations requesting and officer demeanor. service and police question- ing of residents regarding Informal contact with police possible crimes. These for- had a significant effect on mal contacts also included job approval ratings, even arrests of 1 percent of the when considering residents’ respondents. perceptions of the level of disorder in their neighbor- Forty-seven percent of the hood. For residents who respondents reported infor- reported low levels of disor- mal contacts with police. der, job approval ratings These informal contacts ranged from 71 percent for included conversations with residents with only formal police officers on patrol and contact with police to almost interactions with police at 90 percent for those with

5 RESEARCH FOR PRACTICE / JUNE 03

Exhibit 2. Percentage of residents who approve of police, by type of contact they had with local officers

Officer demeanor Job approval

Percentage

100

0 Formal only Informal only Both Neither (n = 140) (n = 134) (n = 166) (n = 200) Type of contact

either no contact or only Prior crime informal contact. Residents victimization who reported high levels of neighborhood disorder had a Prior victimization, especially wide range of opinion about violent crime victimization, job performance—35 percent significantly lowered resi- who had only formal contact, dents’ approval of the police. 49 percent who had both for- As exhibit 3 shows, 57 per- mal and informal contact, and cent of respondents who 85 percent who had only were violent crime victims informal contact approved of and 70 percent who were police performance. property (but not violent)

6 F ACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE

Exhibit 3. Percentage of residents who approve of police, by type of victimization

Job approval Officer demeanor Percentage

100

0 Violent crime Property crime only Not a victim (n = 103) (n = 300) (n = 220) Type of victimization

crime victims approved of for crime victims are consis- police performance. In com- tent with past research.4 parison, 85 percent of residents who were not crime victims approved of the job Demographic their local police were doing. characteristics This pattern held for respondents’ opinions of officer Much past literature focuses demeanor, although the dif- on the association of race ference in perceptions be- and ethnicity with public tween crime victims and opinion of the police. Studies nonvictims was not as great. have found that ethnic minor- The lower approval ratings ities, particularly blacks,

7 RESEARCH FOR PRACTICE / JUNE 03

Exhibit 4. Percentage of residents who approve of police, by race/ethnicity

Job approval Officer demeanor

Percentage

100

0 Asian and other Black Latino White (n = 79) (n = 84) (n = 198) (n = 245) Race/ethnicity

report less favorable attitudes However, once respondents toward the police than whites, are categorized further by possibly because of their per- the level of perceived disorder ception that minorities are in their neighborhood, the mistreated more often by racial/ethnic-based differences police.5 At first glance, the in approval of job performance results of the current study disappear (see exhibit 5). Al- seem to confirm the findings though whites in low-disorder of these studies. As exhibit 4 neighborhoods appear to have shows, whites express high- a higher opinion of police per- er opinions of police perform- formance, disorder is clearly ance and demeanor than any the main influence. Converse- other race/ethnic group. ly, residents’ opinions about

8 F ACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE

Exhibit 5. Percentage of residents who approve of police, by race/ethnicity and neighborhood disorder

Job approval Officer demeanor

Percentage

100

0 High Low High Low High Low High Low Asian and other Black Latino White (n = 79) (n = 84) (n = 198) (n = 245) Race/ethnicity and level of neighborhood disorder

officer demeanor were more Role of the media affected by their race and ethnicity, even though disorder Police are particularly con- remained important. Blacks cerned about the media’s were less likely to think that influence on attitudes toward local police were trustworthy, them. Several Los Angeles fair, helpful, concerned, and police supervisors who were respectful of others in both interviewed before the sur- orderly and disorderly vey was conducted stated neighborhoods. that they believed that a few highly publicized incidents

9 RESEARCH FOR PRACTICE / JUNE 03

might have a widespread THE NEED FOR MORE SURVEYS negative influence on residents’ view of the police. Surveying and understanding the public’s opinion of police performance is important for several reasons: In this survey, 65 percent of the respondents indicated ■ Public acceptance of police authority is essential to that personal experience maintain public order. (including respondents’ expe- ■ Public confidence in police can lead to cooperation rience with other police agen- that is needed for effective policing. cies and their secondhand ■ Public support is crucial to maintaining and expanding knowledge of the experience police funding. of others) most shaped their opinions of the LAPD. ■ Sensitive measurement of public opinion is an important Thirty-five percent were most way to monitor local police activities. influenced by mass media ■ Exchange of information fosters police-community (including newspapers, radio, partnerships that can be tailored to specific community and television). The study concerns. found that residents who rely Perceptions of police activities, crime, and fear and the most on the media did not identification of neighborhood problems and strengths report less favorable opinions represent important aspects of the community context regarding the overall job per- of policing. formance and demeanor Regular surveys are essential for understanding these of the police than those issues further. Such surveys should be conducted by inde- most influenced by personal pendent organizations to avoid biased responses. Con- experience. ducting the surveys over a number of years would allow for greater comprehension of the public’s opinion. An annual survey with standard samples and procedures Implications for police crafted to examine trends would be most meaningful. practice Then, police officials and community representatives would notice anomalies in the pattern of opinion as influ- According to this study, disor- enced by respondents’ experiences and the neighborhood derly neighborhoods and sense of community. Community stakeholders could track neighborhoods with poor indicators of issues that can be targets of intervention, social cohesion and control such as neighborhood conditions or people’s lack of infor- present a challenge for officer- mal contact with officers. community relations. Local Surveying community residents can provide valuable community surveys can help information about the public’s assessment of police activi- police to identify and address ties. As this study demonstrates, such a survey can pro- residents’ situations and con- vide surprisingly rich information about the relationship cerns (see “The Need for between communities and the police. This information can More Surveys”). Communi- direct police and community change and help monitor the ties also may benefit from effectiveness of interventions. community policing strate- gies that increase informal 10 F ACTORS THAT INFLUENCE PUBLIC OPINION OF THE POLICE

contact between local offi- perceived disorder in one’s cers and residents. neighborhood, and higher neighborhood cohesion and The findings confirm and control. Race and ethnicity expand on earlier studies that were not shown to significant- suggested informal contact ly determine public opinion raises public opinion of the of police performance once police. In this survey, speak- other factors (such as per- ing to officers on patrol or at ceived neighborhood disorder) community events was asso- were considered. Residents’ ciated with a positive opinion trust in the police, however, of police performance, was influenced by race and whether or not the resident ethnicity. lived in a disorderly neighborhood. It could be that those Police and others often per- predisposed to be more ceive the media as having favorable are likelier to initi- a significant effect on the ate informal contact with offi- public’s opinion of police cers. In any event, it seems performance. According to likely that promoting informal this study’s findings, the social contact may favorably media did not appear to be a influence public opinion. source of negative opinion of Police departments may the LAPD. Instead of relying want to evaluate the ways on the media for their opin- that they encourage or dis- ions, respondents appeared courage informal contact to react primarily to their own with residents. Police- experiences and expecta- community partnerships are tions in forming opinions of more effective when they their local police. incorporate greater informal contact with residents. Notes Although past literature relies 1. For more information about this heavily on demographic char- report, see Hennigan, K., C. Maxson, acteristics to explain public D. Sloane, and M. Ranney, “Commu- opinion, this study found that nity Views on Crime and Policing: the most important factors Survey Mode Effects on Bias in influencing favorable opinion Community Surveys,” Justice Quar- terly 19 (3), 2002, 565–587. of the police were greater informal contact with police, 2. Skogan, W., Disorder and Decline: less victimization, less fear Crime and the Spiral of Decay in of crime, lower perceived American Neighborhoods, New York: level of violent crime, lower Free Press, 1990; Skogan, W., and S. Harnett, Community Policing

11 RESEARCH FOR PRACTICE / JUNE 03

Chicago Style, New York: Free Press, Police Performance,” Public Admini- 1997. stration Review 43 (1983): 50–58; Carter, D.L., “Hispanic Perception of 3. Social scientists often refer to the Police Performance: An Empirical combination of neighborhood social Assessment,” Journal of Criminal cohesion and informal social control Justice 13 (1985): 487–500; Jesilow, as “collective efficacy.” Sampson, P., J. Meyer, and N. Namazzi, “Public R.J., and S.W. Raudenbush, “Sys- Attitudes Toward the Police,” Ameri- tematic Social Observation of Public can Journal of Police 14 (2) (1995): Spaces: A New Look at Disorder in 67–88. Urban Neighborhoods,” American Journal of Sociology 105 (3) (Novem- 5. Flanagan, T.J., and M.S. Vaughn, ber 1999): 603–651; Sampson, R.J., “Public Opinion and Police Abuse S.W. Raudenbush, and F. Earls, and Force,” in W.A. Gelban and H. “Neighborhoods and Violent Crime: Toch, eds., Police Violence, New A Multilevel Study of Collective Effi- Haven: Yale University Press, 1996; cacy,” Science 277 (1077): 918–924; Huang, W.S., and M.S. Vaughn, Sampson, R.J., S.W. Raudenbush, “Support and Confidence: Public and F. Earls, Neighborhood Collective Attitudes Toward the Police,” in Efficacy: Does It Help Reduce Vio- W.A. Gleban and H. Toch, eds., lence? Research Preview, Washing- Police Violence, New Haven: Yale ton, DC: U.S. Department of Jus- University Press, 1996; Sacco, V., tice, National Institute of Justice, “Evaluating Satisfaction,” in J.P. April 1998, NCJ 184377. Brodeur, ed., How to Recognize Good Policing: Problems and Issues, 4. Kansas City Police Department, Newbury Park, CA: Sage Publica- 1977, studies cited in S.G. Brandl and tions, 1998; Waddington, P.A.J., and F. Horvath, “Crime-Victim Evaluation Q. Braddock, “Guardians or Bullies? of Police Investigative Performance,” Perceptions of the Police Among Journal of Criminal Justice 19 (1991): Adolescent Black, White, and Asian 293–305; Brown, K., and P.B. Coulter, Boys,” Policing and Society 2 (1991): “Subjective and Objective Measure of 31–45.

12 The National Institute of Justice is the research, development, and evaluation agency of the U.S. Department of Justice. NIJ provides objective, independent, nonpartisan, evidence-based knowledge and tools to enhance the administration of justice and public safety.

NIJ is a component of the Office of Justice Programs, which also includes the Bureau of Justice Assistance, the Bureau of Justice Statistics, the Office of Juvenile Justice and Delinquency Prevention, and the Office for Victims of Crime.