There is a rich tradition of research on attitudes toward the police (ATP) in American police scholarship. The principal focus has been on the attitudes of adult citizens, but a smaller body of ATP research has been conducted with juveniles. Researchers in this area commonly use community surveys as their main strategy for data gathering in studying ATP, although other methods such as in-school, self-administered surveys and focus groups are not uncommon. Studies in this area tend to define ATP loosely: ATP can encompass citizen perceptions of police performance, beliefs about police demeanor, assessments of police officer characteristics, and citizen preferences for police duties and practices. In addition, ATP can be global or use very general referents to the police, or it can be very specific and refer to specific incidences and experiences in police-citizen encounters. In the former case, researchers often ask citizens to respond to attitudinal survey items such as ”Most police officers are usually courteous.” In the latter case researchers might ask citizens to respond to specific situations such as ”How satisfied were you with the way the officer treated you when you were stopped?”
Conceptually, attitudes can be thought of as individual cognitive evaluations of objects that have a positive or negative valence attached to them. In the case of ATP, the positive or negative cognitive evaluations are of various police-related attributes such as fairness and courteousness when dealing with minorities. It is generally thought that attitudes toward objects (e.g., the police) and behaviors toward those objects (e.g., cooperation) are positively related. In other words, individuals with positive ATP are more likely to engage in positive behaviors toward and with the police than are individuals with negative ATP. ATP should not be thought of as discreet phenomenon, rather it falls along a continuum ranging from positive to negative.
Understanding ATP and ATP research is important for several reasons. Decker (1981) points out that police organizations need public support, and positive attitudes toward the police are especially important in urban societies where the police are primarily reactive and depend on the public to initiate police activity. The success of the police in carrying out their duties depends heavily on the cooperation of the public, and the absence of cooperation and support makes it difficult if not impossible for the police to perform effectively (Rosenbaum et al. 2005). Other police scholars have noted that residents from different cultural backgrounds have different values, attitudes, and beliefs regarding the appropriateness of police practices, and ATP research is important for detecting these cultural differences. ATP research became even more important with the community-oriented policing movement that started in the 1980s and continues through the present time. Community-oriented policing requires the community and police to work closely together to develop and implement public safety strategies, and ATP is an important determinant of the ability of citizens and police to work closely together. Improved understanding of what shapes citizen ATP can enable police administrators and policy makers to develop strategies for improving ATP that leads to citizen-police cooperation and the coproduction of public safety.
A major impetus for much of the early ATP research was the police response to the civil unrest and disorder associated with the civil rights movement and anti-Vietnam War movement of the 1960s. Graphic media images of police brutality in responding to urban rioting, along with increased concern over the negative state of police-community relations, led to an expansion of research interest in ATP. The National Advisory Commission on Civil Disorders (known as the Kerner Commission) of 1968 carried out the first national, multicity studies of ATP. A cycle of renewed interest in ATP and ATP research tends to be repeated whenever high-profile, nationally visible incidents of police brutality occur such as the Rodney King incident involving the Los Angeles Police Department in the early 1990s and the torture of Abner Louima by several New York City police officers in the late 1990s.
General ATP Research Findings
One of the most persistent findings of ATP research is that citizens generally hold positive attitudes toward the police.
This finding has been replicated in several studies and in national opinion polls that consistently indicate that 70% or more of the general public have positive attitudes toward the police. However, in addition to generalizing to the American public, much of the early ATP research was aimed at identifying the specific determinants of ATP and focused on racial and ethnic differences in ATP. Most of the research carried out in this tradition has focused on differences between African Americans and whites with considerably less emphasis placed on Hispanics. This research consistently demonstrates that ATP among African Americans is substantially less positive than that of whites. Although more limited in number, studies show that ATP of Hispanics is less favorable than that of whites, but more favorable than that of African Americans. In other words, ATP of Hispanics falls in between that of African Americans and whites.
In addition to exploring the role of race and ethnicity in determining ATP, several studies have examined the connection between individual-level characteristics and ATP. In addition to race and ethnicity, three domains of individual-level characteristics or variables have been identified as important predictors of ATP (Webb and Marshall 1995). These include socioeconomic status (SES), age, and gender.
The research evidence on the connection between age and ATP is mixed and less consistent than the research linking race and ethnicity to ATP. However, several studies have found a relationship between age and ATP with older citizens being more positive about the police than younger persons.
Research on the connection between SES and ATP has produced findings that are more equivocal than those on race and ethnicity and ATP or those on gender and ATP. Some studies show a connection while others fail to produce evidence demonstrating a relationship between SES and ATP. The lack of strong evidence in support of such a relationship could very well be a result of the difficulty in disentangling the interrelatedness of SES with race and ethnicity, and varying neighborhood characteristics such as crime rates and differences in the need for police service.
Research evidence on the relationship between gender and ATP tends not to find a significant relationship between the two, and gender is usually not thought of as a predictor of ATP. However, as with age and SES, exceptions do exist; some research shows that females are more positive toward the police than males (Webb and Marshall 1995).
Models of Attitudes toward the Police
The early ATP findings were largely descriptive and characterized by a somewhat oversimplified conceptualization, but over time some relatively distinct “theoretical” models of ATP have evolved. Reisig and Parks (2000) have identified three different models of ATP. They refer to the first model as “the experience with police model.” This model is premised on the idea that ATP is largely a function of one’s personal experience with the police. The second model is the “quality-of-life model,” which sees ATP as being largely determined by the perceptions of one’s neighborhood. The third model is the “neighborhood context model.” In this model, ATP is a function of the objective conditions of neighborhoods such as crime and poverty, and these conditions are thought to be linked to ATP.
A good deal of ATP research has been focused on examining the relationship between actual experience with the police and attitudes toward the police, hence the “experience with police model.” The research in this area indicates that the nature of citizen-police contact plays an important role in shaping ATP. Research focusing on actual citizen-police contact has attempted to determine if the basis for police contact, as well as the manner by which the police handle the contact situation, influences satisfaction with the police. Research findings related to this model show that ATP (i.e., satisfaction with the police) is more positive when citizens initiate the contact than when the police initiate the contact. However, the type of contact is also related to ATP, with negative contacts (e.g., traffics stops and searches) being associated with negative ATP. Negative ATP also results when the police fail to meet citizen expectations, for example, by failing to respond promptly to calls for service.
Research focusing on quality of life and ATP is more recent, and it is largely the result of the influence of the ”broken-windows” hypothesis, which maintains that crime stems from physical and social disorder at the neighborhood level. The body of research in this area, although limited, supports the notion that citizens who perceive quality-of-life problems in their neighborhood tend to hold the police accountable for these problems and as a result they are less satisfied with their police. However, there is also evidence that even in the face of disorder, citizen satisfaction with the police increases when citizens believe that they are living in socially cohesive neighborhoods where neighbors help each other.
Research carried out within the framework of the ”neighborhood context model,” focuses on the neighborhood itself as the primary source of ATP. The idea is that neighborhoods develop relatively distinct cultures that include an ATP component. In other words, ATP depends in part on the neighborhood and its culture and not just the characteristics of the individuals living there or neighborhood conditions. Relatively little research has examined the ”neighborhood context model.” One of the first studies to do so was carried out by Dunham and Alpert (1988) in Dade County, Florida. They conducted research in racially distinct neighborhoods and concluded that the characteristics of the neighborhoods produced relatively unique cultures that shaped ATP. The research of Reisig and Parks (2000) also attempted to assess this model.
Reisig and Parks (2000) were interested in assessing the ability of these three models to predict ATP. Using a complex but sophisticated research strategy, they concluded that the ”quality-of-life model” performed the best in terms of explaining differences in the ATP of the citizens they studied. In other words, citizens who perceived quality-of-life problems in their neighborhoods were more negative in their ATP. However, they also found that the neighborhood itself made a difference, and that the differences in ATP from one neighborhood to the next are a function of a combination of economic conditions and the racial composition of neighborhoods. Citizens in neighborhoods that are at the greatest economic and social disadvantage tend to have more negative ATP. This also helps to explain why most studies find that the ATP of African Americans is more negative than that of whites or Hispanics: African Americans are disproportionately the residents of the most disadvantaged neighborhoods.
Global versus Specific Attitudes toward the Police
Police researchers have expanded insight into ATP by focusing on its structure. Specifically, Brandl and colleagues (1994) attempted to disentangle the interrelationships between global and specific attitudes toward the police. Global attitudes can be thought of as attitudes about the general institution of the police and specific attitudes are those about specific police-related experiences. Prior to their efforts, research had generally looked at and supported the notion that specific ATP influences global ATP. In other words, citizen attitudes based on specific individual experiences with the police were found to influence more general attitudes toward the police. Citizens with negative specific ATP were likely to exhibit less positive global ATP than those with positive specific ATP. Brandl et al. drew on stereotyping theory and suggested just the opposite, that global ATP influenced specific ATP. They suggested that citizen processing of information related to specific encounters with the police would reflect the biases inherent in citizen stereotypes of police. They found that citizens who had contacts with the police tended to have more positive attitudes about their individual contacts with the police than they did about the police in general. This study also indicated that global and specific ATP are causally and asymmetrically reciprocal. This means that one influences the other: Global ATP influences specific ATP and specific ATP influences global ATP.
Frank, Smith, and Novak (2005) explored global and specific ATP in greater detail by examining the reasons citizens gave for their ATP. Consistent with the larger body of ATP research, they found that the majority of citizens had positive ATP and that contact with the police had an important influence on ATP. In addition, they found that the explanations that citizens gave for their global ATP were related to their specific experiences with the police. Perceived police officer or encounter attributes such as friendly versus hostile, polite versus disrespectful, or quick versus slow response time were positively related to global attitudes about the police. The formation of specific ATP is largely a function of citizen processing of information gleaned from their contact with the police. One important implication of this finding is that changes in the nature of citizen-police contacts can produce not only more positive specific ATP, but also more positive global ATP. Police administrators, through training and or other means, can affect citizen ATP if they can get officers to engage in encounters with citizens that are perceived by citizens to be positive. However, Frank et al. (2005) found that citizen ATP sometimes appeared to be inconsistent, with citizens sometimes reporting positive global ATP even after having experienced negative police encounters. They suggested that citizens may have a “reservoir of support” for the police that they dip into and are able to maintain positive ATP in spite of negative experiences.
Juvenile Attitudes toward the Police
In comparison to the extensive research carried out on adult ATP, relatively little is known about juvenile ATP. This paucity of research on juveniles is probably in large part a function of the methods used to study ATP. Most ATP studies rely on citizen surveys, and typically these surveys, regardless of whether they use telephone, mail, or in-person techniques, limit respondents to persons eighteen years of age or older.
There is a small body of evidence indicating that the most important factor affecting juvenile ATP is the nature of juvenile-police interactions. Positive interactions are related to positive ATP and negative interactions are related to negative ATP. However, the nature of police interactions with African American juveniles seems to have less of an impact on their ATP. In contrast to white juveniles, positive interactions in police encounters seem to be less influential in reducing negative ATP. While the nature of police-juvenile encounters is an important source of influence for ATP, it appears that in the case of juveniles, minority status itself has an important independent influence on ATP. This may be a function of socialization processes in disadvantaged neighborhoods and among disadvantaged groups.
It is not surprising that researchers have hypothesized that juveniles who engage in delinquency are more likely to have negative ATP in comparison to law-abiding juveniles. There is some limited evidence to support this hypothesize, but it is indirect and incomplete due to the fact that most studies on the topic have relied almost entirely on samples of juvenile delinquents. As a result they have not been able to fully examine potential ATP differences between juvenile delinquents and nondelinquents. Nevertheless, ATP studies of juvenile delinquents have contributed to understanding the development of juvenile ATP. Lieber, Nalla, and Farnworth (1998) studied a sample of adjudicated or accused juvenile delinquents in order to determine if their ATP was primarily a function of contact with the police or socialization in their community or neighborhood. Although they were not able to compare the ATP of the delinquent sample with a nondelinquent sample, they found that social environmental variables such as race, economic status, and neighborhood were a bigger source of influence on ATP than was contact with the police. Additionally, they found that in comparison to white juvenile delinquents, minority juvenile delinquents were more negative in their ATP. They also found that race and ethnicity were the strongest predictors of ATP, and juvenile delinquents who were most committed to delinquent values had the most negative ATP. This later finding provides support for the contention that, in contrast to law-abiding youth, juvenile delinquents have less favorable ATP, although the causal direction of the relationship remains unclear. Does negative ATP, for example, disrespect for the police, lead to delinquency or does delinquency lead to negative contact with the police that in turn leads to negative ATP? Although additional research needs to be carried out in order to answer this question, the evidence to date seems to suggest that at least for minority juveniles, global ATP is a greater source of influence than is their contact with the police. This means that even positive contact with the police might be insufficient to outweigh the negative ATP resulting from socialization within minority neighborhoods and communities.
Societal Change and Change in Attitudes toward the Police
Very little of the research on ATP conducted during the past forty to fifty years has considered the possibility of change in ATP resulting from larger social changes in American society. Most ATP research assumes stability both in the sources of influence on the development of ATP as well as ATP itself. Few scholars have considered the possibility that demographic changes in America’s communities could lead to changes in global ATP. An important exception is the research carried out by Frank and colleagues (1996) in Detroit. These scholars pointed out that early ATP studies were carried out when there were no major American cities led by minority mayors. Additionally, the composition of local police forces, including police leadership, was overwhelmingly white. Several major American cities have undergone dramatic political and demographic changes since these early studies were carried out. For example, cities such as Atlanta, Detroit, and Washington, D.C., have been racially transformed and African Americans make up the majority of their populations. These cities have also developed minority mayoral and police leadership. In their study of Detroit, Frank et al. found that African Americans had more favorable ATP than whites. They suggest that this could be a function of changes over time in the racial composition of the city since African Americans went from being a minority to the majority, and the numbers of African American police officers and police and municipal administrators have also become substantial. At best, these findings are tentative and suggestive since their study could not track specific changes longitudinally over time.
The Future of Research on Attitudes toward the Police
Increasing knowledge about ATP will continue to be important for the same reasons that gave rise to ATP research in the first place. Policy makers, police practitioners, and academics alike recognize that effective crime prevention and control strategies depend on citizen-police cooperation, and that ATP is an important determinant of the extent and nature of citizen-police cooperation. The history of ATP research demonstrates that understanding ATP is a challenging task, and disentangling the complex web of influences that shape ATP will be necessary if effective strategies for improving ATP and consequently citizen-police cooperation.
