Socialization and Self-Selection

Abstract

To what extent are bureaucrats’ views shaped by the organizational contexts that they enter? What role do preorganizational influences have? This study provides some answers to these questions by testing hypotheses drawn from organization and personality theories. Using survey and interview data from a longitudinal study, this article examines how police officers develop their attitudes about using force during their first 2 years on the job. This study shows that development is a mixture of socialization and self-selection. However, it moves beyond the “it’s both” explanation by showing which influences mattered most and at which times.

Keywords

police socialization use of force street-level bureaucracy

Introduction

What shapes bureaucrats’ attitudes, identities, and motivations? One answer, drawn from continuity theory, personality studies, and a variety of public administration literatures (public service motivation, representative bureaucracy, psychodynamic bureaucracy, and street-level bureaucracy) is that bureaucrats are strongly affected by preorganizational experiences (Atchley, 1999; Diamond, 1985; Perry, 1997; Selden, Brudney, & Kellough, 1998). As a result, a significant portion of workers’ views are set by the time that they are recruited or seek out agency work; as such, much of organizational thought and behavior would appear to be tied to self-selection or extraorganizational influences. Another answer is that organizations shape the people who enter them (Kaufman, 1960; Merton, 1940; Rubenstein, 1973; Sandfort, 2000; Van Maanen, 1975; Wilkins & Williams, 2009). This approach to understanding personnel tries to identify the organizational processes, mechanisms, and characteristics that enable organizations to bend individuals to their needs. In this way, organizations are able to replicate themselves, maintain relative continuity, and ensure the consistency of decisions made by actors in different venues.

Of course, explaining bureaucratic development is likely not a case of “either or.” Rather, personnel are affected by outside factors and organizational situations (Argyris, 1993; Diamond & Allcorn, 1986; Sandfort, 2000). Still, saying “it’s both” is unsatisfying from a theoretical and practical standpoint. Which influences matter most? At which points do particular influences matter? How do they matter? Unfortunately, few works have conducted rigorous, longitudinal studies of entering public workers so it is difficult to answer these questions. This manuscript provides some answers by studying how one group of entrants developed key views during their first 2 years inside their organization. The participants of this study are police officers, and I focus on their attitudes about using force. In this article I ask, “At what times, and to what extent, are police officers’ attitudes about force associated with external influences? At what times, and to what extent, are officers’ attitudes associated with intraorganization influences?”

This study suggests that development is a mixture of socialization and self-selection. However, it moves beyond the “it’s both” explanation by showing which influences mattered most and at which times. As socialization theories would expect, most officers shifted their attitudes somewhat during development. However, in most cases, this shift was not radical. Also, there were few strong organization predictors of officers’ views and none after their first 3 months in the department. In contrast, throughout the study, extraorganizational variables, such as family income, entering views, and race were frequently strong predictors of officers’ views.

The Attitude-Behavior (A-B) Connection

Before discussing the participants and content of this study, it is important to define “attitude”¹ and discuss the connection between attitudes and behavior. An attitude is the summary evaluation that an individual makes about a particular thing (Ajzen, 2001). For many years, scholars have investigated the A-B link. Schuman and Johnson (1976), in a review article, conclude that “most A-B studies yield positive results. The correlations that do occur are large enough to indicate that important causal forces are involved, whatever one’s model of the underlying causal process may be.” Although they note that there are “important areas where attitudinal measures are largely unrelated to behavior,” they suggest these are “interesting cases that reveal something special about social life” (p. 670). Maio, Olson, Bernard, and Luke (2006), in a more recent review article, buttress this finding and argue that there is “abundant social psychological research” (p. 284) supporting the contention that attitudes, along with ideologies and values, influence behavior.

In the public administration and decision-making literatures, there is also support for the existence of an A-B connection: Workers’ loyalties, identities, and motivations appear associated with how they perform on the job (Kaufman, 1960; Maynard-Moody & Musheno, 2003; Van Maanen, 1975). For example, federal workers motivated by protecting the public interest were more likely to be whistleblowers when they witnessed government waste or fraud (Brewer & Selden, 1998). Similarly, social service directors’ attitudes about welfare claiming influenced the policy choices that they made in the wake of welfare reform (Reingold & Liu, 2009). Some police studies explicitly suggest that officers’ views of their jobs and the people they police are associated with how they act (Maynard-Moody & Musheno, 2003; Muir, 1977; Rubenstein, 1973; Skolnick, 1966; Van Maanen, 1974; Weisburd, Greenspan, Hamilton, & Williams, 2000; Worden, 1995). For example, police may rely on their perceptions of the deservingness of an individual: A “hard worker” who is contributing to the neighborhood may be dealt with more favorably than a deviant (Rubenstein, 1973). Similarly, police officers who view citizens negatively, as well as those who view the use of force positively, may be more likely to use force (Worden, 1996).

Although there are signs of an A-B connection across a range of literatures, there is no consensus about its strength. For instance, Ajzen and Fishbein (1977) argue that there is a connection between attitudes and behavior but that it is rarely strong. Also, some police studies suggest that attitudes are unrelated or weakly related to behavior (Smith & Visher, 1981; Terrill & Mastrofski, 2002; Worden, 1989; Worden & Shepard, 1996).

As there is some disagreement in the literature, I do not take a position about the strength of the A-B connection. Rather, like Maio et al. (2006), I suggest that there is evidence that attitudes are one of a variety of psychological influences that shape how people act. This approach makes practical sense because it matches the perspective and actions of police departments: Most departments carefully screen officers psychologically and then expend significant resources to ensure that police cadets learn to “think like a police” (Davis, Rostow, Pinkston, & Cowick, 1999; Grant & Grant, 1996; Rubenstein, 1973; Van Maanen, 1974). Thus, I assume that police officers’ views about force are associated with how they use it. However, this assumption is not tested in this article, and I do not make any behavioral claims; rather, the A-B connection provides a compelling rationale for this study.

Entrants and Organization Socialization

As a result of a basic understanding that psychology and behavior are linked, there have been many studies about bureaucrats’ attitudes, identities, and motivations (Aberbach & Rockman, 2000; Golden, 2000; Katz & Kahn, 1966; Lipsky, 1980; Merton, 1940; Perry, 1997; Prottas, 1979; Sandfort, 2000). However, with very few exceptions, these studies have used cross-sectional research designs of experienced workers (i.e., not entrants). Although useful for providing a snapshot of a group of workers at a particular moment in time, these studies provide little evidence about why workers hold certain views. Did they enter with their views? Did they alter their views after entering their organizations? What is the role of organization socialization—the process by which organizational entrants adjust to their surroundings and acquire the knowledge and ability to operate as effective organization members?

Although there has been relatively little attention paid to these questions in the various public affairs literatures (public administration, political science, and public policy), there has been considerable research on the general topic of organization socialization in the business literature (Jones, 1986; Morrison, 2002; Saks & Ashforth, 1997; Saks, Uggerslev, & Fassina, 2007; Van Maanen & Schein, 1979; Wanous, 1992). This literature shows that during socialization, individuals acquire the attitudes, patterns of behavior, and knowledge to participate as organization members. It also highlights the efficacy of particular tactics—like the manner of training—and that intra- and extraorganizational factors shape socialization outcomes.

The organization socialization literature provides useful guidance for this inquiry. However, in addition to borrowing from this literature, this project will contribute to it in two ways. First, the outcomes that the organization socialization literature typically studies are role ambiguity, role conflict, organization commitment, job satisfaction, and intention to quit (Saks et al., 2007). Although these outcomes are important to study, decision-making theories indicate that choices may be driven by an array of psychological factors, including identities, attitudes, and motivations (March, 1994). As attitudes are the focus of this project, I contribute by broadening the types of outcomes that are studied during socialization. Second, the literature tends to focus on workers entering private firms (Wanous, 1992). As there is evidence that public workers differ from private workers (Crewson, 1997; Goodsell, 2004; Houston, 2000), this project contributes by focusing on the socialization of public workers.

Before ending this section, it is important to discuss my temporal conception of organization socialization. Research in the organization socialization literature varies on the length and stages of socialization (Van Maanen, 1975; Wanous, 1992). However, many studies suggest that the process takes 6 months to a year (Morrison, 2002; Saks & Ashforth, 1997; Wanous, 1992). This study examines workers during their first 2 years on the job. Although I do not expect that workers are fully formed after 2 years, this period provides enough time after the completion of training for entrants to move from feeling like an outsider to an insider.

Police Officers and Using Force

To study bureaucratic socialization, this project focuses on a group of entering police officers and their views of using force. The officers discussed here were municipally employed patrol officers not yet eligible for promotion or special assignment; during the study, they patrolled neighborhoods by foot, bike, or car and responded to radio calls. At an abstract level, police officers have the following goals: maintaining social order, enforcing the law, and preventing crime; as social actors, they are unique because they have a monopoly on the legal use of force (Sherman, 1980). Although definitions vary somewhat, the National Institute of Justice defines force as the “amount of effort required by police to compel compliance by an unwilling suspect.” In practice, scholars have defined force as a continuum of actions from verbal commands, to basic physical restraint, to less-lethal force (pepper spray, stun guns, etc.), to lethal force (Garner, Schade, Hepburn, & Buchanan, 1995; Mesloh, Henych, & Wolf, 2008; National Institute of Justice, 2010; Terrill & Mastrofski, 2002). As verbal commands are harder to monitor, and may not be seen as force by officers or citizens, I define force as follows: the continuum of legal “physical action[s] taken to control the movement or freedom of another individual” (Alpert & Dunham, 2004, p. 20).

Why focus on police and their views of force to explore bureaucratic socialization? I argue that this is a crucial case because police play a powerful symbolic role in society; police socialization is depicted as strong; force is a defining element of police power; police have significant discretion and, therefore, power; and police behavior can galvanize social unrest and teach citizens about their place in society (Alpert & Dunham, 2004; Bazley, Lersch, & Mieczkowski, 2006; Bittner, 1970; Lipsky, 1980; Theobald & Haider-Markel, 2008; Wilkins & Williams, 2009). Perhaps due to these factors, many bureaucracy studies have relied on case studies of police (Bardach & Kagan, 1982; Lipsky, 1980; Maynard-Moody & Musheno, 2003; Wilson, 1989).

One of their defining characteristics, and the most politically sensitive aspect of their work, is their ability to use force (Bittner, 1970; McEwan, 1996). As such, many studies have examined what influences the use of force. To begin, there is consensus that the use of force is rare (Adams et al., 1999; Alpert & Dunham, 2004; Bayley & Garofalo, 1989; Langan, Greenfield, Smith, Durose, & Levin, 2001). Also, most instances of force are weaponless, low-level interactions like pushing or grabbing (Adams et al., 1999; Bazley et al., 2006; Gallo, Collyer, & Gallagher, 2008; Garner & Maxwell, 2002).

What influences officers’ use of force? In answering this question, scholars have divided influences into three types: personal or psychological, situational or sociological, and organizational (Friedrich, 1980; Worden, 1996). Personal explanations include, among others, psychological outlook, attitudes, education, perceptions of role orientation, and perceptions of citizens. Situational explanations include, among others, the neighborhood, time of day, the demeanor of the suspect, and the size of the crowd of onlookers. Finally, organizational explanations seek to explain differences across jurisdictions by the varying approaches to force taken by different departments.

In general, scholars agree that situational factors are important. In particular, there is evidence that officers use force when the suspect resists or appears intoxicated or mentally ill (Adams et al., 1999; Terrill & Mastrofski, 2002; Worden, 1996), is young or poor (Terrill & Mastrofski, 2002), is disrespectful or antagonistic (Friedrich, 1980; Gallo et al., 2008), and is a racial or ethnic minority (Alpert & Dunham, 2004; Fyfe, 1982; Garner & Maxwell, 2002; Meyer, 1980; Worden, 1996).² There is also evidence that officers use force when there are more police or citizens present (Friedrich, 1980; Worden, 1996); the bystanders have an antagonistic demeanor toward the police (Garner & Maxwell, 2002); there are more police at the scene (Garner & Maxwell, 2002); the encounter with a suspect occurs in a disadvantaged neighborhood (Terrill & Reisig, 2003); the police initiate an interaction, are responding to a priority call, use their lights and sirens, or call for backup (Garner & Maxwell, 2002).

Also, comparisons of force in different jurisdictions suggest that organizational factors may matter (Alpert & Dunham, 2004; Fyfe, 1982; Garner & Maxwell, 2002). Although these jurisdictional differences may reflect organizational differences, these studies give little indication about which factors are associated with the use of force. In fact, evidence about which intraorganizational influences affect the use of force is scant; however, one study showed that officers working in more highly bureaucratized police departments (i.e., more officers, levels, and specialization) were more likely to use reasonable force (Worden, 1996).

In contrast to situational and organizational explanations, police scholars commonly assert that there is very little evidence supporting personal, psychological explanations about police behavior generally and the use of force specifically (Adams et al., 1999; Alpert & Dunham, 2004; Worden, 1989). However, there are at least three reasons that question this claim. First, many theoretical accounts of police behavior emphasize the connection between the psychological and the behavioral (Brown, 1981; Lester, 1996; Muir, 1977; Worden, 1995). In effect, these works theorize a psychological and behavioral typology of police. For instance, Worden (1995) argues that police can be combined into five types: “professional,” “tough cop,” “clean-beat, crime-fighter,” “problem solver,” and “avoider.” “Tough cops” have cynical outlooks, see people as self-interested and hostile to police, and are more likely to use force improperly than other types of police.

Second, many empirical studies of police demonstrate statistically significant connections between personal characteristics and officer behavior (Alpert & Dunham, 2004; Bayley & Garofalo, 1989; Garner & Maxwell, 2002; Worden, 1989; Worden & Shepard, 1996). For instance, less-educated and experienced officers may use more force than more educated and experienced officers (Terrill & Mastrofski, 2002; Worden 1996). Also, officers who have negative views about citizens or positive views about using force may be more likely to use force (Worden, 1996). A review of the use of force by the Los Angeles Police Department showed that certain officers used force disproportionately and concluded that individual-level characteristics may make some officers more violence prone (Toch, 1996). Although personal factors may not be as powerful at predicting police behavior as situational characteristics (Worden, 1996), it is not rare for personal factors to be statistically significant predictors of officer behavior.

A final reason to question the view that personal or psychological characteristics are less important is that scholars have been unduly narrow in conceiving of the individual and his or her psychological approach to an interaction. In fact, many of the traditional “situational” factors rely on officers’ perceptions. Whether a person, or a group of bystanders, is being “antagonistic” or “disrespectful” is, to some degree, a matter of individual perception. Likewise, though the race and income level of suspects are seen as objective realities in police scholarship, determining the racial or class identity of an individual is a perceptual challenge: It is not clear that different police interpret the same identity or situation in the same way. In fact, sequential step research, which examines the police–suspect encounter as a series of interactions and responses, depicts interactions as particular because they are based on assessments of individuals and situations (Alpert & Dunham, 2004; Sykes & Brent, 1983). As such, each step of an interaction between the police and a suspect may require psychological interpretation and analysis.

Hypotheses

To understand how police officers develop their views about force, this manuscript discusses three different theoretical approaches to understanding organization psychology. The rational systems approach defines organizations as collectivities that pursue relatively specific goals using formalization—attempts to routinize or structure employee behavior (Scott, 2003). This perspective is useful because it suggests that organization leaders are likely to use formal influences, tools that are relatively in their control like training, monitoring, and supervision, to create a “psychological environment” that influences workers’ perceptions and behavior (Simon, 1997). Indeed, in the context of the police department, there is reason to believe that formal influences can have a strong effect on entrants (Alpert & Dunham, 2004; Fyfe, 1996; Klockars, 1996; Rubenstein, 1973; Van Maanen, 1974; Worden, 1996). For example, training is the place where cadets first encounter the mores and rhythms of the department. According to some accounts, this is a highly formative experience (Van Maanen, 1974). The most important actors at training are instructors—typically former police officers—who regale cadets with “war stories” and lessons about “what it’s really like out there.” In addition to training, monitoring how police officers use force may be an effective way to minimize and control excessive force (Klockars, 1996). Formalization may also be accomplished via department superiors. When they wish to structure officer conduct, department elites communicate through precinct commanders and sergeants; there is some indication that these formal influences shape police officers’ views of force (Alpert & Dunham, 2004; Lawrence, 2000).

Based on the rational systems approach, I expect that the police department will make a strong effort to shape incoming police cadets’ views of force using formalization. However, it is difficult to state a directional hypothesis because there are conflicting accounts about the effects of formal influences on police. Some argue that these factors contribute to encouraging the use of force (Lawrence, 2000), whereas others argue that they reduce the amount and types of force used by police (Alpert & Dunham, 2004; Klockars, 1996). Also, police departments may send mixed signals about the use of force via formal communication channels (Toch, 1996), and these messages may vary by city and department (Fyfe, 1996). Like most jurisdictions, the city that I studied in this project had written guidelines meant to regulate the use of force. In response to complaints about excessive force, prior to the study, the city reviewed its use of force policy and issued stricter regulations about some aspects of the use of force. However, it is not clear whether these policies were being conveyed, reinforced, or counteracted by other formal influences. As such, I adopt a nondirectional hypothesis:

Hypothesis 1: Police officers’ views of force will be associated with formal organization influences like training, monitoring, and supervision.

It is also useful to consider how the natural systems perspective on organizations might help explain police development. This perspective defines organizations as porous, multigoaled collectivities that attempt to adapt to their surroundings (Scott, 2003). With a more expansive definition, this perspective directs attention to the importance of intra- and extraorganizational influences on organization functioning. The natural systems perspective does not deny the importance of formalization but argues that it is also important to understand internal, informal influences, such as culture and peer networks. Empirical research reinforces this general expectation. For example, one study looked across a host of bureaucracies and tried to understand workers’ decisions to work, shirk, or sabotage (Brehm & Gates, 1997). The study found that an important factor influencing employees was their solidary preferences—their feeling of acceptance inside their workplace based on their interactions with peers.

In the context of the police department, there are reasons to expect that informal organization influences may shape officers’ views (Worden, 1996). For example, ethnographies of police socialization suggest that the most important influence on entering workers is experienced officers (Rubenstein, 1973; Van Maanen, 1974). After academy graduation, and concluding their formal training, police in the research city are dispatched to their precincts where they receive informal training from veterans. Veterans teach rookies how to do the job and end up suffering their errors (Rubenstein, 1973). According to Lester (1996), police are rewarded by their colleagues when they take aggressive and forceful action and punished when they take caution. As such, perhaps veterans encourage newcomers to use more force. Alternatively, because departments have many different types of veteran officers (Worden, 1995), perhaps the messages sent by veterans about force vary considerably by officer and location.

The culture of a police department—“the unwritten rules, mores, customs, codes, values and outlooks” (Chemerinsky, 2000)—may also be an important informal influence if it affects the choices that officers make (Goldsmith, 1990; Worden, 1996). There is general agreement about the existence of a police occupational culture that links officers across geographical spaces and, perhaps, time (Walker & Katz, 2008). However, all departments are not alike: Some have argued that each department has its own unique culture (Chemerinsky, 2000; Walker & Katz, 2008); these cultures, to the extent that individual officers are affected by them, may send particular messages about the use of force (Toch, 1996; Walker & Katz, 2008).

Based on these accounts, it appears important to focus on informal organization influences. However, as with the formal influences, it is not clear how informal influences will shape officers’ views of force: They may push new officers toward or away from a more enthusiastic approach to force. As such, I adopt a nondirectional hypothesis:

Hypothesis 2: Police officers’ views of force will be associated with informal organization influences like peers, culture, and veterans.

In contrast to these two organization theories, continuity theory expects that psychological characteristics, such as personality, ideas, and beliefs, persist throughout life (Atchley, 1999). The theory suggests that, as they have new experiences, people use psychological approaches that have worked for them previously. There is empirical support for this approach in personality studies (De Fruyt et al., 2006; Hampson & Goldberg, 2006), the representative bureaucracy literature (Selden et al., 1998), and the public service motivation literature (Perry, 1997). Although these perspectives do not suggest that organizations are impotent, or that individuals are static, they highlight the importance of pre- or extraorganization influences.

In police scholarship, there has been significant attention paid to the personality tendencies of officers. Since the 1970s, many police departments have screened applicants using psychological tests like the Minnesota Multiphasic Personality Inventory. Although the utility of these tests in predicting successful policing is questionable (Grant & Grant, 1996), they remain popular and indicate the importance that departments place on understanding the officers that they accept. Also, scholars have tried to understand whether people with particular personality types are drawn to police work (Skolnick, 1966; Walker & Katz, 2008). In particular, there is interest in whether recruits are more authoritarian, violence prone, or attracted to rules than the general public. In general, there is little evidence suggesting that particular types of people are drawn to work as police or that police are psychologically different from the general public.

However, even if police are not all of one psychological type, at the individual level, we may expect continuity even after organization entry. Just as some workers have tendencies toward rule-following, whereas others tend to think and act in “unbureaucratic” ways (DeHart-Davis, 2007), perhaps some police are more enthusiastic about force than others based on their preorganization experiences and demographic characteristics. For example, because many minorities believe that police misuse force (Weitzer & Tuch, 2004), perhaps minority police are less likely to view using force as an effective way to keep the peace. Also, historically, police departments have recruited heavily from the military; as using force is a key aspect of military service, perhaps former soldiers have different views about using force. Based on continuity theory, and related empirical research, I expect that

Hypothesis 3: Police officers’ views of force will remain associated with their entering views, prior experiences, and demographic characteristics.

Research Design and Data

The above hypotheses point to different explanations about how police officers develop their views of force. I test these expectations using data from a longitudinal case study of entering police officers in a single city during their first 2 years on the job. To protect the identities of the research participants, the city’s name will remain confidential. As police departments are typically controlled by municipalities, there is probably no single U.S. police department that is representative of all departments. As such, there may be aspects of the project’s findings that are location specific. However, as this project aims to generalize theoretically rather than empirically, nonrepresentativeness is not a major weakness (Yin, 2003). Although I do not argue that these findings can be empirically generalized, a variety of statistics suggest that the research city is comparable with other large cities.³

In early 2006, I recruited an entering class of approximately 80 police cadets for this project. I administered surveys to all entrants at five times (T1-T5): at the outset of training, after 3 months of training, after 6 months of training, after 1 year inside the department, and after 2 years inside the department. Although they were at the academy, I distributed surveys by hand; after graduation, I sent surveys in the mail. I avoided some sampling issues because all class members had the opportunity to complete surveys at each time; however, it is important to explore the possibility of response or attrition bias. At the police academy such inquiries are unnecessary because more than 97% of cadets participated at each time (T1-T3). When I shifted to mailing surveys (T4-T5), the response rates dropped to 57% and 74%, respectively. To search for response bias, I estimated workers’ likelihood of responding based on a host of control variables. At T4, there were no statistically significant patterns in the data; at T5, the data were generally free from response bias except that males were more likely to respond than females. As a result, gender will be excluded from this discussion and analysis. I also searched for response bias by comparing the entering attitudes of both sets of participants who did not respond with those who did; similarly, I searched for attrition bias by comparing the entering responses of those who exited the department with those who did not. For all comparisons, I found no statistically significant differences. These findings suggest that attrition and response bias were not substantive problems for this analysis.

Complementing the surveys, I conducted in-depth interviews with a subset of 15 officers at three times (T1, T4, and T5). In total, I conducted 39 interviews.⁴ On the first round of surveys, I asked police cadets whether they were willing to be interviewed; in exchange for their participation, I offered them a small gift or meal. Nearly two-thirds of police cadets (63%) indicated a willingness to participate. In choosing interviewees, I balanced an interest in selecting individuals who reflected the demographic attributes of the entering group with an interest in having some variation along key demographic variables (race, sex, etc.). By and large, I achieved this goal: Although interviewees were not representative of entrants in all ways, there were few differences between interviewees and the general group of entrants. I conducted interviews in person or by phone. In form, the interviews were semistructured: I used a uniform interview protocol so that all interviews followed the same pattern and included a basic set of questions; however, I also asked nonstandardized follow-up questions in response to interviewee answers. Interviews were recorded and typically lasted from 30 to 45 min.

As this study relies on surveys and interviews, it is important that the reader understands this study’s parameters in relation to other studies of police. Many studies in the police literature rely on data gathered by observation and administrative records (Alpert & Dunham, 2004). Although these studies are vulnerable to an array of epistemological critiques,⁵ they are useful tools for studying police behavior. As conducting observation and accessing the department’s administrative records were not possible for this project, like other studies, I focus on officers’ views of force (Lester, 1996; Weisburd et al., 2000). Although there is sufficient evidence to assume that attitudes and behavior are connected, this article does not show how socialization affected police behavior. Future efforts should study police socialization and use of force using systematic observation and analysis of administrative records.

Independent Variables

To test this study’s hypotheses, I gathered organizational and personal data from individual officers. Organizational questions can be found in the appendix. Although the organizational questions were theoretically independent, in practice, there were some strong relationships (correlated above .50). As a result, I dropped supervision and peers from this analysis. Also, though I made an effort to control for the level of crime that officers faced, the department does not publish crime statistics by district. As such, it is unclear how the crime environment that officers entered affected their views of force.

On the personal side, police officers’ entering views about force (discussed later in the article under “Dependent Variables”) will be used to predict their subsequent views. I also measure the association between officers’ prior experiences and demographic characteristics and their views of force: Racial identity is a dummy variable for minority or White, educational attainment is coded from 1 (high school diploma or general equivalency degree) to 6 (graduate degree), political views is coded from 1 (liberal) to 7 (conservative), and preorganizational family incomes is coded from 1 (US$19,999 and below) to 7 (US$70,000 and above).

Dependent Variables

Like other studies in the police literature, I seek to understand how police interpret and understand their roles and powers (Glaser & Denhardt, 2010; Maynard-Moody & Musheno, 2003; Muir, 1977; Rubenstein, 1973; Weisburd et al., 2000); as such, this article’s dependent variables are police officers’ attitudes about the use of force. In particular, there were two aspects of officers’ views about force that I examined. First, like other attitudinal studies of the use of force, I sought to understand how police viewed the amount of force that they had (Corbett, Meyer, Baker, & Rudoni, 1979; Weisburd et al., 2000). Did police believe that they should be able to use whatever force is necessary to do their jobs or did they think that there were limits to about how much force police should have? In other words, I wanted to understand how much police supported a “strong police role” (Corbett et al., 1979). To measure this perception, I asked them how much they agreed that “cops should be allowed to use all necessary force to do their jobs.” Second, prior studies have inquired about police officers’ views about the utility of force in doing their jobs (Worden, 1996). The theory behind this question is that police who do not view force as an effective tool may be less inclined to use it regularly. For this area, I asked them how much they agreed that “using force is an effective technique for keeping the peace.”⁶ Officers responded to these questions by choosing an answer from a 5-point scale from 1 (strongly disagree) to 5 (strongly agree); the middle point of the scale (3) was labeled “neutral.”

Survey Findings

Looking at the mean and standard deviation of the dependent variables in Table 1, officers’ views of the effectiveness and necessity of force developed differently. There is a steady increase in agreement about the necessity of force: In each wave, the average response moves closer toward “strongly agree,” and the spread diminishes. Combining the answers “agree” and “strongly agree,” by the end of their 2nd year, 96% agreed with this statement. However, officers’ views about the effectiveness of force reveals a different pattern: For the 1st year, after entrance, officers’ responses moved toward the view that force was effective; however, at the final time, there was a large decrease in the mean response. In addition, the spread of responses grew over time, indicating that there were diverging views about the effectiveness of force. Combining the answers “agree” and “strongly agree,” initially more than half of the cadets agreed that force was effective. After entering the academy (T2 and T3), the percentage that saw force as effective rose (to around 75%) and stayed roughly constant through the end of their 1st year (T4); at the end of their 2nd year, this figure dropped to approximately 60%.

Table 1.

Summary Statistics and Dependent Variable Cross Tabulation

	n	M	SD	Minimum	Maximum	Cronbach’s α
Independent variables
Prior service in armed forces	84	0.35	0.48	0	1	—
White	81	0.67	0.47	0	1	—
Political views	81	4.06	1.29	1	7	—
Education	83	2.71	1.31	1	5	—
Age	74	26.39	4.81	19	39	—
Family income	82	4.87	1.68	1	7	—
T1 view (force necessity)	83	3.90	1.09	1	5	—
T1 view (force effectiveness)	80	2.51	0.83	1	5	—
Training T2	75	22.16	2.28	16	25	.80
Culture T2	74	4.29	0.70	2	5	—
Training T3	72	21.00	2.95	12	25	.79
Culture T3	73	4.00	0.90	1	5	—
Training T4	39	6.56	1.87	3	10	.69
Veterans T4	40	3.87	1.01	1	5	—
Culture T4	41	3.37	0.94	1	5	—
Monitoring T4	41	24.15	5.17	11	35	.83
Training T5	48	6.33	1.94	2	10	.69
Veterans T5	50	3.16	1.20	1	5	—
Culture T5	49	3.65	1.01	1	5	—
Monitoring T5	50	23.26	6.64	10	35	.86
Dependent variables
Necessary force T1	83	3.90	1.09	1	5
Necessary force T2	75	4.25	0.95	1	5
Necessary force T3	72	4.22	0.88	2	5
Necessary force T4	40	4.32	0.91	2	5
Necessary force T5	50	4.50	0.58	3	5
Effective force T1	80	3.49	0.83	1	5
Effective force T2	75	3.92	0.93	1	5
Effective force T3	73	3.93	0.86	2	5
Effective force T4	38	4.08	0.88	2	5
Effective force T5	49	3.63	1.17	1	5
Dependent variable cross tabulation
	Necessary force			Effective force
	5-point scale		3-point scale	5-point scale		3-point scale
T2	47% (35/74)		68% (50/74)	39% (28/72)		54% (39/72)
T3	42% (30/71)		68% (48/71)	39% (27/70)		53% (37/70)
T4	45% (18/40)		73% (29/40)	34% (13/38)		63% (24/38)
T5	37% (18/49)		67% (33/49)	38% (18/47)		51% (24/47)

Note: Cross-tabulation analysis examines continuity by comparing initial and subsequent views.

Interview Findings

This section discusses how police in this project viewed the use of force and how, if at all, their views changed. To begin, two key force themes emerged during project interviews. First, police appeared to view force through a prism of safety: their own and that of their colleagues. If force needed to be used to protect oneself or fellow police, one should do it. For example, consider the comments of one officer in describing the use of force:

No one else can tell you when you feel safe or if it was dangerous or not dangerous. Therefore if you need to use a certain amount of force to ensure your safety or another officer’s safety that’s what you [do]. Even if it’s something as simple as grabbing someone and handcuffing them and restraining them until your investigation is complete … nobody else [can] tell you, “No, you shouldn’t have done that.” Then there’s the extreme if you’re in a situation where you have to use your baton or gotta use deadly force … nobody else [can] tell you, “Well, no, you don’t have to shoot there” or “No, you don’t have to hit ‘em with the baton.” In the end you got a goal and hopefully you get back home the same you came to work.

Although police use of force is sometimes depicted as haphazard and deliberately confrontational, these data indicate that officers saw force as a defensive measure.

The second theme emerging from these interviews had to do with officers’ views about the effect of force on a situation; this finding may offer a clue for explaining the variation (in the survey data) in officers’ views about the effectiveness of force. Although all officers described volatile situations as difficult, they differed about the effect that force had on such situations. One group appeared to see force as the way to show that they were dominant and defuse a tense situation. Another group saw force as a factor that could make a bad situation much worse. Representing the latter group, one officer commented,

You know I thought I’d have to go in and pull my stick out and yell at people. You can handle all the situations a million different ways. And some people like to go in screaming and yelling and swinging a stick and beating everybody up. But if you go in and don’t swing the stick and don’t have [to] yell and get the same amount accomplished. It’s all how you want to go about doing it. Some instances you gotta go in swinging and some instances you gotta go in and grab people but as long as you don’t let your emotions take control of what your actions are gonna be. Most of the time you get a lot done with just being calm and talking in a normal tone of voice.

In addition to understanding how officers generally saw the use of force, I analyzed officers’ interview answers over time to detect whether or not they adopted new views about the use of force. Of the 15 interview participants, 9 discussed force in multiple waves. Comparing officers’ views across many interviews requires a certain amount of subjective interpretation. To guide my analysis, I examined their comments along three dimensions: tone (how officers presented the use of force in interviews), content (what type of force officers used), and context (in what type of situations officers used force).

My analysis of officer interviews showed that there was relative continuity over time in eight of the nine cases in which officers discussed force in multiple waves. Although there is not enough space to discuss each of these instances in sufficient depth, this section highlights the views of one officer who, although not necessarily representative in content, demonstrates the continuity that was evident in most interviews. In our first interview, while Officer Stevens (a pseudonym) was at the police academy, he was resisting the view that he had heard from an instructor at the academy that officers cannot work too closely with the neighborhood. He was resisting this view because of his interest in reducing the use of force in an effort to improve relations with the community.

Putting on the uniform is going to give me a certain appearance to the public they’re gonna act a lot different to me than just a regular person out on the street. So I’m curious to see the feedback when I’m on the street in uniform … And to see if I have the ability to kind of tone down some of the aggression that they have toward police officers … the sergeant talked to us about how you can’t, you can’t be integrated into the community so much. Well I’d like to see how much of a, how, what type of reaction I get from them and how I react.

At T2, after he had completed the academy and worked on the streets for 6 months, he discussed his approach to dealing with difficult, force-potential situations. Again, we see elements of his interest in easing tensions.

I’d describe [my style] as wants to see who needs the most calming down because that’s the majority of the situations you go in on are high tension or … I would say that you know I would be calm about going into a situation to try to calm them down because most of the time I’ve found when you handle someone real calmly they try to calm themselves down to meet the level that you’re on communicating with them versus seeing someone arguing with someone and they’ll continue arguing with you so you try to control their emotions by your level of interaction.

Finally, at the end of his 2nd year in the department, he described how his approach impacted the situations he was trying to deal with on the streets.

Depending on how you show if you’re showing aggression if you’re being friendly all of your presentation why you’re doing what you have to do—you’re explaining to them. That’s what I found out. I do the same type of work that some officers do that they’ll take their work and they’ll get assault on police and I’ll have those same type of people thanking me for doing my job. And now I’m finding that just on how you conduct yourself it depends on it on what the outcome will be.

As these selections show, Officer Stevens saw his role as a police officer as an ameliorator of tension. Although he understood the powers that he had, and the importance of being authoritative, like most officers that I interviewed, his tone about aggression and calmness was remarkably similar throughout the 2 years.

Analysis

Thus far, I have explored and described police officers’ views about force. Now I return to the question posed at the outset: At what times, and how, are officers’ attitudes about force associated with intra- and extraorganization influences? To test the hypotheses developed previously, this section analyzes officer survey data. It begins with a cross-tabulation analysis (bottom of Table 1), which provides some initial evidence about the third hypothesis; by comparing officers’ initial and subsequent views about force, I can determine the extent to which individuals altered their views. For each question, there are two columns. The first column (5-point scale) indicates the percentage of respondents who chose the exact same answer in a subsequent wave that they chose initially; the second column (3-point scale) indicates the percentage who chose the same general answer (agree, neutral, or disagree). Looking at the 5-point scales for both questions, we see a drop from T2 to T5: By the end of the 2nd year, less than 40% of officers maintained the exact same view about the necessity and utility of force. The corollary, of course, is that more than 60% of officers shifted their views about force. However, when we collapse their answers into three categories (agree, neutral, and disagree), we see that more than two-thirds maintained their initial, general view about the necessity of force and more than half maintained their initial, general view about the effectiveness of force. In comparing the two questions, these results show that fewer officers radically altered their views about necessary force. Taken as a whole, this table paints a picture of change and stability: Many officers shifted their views somewhat, but few officers adopted substantially new positions.

Next, I shift to analyzing the survey data using a multivariate model that includes organizational, demographic, and attitudinal variables.⁷ This approach permits a rigorous examination of the survey data to test each of the three hypotheses. In T1, I use a model that examines whether any demographic or attitudinal variables were associated with officers’ views of force. After T1, the model contains officers’ preorganization views of force, demographic and attitudinal variables, and organizational variables (formal and informal influences). Table 2 summarizes the results of this analysis at each time.

Table 2.

Statistical Analysis of Officers’ Views

	Necessary					Effective
	T1	T2	T3	T4	T5	T1	T2	T3	T4	T5
	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)
Extraorganizational influences
Armed services	−0.285 (0.294)	0.462 (0.373)	−0.357 (0.386)	0.847 (0.829)	0.554 (0.495)	−0.214 (0.295)	−0.419 (0.345)	0.207 (0.376)	2.017* (0.828)	0.071 (0.410)
White	0.134 (0.296)	0.068 (0.352)	0.100 (0.352)	0.835 (0.764)	−0.461 (0.507)	0.777* (0.311)	0.237 (0.345)	0.039 (0.347)	−0.406 (0.861)	1.132* (0.441)
Age	−0.059 (0.032)	−0.012 (0.039)	−0.016 (0.040)	−0.130 (0.078)	−0.113 (0.060)	0.018 (0.031)	−0.123* (0.038)	−0.040 (0.037)	−0.362* (0.104)	−0.019 (0.048)
Education	−0.182 (0.142)	−0.051 (0.158)	0.031 (0.162)	−0.035 (0.253)	0.246 (0.214)	0.002 (0.141)	0.148 (0.158)	0.059 (0.158)	0.322 (0.270)	0.146 (0.183)
Family income	0.038 (0.088)	0.008 (0.102)	−0.211* (0.105)	−0.221 (0.193)	−0.224 (0.148)	−0.072 (0.085)	−0.214* (0.097)	−0.195* (0.097)	−0.509* (0.209)	−0.008 (0.124)
Conservative	0.298* (0.116)	−0.211 (0.134)	−0.102 (0.140)	−0.096 (0.279)	0.224 (0.194)	−0.140 (0.114)	0.106 (0.121)	0.058 (0.132)	0.309 (0.252)	−0.091 (0.165)
T1 views of force		0.525* (0.155)	0.576* (0.168)	0.523 (0.274)	0.352 (0.222)		0.331 (0.192)	0.205 (0.205)	0.757* (0.345)	0.514* (0.237)
Organizational influences
Training		−0.099 (0.078)	0.050 (0.066)	0.171 (0.148)	0.006 (0.116)		−0.164* (0.077)	0.004 (0.068)	0.174 (0.155)	0.155 (0.105)
Culture		0.360 (0.256)	−0.011 (0.269)	0.270 (0.272)	0.426 (0.260)		0.548* (0.267)	0.444 (0.300)	0.331 (0.306)	−0.013 (0.209)
Veterans				−0.290 (0.337)	−0.482 (0.260)				−0.201 (0.355)	0.002 (0.214)
Monitoring				0.033 (0.054)	0.080 (0.043)				0.040 (0.054)	0.044 (0.035)
n	69	62	59	34	42	68	62	60	33	42
Likelihood Ratio Chi (df)	16.21	19.20	18.87	20.35	14.55	7.55	28.30	10.92	29.00	23.36
p > χ²	.013	.024	.026	.041	.204	.273	.001	.281	.002	.016
Pseudo R²	.085	.132	.137	.266	.204	.046	.189	.079	.362	.189

p > .05.

On the left side of this table, we see that at T1, cadets who were more conservative were more likely to agree that police should be allowed to use all necessary force to do their jobs. Postestimation analysis shows that conservative officers were 42 percentage points more likely to strongly agree with this statement. After they began training, this relationship ceased being significant for the remainder of their 2 years in the study. While at the police academy (T2 and T3), the cadets’ views remained strongly associated with their entering views. At T2, cadets who initially agreed that police should be able to use all necessary force were 56 percentage points more likely to strongly agree; at T3, this group was 55 percentage points more likely to strongly agree. In addition, as they neared graduation (T3), cadets who lived in less prosperous households were 31 percentage points less likely to strongly agree with this statement. After graduating from the police academy, becoming full-fledged police officers, and beginning work on the streets, there were no strong predictors of officers’ views about necessary force.

On the right side of the table, we see that initially cadets’ views were tied to their racial identities: Postestimation analysis reveals that Whites were 8 percentage points more likely to strongly agree that force was effective. This relationship diminished while cadets were enrolled at the academy suggesting the potency of some organizational influences at that time. Further solidifying that view, at T2, we see that two organizational factors were significant predictors of officers’ views. Cadets who were strongly influenced by the academy’s culture were 28 percentage points more likely to strongly agree that force was effective; cadets who were strongly influenced by instructors were 24 percentage points less likely to strongly agree that force was effective. In addition, two demographic variables were strong predictors of cadets’ views: Older cadets were 32 percentage points less likely to strongly agree, and cadets from higher income families were 24 percentage points less likely to strongly agree.

As the academy progressed, and cadets approached graduation, there was no longer any strong relationship between organizational variables and cadets’ views about the effectiveness of force. Rather, the only statistically significant predictor of officers’ views about the effectiveness of force was their level of family income; at T3, those from higher income families were 24 percentage points less likely to strongly agree that force was effective. After graduating from the academy and working on the streets for 6 months (T4), we see that four extraorganizational influences were strong predictors of officers’ views. Those who initially saw force as effective were 70 percentage points more likely to strongly agree that force was effective; those who had previously served in the armed forces were 68 percentage points more likely to strongly agree that force was effective. Also, older officers were 78 percentage points less likely to strongly agree, and officers from higher income families were 61 percentage points less likely to strongly agree. Finally, after 2 years inside the organization, and a year of working on the streets (T5), two extraorganizational influences were significant: Officers who initially saw force as effective were 34 percentage points more likely to strongly agree that force was effective, and White officers were 18 percentage points more likely to strongly agree that force was effective.

Conclusion

To what extent are public workers shaped by the organizations that they join and to what extent are they shaped by outside factors? What roles do socialization and self-selection play in creating public workforces? Before discussing the findings and implications of this study, it is important to restate some of its characteristics. To begin, this study examined one group of public servants in one city at one period of time. Although the city in which the police trained and worked was not remarkably different from other large cities, there may be aspects of this project that are location specific. Also, the police are not representative of all public servants; rather, they are a crucial case to examine because prior research suggests that socialization in police departments is particularly strong. In addition, though I assume they are connected, I have focused on workers’ attitudes not behavior. Finally, I have focused on one aspect of development: officers’ views about using force. As such, I do not argue that the trends observed in this study mirror other aspects of their psychological development.

This analysis was guided by three hypotheses: The rational systems perspective indicated that officers’ views would be associated with formal influences, the natural systems perspective suggested that officers’ views would be associated with informal influences, and continuity theory indicated that officers’ views would be associated with outside influences. In sum, the findings here suggest some support for the first and second hypotheses and significant support for the third hypotheses. In Table 2, we see that there was one instance in which a formal organizational influence (training) was strongly associated with officers’ views and one instance in which an informal organizational influence (culture) was strongly associated. Both occurred early in training and then faded away at subsequent times. Veterans and monitoring, two organization influences that we expected to affect officer development, were never statistically significant predictors of officers’ views. Although these findings imply a limited role for the organization, there is evidence that workers altered their views over the course of entry. First, for the necessary force question, we saw a steady reduction in the spread of views as officers’ consolidated around the view that police should be able to use all necessary force. Second, at the bottom of Table 1, we see that, though the shift was rarely radical (defined as 2 points or more), more than 60% of officers shifted their view on using force by the end of their 2nd year.

Turning to the third hypothesis, there was support for the view that outside factors continued to matter even after socialization. Table 2 shows that extraorganizational variables were often strong predictors of officers’ views. During their first 6 months inside the department, cadets’ views about necessary force remained tied to their entering views and, in one instance, preacademy family income. We see that throughout the full 2 years officers’ views about the effectiveness of force remained strongly associated with outside influences. Also, project interviews suggested that there was more continuity than change in officers’ views about the usage of force.

These findings have four implications for our understanding of socialization and personnel management. First, they reinforce the view that organizations shift workers’ attitudes and approaches to the job. Temporally, these findings suggest that the best time for organizations to affect how workers think is early in socialization (that was when two organization influences were associated with officers’ views). However, more work is needed to understand how public organizations shape workers’ views. Although I included the variables that the police, public administration, and organization socialization literatures suggest are most important in shaping newcomers, only two organization factors were strongly associated with workers’ attitudes and both of these occurred early in training. Perhaps these nonfindings resulted from how I measured organization influences; nonetheless, future works should revisit these questions and show more clearly how public organizations affect newcomers’ views.

Second, though most of the officers in this study shifted their views, in most cases, it was not a major shift. Prior studies have argued that organization socialization, generally, and police socialization, particularly, explain much about how public servants think and act. Van Maanen (1975) argues that “the police culture can be viewed as molding the attitudes—with numbing regularity—of virtually all who enter.” Wilkins and Williams (2009, p. 215) suggest that socialization may “hinder the link between passive and active representation” for minority police. This project’s findings imply a different view. Organizations are certainly an important influence on incoming police but, as the public service motivation and representative bureaucracy literatures argue, so are their preorganizational views and experiences. The findings here suggest that entering officers may have thought like police before they entered their organizations so the distance between newcomer and veteran was minimal.

This continuity is also important for our understanding of use of force: Although the individual-level approach to understanding force is usually portrayed as explaining little about how police think or act, these findings lend credence to this approach. This is not to suggest that organizations and situations are secondary; rather, as scholars try to understand police behavior, the findings here suggest they would be well served by paying attention to what views and experiences individuals bring with them into the department.

Third, these findings suggest that socialization is more complicated than it is sometimes portrayed. In particular, socialization, even in a particular realm like attitudes, may succeed in some ways and fail in others. Put simply, socialization may be content dependent. This was evident when I looked at my two outcomes individually: There was more evidence that the police department affected officers’ views about necessary force—the groups’ views tightened around the agreed answer, and though initially officers’ political views and class predicted their views, this relationship receded. For the effectiveness question, the groups’ views never tightened around the mean, and officers’ views remained tied to outside factors. This finding suggests that observers of organizations need to be specific when they make claims about socialization and should resist the tendency to make blanket statements about socialization. Related to this point, the findings here demonstrate the importance of broadening the types of outcomes studied by organization socialization scholars. Although the typical outcomes (role ambiguity, role conflict, organization commitment, job satisfaction, etc.) are important to study, this literature would be enhanced by looking at an array of other outcomes like workplace attitudes and identities.

Finally, this study shows the importance of studying socialization for longer periods of time. Although the organization socialization literature tends to focus on the first 6 months to a year of a worker’s career, this study shows there can be changes after this time. In particular, when we look at the effectiveness question, we see that agreement rose during the academy and then fell off after police began work. Such data would never be captured if we assumed that workers were socialized by the end of the training period. To develop a fuller portrait of officer development, this literature needs to track incoming workers for longer periods of time.

Footnotes

Appendix

Questions Used to Determine Organizational Independent Variables

Training T2-T3 (index)

In general I don’t feel like our training has been sufficient. (reversed)

Communication between the cadets and the trainers is pretty good.

Academy instructors pay close attention to our development.The instructors’ stories are an important aspect of learning this job.

It is easy to ask instructors questions when I feel confused.

Training T4-T5 (index)

In general I don’t feel like our training was sufficient. (reversed)

Thinking about how you do your job, how much [does what you learned at the Academy] influence you?

Monitoring T4-T5 (index)

How closely are you monitored on the following actions?

• Making arrests

• Writing tickets

• Walking or driving a beat

• Keeping up with paperwork/reports

• Interacting with civilians face-to-face

• Using a weapon

• Issuing warnings

Culture T2-T3

Just being around the Academy environment has had an effect on me.

Culture T4-T5

The district environment has had an effect on me.

Veterans T4-T5

When you think about how you do your job, how much [do veteran officers in your district] influence you?

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author received generous financial support for this research from the Institute for Research on Poverty at the University of Wisconsin-Madison, The City College of New York, and Haverford College.

Notes

Bio

Zachary W. Oberfield is an assistant professor of political science at Haverford College. He studies organization socialization, diversity, and representative bureaucracy. He has previously published in the Journal of Public Administration Research and Theory.

References

Aberbach

Rockman

(2000). In the web of politics. Washington, DC: Brookings Institution.

Adams

Alpert

Dunham

Garner

Greenfeld

Henriquez

. . . Smith

(1999). Use of force by police. Washington, DC: National Institute of Justice and Bureau of Justice Statistics.

Ajzen

(2001). Nature and operation of attitudes. Annual Review of Psychology, 52, 27-58.

Ajzen

Fishbein

(1977). Attitude-behavior relations. Psychological Bulletin, 84, 888-918.

Alpert

Dunham

(2004). Understanding police use of force. New York, NY: Cambridge University Press.

Argyris

(1993). Integrating the individual and the organization. New Brunswick, NJ: Transaction.

Atchley

(1999). Continuity and adaptation in aging. Baltimore, MD: Johns Hopkins University Press.

Bardach

Kagan

(1982). Going by the book. Philadelphia, PA: Temple University Press.

Bayley

Garofalo

(1989). The management of violence by police patrol officers. Criminology, 27, 1-25.

10.

Bazley

Lersch

Mieczkowski

(2006). Police use of force: Detectives in an urban police department. Criminal Justice Review, 31, 213-229.

11.

Bittner

(1970). The functions of the police in modern society. Washington, DC: U.S. Government Printing Office.

12.

Brehm

Gates

(1997). Working, shirking, and sabotage. Ann Arbor: University of Michigan Press.

13.

Brewer

Selden

(1998). Whistleblowers in the federal civil service. Journal of Public Administration Research and Theory, 8, 413-439.

14.

Brown

(1981). Working the street. New York, NY: SAGE.

15.

Chemerinsky

(2000). An independent analysis of the Los Angeles Police Department’s Board of Inquiry Report on the Rampart scandal. Los Angeles, CA: Police Protective Society.

16.

Corbett

Meyer

Baker

Rudoni

(1979). Conflict, consensus, and perceptions of support among the police. Journal of Police Science and Administration, 7, 449-458.

17.

Crewson

(1997). Public service motivation: Building empirical evidence of incidence and effect. Journal of Public Administration Research and Theory, 7, 499-518.

18.

Davis

Rostow

Pinkston

Cowick

(1999). An investigation into the usefulness of the MMPI and MMPI-2. Journal of Police and Criminal Psychology, 14, 100-106.

19.

De Fruyt

Bartels

Van Leeuwen

De Clercq

Decuyper

Mervielde

(2006). Five types of personality continuity in childhood and adolescence. Journal of Personality and Social Psychology, 91, 538-552.

20.

DeHart-Davis

(2007). The unbureaucratic personality. Public Administration Review, 67, 892-903.

21.

Diamond

(1985). The social character of bureaucracy. Political Psychology, 6, 663-679.

22.

Diamond

Allcorn

(1986). Role formation as defensive activity in bureaucratic organizations. Political Psychology, 7, 709-732.

23.

Friedrich

(1980). Police use of force. Annals of the American Academy of Political and Social Science, 452, 82-97.

24.

Fyfe

(1982). Blind justice: Police shootings in Memphis. Journal of Criminal Law and Criminology, 73, 707-722.

25.

Fyfe

(1996). Training to reduce police-civilian violence. In Geller

Toch

(Eds.), Police violence. (pp. 165-179) New Haven, CT: Yale University Press.

26.

Gallo

Collyer

Gallagher

(2008). Prevalence of force by police in Rhode Island jurisdictions. Criminal Justice Review, 33, 480-501.

27.

Garner

Maxwell

(2002). Understanding the use of force by and against police in six jurisdictions. Washington, DC: Department of Justice.

28.

Garner

Schade

Hepburn

Buchanan

(1995). Measuring the continuum of force used by and against police. Criminal Justice Review, 20, 146-168.

29.

Glaser

Denhardt

(2010). Community policing and community building: A case study of officers’ perceptions. American Review of Public Administration, 40, 309-325.

30.

Golden

(2000). What motivates bureaucrats? New York, NY: Columbia University Press.

31.

Goldsmith

(1990). Taking police culture seriously. Policing & Society, 1, 91-114.

32.

Goodsell

(2004). The case for bureaucracy. Washington, DC: CQ Press.

33.

Grant

(1996). Officer selection and the prevention of abuse of force. In Geller

Toch

(Eds.), Police violence. (pp. 150-164) New Haven, CT: Yale University Press.

34.

Hampson

Goldberg

(2006). A first large cohort study of personality trait stability. Journal of Personality and Social Psychology, 91, 763-779.

35.

Houston

(2000). Public service motivation. Journal of Public Administration Research and Theory, 10, 713-727.

36.

Jones

(1986). Socialization tactics, self-efficacy, and newcomers’ adjustments to organizations. Academy of Management Journal, 29, 262-279.

37.

Katz

Kahn

(1966). The social psychology of organizations. New York, NY: John Wiley.

38.

Kaufman

(1960). The forest ranger. Washington, DC: Resources for the Future.

39.

Klockars

(1996). A theory of excessive force. In Geller

Toch

(Eds.), Police violence. (pp. 1-22) New Haven, CT: Yale University Press.

40.

Langan

Greenfield

Smith

Durose

Levin

(2001). Contacts between police and the public: Findings from the 1999 national survey. Washington, DC: U.S. Department of Justice.

41.

Lawrence

(2000). The politics of force. Berkeley: University of California Press.

42.

Lester

(1996). Officer attitudes toward police use of force. In Geller

Toch

(Eds.), Police violence. (pp. 180-190) New Haven, CT: Yale University Press.

43.

Lipsky

(1980). Street-level bureaucracy. New York, NY: Russell Sage.

44.

Maio

Olson

Bernard

Luke

(2006). Ideologies, values, attitudes, and behavior. In Delamater

(Ed.), Handbook of social psychology. (pp. 283-308) New York, NY: Springer.

45.

March

(1994). A primer on decision making. New York, NY: Free Press.

46.

Maynard-Moody

Musheno

(2003). Cops, teachers, counselors. Ann Arbor: University of Michigan Press.

47.

McEwan

(1996). National data collection on police use of force. Washington, DC: Department of Justice.

48.

Merton

(1940). Bureaucratic structure and personality. Social Forces, 18, 560-568.

49.

Mesloh

Henych

Wolf

(2008). Less lethal weapon effectiveness, use of force, and suspect and officer injuries. Washington, DC: National Criminal Justice Referencing Service.

50.

Meyer

(1980). Police shootings at minorities: The case of Los Angeles. Annals of the American Academy of Political and Social Science, 452, 98-110.

51.

Morrison

(2002). Newcomers’ relationships: The role of social network ties during socialization. Academy of Management Journal, 45, 1149-1160.

52.

Muir

(1977). Police: Streetcorner politicians. Chicago, IL: University of Chicago Press.

53.

National Institute of Justice. (2010). Police use of force. Retrieved from http://www.ojp.usdoj.gov/nij/topics/law-enforcement/officer-safety/use-of-force/welcome.htm

54.

Perry

(1997). Antecedents of public service motivation. Journal of Public Administration Research and Theory, 7, 181-208.

55.

Prottas

(1979). People processing. Lexington, MA: Lexington Books.

56.

Reingold

Liu

(2009). Do poverty attitudes of social service agency directors influence organizational behavior? Nonprofit and Voluntary Sector Quarterly, 38, 307-322.

57.

Rubenstein

(1973). City police. New York, NY: Farrar, Straus and Giroux.

58.

Saks

Ashforth

(1997). Organizational socialization. Journal of Vocational Behavior, 51, 234-279.

59.

Saks

Uggerslev

Fassina

(2007). Socialization tactics and newcomer adjustment: A meta-analytic review and test of a model. Journal of Vocational Behavior, 70, 413-446.

60.

Sandfort

(2000). Moving beyond discretion and outcomes. Journal of Public Administration Research and Theory, 10, 729-757.

61.

Schuman

Johnson

(1976). Attitudes and behavior. Annual Review of Sociology, 2, 161-207.

62.

Scott

(2003). Organizations: Rational, natural, and open systems. New York, NY: Prentice Hall.

63.

Selden

Brudney

Kellough

(1998). Bureaucracy as a representative institution. American Journal of Political Science, 42, 717-745.

64.

Sherman

(1980). Perspectives on police and violence. Annals of the American Academy of Political and Social Science, 452, 1-12.

65.

Simon

(1997). Administrative behavior (4th ed.). New York, NY: Free Press.

66.

Skolnick

(1966). Justice without trial: Law enforcement in democratic society. New York, NY: Wiley.

67.

Smith

Visher

(1981). Street-level justice. Social Problems, 29, 167-177.

68.

Sykes

Brent

(1983). Policing: A social behaviorist perspective. New Brunswick, NJ: Rutgers University Press.

69.

Terrill

Mastrofski

(2002). Situational and officer-based determinants of police coercion. Justice Quarterly, 19, 215-248.

70.

Terrill

Reisig

(2003). Neighborhood context and police use of force. Journal of Research in Crime & Delinquency, 40, 291-321.

71.

Theobald

Haider-Markel

(2008). Race, bureaucracy, and symbolic representation. Journal of Public Administration Research and Theory, 19, 409-426.

72.

Toch

(1996). The violence-prone officer. In Geller

Toch

(Eds.), Police violence. (pp. 94-112) New Haven, CT: Yale University Press.

73.

Van Maanen

(1974). Working the street: A developmental view of police behavior. In Jacob

(Ed.), The potential for reform of criminal justice (pp. 83-129). Beverly Hills, CA: SAGE.

74.

Van Maanen

(1975). Police socialization: A longitudinal examination of job attitudes in an urban police department. Administrative Science Quarterly, 20, 207-228.

75.

Van Maanen

Schein

(1979). Toward a theory of organizational socialization. In Staw

(Ed.), Research in organizational behavior (pp. 209-264). Greenwich, CT: JAI.

76.

Walker

Katz

(2008). The police in America: An introduction. New York, NY: McGraw-Hill.

77.

Wanous

(1992). Organizational entry. New York, NY: Addison-Wesley.

78.

Weisburd

Greenspan

Hamilton

Williams

(2000). The abuse of police authority: A national study of police officers’ attitudes. Washington, DC: The Police Foundation.

79.

Weitzer

Tuch

(2004). Rethinking minority attitudes toward the police. National Criminal Justice Referencing Service. Washington, DC: U.S. Department of Justice.

80.

Wilkins

Williams

(2009). Representing blue. Administration & Society, 40, 775-798.

81.

Wilson

(1989). Bureaucracy: What government agencies do and why they do it. New York, NY: Basic Books.

82.

Worden

(1989). Situational and attitudinal explanations of police behavior: A theoretical reappraisal and empirical assessment. Law & Society Review, 23, 667-711.

83.

Worden

(1995). Police officers’ belief systems. American Journal of Police, 14, 49-81.

84.

Worden

(1996). The causes of police brutality: Theory and evidence on police use of force. In Geller

Toch

(Eds.), Police violence. (pp. 23-51) New Haven, CT: Yale University Press.

85.

Worden

Shepard

R. L.

(1996). Demeanor, crime, and police behavior: A reexamination of the police services study data. Criminology, 34, 83-105.

86.

Yin

(2003). Case study research: Design and methods. Thousand Oaks, CA: SAGE.