Are the Effects of Legitimacy and Its Components Invariant? Operationalization and the Generality of Sunshine and Tyler’s Empowerment Hypothesis

Abstract

Objectives:

To assess whether the relationship between legitimacy and police empowerment is sensitive to the operationalization of legitimacy, and whether the effects of legitimacy and its components on empowerment are invariant. Empowerment is examined in the context of police militarization—public support for the discretionary use of surplus military equipment by law enforcement.

Method:

Using a national sample of 702 American adults and a series of ordinary least squares regressions, the direct and interactive effects of legitimacy and its components on empowerment are examined.

Results:

The composite measure of legitimacy, as well as its individual components, each exert direct effects on police empowerment. Instrumental factors have persisting, albeit weaker, effects relative to normative factors. Interaction terms between the composite legitimacy measure or its components and sociodemographic characteristics were not statistically significant. In short, the effects of legitimacy and its components on empowerment appear invariant.

Conclusions:

Findings provide additional evidence of the generality of Tyler’s process–based model and extend these considerations to the burgeoning literature on public empowerment of police.

Keywords

empowerment obligation to obey police legitimacy police militarization process-based model trust in police

In democratic societies, institutional authorities derive their authority from the public (Locke 1988; Sabine 1937). These authorities must work to achieve and maintain legitimacy among those they serve, as it allows for effective governance. This is especially true for the police (Tyle 2006). Over the past two decades, a sizable body of research has explored the process-based model, or the antecedents and consequences of police legitimacy. This research has focused primarily on the relationship between procedural justice and legitimacy (Mazerolle, Antrobus et al. 2013; Tyler 2017; Wolfe et al. 2016) or two main consequences of legitimacy, compliance with the law, and cooperation with police (Tyler and Jackson 2014; Walters and Bolger 2018). Beyond compliance and cooperation, however, Sunshine and Tyler (2003) identified a third possible consequence of legitimacy: empowerment. When the police are seen as more legitimate, the public is suggested to be more willing to empower them or grant them greater discretion, autonomy, and support in the furtherance of their duties. A growing literature has begun to elaborate on the linkages between legitimacy and empowerment, with preliminary evidence indicating that greater legitimacy increases public willingness to empower the police (see Moule, Burruss, Parry, et al. 2019).

Despite evidence supporting the legitimacy–empowerment link, issues remain in this literature. The current study focuses on two specific issues. First, debates about the appropriate operationalization of legitimacy are ongoing (Huq, Jackson, and Trinkner 2017; Jackson and Bradford 2019; Jackson and Gau 2016; Tankeb 2009), and the operationalization of legitimacy in empowerment research varies considerably. Assessing whether the legitimacy–empowerment relationship is sensitive to differences in the measurement of legitimacy is thus necessary. Second, research finds the effect of procedural justice on the components of legitimacy (trust in police and obligation to obey) to be largely invariant across many sociodemographic characteristics (e.g., age, sex, and race; Wolfe et al. 2016). As Sunshine and Tyler (2003:522, emphasis added) noted, it is “important to consider whether the models being evaluated apply equally well to everyone in a community. In particular, do the members of majority and minority groups consider the same issues when evaluating the police?” Whether the effects of legitimacy or its components have invariant effects on empowerment is unknown. Addressing the issues of operationalization and invariance on the “back end” of the process-based model is necessary for further establishing the generality of the theory (Aiello 2019; Jackson 2018; Wolfe et al. 2016).

The process-based model is a pillar of police–community relations (President’s Task Force on 21st Century Policing 2015). Empowerment gets at the heart of these relations as it recognizes the public’s ability to grant police greater latitude to execute their duties. In the past few years, tensions between the police and the public, particularly among minority community members, have escalated into protests and civil unrest. The current study focuses on one issue that emerged during this time: the acquisition and use of surplus military weapons and equipment by police (Balko 2013; Kraska 2007). This topic is well-suited for examining empowerment, as it involves discretionary police practices which may harm police–community relations and civil liberties (President’s Task Force on 21st Century Policing 2015). Recent research has examined public support for police militarization (e.g., Lockwood, Doyle, and Comiskey 2018) but not whether legitimacy corresponds with greater willingness to empower police for the members of minority and majority groups. Overall, evidence linking legitimacy and empowerment, and of the generality of the process-based model, coupled with Sunshine and Tyler’s explicit suggestion to test the applicability of the theory across social groups, reinforces the importance of examining invariance. Likewise, the potential harm that militarization may do to police–community relations justifies a more in-depth assessment of public support for this behavior.

To that end, the current study assesses whether legitimacy and its components influence public empowerment of police, and whether these effects are invariant across a variety of sociodemographic characteristics. Using a national sample of American adults and ordinary least squares (OLS) regression, we examine whether a composite measure of legitimacy (trust in police and obligation to obey) as well as individual trust in the police and obligation to obey police scales influence empowerment. Next, we examine whether the effects of the composite legitimacy measure and the individual trust in police and obligation to obey police scales are invariant across sociodemographics, as prior research has found perceptions of police differ across specific social characteristics such as race/ethnicity, sex, age, education, contact with police, and victimization experience (e.g., Huebner, Schafer and Bynum 2004; see also Sunshine and Tyler 2003; Wolfe et al. 2016). The overall goal of this article is to extend research on the empowerment hypothesis and the generality of Tyler’s process–based model. We begin by reviewing the process-based model of policing and its consequences for citizen behavior.

The Process-based Model of Policing and Its Consequences for Citizen Behavior

In order to effectively govern, institutional authorities must work to cultivate legitimacy in the eyes of the public. Legitimacy, specifically, is “a property of an authority or institution that leads people to feel that that authority or institution is entitled to be deferred to and obeyed” (Sunshine and Tyler 2003:514; see Weber 1968). Cultivating and maintaining legitimacy is especially important for police, as they rely on voluntary assistance, compliance, cooperation, and support from the public to faithfully execute their duties (e.g., maintaining order and ensuring public safety; Decker 1981; Frank, Smith, and Novak 2005; Huang and Vaughn 1996; Reiss 1971). The process-based model of policing delineates the antecedents of police legitimacy and its consequences for public behavior in relation to police and the law (Tyler 2006; see also Sunshine and Tyler 2003).

Most often, legitimacy is theorized to be built upon a foundation of fair and just treatment of the public by law enforcement (Hinds and Murphy 2007; Lind and Tyler 1988; Mazerolle, Bennett, et al. 2013; Tyler 2006). When police are seen as behaving in a procedurally just manner—such as being respectful and courteous in their interactions with the public, engaging in unbiased decision-making, and allowing citizens to participate in a dialogue during interactions with officers—they foster positive orientations toward the law and their authority (Tyler 2004, 2017; Walters and Bolger 2018). These orientations reflect the legitimacy of the police (Mazerolle, Antrobus et al. 2013; Tyler 2006; Wolfe et al. 2016; Worden and McLean 2017). Higher levels of legitimacy are consequential for a number of citizen behaviors, including complying with the law and cooperating with police (Donner et al. 2015; Jackson et al. 2012; Reisig, Wolfe, and Holtfreter 2011; Tyler 2006).

The behavioral consequences of legitimacy are not limited to compliance with the law or cooperation with authorities. Sunshine and Tyler (2003) also suggested a third consequence of police legitimacy: public willingness to empower police. Recent research has begun examining empowerment in greater depth, although this literature remains underdeveloped relative to the other dimensions and outcomes associated with the process-based model. We discuss the linkages between legitimacy and empowerment, and extant research on the topic, in greater depth in the following section.

Legitimacy and Public Empowerment of Police

Empowerment refers generally to public deference toward authorities and the decisions made or actions undertaken by those authorities (Sunshine and Tyler 2003; Tyler and Mitchell 1994). The “empowerment hypothesis” suggests that as the legitimacy of the police increases, the public is increasingly willing to grant more discretion and autonomy to law enforcement and support police practices, even those that may infringe on civil liberties (Sunshine and Tyler 2003). In contrast, the decisions and actions of law enforcement are subject to increased second-guessing and legal challenges when they are viewed as illegitimate by citizens (Sunshine and Tyler 2003). Studies examining the relationship between the legitimacy of police and public willingness to empower law enforcement, including sample characteristics, focal independent variables, and empowerment-related outcomes, are displayed in Table 1.

Table 1.

Empowerment Studies.

Author(s)	Year	Sample^a	Instrumental and Normative Constructs in Analysis	Empowerment Outcome(s)
Sunshine and Tyler	2003	586 Adult residents of New York City (pre-9/11)	Legitimacy (trust in police + obligation to obey), risk of sanctioning, police effectiveness, distributive fairness, and procedural fairness (procedural justice + quality of treatment + quality of decision-making)	Support for general police discretion
Sunshine and Tyler	2003	1,653 Adult residents of New York City (post-9/11)	Legitimacy (trust in police + obligation to obey), risk of sanctioning, police performance (police effectiveness + neighborhood conditions + fear of victimization), distributive fairness, and procedural fairness (procedural justice + quality of treatment + quality of decision-making)	Support for general police discretion
Gerber and Jackson	2017	206 Americans	Obligation to obey and normative alignment	Support for police use of reasonable and excessive use of force
Lockwood, Doyle, and Comiskey	2018	1,005 American adults	Satisfaction with police, concern about terrorism, concern about drugs and gangs, and harassed by police	Support for police use of military weapons and equipment
Metcalfe and Hodge	2018	411 Jewish Israeli adults in Tel Aviv and Jerusalem	Trust in police, obligation to obey, procedural justice, crime performance, homeland security performance, distributive fairness, distributive equality, and risk of terrorism	Support for police authority and power to handle terrorism
Yesberg and Bradford	2018	12,821 London, UK, residents	Police effectiveness, police fairness, affect toward police, concern about crime, and concern about terrorism	Support for increased arming of London police with firearms
Fox, Moule, and Parry	2018	702 American adults	Legitimacy (trust in police + obligation to obey), legal cynicism, trust in federal government, trust in local police, and fear of crime	Support for police militarization
Moule, Parry, and Fox	2019	702 American adults	Legitimacy (trust in police + obligation to obey)	Support for police use of SWAT
Pryce	2019	304 Adult Ghanian immigrants in Alexandria, VA	Procedural justice (quality of treatment + quality of decision-making), obligation to obey, risk of sanctioning, police effectiveness, and legitimacy of Ghanian police	Support for general police discretion
Moule, Fox, and Parry	2019	702 American adults	Legitimacy (trust in police + obligation of obey) and legal cynicism	Support for police use of surplus military equipment and advanced surveillance tech
Moule, Burruss, Parry, et al.	2019	702 American adults	Legitimacy (trust in police + obligation of obey) and perceived consequences of empowerment	Support for police use of surplus military equipment
Pryce and Grant	2019	163 Jamaican adults	Procedural justice (quality of treatment + quality of decision-making), obligation to obey, risk of sanctioning, and police effectiveness	Support for general police discretion

^aDoes not account for respondents excluded from analyses due to missing data.

Sunshine and Tyler (2003), for example, examined public willingness to give police greater autonomy and discretion in the performance of their duties (e.g., select areas that require greater police presence, conduct “stop and question” with the members of the public, and have the ability to do whatever they feel is necessary to fight crime). Using two samples of New York City residents, Sunshine and Tyler (2003) found that legitimacy (a composite measure of trust in police and obligation to obey) had a moderate, positive effect on empowerment. Other research has demonstrated the breadth of the legitimacy–empowerment link. Among Israelis, Metcalfe and Hodge (2018) found trust in police and perceived obligation to obey independently increased support for allowing the police to fight terrorism. Yesberg and Bradford (2018) found that London residents who more strongly trusted the police were more supportive of arming the police. Together, these studies provide general support for the empowerment hypothesis.

Recent research has also begun to examine public support for the acquisition and use of surplus military weapons and equipment by local law enforcement (B. Fox, Moule, and Parry 2018; Steidley and Ramey 2019). Paralleling the empowerment literature focusing on other discretionary police practices, this growing body of research finds that legitimacy makes the public more supportive of police acquiring surplus military equipment. For example, Lockwood and colleagues (2018) found that individuals who were more satisfied with police were more likely to support police use of military weapons and equipment for a variety of crime-control tasks (e.g., counterterrorism, drug enforcement). Others have shown that higher levels of legitimacy, measured as a composite of trust in police and obligation to obey police, corresponded with more support for allowing police to acquire surplus military weapons, vehicles, and other equipment (Moule, Burruss, Parry, et al. 2019; Moule, Fox, and Parry 2019).

Although this evidence supports a direct link between legitimacy (generally) and empowerment, and identifies mechanisms by which legitimacy can influence empowerment (e.g. evaluations of the costs and benefits associated with police use of surplus military equipment), issues regarding the operationalization of legitimacy and the possible consequences of operationalization for this relationship have not been sufficiently examined. These considerations take on additional significance as research has begun exploring the generality of the process-based model and whether the effects of its key components vary across sociodemographic characteristics. We discuss this in greater depth in the following section.

Operationalization and Invariance of the Legitimacy–Empowerment Relationship

There is a general consensus that legitimacy reflects the right of institutional authorities to exercise power (e.g., Jackson 2018; Tankebe 2009), and a sizable literature demonstrates the influence of legitimacy on an array of citizen behaviors. Despite this consensus and evidence, debates about how to best operationalize legitimacy are ongoing among criminologists (e.g., Bottoms and Tankebe 2012; Jackson 2018; Jackson and Bradford 2019; Wolfe et al. 2016). Two common approaches to the operationalization of legitimacy involve the use of single scales or composite measures. Single scale measures typically focus on either trust in the police (e.g., Nix et al. 2015; Sargeant, Murphy, and Cherney 2014; Tyler 2001) or perceived obligation to obey the police (e.g., Kochel, Parks, and Mastrofski 2013; Pryce 2019; Sun et al. 2017). Composite measures aggregate these two scales together¹ (e.g., Huq, Tyler, and Schulhofer 2011; Reisig, Bratton, and Gertz 2007; Sunshine and Tyler 2003; Tyler 2006; Wolfe 2011).²

Some evidence suggests that differing operationalizations can affect the legitimacy–behavior relationship. For example, Reisig and colleagues (2007) used a composite legitimacy scale which predicted both cooperation and compliance. When the scale was disaggregated into trust and obligation to obey subscales, trust in police predicted compliance and cooperation; obligation to obey was not a significant predictor of either outcome. Others have found obligation to obey to more strongly predict compliance (Tyler 2006). As shown in Table 1, variation in the operationalization of legitimacy exists in the empowerment literature. Paralleling the broader legitimacy literature, this research is split between the use of single scales (e.g., trust in police, Yesberg and Bradford 2018; obligation to obey, Pryce 2019) or composite measures (e.g., Moule, Burruss, Parry, et al. 2019). To our knowledge, no research has examined whether the legitimacy–empowerment relationship is sensitive to these various operationalization strategies. This is important to assess given that the issues of measurement validity can have a profound effect on the body of literature. Varied findings may simply be the result of using composite versus single-scale measures of legitimacy rather than true differences in effects across samples.

Issues of operationalization take on additional significance as research has begun to explore whether the process-based model is a general theory. Wolfe and colleagues (2016) used a sample of 1,681 residents from a southeastern U.S. city to examine whether the effects of procedural justice, distributive justice, and effectiveness were invariant on two components of legitimacy, trust in police and obligation, to obey the police. Controlling for instrumental factors, Wolfe et al. (2016) found procedural justice exerted direct effects on trust in police and obligation to obey. Interaction terms between antecedents of legitimacy and sociodemographic characteristics (age, race, gender, education, prior victimization, police contact, and neighborhood conditions) were then used to assess invariance. There were no significant interactions with procedural justice when predicting obligation to obey, and only one significant interaction when predicting trust. Similarly, only interactions between distributive justice, fear of crime, and neighborhood disorder predicting trust were significant. These results led Wolfe et al. (2016:276) to suggest the effects of competing antecedents on legitimacy are largely invariant.

Might this invariance extend to the “back end” of the process-based model? Some evidence is suggestive of invariance in the legitimacy–empowerment relationship. Sunshine and Tyler (2003:532) found that legitimacy was a significant predictor of a composite measure of compliance, cooperation, and empowerment within various racial/ethnic groups (White, Black, and Hispanic adults, respectively). Owing to this analytic strategy, however, it is unclear whether the effect of legitimacy significantly differed between groups generally, whether this effect might differ specifically for empowerment, or whether alternative operationalizations of legitimacy might exhibit similar effects. Tentative evidence of invariance was shown by Moule, Burruss, Parry, et al. (2019:175), who found the influence of legitimacy on empowerment was not conditioned by political conservatism. Despite this preliminary evidence, additional research is necessary.

Current Study

A sizable literature has explored the consequences of legitimacy, though far less has examined its influence on public empowerment of police. Issues of operationalization pervade the legitimacy literature, including research on empowerment. Against this backdrop, recent research has examined invariance in the “front end” of the process-based model (Wolfe et al. 2016). The current study seeks to further explore the issues of operationalization and invariance on the “back end” of the model, specifically empowerment. To do so, it first asks three questions: (1) Does a composite measure of legitimacy influence empowerment? (2) Does trust in the police influence empowerment? and (3) Does obligation to obey police influence empowerment? If so, we ask three additional questions: (4) Is the effect of a composite legitimacy measure on empowerment invariant? (5) Is the effect of trust in police on empowerment invariant? and (6) Is the effect of obligation to obey police on empowerment invariant? Drawing from prior research on empowerment and preliminary evidence suggestive of invariance, we specify the following hypotheses:

Hypothesis 1: A composite measure of legitimacy (trust in policy and obligation to obey) is positively associated with empowerment.

Hypothesis 2: Trust in police is positively associated with empowerment.

Hypothesis 3: Perceived obligation to obey the police is positively associated with empowerment.

If legitimacy and/or its components exert direct effects on empowerment, we then specify the following hypotheses:

Hypothesis 4: The effect of a composite measure of legitimacy on empowerment will be invariant across sociodemographic characteristics.

Hypothesis 5: The effect of trust in police on empowerment will be invariant across sociodemographic characteristics.

Hypothesis 6: The effect of perceived obligation to obey the police will be invariant across sociodemographic characteristics.

Answering these questions will contribute critical insights into the empowerment hypothesis. We first address whether the legitimacy–empowerment link is sensitive to the operationalization of legitimacy. Second, consistent with the work of Sunshine and Tyler (2003) and the approach of Wolfe et al. (2016), we test whether the effects of legitimacy generalize across age, sex, race/ethnicity, urbanicity, education, contact with police, fear of crime, and victimization status. In the following section, we describe the data and methods used to answer our research questions.

Data and Methods

Data used in the current study consist of a national sample of 702 American adults surveyed about their perceptions of law enforcement, particularly issues relating to police militarization. Data were collected in the Spring of 2017 using Qualtrics’ online survey service.³

Qualtrics works with market research firms and social media sites, among others, to recruit research participants (Qualtrics 2014). The service contains over 13 million diverse users, who are solicited to participate in survey research through multiple methods, and is increasingly being used in social science research (Boas, Christenson, and Glick 2018; Bushman et al. 2012; Wright and Skagerberg 2012). Respondents were selected from Qualtrics’ panel of survey participants using stratified random sampling procedures. Participants were stratified on gender, race, and household income to mirror the composition of American adults from the 2010 U.S. Census.⁴

To determine whether the sample significantly differs from the 2010 Census, Z tests for sample versus population proportions were conducted (VanVoorhis and Morgan 2007). Data from the 2010 U.S. Census and related government sources were used to calculate all population estimates (e.g., American Community Survey 2010a; U.S. Census Bureau, 2010). Z test results suggest the present sample does not significantly differ from the U.S. population in terms of gender, the percentage of White respondents, marital status, veteran status, and most income categories (see Appendix A). Some statistical differences, with regard to the percentages of Black and Hispanic respondents, age distribution, and some income levels, were observed. Racial and ethnic minorities are somewhat overrepresented in the data, relative to the general population. This sample is also slightly more middle-aged when compared to the overall American adult population. Overall, this sample is largely characteristic of the broader U.S. population.

A total of 705 individuals originally completed the instrument; three individuals were removed for failing attention checks embedded in the survey (Oppenheimer, Meyvis, and Davidenko 2009). The remaining 702 participants entered and completed the survey in a satisfactory manner. Responses were required for all questions, resulting in no missing data. Surveys were completed in an average of 18 min, and respondents were compensated above industry standards (US$3 upon completion) to encourage high response and completion rates in the study.

As we detail in the next section, we are able to explore the effects of legitimacy and its components on police empowerment relative to a rich set of competing covariates. Coupled with a demographically and geographically diverse sample, and questions tailored specifically to understand public willingness to allow police to use surplus military equipment, the data are well-suited for addressing invariance in the context of the empowerment hypothesis.

Dependent Variable

Support for police militarization is a 4-item weighted score established through factor analysis to capture support for police use of equipment commonly associated with militarization (e.g., surplus military weapons and equipment, style of dress, or “material militarization”; see Kraska 2007; Lockwood et al. 2018). Respondents were asked whether law enforcement should be able to (a) use surplus military weapons (e.g., assault weapons—AR-15/M4, submachine guns—MP5), (b) use surplus military vehicles (e.g., BearCat armored personnel carrier, mine resistant ambush protected vehicle-MRAP), (c) use surplus military equipment (e.g., computers, tools, generators; see Radil, Dezzani, and McAden 2017), and (d) wear military style uniforms (see Moule, Fox, et al. 2019). Respondents indicated agreement with each statement on a Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree), where higher values indicate more support for police use of surplus military equipment. We note that the phrasing of these questions explicitly focuses on respondent views about whether or not police should be allowed to engage in these behaviors, consistent with the discretionary aspect of the empowerment hypothesis. This dependent variable is a particularly appropriate measure with which to consider the broader conceptual and theoretical questions of empowerment. In contrast to questions assessing support for general police discretion (see Table 1), these items tap into specific behaviors that have been identified to potentially harm police–community relations and infringe on the civil liberties of the populace (President’s Task Force on 21st Century Policing 2015). This measure is also consistent with outcomes examined in prior literature on empowerment (see Table 1). The scale exhibits sound psychometric properties (Cronbach’s α = .87).⁵

Independent Variables

Police legitimacy is measured using 8 items capturing respondents’ normative orientations toward the police, an important foundation from which willingness to empower the law enforcement emerges (Sunshine and Tyler 2003). Measures were adapted from prior research (e.g., Parry, Moule, and Dario 2019; Reisig et al. 2007; Sunshine and Tyler 2003). Respondents were asked the extent to which they agreed with the following items, the first six of which measure trust in police: (a) Most police officers do their job well, (b) I respect the police and their authority, (c) I trust police officers, (d) I am confident in law enforcement, (e) in general, police officers act professionally, and (f) the police can be trusted to make decisions that are best for my community. Two additional items, measuring obligation to obey, are also included in this scale: (g) You should accept police decisions, even if you think they are wrong and (h) You should do what the police tell you even if you disagree. Responses to each item ranged from 1 (strongly disagree) to 4 (strongly agree) and are weighted score created through factor analysis, with higher scores indicating stronger perceptions of police legitimacy. The scale exhibits satisfactory psychometric properties (Cronbach’s α = .91).⁶

Distributive fairness is measured using 3 items adapted from prior research (Reisig et al. 2007; Sunshine and Tyler 2003; Tankebe 2009; see also Sarat 1977). Respondents were asked the extent to which they agreed with the following statements: (a) Police provide the same quality of service to all citizens, (b) police give minorities less help because of their race (reverse coded), and (c) police provide better service to wealthier citizens (reverse coded). Responses to each item ranged from 1 (strongly disagree) to 4 (strongly agree) and are weighted score established through factor analysis, with higher scores indicating a belief that police serve all citizens equally. The scale exhibits acceptable psychometric properties (Cronbach’s α = .80).

Fear of crime is measured using 3 items capturing respondents’ concerns about various forms of crime (e.g., LaGrange and Ferraro 1989; LaGrange, Ferraro, and Supancic 1992; see also Ferraro 1995; Hale 1996). Respondents were asked “In your everyday life, how afraid are you of being the victim of…”: (a) a property crime (e.g., burglary, theft, car break-in)? (b) a violent crime (e.g., assault, rape, robbery, homicide)? and (c) a terrorist attack. Responses for each item ranged from 1 (not at all fearful) to 4 (extremely fearful). The measure is a weighted score established through factor analysis, with higher scores indicating a greater fear of crime. The scale exhibits acceptable psychometric properties (Cronbach’s α = .86).

We also account for demographic and socioeconomic characteristics that might influence willingness to empower the police. Stopped is a dichotomous variable, indicating whether or not respondents had ever experienced a procedurally unjust interaction with law enforcement (1 = yes, 0 = no). Respondents were specifically asked “Have you ever been stopped by the police for no good reason?” (Weitzer and Tuch 2004, 2005). Victimization (= 1, no victimization = 0) is a dichotomous variable indicating whether the respondent had been the victim of a property or violent crime in the three years prior to the survey (Wolfe et al. 2016). Political conservatism is a 2-item scale (Spearman’s ρ = .80), with respondents indicating how fiscally and socially conservative they are. Respondents were asked “How would you describe your political views on (social/fiscal) issues?” Responses for each item ranged from 1 (extremely liberal) to 4 (extremely conservative). Higher scores indicate stronger conservative beliefs. Veteran (= 1, no military service = 0) is a dummy variable indicating whether the respondent reported being a current or former member of the U.S. Armed Forces, including the National Guard or Reserves (White et al. 2012).

We account for respondent age (in years) and whether the respondent is male (= 1, female = 0). With regard to race and ethnicity, we account for whether the respondent is Black (= 1) or Hispanic (= 1, White, non-Hispanic = 0). Respondent residence is measured as a dummy variable, urban (= 1, not urban = 0). Respondent education is measured as a dummy variable indicating whether they possess a college degree (= 1, no college degree = 0). Finally, we account for whether the respondent is currently married (= 1, not married = 0) and whether they are a parent (= 1, no children = 0).

Tests of Measurement Invariance

The degree to which theoretical constructs are invariant between sociodemographic subgroups is important to consider as a foundation for comparative analyses. Measurement invariance is a key concern because scale items may be interpreted differently by individuals from distinct sociodemographic groups, causing systematic differences in item response and biased comparisons. To examine this, we fit nested structural equation models with increasing constraints for our four main constructs (support for police militarization, trust in police, obligation to obey, and legitimacy) and six dichotomous demographic characteristics of interest (sex, race/ethnicity, college graduate, victimization, stopped by police, and urban)—resulting in 24 subgroup comparisons.⁷ All comparisons indicated configural (the same items measure the construct between subgroups) and metric invariance (the factor loadings of the items are equivalent between subgroups). We then tested for scalar invariance, the level of measurement invariance required to justify mean comparisons between subgroups (requiring item intercepts to be equivalent across subgroups, in addition to configural and metric invariance). While 18 of the comparisons were found to be scalar invariant, 6 were not (militarization: White/non-White; trust: White/non-White, stopped/not stopped by police; legitimacy: White/non-White, stopped/not stopped by police, college graduate/not college graduate).

To address this, for each comparison of concern, we began with 2 items which were found to be scalar invariant and added items iteratively until the scale no longer achieved scalar invariance (p < .05; Jaynes and Loughran, 2019). This allowed us to maximize the number of items retained from each original scale and maintain scalar invariance, resulting in a 3-item militarization scale, a 5-item trust scale, and a 5-item legitimacy scale which were scalar invariant across their respective subgroups (p > .05).⁸ Appendix Table B1 indicates scale items which we removed to establish scalar invariance.

The forthcoming comparative analyses were then conducted with the original scales (described in the previous subsection) and the scalar invariant scales. These findings are entirely consistent, suggesting that systematic differences in measurement between subgroups are not responsible for our substantive conclusions (results available upon request). We report findings from the original scales to remain consistent with prior research on empowerment that use these scales. In addition, our iterative approach to establishing scalar invariance could have been influenced by chance characteristics of our sample. It is possible that future scholars may test for scalar invariance with a different sample and identify different problematic items. Therefore, in the interest of replicability, we chose to maintain the original scales.

Analytic Strategy

To examine whether the effects of legitimacy and its components are invariant on public empowerment of the police, our analysis proceeds in four stages. First, summary statistics of all study variables are reported. Second, bivariate correlations between study variables and model diagnostics are examined to ensure collinearity will not bias regression estimates. Third, OLS regressions are used to examine the direct effects of legitimacy, its components, and instrumental constructs on public empowerment of police. Fourth, to assess invariance, a series of OLS regressions incorporating interaction terms between legitimacy or its components and various sociodemographic characteristics are estimated. Here, statistically nonsignificant⁹ interaction terms are indicative of invariance (see Wolfe et al. 2016:266). Together, this strategy provides a holistic and necessary assessment of the factors shaping empowerment and a rigorous test of the generality of Sunshine and Tyler’s (2003) empowerment hypothesis. All regressions are conducted using Stata Version 13.1 (Stata Corp., College Station, TX).

Results

Table 2 contains summary statistics of all study variables. As noted, study scales were computed using weighted factor scores, which are all on a standardized scale where the mean is 0.0 and the standard deviation of the distribution of factors scores is 1.0.

Table 2.

Summary Statistics of all Study Variables.

Variables	Mean/Percentage	SD	Median	Min	Max
Dependent variable
Support for police militarization	0.0	1.0	0.1	−2.0	1.6
Independent variables
Legitimacy	0.0	1.0	−0.4	−3.5	1.5
Trust in police	0.0	1.0	−0.2	−3.5	1.4
Obligation to obey	0.0	1.0	0.1	−2.8	1.8
Distributive fairness	0.0	1.0	−0.1	−2.0	1.8
Fear of crime	0.0	1.0	−0.1	−1.4	2.5
Victimization (1 = yes, 0 = no)	11.5 percent	—		0	1
Stopped for no reason (1 = yes, 0 = no)	16.2 percent	—		0	1
Political conservatism	5.1	1.6	5.0	2.0	8.0
Male (1 = male, 0 = female)	48.7 percent	—		0	1
Black (1 = Black, 0 = non-Black)	16.1 percent	—		0	1
Hispanic (1 = Hispanic, 0 = non-Hispanic)	20.1 percent	—		0	1
Age	47.6	15.2	47.0	18.0	85.0
Urban (1 = yes, 0 = no)	29.1 percent	—		0	1
College graduate (1 = yes, 0 = no)	38.0 percent	—		0	1
Veteran (1 = yes, 0 = no)	11.1 percent	—		0	1
Married (1 = yes, 0 = no)	47.0 percent	—		0	1
Parent (1 = yes, 0 = no)	68.0 percent	—		0	1

With respect to empowerment, there is wide variability in support for police use of surplus military equipment, as the weighted factor scores for the militarization scale ranged from −2.0 to 1.6. This variability allows for the modeling of covariates of empowerment.

Among key independent variables, respondents exhibit fairly high levels of legitimacy (range: −3.5 to 1.5), though much of this is driven by high levels of trust in police (range: −3.5 to 1.4) rather than obligation to obey (range: −2.8 to 1.8). Respondents also believe the police to be somewhat equitable in the distribution of their services, on average (range: −2.0 to 1.8) and are nominally fearful of crime (range: −1.4 to 2.5).

Just over 11 percent of the sample report being the victim of a property or violent crime in the past three years, and approximately one-sixth of the sample reports having ever been stopped by the police for no reason. Nearly two-thirds of the sample is White, with the remainder split between Black (16 percent) and Hispanic (20 percent) respondents. Roughly 49 percent of the sample is male. The average respondent age is approximately 48, but age ranged from 18 to 85 years old. Most respondents live in nonurban (suburban or rural) areas, while 29 percent live in urban locations. Approximately half of the sample is not currently married (52 percent), and two-thirds of respondents (67 percent) have children. Roughly one-third of respondents (38 percent) possess a college degree, and 11 percent identify as current or former members of the U.S. armed forces. Overall, the sample is demographically, geographically, and culturally diverse—ideally suited for examining issues relating to invariance and empowerment of police.

To further contextualize this sample and examine differential perceptions and support for police legitimacy and empowerment, a series of t tests were conducted (results are provided in Appendix Table C1). Results indicate that several mean differences in perceptions of legitimacy, its subcomponents, and police empowerment exist among key focal group subcategories. Non-White participants and those stopped by police for no reason showed significantly lower mean levels across all measures of legitimacy and empowerment compared to Whites and those not stopped by police, respectively (p < .001). Similarly, those who lived in urban areas (relative to those who reside in nonurban areas) had significantly lower levels on trust in police (p < .01), police legitimacy (p < .01), and empowerment (p < .001), while college graduates showed significantly lower on obligation to obey compared to noncollege graduates (p < .05). In addition, relative to females, males showed significantly more support for police militarization (p < .01) but were otherwise statistically indistinguishable (p >.05) from females in their mean level of legitimacy and its subcomponents.

Table 3 presents bivariate correlations between all study variables. Consistent with theoretical expectations, the composite legitimacy measure (r = .52, p < .05), as well as trust in police (r = .51, p < .05) and obligation to obey (r = .37, p < .05), are all positively correlated with support for police use of surplus military equipment. Both instrumental factors, distributive fairness (r = .42, p < .05) and fear of crime (r = .10, p < .05), are also positively correlated with support for police use of surplus military equipment.

Table 3.

Bivariate Correlations between Study Variables.

	Variables	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17
(1)	Support for police militarization	—
(2)	Legitimacy	.52*	—
(3)	Trust in police	.51*	.98*	—
(4)	Obligation to obey	.37*	.68*	.53*	—
(5)	Distributive fairness	.42*	.56*	.54*	.43*	—
(6)	Fear of crime	.10*	−.02	−.04	.05	−.03	—
(7)	Victimization	−.05	−.03	−.04	−.01	−.05	.08*	—
(8)	Stopped for no reason	−.13*	−.23*	−.23*	−.15*	−.19*	.00	.13*	—
(9)	Political conservatism	.37*	.28*	.26*	.24*	.38*	−.05	−.06	−.11*	—
(10)	Male	.13*	−.02	−.04	.02	−.05	−.11*	−.01	.17*	.09*	—
(11)	Black	−.19*	−.27*	−.27*	−.17*	−.28*	.04	−.06	.07	−.16*	−.08*	—
(12)	Hispanic	−.03	−.03	−.03	−.04	−.10	.11*	.03	.03	.00	.17*	−.22*	—
(13)	Age	.18*	.20	.19*	.15*	.14*	−.17*	−.05	−.01	.15*	−.06	−.11*	−.28*	—
(14)	Urban	−.15*	−.12	−.12*	−.07	−.19*	.13*	.08*	.05	−.22*	.08*	.16*	.16*	−.15*	—
(15)	College graduate	−.03	−.03	−.01	−.08*	−.08	−.03	.02	.04	−.11*	.18*	−.04	.03	−.03	.02	—
(16)	Veteran	.13*	.08*	.07	.07	.04	−.07	−.01	.04	.06	.23*	−.06	.02	.17*	−.09*	.08*	—
(17)	Married	.13*	.13*	.14*	.03	.10*	−.05	−.01	−.02	.09	.10*	−.15*	−.00	.11*	−.09*	.18*	.05	—
(18)	Parent	.07	.11*	.10*	.09*	.10*	.01	−.00	−.04	.07	.00	−.02	−.01	.27*	−.03	−.00	.08*	.38*

Note: N = 702.

*p < .05.

Bivariate correlations between independent variables are fairly high, particularly between legitimacy and distributive fairness (r = .56), trust in police and obligation to obey (r = .53), and trust in police and distributive fairness (r = .54). Although these correlations do not exceed the common threshold for multicollinearity (.70, see Hanushek and Jackson 1977; Licht 1995), it is possible these high correlations may still produce multicollinearity and bias regression estimates (e.g., Farrar and Glauber 1967). Additional model diagnostics were thus conducted to determine whether collinearity would bias regression estimates. Variance inflation factors (VIFs) and condition indices did not exceed problematic thresholds (VIFs < 2.0, condition indices < 16.0; Belsley, Kuh, and Welsch 1980; J. Fox 1991). These diagnostics suggest that regression estimates should not be biased by collinearity. Having established preliminary empirical relationships between theoretical variables of interest, we proceed to regression analyses.

Table 4 presents a series of OLS regressions, regressing empowerment on normative and instrumental constructs, and sociodemographic characteristics. Model 1 in Table 4 demonstrates, consistent with theoretical expectations and prior research, that the composite measure of legitimacy is a robust predictor of empowerment (β = .38, p < .001). Further, legitimacy exhibits the strongest standardized effect among independent variables on this support. Instrumental constructs also promote support for police use of surplus military equipment. Both measures of instrumental concerns, distributive fairness (β = .12, p < .01) and fear of crime (β = .17, p < .001), are positively associated with empowerment but have weaker standardized effects than legitimacy.

Table 4.

Ordinary Least Squares Regressions Examining Direct Effects of Legitimacy, Its Components, and Instrumental Constructs on Support for Police Militarization.

Variables	Model 1			Model 2			Model 3
Variables	b	SE	β	b	SE	β	b	SE	β
Legitimacy	.38***	.04	.38	—	—	—	—	—	—
Trust in police	—	—	—	.36***	.04	.36	—	—	—
Obligation to obey	—	—	—	—	—	—	.18***	.04	.18
Distributive fairness	.12**	.04	.12	.13**	.04	.13	.23***	.04	.23
Fear of crime	.17***	.03	.17	.18***	.03	.18	.16***	.03	.16
Victimization	−.06	.10	−.02	−.05	.10	−.02	−.07	.10	−.02
Stopped for no reason	−.06	.08	−.02	−.06	.08	−.02	−.16	.09	−.06
Political conservatism	.11***	.02	.18	.12***	.02	.18	.12***	.02	.19
Male	.28***	.06	.14	.29***	.07	.15	.26***	.07	.13
Black	.00	.09	.00	.00	.09	.00	−.08	.10	−.03
Hispanic	−.04	.08	−.02	−.04	.08	−.02	−.02	.09	−.01
Age	.00	.00	.08	.00*	.00	.08	.01**	.00	.10
Urban	−.12	.07	−.05	−.12	.07	−.05	−.12	.07	−.05
College graduate	−.04	.06	−.02	−.05	.06	−.03	−.02	.07	−.01
Veteran	.16	.10	.05	.17	.10	.05	.19	.10	.06
Married	.10	.07	.05	.09	.07	.04	.14*	.07	.07
Parent	−.08	.07	−.04	−.08	.07	−.04	−.09	.08	−.04
Constant	−.88	.16	—	−.91	.16	—	−.10	.17	—

Note: N = 702. Model 1, F: 28.81***, R ²: .38; model 2, F: 28.30***, R ²: .38; model 3, F: 21.55***, R ²: .32. When variables are standardized factor scores, (e.g., legitimacy) b and β coefficients are equivalent.

*p < .05. **p < .01. ***p < .001.

Model 2 replaces the legitimacy scale with only the trust in police subscale. Consistent with model 1, trust in police is significantly, positively associated with empowerment (β = .36, p < .001). In this model, distributive fairness and fear of crime remained significant.

Model 3 replaces the legitimacy scale with the obligation to obey subscale. Obligation to obey is a significant but weaker predictor of empowerment (β = .18, p < .001). Consistent with previous models, both distributive fairness and fear of crime were positively associated with support for police use of surplus military equipment. Notably, the standardized effect of obligation to obey is lower or just slightly above that of these instrumental concerns.

There are two substantive takeaways from the results shown in Table 4. First, legitimacy and its components are each significant, robust predictors of empowerment of police. Second, although these constructs exert are influential for empowerment, much of the effect of legitimacy appears to be driven by trust in police. This is demonstrated by the lower R ² value of model 3, which omits the trust in police measure. Overall, these set of findings allow us to answer our first three research questions affirmatively. Having done so, we next examine whether the effects of legitimacy and its components on empowerment are invariant.

Table 5 contains a number of OLS regressions examining the interactive effects between legitimacy or its components and various sociodemographic characteristics. For ease of presentation, we present only the main effect variable of interest and the interaction terms from each set of regressions; models also contain each of the variables displayed in Table 4.

Table 5 contains interactions between legitimacy and its components, gender (model 1) and race/ethnicity (models 2 and 3).¹⁰ As each set of models demonstrates, none of these interactions are statistically significant. This indicates that the effects of legitimacy and its components on empowerment do not differ by sex, race, or ethnicity.

Furthermore, Table 5 contains interactions between legitimacy and its components, age (model 4), urbanicity (model 5), and possessing a college degree (model 6). As before, none of the interaction terms are statistically significant. Thus, the influence of legitimacy or its components on empowerment is consistent across these individual characteristics. Finally, Table 5 contains interactions between legitimacy and its components, having been stopped by the police for no reason (model 7), being the victim of a crime (model 8), and fear of crime (model 9). No interactions involving being stopped or being fearful of crime are statistically significant. The interaction term between trust in the police and being the victim of a crime was initially marginally significant (p < .10); however, after a Bonferroni post hoc α correction was applied (Benjamini and Hochberg 1995), this interaction was no longer statistically significant.¹¹

Table 5.

Ordinary Least Squares Regressions Examining the Effect of Legitimacy and Its Components on Empowerment across Sex, Race/Ethnicity, Age, Urban, College Graduate, Being Stopped by Police, Victimization, and Fear of Crime.

Interaction Variables	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7		Model 8		Model 9
	(Male)		(Black)		(Hispanic)		(Age)		(Urban)		(College Graduate)		(Stopped)		(Victim)		(Fear of Crime)
	b	SE	b	SE	b	SE	b	SE	b	SE	b	SE	b	SE	b	SE	b	SE
Legitimacy × (Variable)	.03	.06	−.01	.08	.05	.08	−.00	.00	−.07	.07	−.04	.06	−.05	.07	−.15	.10	.03	.03
Legitimacy	.36***	.05	.38***	.04	.37***	.04	.38***	.03	.40***	.04	.40***	.04	.39***	.04	.39***	.04	.38***	.04
Trust in Police × (Variable)	.03	.06	−.02	.08	.06	.08	−.00	.00	−.08	.07	−.03	.06	−.03	.07	−.18†	.09	.04	.03
Trust in police	.35***	.05	.37***	.04	.35***	.04	.36***	.04	.39***	.04	.38***	.04	.37***	.04	.38***	.04	.36***	.04
Obligation to Obey × (Variable)	.05	.06	−.01	.08	−−.04	.08	−.00	.01	−.06	.07	−.01	.07	−.05	.08	−.05	.10	−.00	.03
Obligation to obey	.16**	.05	.18***	.04	.19***	.04	.18***	.08	.20***	.04	.18***	.04	.19***	.04	.19***	.04	.18***	.04

Note: Each model contains all other variables presented in Table 4 (e.g., distributive fairness through parent). These other variables have been omitted for space. Legitimacy × Male, F: 27.00***, R ²: .39; Legitimacy × Black, F: 26.97***, R ²: .39; Legitimacy × Hispanic, F: 27.01***, R ²: .39; Legitimacy × Age, F: 26.98***, R ²: .39; Legitimacy × Urban, F: 27.10*** R ²: .39; Legitimacy × College Graduate, F: 27.02***, R ²: .39; Legitimacy × Stopped, F: 27.01***, R ²: .39; Legitimacy × Victimization, F: 27.23***, R ²: .39; Legitimacy × Fear of Crime, F: 27.09***, R ²: .39; Trust in Police × Male, F: 26.51***, R ²: .38; Trust in Police × Black, F: 26.50***, R ²: .38; Trust in Police × Hispanic, F: 26.56***, R ²: .38; Trust in Police × Age, F: 26.52***, R ²: .38; Trust in Police × Urban, F: 26.65***, R ²: .38; Trust in Police × College Graduate, F: 26.52***, R ²: .38; Trust in Police × Stopped, F: 26.51***, R ²: .38; Trust in Police × Victimization, F: 26.84***, R ²: .39; Trust in Police × Fear of Crime, F: 26.64***, R ²: .38; Obligation to Obey × Male, F: 20.23***, R ²: .32; Obligation to Obey × Black, F: 20.18***, R ²: .32; Obligation to Obey × Hispanic, F: 20.20***, R ²: .32; Obligation to Obey × Age, F: 20.21***, R ²: .32; Obligation to Obey × Urban, F: 20.24***, R ²: .32; Obligation to Obey × College Graduate, F: 20.18***, R ²: .32; Obligation to Obey × Stopped, F: 20.20***, R ²: .32; Obligation to Obey × Victimization, F: 20.20***, R ²: .32; Obligation to Obey × Fear of Crime, F: 20.17***, R ²: .32.

†p < .10. *p < .05. **p < .01. ***p < .001.

These set of findings allow us to answer our final three research questions in the affirmative; it appears that the influence of legitimacy and its components is largely invariant with respect to empowerment. Across a number of sociodemographic characteristics, only one possible interaction was found to be significant, and there is reason to suspect this finding was a statistical artifact. We reserve further discussion of these findings for the following section.

Discussion

For two decades, a sizable body of research has examined dimensions of the process-based model of policing (Mazerolle, Antrobus et al. 2013; Tyler 2006, 2017). Much of this research has concentrated on the relationship between procedural justice and legitimacy (e.g., Tyler 2017; Wolfe et al. 2016) or the influence of legitimacy on citizen compliance and cooperation with the police (e.g. Walters and Bolger 2018). A third consequence of legitimacy, the empowerment of police, has received comparatively little attention (Pryce 2019; Sunshine and Tyler 2003). Growing interest in empowerment has coincided with debates about the appropriate operationalization of legitimacy (see Jackson 2018, for a review) and the generality of the process-based model (e.g. Wolfe et al. 2016). The current study sought to better understand the influence of legitimacy and its components (trust in police and obligation to obey) on empowerment and whether the effects of these characteristics are invariant across social groups (Sunshine and Tyler 2003). This is particularly relevant given known variation in the perceptions of police across these groups (Huebner et al. 2004) and recent tensions between the police and the public, especially within minority communities (President’s Task Force on 21st Century Policing 2015). To do so, it used a national sample of American adults and a series of OLS regressions. Our results warrant three broader points of discussion.

First, legitimacy and its components exerted direct effects on empowerment, net of a variety of demographic features, instrumental concerns, and experiences with the police (see Table 4). Further, the standardized effects of both a composite measure of legitimacy (β = .38) and trust in police (β = .36) were approximately twice as powerful as the next most influential covariates (see Table 4). Obligation to obey had weaker standardized effects but was still consistently, positively associated with empowerment. These findings comport with prior research on the legitimacy–empowerment link and provide additional evidence of the role that legitimacy and its components may play in fostering empowerment (e.g., Pryce 2019; Sunshine and Tyler 2003). Future research should examine whether alternative measures and approaches to legitimacy influence empowerment. For example, some researchers have considered legitimacy as involving “normative alignment,” or the extent to which police are seen as sharing the values of the public (e.g., Huq et al. 2017; more generally, see Jackson 2018).

In addition to the role that normative orientations play in fostering empowerment, instrumental concerns are relevant for public empowerment of police. Both distributive fairness and fear of crime were consistently, positively associated with empowerment, regardless of how legitimacy was operationalized. The standardized effects of these instrumental concerns are approximately one-half to one-third of either the composite legitimacy measure or trust in the police, though were stronger in the models only containing obligation to obey. Despite this, the persisting effects of these concerns suggest a continuing need to understand how instrumental considerations shape empowerment. Future research would do well to consider how alternative instrumental considerations, such as the perceptions of police effectiveness (e.g., Tankebe 2009), considerations of civil liberties (Davis and Silver 2004; Moule, Burruss, Gifford, et al. 2019), or fear in regard to other dimensions of social life (e.g., social change; Blalock 1967; Wang 2012; see also Glassner 1999), might be relevant for public willingness to empower law enforcement.

Second, although there were mean differences in legitimacy and its subcomponents across sample subgroups (most notably, differences were found by race/ethnicity, urban, and stopped for no reason), the effects of legitimacy (and its subcomponents) on support for police militarization were found to be similar across age, race, ethnicity, gender, education, urbanicity, prior experience with the police, and fear of crime (see Table 5). This finding was robust whether the composite measure of legitimacy or individual subscales of trust or obligation to obey was used. Thus, the legitimacy–empowerment relationship does not appear sensitive to differing operationalizations of legitimacy. These findings extend recent work by Wolfe and colleagues (2016) and provide additional evidence of the generality of the process-based model, at least with regard to one of the consequences of legitimacy. Future research should examine whether the legitimacy–empowerment relationship is sensitive to other operationalizations of legitimacy. For example, some argue legitimacy is a latent construct comprised of procedural fairness, distributive justice, lawfulness, and effectiveness (Tankebe 2013). Other research notes the importance of bounded authority, where police are viewed as respecting legal limits on their powers (Trinkner, Jackson, and Tyler 2018). Public perceptions of bounded authority shape police legitimacy and influence compliance with the law. We anticipate that bounded authority would strongly influence public willingness to empower police. More generally, future research would do well to focus on the possible invariant effects of legitimacy on cooperation with police and compliance with the law and the other dimensions of the process-based model (Aiello 2019; Sunshine and Tyler 2003; Tyler 2006; Zahnow, Mazerolle, and Pang 2019).

Third, the findings from the current study have important implications for policy. Accumulating evidence suggests police can enhance their legitimacy through treating citizens fairly, demonstrating effective crime control practices, distributing resources equitably, and respecting legal constraints on their authority (e.g., Huq et al. 2017; Tyler 2006; Worden and McLean 2017). Pursuing each avenue for the cultivation of legitimacy (e.g., procedural justice, effectiveness, lawfulness, and distributive fairness) should be a priority for police agencies.

Findings from the current study reaffirm the need for police to cultivate legitimacy among the members of the public, insofar as it appears this will make the public more deferential to police decision-making and supportive of police discretionary practices. That the positive effects of legitimacy and its components appear invariant across a number of sociodemographic characteristics further reinforce the importance of doing so.

Of course, the current study is not without limitations. First, the data are cross-sectional. This is common in the empowerment literature, and longitudinal work should be a priority for future studies on empowerment. Second, we accounted for a number of factors which influence empowerment, though additional factors may also be relevant. Wolfe et al. (2016) found perceptions of neighborhood conditions matter for perceptions of police legitimacy; so too might they influence empowerment. Similarly, there is scant research on how contextual factors contribute to empowerment. Exploring how these characteristics influence empowerment would be worthwhile. Third, we focused only on support for police use of surplus military equipment. Assessing alternative measures of empowerment, such as general discretion, deference, and autonomy, is a necessary next step for this literature (e.g., Pryce 2019; Sunshine and Tyler 2003). These limitations notwithstanding, this study reflects an early attempt to explore the generality of the legitimacy–empowerment relationship and the invariance of this relationship.

Concerns about online sampling and generalizability must also be addressed. Online samples offer a timely and cost-effective method of conducting social science research. Although traditional convenience samples from online platforms may underrepresent minority individuals, one of the best ways to combat this, as was done here, is the use of survey panels where sample characteristics are specified by researchers. Compared to other online survey platforms, Qualtrics is best able to provide samples which closely resemble the actual population (Boas et al. 2018; Heen et al., 2014). Comparisons between 2010 Census information and sample data (see Appendix A) suggest the data are fairly representative. A second concern of online surveys is that inattentive respondents might hurry through the survey, resulting in invalid responses. Instructional manipulation checks were included in the survey to identify suspect responses and exclude the few respondents (n = 3) who failed any checks. Taken together, we are confident in the validity of our data and findings; nonetheless, our results should be interpreted with care until replicated using alternative samples and methods.

In the end, understanding the front and back ends of the process-based model have important implications for the relationship between the public and the police (Tyler 2006; President’s Task Force on 21st Century Policing 2015). The outcomes that researchers have focused on largely involve compliance and cooperation. A third process, empowerment, continues to require greater attention. The current study focused on the linkages between legitimacy, its components (trust and obligation to obey), and empowerment. Results indicated that the effects of legitimacy and its components on empowerment are largely invariant, at least in relation to public support for police use of surplus military equipment. We encourage social scientists to continue exploring the generality of the process-based model and its consequences for police–community relations.

Footnotes

Appendix A

Appendix B

Table B1.

Items Included in Study Scales and Results of Tests of Scalar Invariance.

Support for Police Militarization (Cronbach’s α = .87)

Law enforcement should be able to use surplus military weapons (e.g., assault weapons—AR-15/M4 and submachine guns—MP5)

Law enforcement should be able to use surplus military vehicles (e.g., BearCat armored personnel carrier and mine resistant ambush-protected vehicle)

Law enforcement should be able to use surplus military equipment (e.g., computers, tools, generators)^a

Law enforcement should be able to wear military-style uniforms

Trust in police (Cronbach’s α = .93)

Most officers do their job well

I respect police and their authority

I trust police officers^a

I am confident in law enforcement

Police officers act professionally

Police officers can be trusted to make decisions that are best for my community

Obligation to obey (Cronbach’s α = .70)

You should accept police decisions, even if you think they are wrong

You should do what the police tell you, even if you disagree

Legitimacy (Cronbach’s α = .91)

Most officers do their job well

I respect police and their authority

I trust police officers^a

I am confident in law enforcement

Police officers act professionally^a

Police officers can be trusted to make decisions that are best for my community^a

You should accept police decisions, even if you think they are wrong

You should do what the police tell you, even if you disagree

^aItems that were not included in the scalar invariant scales due to lack of scalar invariance.

Appendix C

Table C1.

Baseline Mean Comparisons on Perceptions of Legitimacy and Empowerment within Focal Groups.

Variables	Legitimacy	Trust in Police	Obligation to Obey	Support for Police Militarization
Female	0.03	0.03	−0.02	−0.12
Male	−0.03	−0.04	0.02	0.12
t =	−0.73	−0.93	0.47	3.25**
Non-White	−0.32	−0.30	−0.22	−0.22
White	0.18	0.17	0.13	0.13
t =	6.50***	6.25***	4.53***	4.54***
Not college graduate	0.02	0.01	0.06	0.03
College graduate	−0.04	−0.02	−0.10	−0.04
t =	−0.85	−0.34	−2.16*	−0.88
Not victimized	0.01	0.01	0.00	0.02
Victimized	−0.10	−0.11	−0.03	−0.14
t =	−0.92	−1.02	−0.25	−1.33
Not stopped for no reason	0.10	0.10	0.06	0.06
Stopped for no reason	−0.52	−0.52	−0.34	−0.30
t =	−6.29***	−6.29***	−3.97***	−3.53***
Nonurban	0.08	0.08	0.04	0.10
Urban	−0.19	−0.19	−0.10	−0.24
t =	−3.33**	−3.24**	−1.77	−4.14***

Note: N = 702. Mean scores on weighted factor scores for each construct by focal group conducted using independent sample t tests.

*p < .05. **p < .01. ***p < .001.

Acknowledgment

The authors thank Editor Chris Sullivan and three anonymous reviewers for their valuable comments, which helped to improve this article.

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(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Bryanna Fox

Richard K. Moule, Jr.

Chae M. Jaynes

Notes

References

Aiello

Michael F

. 2019. “Legitimacy Invariance and Campus Crime: The Impact of Campus Police Legitimacy in Different Reporting Contexts.” Police Practice and Research. doi: 10.1080/15614263.2019.1570849.

American Community Survey. 2010a. “Marital Status, 2008-2010 American Community Survey 3-Year Estimates.” Retrieved April 17, 2018 (https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_10_3YR_S1201&prodType=table).

American Community Survey. 2010b. “Veteran Status, 2008-2010 American Community Survey 3-Year Estimates.” Retrieved April 17, 2018 (https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_10_3YR_S2101&prodType=table).

Balko

Radley

. 2013. Rise of the Warrior Cop: The Militarization of America’s Police Forces. New York: Public Affairs.

Belsley

David A.

Kuh

Edwin

Welsch

Roy

. 1980. Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. New York: John Wiley & Sons.

Benjamini

Yoav

Hochberg

Yosef

. 1995. “Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing.” Journal of the Royal Statistical Society 57:289–300.

Blalock

Hubert M.

1967. Toward a Theory of Minority-group Relations. New York: John Wiley & Sons.

Boas

Taylor C.

Christenson

Dino P.

Glick

David M.

. 2018. “Recruiting Large Online Samples in the United States and India: Facebook, Mechanical Turk, and Qualtrics.” Political Science Research and Methods. doi: 10.1017/psrm.2018.28.

Bottoms

Anthony

Tankebe

Justice

. 2012. “Beyond Procedural Justice: A Dialogic Approach to Legitimacy in Criminal Justice.” Journal of Criminal Law & Criminology 102:119–70.

10.

Bushman

Brad J.

Moeller

Scott J.

Konrath

Sara

Crocker

Jennifer

. 2012. “Investigating the Link between Liking versus Wanting Self-esteem and Depression in a Nationally Representative Sample of American Adults.” Journal of Personality 80:1453–69.

11.

Cao

Liqun

Graham

Amanda

. 2019. “The Measurement of Legitimacy: A Rush to Judgement?” Asian Journal of Criminology 14: 291–99.

12.

Davis

Darren W.

Silver

Brian D.

. 2004. “Civil Liberties vs. Security: Public Opinion in the Context of the Terrorist Attacks on America.” American Journal of Political Science 48:28–46.

13.

Decker

Scott H

. 1981. “Citizen Attitudes toward the Police: A Review of Past Findings and Suggestions for Future Policy.” Journal of Police Science and Administration 9:80–87.

14.

Donner

Chirstopher

Maskaly

Jon

Fridell

Lorie

Jennings

Wesley G.

. 2015. “Policing and Procedural Justice: A State-of-the-art Review.” Policing: An International Journal of Police Strategies & Management 38:153–72.

15.

Farrar

Donald E.

Glauber

Robert R.

. 1967. “Multicollinearity in Regression Analysis: The Problem Revisited.” Review of Economics and Statistics 49:92–107.

16.

Ferraro

Kenneth F.

1995. Fear of Crime: Interpreting Victimization Risk. Albany: State University of New York Press.

17.

Fox

Bryanna

Moule

Richard K.

Jr. Parry

Megan M.

. 2018. “Categorically Complex: A Latent Class Analysis of Public Perceptions of Police Militarization.” Journal of Criminal Justice 58:33–46.

18.

Fox

John

. 1991. Regression Diagnostics: An Introduction. Newbury Park, CA: Sage.

19.

Frank

James

Smith

Brad W.

Novak

Kenneth J.

. 2005. “Exploring the Basis of Citizens’ Attitudes toward the Police.” Police Quarterly 8:206–28.

20.

Glassner

Barry

. 1999. The Culture of Fear: Why Americans Are Afraid of the Wrong Things. New York: Basic Books.

21.

Hale

Chris

. 1996. “Fear of Crime: A Review of the Literature.” International Review of Victimology 4:79–150.

22.

Hanushek

Eric A.

Jackson

John E.

. 1977. Statistical Methods for Social Scientists. San Diego, CA: Academic Press.

23.

Heen

M. S. J.

Lieberman

J. D.

Miethe

T. D.

(2014). “A Comparison of Different Onlinesampling Approaches for Generating National Samples.” Center for Crime and Justice Policy. Las Vegas, NV: University of Nevada, Las Vegas. Retrieved April 17, 2018 (http://www.unlv.edu/sites/default/files/page_files/27/ComparisonDifferentOnlineSampling.pdf).

24.

Hinds

Lyn

Murphy

Kristina

. 2007. “Public Satisfaction with Police: Using Procedural Justice to Improve Police Legitimacy.” Australian & New Zealand Journal of Criminology 40:27–42.

25.

Huang

Wilson

Vaughn

Michael S.

. 1996. “Support and Confidence: Public Attitudes toward the Police.” Pp. 31–45 in Americans View Crime and Justice: A National Public Opinion Survey, edited by Flanagan

Timothy J.

Longmire

Dennis R.

. Thousand Oaks, CA: Sage.

26.

Huebner

B.M.

Schafer

J.A.

Bynum

T.S

. 2004. “African American and White Perceptions of Police Services: Within- and Between-group Variation.” Journal of Criminal Justice 32:123–35.

27.

Huq

Aziz Z.

Jackson

Jonathan

Trinkner

Rick

. 2017. “Legitimating Practices: Revisiting the Predicates of Police Legitimacy.” British Journal of Criminology 57:1101–22.

28.

Huq

Aziz Z.

Tyler

Tom R.

Schulhofer

Stephen J.

. 2011. “Why Does the Public Cooperate with Law Enforcement? The Influence of the Purposes and Targets of Policing.” Psychology, Public Policy, and Law 17:419–50.

29.

Jackson

Jonathan

. 2018. “Norms, Normativity, and the Legitimacy of Justice Institutions: International Perspectives.” Annual Review of Law and Social Science 14:145–65

30.

Jackson

Jonathan

Bradford

Ben

. 2019. “Blurring the Distinction between Empirical and Normative Legitimacy? A Commentary on ‘Police Legitimacy and Citizen Cooperation in China’.” Asian Journal of Criminology 14:265–89.

31.

Jackson

Jonathan

Bradford

Ben

Hough

Mike

Myhill

Andy

Quinton

Paul

Tyler

Tom R.

. 2012. “Why Do People Comply with the Law? Legitimacy and the Influence of Legal Institutions.” British Journal of Criminology 52:1051–71

32.

Jackson

Jonathan

Gau

Jacinta M.

. 2016. “Carving Up Concepts? Differentiating between Trust and Legitimacy in Public Attitudes toward Legal Authorities.” Pp. 49–70 in Interdisciplinary Perspectives on Trust: Towards Theoretical and Methodological Integration, edited by Shockley

Neal

T. M. S.

Pytlik Zillig

Bornstein

. New York: Springer.

33.

Jaynes

C. M.

Loughran

T. A.

(2019). “Social Preferences as an Individual Difference in Offender Decision-Making.” Journal of Research in Crime and Delinquency 56(1):129–169.

34.

Kochel

Tammy Rinehart

Parks

Roger

Mastrofski

Stephen D.

. 2013. “Examining Police Effectiveness as a Precursor to Legitimacy and Cooperation with Police.” Justice Quarterly 30:895–925.

35.

Kraska

Peter B.

2007. “Militarization and Policing—Its Relevance to 21st Century Police.” Policing: A Journal of Policy and Practice 1:501–13.

36.

LaGrange

Randy L.

Ferraro

Kenneth F.

. 1989. “Assessing Age and Gender Differences in Perceived Risk and Fear of Crime.” Criminology 27:697–720.

37.

LaGrange

Randy L.

Ferraro

Kenneth F.

Supancic

Michael

. 1992. “Perceived Risk and Fear of Crime: Role of Social and Physical Incivilities.” Journal of Research in Crime and Delinquency 29:311–34

38.

Licht

Mark H.

1995. “Multiple Regression and Correlation.” Pp. 19–64 in Reading and Understanding Multivariate Statistics, edited by Grimm

L. G.

Yarnold

P. R.

. Washington, DC: American Psychological Association.

39.

Lind

E. Allan

Tyler

Tom R.

. 1988. The Social Psychology of Procedural Justice. New York: Plenum.

40.

Locke

John

. 1988. Two Treatises of Government. Cambridge: Cambridge University Press.

41.

Lockwood

Brian

Doyle

Matthew D.

Comiskey

John G.

. 2018. “Armed, but too Dangerous? Factors Associated with Citizen Support for the Militarization of the Police.” Criminal Justice Studies 31:113–27.

42.

Mazerolle

Lorraine

Antrobus

Emma

Bennett

Sarah

Tyler

Tom R.

. 2013. “Shaping Citizen Perceptions of Police Legitimacy: A Randomized Field Trial of Procedural Justice.” Criminology 51:33–63.

43.

Mazerolle

Lorraine

Bennett

Sarah

Davis

Jacqueline

Sargeant

Elise

Manning

Matthew

. 2013. “Procedural Justice and Police Legitimacy: A Systematic Review of the Research Evidence.” Journal of Experimental Criminology 9:245–74.

44.

Metcalfe

Christi

Hodge

Olivia

. 2018. Empowering the Police to Fight Terrorism in Israel. Criminology and Criminal Justice 18:585–603.

45.

Moule

Richard K.

Jr. Burruss

George W.

Parry

Megan M.

Fox

Bryanna

. 2019. “Assessing the Direct and Indirect Effects of Legitimacy on Public Empowerment of Police: A Study of Public Support for Police Militarization in America.” Law and Society Review 53:77–107.

46.

Moule

Richard K.

Jr. Burruss

George W.

Gifford

Faith E.

Parry

Megan M.

Fox

Bryanna

. 2019. “Legal Socialization and Subcultural Norms: Examining Linkages between Perceptions of Procedural Justice, Legal Cynicism, and the Code of the Street.” Journal of Criminal Justice 61:26–39.

47.

Moule

Richard K., Jr.

Fox

Bryanna

Parry

Megan M.

. 2019. “The Long Shadow of Ferguson: Legitimacy, Legal Cynicism, and Public Perceptions of Police Militarization.” Crime and Delinquency 65:151–82.

48.

Moule

Richard K.

Jr. Parry

Megan M.

Fox

Bryanna

. 2019. “Public Support for Police Use of SWAT: Examining the Relevance of Legitimacy.” Journal of Crime and Justice 42:45–59

49.

Nakagawa

Shinichi

. 2004. “A Farewell to Bonferroni: The Problems of Low Statistical Power and Publication Bias.” Behavioral Ecology 15:1044–45.

50.

Nix

Justin

Wolfe

Scott E.

Rojek

Jeff

Kaminski

Robert J.

. 2015. “Trust in the Police: The Influence of Procedural Justice and Perceived Collective Efficacy.” Crime & Delinquency 61:610–40.

51.

Oppenheimer

Daniel M.

Meyvis

Tom

Davidenko

Nicolas

. 2009. “Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power.” Journal of Experimental Social Psychology 45:867–72.

52.

Parry

Megan M.

Moule

Richard K.

Jr. Dario

Lisa M.

. 2019. “Technology-mediated Exposure to Police–citizen Encounters: A Quasi-experimental Assessment of Consequences for Citizen Perceptions.” Justice Quarterly 36:412–36.

53.

Perneger

Thomas V.

1998. “What’s Wrong with Bonferroni Adjustments.” British Medical Journal 316:1236–38.

54.

President’s Task Force on 21st Century Policing. 2015. Final Report of the President’s Task Force on 21st Century Policing. Washington, DC: Office of Community Oriented Policing Services.

55.

Pryce

Daniel K.

2019. “The Relative Effects of Normative and Instrumental Models of Policing on Police Empowerment: Evidence from a Sample of Sub-Saharan African Immigrants.” Criminal Justice Policy Review 30:428–50.

56.

Pryce

Daniel K.

Grant

Lorna

. 2019. “The Relative Impacts of Normative and Instrumental Factors of Policing on Willingness to Empower the Police: A Study from Jamaica.” Journal of Ethnicity in Criminal Justice. doi: 10.1080/15377938.2019.1681046.

57.

Qualtrics. 2014. “ESOMAR 28: 28 Questions to Help Research Buyers of Online Samples.” Retrieved April 17, 2018 (https://success.qualtrics.com/rs/qualtrics/images/ESOMAR%2028%202014.pdf).

58.

Radil

Steven M.

Dezzani

Raymond J.

McAden

Lanny D.

. 2017. “Geographies of US Police Militarization and the Role of the 1033 Program.” The Professional Geographer 69:203–13.

59.

Reisig

Michael D.

Bratton

Jason

Gertz

Marc G.

. 2007. “The Construct Validity and Refinement of Process-based Policing Measures.” Criminal Justice and Behavior 34:1005–28.

60.

Reisig

Michael D.

Wolfe

Scott E.

Holtfreter

Kristy

. 2011. “Legal Cynicism, Legitimacy, and Criminal Offending: The Nonconfounding Effect of Low Self-control.” Criminal Justice and Behavior 38:1265–79.

61.

Reiss

Albert J.

1971. The Police and the Public. New Haven, CT: Yale University Press.

62.

Sabine

George

. 1937. A History of Political Theory. New York: Holt, Rinehart and Winston.

63.

Sarat

Austin

. 1977. “Studying Legal Culture: An Assessment of Survey Evidence.” Law and Society Review 11:427–88.

64.

Sargeant

Elise

Murphy

Kristina

Cherney

Adrian

. 2014. “Ethnicity, Trust and Cooperation with Police: Testing the Dominance of the Process-based Model.” European Journal of Criminology 11:500–24.

65.

Steidley

Trent

Ramey

David M.

. 2019. “Police Militarization in the United States.” Sociology Compass 13:e12674.

66.

Sun

Ivan Y.

Yuning

Rong

Farmer

Ashley K.

. 2017. “Procedural Justice, Legitimacy, and Public Cooperation with Police: Does Western Wisdom Hold in China?” Journal of Research in Crime and Delinquency 54:454–78.

67.

Sunshine

Jason

Tyler

Tom R.

. 2003. “The Role of Procedural Justice and Legitimacy in Shaping Public Support for Policing.” Law and Society Review 37:513–48.

68.

Tankebe

Justice

. 2009. “Public Cooperation with the Police in Ghana: Does Procedural Fairness Matter?” Criminology 47:1265–93.

69.

Tankebe

Justice

. 2013. “Viewing Things Differently: The Dimensions of Public Perceptions of Police Legitimacy.” Criminology 51:103–35

70.

Trinkner

Rick

. 2019. “Clarifying the Contours of the Police Legitimacy Measurement Debate: A Response to Cao and Graham.” Asian Journal of Criminology 14:309–35.

71.

Trinkner

Rick

Jackson

Jonathan

Tyler

Tom R.

. 2018. “Bounded Authority: Expanding ‘Appropriate’ Police Behavior beyond Procedural Justice.” Law and Human Behavior 42:280–93.

72.

Tyler

Tom R.

2001. “Public Trust and Confidence in Legal Authorities: What Do Majority and Minority Group Members Want from the Law and Legal Institutions?” Behavioral Sciences & the Law 19:215–35.

73.

Tyler

Tom R.

2003. “Procedural Justice, Legitimacy, and the Effective Rule of Law.” Crime and Justice 30:283–357.

74.

Tyler

Tom R.

2004. “Enhancing Police Legitimacy.” Annals of the American Academy of Political and Social Science 593:84–99.

75.

Tyler

Tom R.

2006. Why People Obey the Law. Princeton, NJ: Princeton University Press.

76.

Tyler

Tom R

. 2017. “Procedural Justice and Policing: A Rush to Judgment?” Annual Review of Law and Social Science 13:2.1–2.25.

77.

Tyler

Tom R.

Jackson

Jonathan

. 2014. “Popular Legitimacy and the Exercise of Legal Authority: Motivating Compliance, Cooperation, and Engagement.” Psychology, Public Policy, and Law 20:78–95.

78.

Tyler

Tom R.

Mitchell

Gregory

. 1994. “Legitimacy and the Empowerment of Discretionary Legal Authority: The United States Supreme Court and Abortion Rights.” Duke Law Journal 43:703–814.

79.

U.S. Census Bureau. 2010. Profile of General Population and Housing Characteristics: 2010, 2010 Demographic Profile Data. Retrieved April 17, 2018 (https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=DEC_10_DP_DPDP1&src=pt).

80.

U.S. Census Bureau. 2011a. Age and Sex Composition: 2010. Retrieved April 17, 2018 (https://www.census.gov/prod/cen2010/briefs/c2010br-03.pdf).

81.

U.S. Census Bureau. 2011b. “Distribution of Annual Household Income in the United States, 2010 Estimate. Current Population Survey, 2011 Annual Social and Economic Supplement.” Retrieved April 17, 2018 (https://www.census.gov//hhes/www/cpstables/032011/hhinc/new06_000.htm).

82.

VanVoorhis

Carmen R. Wilson

Morgan

Betsy L.

. 2007. “Understanding Power and Rules of Thumb for Determining Sample Sizes.” Tutorials in Quantitative Methods for Psychology 3:43–50.

83.

Walters

Glenn D.

Colin Bolger

. 2018. “Procedural Justice Perceptions, Legitimacy Beliefs, and Compliance with the Law: A Meta-analysis.” Journal of Experimental Criminology. doi: 10.1007/s11292-018-9338-2.

84.

Wang

Xia

. 2012. “Undocumented Immigrants as Perceived Criminal Threat: A Test of the Minority Threat Perspective.” Criminology 50:743–76.

85.

Weber

Max

. 1968. Economy and Society. Edited by Roth

Guenther

Wittich

Claus

. New York: Bedminster Press.

86.

Weitzer

Ronald

Tuch

Steven A.

. 2004. “Race and Perceptions of Police Misconduct.” Social Problems 51(3):305–325.

87.

Weitzer

Ronald

Tuch

Steven A.

. 2005. “Racially Biased Policing: Determinants of Citizen Perceptions. Social Forces 83:1009–1030.

88.

White

Michael D.

Mulvey

Philip

Fox

Andrew M.

Choate

David

. 2012. “A Hero’s Welcome? Exploring the Prevalence and Problems of Military Veterans in the Arrestee Population.” Justice Quarterly 29:258–86.

89.

Wolfe

Scott E.

2011. “The Effect of Low Self-control on Perceived Police Legitimacy.” Journal of Criminal Justice 39:67–74.

90.

Wolfe

Scott E.

Nix

Justin

Kaminski

Robert

Rojek

Jeff

. 2016. “Is the Effect of Procedural Justice on Police Legitimacy Invariant? Testing the Generality of Procedural Justice and Antecedents of Police Legitimacy.” Journal of Quantitative Criminology 32:253–82.

91.

Worden

Robert E.

McLean

Sarah J.

. 2017. “Research on Police Legitimacy: The State of the Art.” Policing: An International Journal 40:480–513.

92.

Wright

Daniel B.

Skagerberg

Elin M.

. 2012. “Measuring Empathizing and Systemizing with a Large US Sample.” PLoS One 7:e31661.

93.

Yesberg

Julia A.

Bradford

Ben

. 2018. “Affect and Trust as Predictors of Public Support for Armed Police: Evidence from London.” Policing and Society. doi: 10.1080/10439463. 2018.1488847.

94.

Zahnow

Renee

Mazerolle

Lorraine

Pang

Alicia

. 2019. “Do Individual Differences Matter in the Way People View Police Legitimacy? A Partial Replication and Extension of Invariance Thesis.” Policing: A Journal of Policy and Practice. doi: 10.1093/police/paz06.