Keywords

Introduction

The code of silence has been a topic of scholarly work since the 1970s (e.g., Bittner, 1970; Klockars et al., 2000, 2004, 2006; Westley, 1970). Since the 1990s, the theory of police integrity and the related methodology have been used to measure empirically the contours of the code of silence across the world (for an overview, see Kutnjak Ivković, 2015; see also Klockars et al., 2004; Kutnjak Ivković & Haberfeld, 2015, 2019). Extant research on police integrity has established that police integrity measures, such as perceptions of misconduct seriousness, views about expected discipline, and the anticipated peers’ willingness to report misconduct, are strongly related to the police officers’ (un)willingness to stick to the code of silence (see Chap. 2). We have also demonstrated that the perceptions of disciplinary fairness seem to affect the respondents’ determination whether to report misconduct (see Chap. 3). In further extension of police integrity work, we have incorporated measures of organizational justice (see Chap. 4) and self-legitimacy (see Chap. 5) to assess their relationship with the code of silence, while controlling for the police integrity measures. Up to this point, we have not yet combined these different approaches to obtain a full model of factors that affect the code of silence. Such an approach would allow us to evaluate the complexity of the ways in which these factors relate to the code of silence and to assess their individual strength in this multivariate space.

During multivariate statistics courses, most students are repeatedly told some form of the adage that “we do live in a multivariate world.” This comment is meant to imbue future analysts with the belief that there is often more than a singular explanation for the same behavior, especially in the social sciences. After all, human behavior is thought to be infinitely complex. However, this realization—and analytic consequences—of this adage must be weighed against the competing concern for parsimonious theoretical explanations. In fact, students are often taught that parsimony is one of the criteria under which a criminological explanation can be judged (Akers & Sellers, 2013). This means that in our quest to explain any phenomenon and develop theoretical models, we must balance the competing concerns of completeness and parsimony.

Essentially, scholars are trying to balance these competing concerns. Neglecting important variables in a theoretical explanation may be inadvertently causing two problems. First, by ignoring key variables from the model we may end up identifying a spurious relationship between two variables that are in fact caused by another common factor, which is often excluded from the model (Weisburd et al., 2016). This error has at least two consequences. For scholars, this may lead to unnecessarily squandered research efforts to identify theoretical explanations of behavior. For practitioners, it could result in the wrong factors being manipulated to control the behavior in question. In other words, as it relates to the topic of this book, such an approach could result in policing scholars focusing on inappropriate factors to explain why officers adhere to the code of silence and can lead police executives to manipulate the wrong factors to control this behavior. Both outcomes are deeply problematic for police integrity research, which is difficult enough to begin with (e.g., Klockars et al., 2004; 2006).

The second problem associated with trying to balance the multivariate world and the parsimonious theoretical explanations is related to the concept of omitted variable bias. The problem of omitted variable bias, an issue frequently neglected in much of the criminological research (Weisburd et al., 2016), speaks to the fact that omitting an important explanatory variable from a regression model may yield problematic parameter estimates (Weisburd, 2010). Specifically, if a key variable that is correlated with the dependent variable (i.e., adhering to the code of silence) is omitted from the model, it could make the results of the entire model highly questionable, especially when correlated with another independent variable in the model (Angrist and Pischke, 2008). Failing to include the appropriate variables in our theoretical explanations of social behavior can be deeply problematic.

In short, this means that our explanations of the factors affecting the code of silence explored in this book so far may have been incomplete. Our primary argument here is that, while police integrity theory, which has been rigorously and robustly tested (Klockars et al., 2006; Maskàly et al., 2019), offers one type of explanation for the officer’s decision to adhere to the code of silence, the theoretical argument could potentially be furthered with the inclusion of other factors. In prior chapters, we have looked at perceptions of disciplinary fairness (Chap. 3), organizational justice (Chap. 4), and self-legitimacy (Chap. 5) and their link with the police offices’ adherence to the code of silence. However, each of these factors was independently added to the traditional police integrity variables in prior chapters and, as the discussion on omitted variable bias above suggests, the results from prior models may present an incomplete explanation of an officer’s decision to adhere to the code of silence. This is an especially pressing concern given the potentially strong theoretical and empirical relationships between the factors that were independently added in prior chapters.

In this chapter, we briefly explain the links between these additional theoretical concepts (i.e., discipline fairness, organizational justice, self-legitimacy) and the traditional police integrity measures (i.e., perceptions of policy violation, other police officers’ willingness to report, and perceptions of seriousness). We start by drawing together the traditions of disciplinary fairness and organizational justice. We continue by connecting the theoretical link between organizational justice and self-legitimacy. While the concepts of disciplinary fairness and self-legitimacy are almost certainly related, given that disciplinary fairness falls under the umbrella of organizational justice, to our knowledge the link has never been expressly tested nor theoretical linkages previously established. After presenting the theoretical arguments for the connections, we present the results from the full models that attempt to explain police officers’ decision to adhere to the code of silence using the traditional police integrity variables, perceptions of disciplinary fairness, organizational justice, and self-legitimacy.

Disciplinary Fairness and Organizational Justice

One of the recurrent arguments in philosophy and social scientific literature revolves around the related concepts of fairness and justice, which tend to be indiscriminately used interchangeably (Colquitt et al., 2005). Humans living in social settings have long been overly concerned about being treated fairly by other members in the social group. Furthermore, this idea of fair treatment has several subdomains, which are frequently encapsulated under the larger umbrella of organizational justice, comprised of distributive, procedural, and interpersonal justice (Greenberg, 1987).

The first type of justice was distributive justice, which can be traced back to the writings of ancient Greek philosophers (e.g., Plato). The primary argument of distributive justice is that people are concerned with the fairness of various types of outcomes (Adams, 1963). Initially, much of the work in this area emphasized that individuals are concerned about receiving their fair share of outcomes, rewards, and promotions. There have been many iterations of distributive justice theories, but one of the common elements is that perceptions of distributive justice are not as much predicated on the absolute level of the outcome as they are on the relative comparisons to what is known about the outcomes that others in the same social group are receiving (Festinger, 1954).

Initial conceptualizations of distributive justice focused on the distribution of rewards, but research soon recognized that punishments—or discipline—generate the same pattern of concerns (Colquitt et al., 2005). Essentially, people may be willing to accept punishment for engaging in certain types of behavior, although they want to make sure that the distribution of this punishment is fair. There are three potential concerns: equity, equality, and need (Deutsch, 1975). The concept of equity is based on the initial construction of distributive justice, that a person receives a fair outcome based on their contribution (Adams, 1963, 1965). The concept of equality focuses on the fact that all people who do the same thing will receive the same distribution (Deutsch, 1985). This is akin to the concept of proportionality in criminal punishments, whereby all similarly situated defendants will be punished in the same way for the same crime (Feeley, 1978). The final concept—need—is the least developed of three concepts. It focuses on what sorts of aggravating and mitigating factors a person may have that affect his/her distribution (Lerner, 1977). Determining which of these specific motivations is present in the current data is beyond the scope of this book. Instead, we use this to highlight the link between organizational justice and disciplinary fairness.

Obviously, there is a direct link between organizational justice and discipline fairness through the lens of distributive justice, but the theoretical links between these two concepts do not stop there. Instead, the issue is more nuanced when looking at the literature on effective discipline in organizations, which suggests that the amount of discipline administered may be an important factor for the way employees feel about their organizations (Arvey & Ivancevich, 1980). If the discipline is too harsh, it may have unintended consequences because members of the organization become afraid of the discipline, rather than seeing discipline as a tool to enhance learning (Parke, 1972). Furthermore, employees are concerned that the organization consistently disciplines all employees for the same behaviors (Gary, 1971), which, in turn, enhances the effectiveness of discipline. The same effect is seen for individual managers consistently disciplining their subordinates (Rosen and Jardee, 1974). Finally, we see that employees tend to be concerned with the fact that all managers will issue the same type of discipline for the same types of behaviors to their subordinates (Walster and Walster, 1975). Each of these concerns can be lumped together under the umbrella of distributive justice, as each of them represents employee concerns that the discipline reflects equity and equality. There are also concerns of need that have been identified in the literature, although the implications and consequences are not nearly as fleshed out in the extant empirical research.

Additionally, research suggests the nature of the relationship to the person administering the punishment/discipline is also an important factor in determining how effective punishment/discipline can be. Specifically, discipline tends to be best received when there is a relatively close relationship between the person administering the discipline and the person being disciplined (Parke, 1972). Essentially, this type of situation leads the person who made the mistake to feel the discipline more acutely because of the positive relationship with the person delving out the discipline. This speaks to the concept of interactional justice within organizations, whereby the disciplined person is more likely to feel that they were treated with dignity and respect while being disciplined (Greenberg, 2009).

Finally, the process through which discipline is meted out also has important effects. Specifically, research suggests that conveying trustworthy motives in the form of a rationale for the punishment is particularly effective in enhancing the efficacy and perceived fairness of punishment (Parke, 1972). This is particularly true when there is a substantial gap in time between the behavior and the discipline (Aronfreed, 1965). This concept can easily be applied to policing, given the often-substantial amount of time that it takes to conduct internal investigations related to lapses in police integrity. Again, the procedure in which the rationale for punishment is clearly outlined demonstrates to the employee being punished that the punishment is being implemented for trustworthy reasons. This is a key element of procedural justice, the final subcomponent of organizational justice.

Organizational Justice and Self-Legitimacy

The theoretical relationship between organizational justice and self-legitimacy has been fleshed out relatively recently. Bradford and Quinton (2014, p. 1023) succinctly identify the nature of the relationship by stating, “…organizational configurations—officers’ sense of their place in their organization and the relationships they have with both it and external stakeholders—can serve to promote or inhibit normative modes of policing.” Self-legitimacy is a concern in policing because of the dialogic conversation between power holders and those over whom they exert their power (Bottoms & Tankebe, 2012). An officer’s sense of self-legitimacy is important because it has been shown to explain differences in officer behavior and attitudes (e.g., Bradford & Quinton, 2014; Tankebe & Meško, 2015). Therefore, self-legitimacy may play an important role in explaining police integrity (see Chap. 5).

Relatively little is known about how self-legitimacy is developed. Tankebe (2014, p. 5) stated that “direct interactions with others provide information about confidence in self-legitimacy,” but did not define who precisely the others are. Nix and Wolfe (2017) suggest that these conversations may be conducted with at least three primary groups: supervisors, colleagues, and the public. Our interest here revolves around the effects of those conversations that officers are having with their supervisors, specifically the organizationally just nature of those conversations. The theoretical argument suggests that officers’ self-legitimacy may be eroded or otherwise inhibited from forming in organizations that lack organizational justice.

The theoretical antecedents to self-legitimacy have been sparsely tested, although the limited evidence suggests organizational justice—and its components—are generally positively associated with self-legitimacy. For example, Nix and Wolfe (2017) found that perceptions of organizational justice are positively associated with self-legitimacy, even after considering the effects of negative publicity. This implies that, while certain types of negative publicity can certainly erode the self-legitimacy of officers, this effect can be relatively easily overcome by enhancing the perceptions of organizational justice within the organization. Moreto et al. (2021) found a similar effect for conservation rangers’ feelings of self-legitimacy. Moreover, Meško et al. (2017) also found that tenets of organizational justice affect the self-legitimacy of prison officers in Slovenia, where perceptions of self-legitimacy were associated with better treatment of inmates. This emerging body of literature suggests that organizational justice plays a role in shaping police officers’ self-legitimacy, although no scholarship of which we are aware directly links self-legitimacy and organizational justice to police integrity.

When considering the relationship between organizational justice and self-legitimacy, an open question revolves around the theoretical specification and operationalization of these concepts. There seems to be measurement ambiguity associated with both organizational justice and self-legitimacy, but for different reasons. The trouble with organizational justice stems from the potential theoretical redundancy of organizational justice with other measures, namely organizational commitment. Some scholars argue that the relationship between organizational justice and other factors is mediated by organizational commitment (Cohen-Charash and Spector, 2001). This may suggest that the effects ascribed to organizational justice may, in fact, truly belong to organizational commitment. Bradford and Quinton (2014) found that organizational identity—an analogous indicator to organizational commitment—mediated the relationship between procedurally fair treatment from supervisors and self-legitimacy. Research shows that the effect of organizational justice and a measure of police integrity (i.e., perceptions of seriousness) were partially mediated by organizational commitment (Fridell et al., 2021).

The second open theoretical and methodological problem revolves around the ambiguous theoretical operationalization of self-legitimacy. As Gau and Paoline (2020) point out, there have been two distinct discussions on the role of self-legitimacy: one that is inwardly focused and the other that involves public perceptions. The inward-focused theoretical notion of self-legitimacy puts the focus on the holder’s assessment of their legitimacy, regardless of other factors (Barker, 2001). Conversely, the externally focused theoretical notion of self-legitimacy says that an officer’s claim of self-legitimacy involves an ongoing dialogue with the public (Tankebe, 2010). This distinction is important because, as Gau and Paoline (2020) show, measures of organizational justice exert differential effects. Notably, the scholars include two pseudo-measures of organizational justice analogous to organizational justice measures that are targeted at the officer’s direct supervisors and the organization’s top management. The results from Gau and Paoline’s study (2020) show that officers’ self-legitimacy is positively related, although not significantly, to supervisory support, and negatively and significantly related to top management support, whereas the opposite effects, although with the same significance interpretation, are observed for externally generated self-legitimacy. Furthermore, there was no significant bivariate relationship between the two measures of self-legitimacy. Taken as a whole, this would suggest that the relationship between organizational justice and self-legitimacy—if any—may be highly dependent upon the operational definition of self-legitimacy and the operational definition of organizational justice.

Disciplinary Fairness and Self-Legitimacy

The link between disciplinary fairness and self-legitimacy is arguably the most theoretically underdeveloped among the independent variables in this book. However, there are at least two reasons to suspect that there might be a theoretical relationship between these two concepts. The first is the transitive property of logic. This property states that, if A → B and B → C, then A → C. Thus, as shown above, because disciplinary fairness is related to organizational justice and organizational justice is related to self-legitimacy, it follows that self-legitimacy should also be related to disciplinary fairness. At the same time, the extant literature does not provide enough evidence to perform a robust literature review on the topic. Instead, we offer some early evidence that seeks preliminarily to tie these two constructs together.

We argue that the link between self-legitimacy and disciplinary fairness can be found by examining the operational definitions of each. Another definition of self-legitimacy is “power-holders’ recognition of, or confidence in, their own individual entitlement to power” (Tankebe, 2014, p. 3). Some scholars argue that officers with a higher degree of self-legitimacy are better able to perform their duties because they do not fear being unfairly disciplined for their actions (Nix and Wolfe, 2017), which is the key for linking this concept with disciplinary fairness.

Research finds that overly harsh punishments can have a paralyzing effect on the decision-making and learning of employees in organizations (Parke, 1972). This would suggest that, when police organizations harshly discipline officers who make mistakes, there may be an unintended consequence in the form of eroding officers’ self-legitimacy such that they no longer have confidence in the power bestowed upon them. Conversely, other research suggests that, when discipline is too lenient, this can also have deleterious effects because such discipline likely fails to change or stop the adverse behaviors (Weinstein, 1969). This failure to change behaviors and attitudes could then expose the disciplined officers to adverse treatment by their peers. An officer’s peers represent one of the three constituencies with whom the officer engages in the dialogic conversation to build self-legitimacy. Therefore, discipline that is too harsh may lead to the officers’ erosion of self-legitimacy because the officers feel that they are unqualified/incapable of holding the power vested in them. On the other hand, discipline that is too lenient may lead to the officer’s peers questioning the officer’s legitimacy as a police officer. Therefore, there is indeed a relationship between self-legitimacy and disciplinary fairness, although the precise connection could use additional theoretical refinement and empirical assessment.

This Chapter

Police integrity theory has a storied and consistent ability to explain adherence to the code of silence around the world. In this chapter, we seek to determine whether the inclusion of several theoretically relevant factors adds explanatory power to the traditional police integrity variables. Specifically, we investigate whether simultaneous addition of measures of disciplinary fairness, organizational justice, and self-legitimacy changes our ability to predict police officers’ adherence to the code of silence in the scenarios depicting instances of various types of misconduct. Since each of these other constructs—especially organizational justice and self-legitimacy—have been used by policing scholars previously, they have proven to be quite successful in explaining various facets of police officers’ attitudes and behaviors, but there were no direct tests of their effectiveness in explaining police integrity. In prior chapters, we have seen how each of these factors operates independently when added to the traditional police integrity variables. Given the strong theoretical links between each of these additional constructs, it is possible that failing to include all of them in the same model may lead to omitted variable bias.

Methodology

Sample

In late 2018/2019, we surveyed police officers from a medium-sized municipal police agency in the United States. The sample consists of 148 officers that serve as sworn police officers in this agency. For a more detailed description of the sampling method and sample characteristics, please see Chaps. 1 and 2.

Measures

As with previous chapters, we are interested in explaining police officers’ decisions to adhere to the code of silence using traditional police integrity variables and the other independent variables described in previous chapters (Chaps. 3, 4, and 5). All measures were included in the newest version of the police integrity questionnaire designed by Kutnjak Ivkovicć and colleagues (2019). The central focus of the police integrity questionnaire are 12 scenarios depicting four categories of police misconduct: police corruption, use of excessive force, organizational deviance, and interpersonal deviance—each represented by three scenarios (see Chap. 2). For more details, please see Chaps. 1 and 2.

Dependent Variable

Our measure of the respondents’ adherence to the code of silence is based on their own willingness to report misconduct described in each scenario. Respondents were asked to indicate their willingness to report the behavior using a 5-point Likert scale that ranged from 1 (definitely would not report) to 5 (definitely would report). Given that we are interested in adherence to the code of silence (i.e., unwillingness to report), we collapsed the answers into two categories, with values of 1 and 2 coded as 1 (i.e., adhere to the code of silence) and the remaining values coded as 0. There are two exceptions to this rule for two of the most serious scenarios (i.e., theft from a burglary scene and shooting a suspect in the back), where reporting was much more ubiquitous than for other scenarios. In these cases, we assigned a value of 1 to those who initially indicated 1 through 3 on the initial Likert scale, with the other two values (4 and 5) coded as 0. For additional details, please see Chap. 2.

Organizational Independent Variables

We incorporated several traditional police integrity variables into multivariate models. They include police officers’ own estimates of misconduct seriousness, their evaluations of misconduct as a violation of official rules, their views about what the appropriate discipline is and what the expected discipline would be, and their estimates of how willing other police officers in their agency would be to report misconduct. For details, please see Chap. 2.

Disciplinary Fairness Independent Variables

In addition to the above variables, we have included a measure of disciplinary fairness. For details, please see Chap. 3.

Organizational Justice Independent Variables

The multivariate models of the respondents’ adherence to the code of silence include the global measure of organizational justice. For details, please see Chap. 4.

Self-Legitimacy Independent Variables

The models that we have developed incorporate two measures of self-legitimacy. The results of the factor analysis showed that they are loading together as a simple construct of self-legitimacy. For details, please see Chap. 5.

Individual Independent Variables

The models also include several measures of the respondents’ demographic characteristics: length of service, gender, assignment, supervisory status, and education. We utilize these demographic characteristics as the control variables. For details, please see Chap. 2.

Analytic Strategy

The analyses build on the multivariate models of the code of silence presented in prior chapters. We predict adherence to the code of silence for all 12 scenarios using the traditional police integrity variables, the measure of disciplinary fairness, the measure of organizational justice, and the measure of self-legitimacy. We group together the results by misconduct type (i.e., corruption, use of excessive force, organizational deviance, and interpersonal deviance). Again, we continue to use the LASSO modeling procedure to estimate the multivariate effects, net of the control variables. For details, please see Chap. 2.

Results

The Effects of Police Integrity, Distributive Fairness, Organizational Justice, and Self-Legitimacy on the Code of Silence

Police Corruption

After including all independent variables in the model (Table 6.1), we see that the effects of the traditional police integrity measures (e.g., others’ adherence to the code of silence, violation of agency rules, perceptions of seriousness) are largely consistent with the results from prior research. In all scenarios, officers who feel their colleagues are unlikely to report the misconduct are also more likely to say that they would adhere to the code of silence. Additionally, knowledge of organizational rules (violation of policy) and perceptions of seriousness are both negatively associated with adherence to the code of silence, again consistent with prior research. Furthermore, compared to the respondents who expected some intermediate discipline, the respondents who expected no discipline tended to be more likely to say that they would not report (Table 6.1). Finally, compared to the respondents who expected intermediate discipline, the respondents who expected dismissal were more likely to say that they would adhere to the code of silence in two out of three scenarios than the respondents who expected intermediate discipline (Table 6.1).

Table 6.1 Full model estimates of adherence to the code of silence
Full size table

Comparing the results of the traditional police integrity variables to those from the unconditional models estimated in Chap. 2, the results are substantively the same. Some of the effect sizes were attenuated by the inclusion of all the independent variables in the model, although some of the effects strengthened, especially in the scenario depicting supervisory corruption. Overall, these results suggest that the addition of these new variables is not redundant with the traditional police integrity variables and that they enhance the explanation of the police officers’ adherence to the code of silence.

The effects of disciplinary fairness remain substantively the same as they did in Chap. 3. However, the inclusion of the additional independent variables leads to harsh discipline—relative to a fair discipline—being statistically significant for the scenario of accepting free gifts (OR = 0.49, p < 0.01), unlike the models estimated in Chap. 3. The same pattern is observed for the same variable in the supervisory corruption scenario (OR = 0.45, p < 0.01). This would suggest that, while disciplinary fairness may be theoretically related to organizational justice—especially the distributive justice subcomponent—this construct offers a unique addition to our ability to explain adherence to the code of silence.

A very different pattern is seen for organizational justice in these corruption scenarios. After including the other independent variables, we see that organizational justice significantly reduces the likelihood that officers would adhere to the code of silence in each of these three scenarios. These results are similar to the results from Chap. 4, although the effects are now significant for all scenarios. This consistent organizational justice effect is in accordance with theoretical expectations and prior research.

Finally, the results for self-legitimacy universally show that officers who have a higher sense of self-legitimacy are more likely to adhere to the code of silence in these corruption scenarios, net of all of the other factors. This is consistent with the results from the unconditional models of self-legitimacy presented in Chap. 5. However, compared to the results reported in Chap. 5, the effects in current models have been substantially strengthened for two of the scenarios (i.e., theft from a burglary scene and supervisory corruption) and reversed for the scenario depicting an officer accepting free gifts. Given the strength of the relationship between self-legitimacy and the other independent variables, such a stark change in the effect size is not surprising. This is the embodiment of the problems associated with omitted variable bias. These results challenge the emerging lore surrounding the importance of officers developing self-legitimacy.

Use of Excessive Force

The results from the full models for the three scenarios depicting the use of excessive force (Table 6.1) show that the traditional police integrity variables are again consistent with prior research and largely remain substantively unchanged relative to the results from the unconditional models presented in Chap. 2. The only substantive changes are the changes in the effect of the type of discipline that an officer who engaged in this type of behavior would likely receive. The interpretation of the effects remains substantively unchanged in this full model, and the difference in the statistical significance of these two variables is largely attributable to the standard errors becoming more stable with the inclusion of the additional independent variables. In other words, the inclusion of these independent variables accounted for some amount of noise in the estimates of the traditional police integrity variables, especially as related to the expected type of discipline. The same is also true for the results assessing the effect of the measures of disciplinary fairness in these excessive force scenarios.

We now turn to the results from the measure of organizational justice. Although in Chap. 4 organizational justice was not significantly associated with adherence to the code of silence in the excessive force scenarios, a more complex pattern of the effect of organizational justice emerges after other independent variables are included in the multivariate models. Specifically, for these scenarios depicting the use of excessive force, organizational justice increases adherence to the code of silence (failing to report a beating; OR = 1.37, p < 0.001), reduces adherence to the code of silence (verbally abusing motorist; OR = 0.60, p < 0.01), and exerts no significant effect (shooting suspect in the back; OR = 1.33, p > 0.05). It is unclear what drives these differential effects here, although the two scenarios where organizational justice is positively associated with adherence to the code of silence—regardless of whether the effect is significant—involve the use of physical force, and the other scenario does not.

Likewise, we see a complex set of findings emerge for self-legitimacy. In Chap. 5, the effects of self-legitimacy on the code of silence were complicated. This pattern of findings repeats itself here. Just like organizational justice, self-legitimacy increases the likelihood of adhering to the code of silence in one scenario (verbally abusing a motorist; OR = 1.61, p < 0.01), reduces the likelihood in one scenario (failing to report a beating; OR = 0.54, p < 0.001), and has no significant effect on the third (OR = 4.27, p > 0.05).

Organizational Deviance

The effects of the traditional police integrity variables in the organizational deviance scenarios (Table 6.1) are largely consistent with prior police integrity research. Furthermore, apart from reducing the effect sizes, the results are substantively similar to those presented in Chap. 2. Again, this suggests that the independent variables are adding a unique explanatory power to the model, rather than being redundant with the police integrity theory variables.

Next, we turn to perceptions of disciplinary fairness for each of these three scenarios. The results here are substantively identical to those presented in Chap. 3. The key differences are seen in the reduction in the magnitude of the effect for each of these variables after the inclusion of the other independent variables. This would suggest that there is some redundancy with the disciplinary fairness and the other independent variables included in the model. The potential redundancy here is not entirely surprising, given the theoretical and empirical relationship between the constructs. A potential explanation for this effect may stem from the effects of organizational justice. Organizational justice is theoretically and—to varying degrees—empirically associated with disciplinary fairness. Including both variables in the model can lead to some redundancy in the estimates, which seems to be the case here. In the full models, the effects of organizational justice are strengthened compared to the results in Chap. 4, and the interpretation of the effects remains substantively unchanged. These results suggest that there is some shared variance between disciplinary fairness and organizational justice, although not as much as might be expected.

Conversely, the results from self-legitimacy are quite different from the results presented in Chap. 5. In the models with only self-legitimacy and traditional police integrity measures, we see either a null or negative effect of self-legitimacy on the adherence to the code of silence (see Chap. 5). However, with the inclusion of the other independent variables, the pattern of findings becomes more complex. We see that self-legitimacy reduces the likelihood that an officer would adhere to the code of silence in covering a DUI crash of a fellow officer (OR = 0.30, p < 0.001), but significantly increases adherence to the code of silence in the other two scenarios: falsely calling in sick (OR = 1.67, p < 0.05) and falsely reporting overtime (OR = 1.63, p < 0.05).

Interpersonal Deviance

Finally, we look at the results for the interpersonal deviance scenarios (Table 6.1). Again, the results from the traditional police integrity variables presented here are substantively identical to those presented in the unconditional models in Chap. 2. The same is true for the disciplinary fairness measures. Again, this would suggest that both the traditional police integrity variables and the disciplinary fairness measures exert independent effects on the adherence to the code of silence in these interpersonal deviance scenarios.

We notice some differences when examining the effect of perceptions of organizational justice in these interpersonal deviance scenarios. The results from Chap. 4, which only added organizational justice and traditional police integrity measures to the model, show that organizational justice is only associated with increased adherence to the code of silence for the scenario depicting the spreading of false rumors about a coworker (OR = 1.50, p < 0.001). However, after including all of the independent variables in this chapter, the size of this effect has nearly doubled (OR = 2.86, p < 0.001). Additionally, we see that, unlike in Chap. 4, organizational justice now exerts significant—although differential—effects on adherence to the code of silence for the scenario depicting the telling of sexist jokes (OR = 0.63, p < 0.001) and the scenario depicting yelling at coworkers (OR = 2.77, p < 0.001).

Again, it is not entirely clear what is different about these scenarios that causes the differential effects of organizational justice. However, these varying results may suggest that officers are not simply “pencil whipping” the survey, and instead are making independent judgments about each scenario. This contention seems to be supported by the pattern of results for self-legitimacy, which shows substantively opposite effects compared to organizational justice in these scenarios. In other words, when organizational justice increases adherence to the code of silence, self-legitimacy reduces it. Finally, again, we note that the inclusion of the other independent variables reduced the noise in the parameter estimates for self-legitimacy, a largely expected outcome in light of the theoretical and conceptual overlap of these constructs.

Conclusion

Because of the police agencies’ semi-military organization, the code of silence has developed in each police agency as part of police culture to protect police officers (e.g., Bittner, 1970; Klockars et al., 2000). As Westley (1970) and Bittner (1970) hypothesized, police recruits are socialized into police culture—both at the police academy and during their field training—and taught that they can rely only on their fellow officers. One mechanism used to acquire this trust is adhering to the code of silence and, hence, protecting police misconduct committed by their fellow officers.

Therefore, it comes as no surprise that the empirical analyses in this book (e.g., Chap. 2) reveal the presence of the code of silence among the police officers in the police agency we study. Specifically, for every scenario, including the most serious ones (i.e., stealing from a crime scene, abusing deadly force), there is a certain number of police officers who said that they would not report misconduct. With the exception of these two scenarios, at least one out of five police officers, if not more, said that they would not report any type of misconduct described in our study. In fact, in some scenarios, such as accepting gratuities, verbally abusing citizens, or covering up DUI crash, the majority of the police officers in our study seemed to be willing to adhere to the code of silence.

As various independent commissions (e.g., Knapp Commission, 1972; Mollen Commission, 1994) demonstrate, the code of silence—tolerance of police misconduct without reporting it—is a serious obstacle in the efforts to curtail police misconduct. Whereas assessing the contours of the code of silence is a critical step in such efforts (e.g., Klockars et al., 2000; 2004; Kutnjak Ivković & Haberfeld, 2015, 2019), it is not the only step. The goal of this book has been to determine the factors associated with the strong code of silence and offer evidence-based approach to curtailing the code of silence and, ultimately, police misconduct.

The results of our full models confirm the strong effect of the traditional police integrity measures (e.g., others’ adherence to the code of silence, violation of agency rules, perceptions of seriousness) on the police officers’ decision to adhere to the code of silence. To begin with, we found the effects of police culture at play; in all 12 scenarios, officers who assessed that their colleagues would not report police misconduct are also more likely to say that they would not report either. The perceptions of police culture tolerant of police misconduct have the strongest and most consistent effect on the respondents’ own expressed willingness to adhere to the code of silence. This finding, aligned with the extant literature (Hickman et al., 2016; Kutnjak Ivković et al., 2018, 2019; Long et al., 2013; Lim & Sloan, 2016; Peacock et al., 2020; Van Droogenbroeck et al., 2019), demonstrates the important effect that the police culture has on the way police officers think about police misconduct. Police administrators interested in curtailing the code of silence should learn how accurate the police officers’ perceptions of their peer culture are. If police officers’ views indicate overconfidence in their peers’ willingness to adhere to the code, police administrators should share with police officers the more accurate view of their peer culture. On the other hand, if police officers are accurately “reading” their peers’ willingness to not report, then the more challenging task that the administrators have is to change the peer culture by conveying an unambiguous message about the importance of official rules and their consistent enforcement, as well as by providing the rewards for reporting.

This research shows that the more serious police officers view police misconduct, the less likely they are to adhere to the code of silence. The result is strong and consistent across different forms of police misconduct. This strong negative relationship between the expressed willingness to adhere to the code of silence and perceptions of misconduct seriousness persists even in the presence of other potential factors, such as perceptions of organizational justice or self-legitimacy. These results fit the large body of extant police integrity literature (Cheloukhine et al., 2015; Haberfeld, 2004; Hickman et al., 2016; Khechumyan & Kutnjak Ivković, 2015; Klockars et al., 1997, 2004, 2006; Kutnjak Ivković et al., 2013, 2016, 2018, 2019; Kutnjak Ivković & Khechymian, 2013; Kutnjak Ivković & Sauerman, 2013; Lobnikar & Meško, 2015; Long et al., 2013; Lim & Sloan, 2016; Maskàly et al., 2019; Pagon & Lobnikar, 2000; Peacock et al., 2020; Porter & Prenzler, 2016; Vallmüür, 2015; Westmarland, 2006; Wu & Makin, 2019). They also open another avenue for police administrators willing to curtail the code of silence. Specifically, police officers should be (re)taught to view police misconduct as a serious matter. Nevertheless, police administrators should follow up and demonstrate—through the enforcement of official rules and administration of discipline for their violations—that they perceive police misconduct as a serious matter as well.

How serious police misconduct is evaluated typically has been closely tied with the assumption that misconduct violates official rules and that discipline would result in such rule violations (e.g., Klockars et al., 2000, 2004, 2006; Kutnjak Ivković & Haberfeld, 2015, 2019). In about one-half of the scenarios, our full models do not link the knowledge that police misconduct described in the scenarios violates official rules with the respondents’ expressed willingness to adhere to the code of silence. On the other hand, in most scenarios in which the relationship is statistically significant, there is a clear negative relationship between the respondents’ recognition that the behavior violates official rules and their willingness to stick to the code of silence. These findings demonstrate the importance of establishing clear official rules prohibiting misconduct (e.g., Klockars et al., 2000, 2006).

However, establishing the official rules is a necessary, but not a sufficient condition. Even the most carefully designed official rules need to be enforced consistently and fairly to be effective. Severity of expected discipline (or the lack of any discipline) is also related to how willing police officers are to report misconduct. Our findings show that, compared to the respondents who expected some intermediate discipline, the respondents who expected no discipline tended to be more likely to say that they would not report misconduct. When the police officers perceive that the police administration will not mete out any discipline or will mete out only very mild discipline in the form of a verbal reprimand, these police officers would be more likely to say that they would not report such misconduct—why bother if there will be no discipline at all? Assuming that the police officers accurately predict what the expected discipline will be, the message for police administrators is very clear: if particular behavior is defined as police misconduct, then the administration should mete out discipline in all the cases in which police officers engage in such rule-violating behavior. On the other hand, if the police officers are underestimating the severity of the disciplinary threat that the police administration is actually making, then the administrators need to teach police officers what the accurate disciplinary threat is.

The situation with dismissal as the expected discipline is more complex. In particular, when police misconduct includes very serious forms, such as stealing from a crime scene or abusing deadly force, the police officers expected dismissal and, if they perceive that the police administration will administer some intermediate discipline instead, they would not be willing to report. In all other examples of police misconduct in our questionnaire, the expectation of a dismissal compared to any intermediate discipline resulted in police officers’ increased willingness to adhere to the code of silence. Police officers viewed dismissal in such cases as an inappropriate discipline and, in turn, decided that they should not report a fellow police officer for an act or omission that might result in dismissal. Indeed, in many of these cases, including those tied to organizational deviance and interpersonal deviance, dismissal is an unlikely outcome for a first-time rule-violator. A lesson for police administrators is that they should weigh very careful which violations of official rules should be career-ending. “Getting it right” and communicating so to the police officers should help them curtail the code of silence.

Our results demonstrate that the severity of the expected discipline matters, but it is not the only consideration that matters; its perceived fairness does too. When the expected discipline is perceived as too harsh, police officers might be more willing to adhere to the code of silence. Our findings show substantial evidence of the theoretical model of simple justice at work, fitting well with the results of extant research (e.g., Datzer et al., 2019; Kutnjak Ivković & Klockars, 1998; Kutnjak Ivković et al., 2016; Kutnjak Ivković & Shelley, 2007, 2010). The first question for police administrators to answer is whether police officers are accurately predicting the discipline that their police agency would mete out. In case they are, then the discussion should be focused on why they perceive such discipline as too harsh. Police administrators may discover that the conditions have changed since the official rules were established and that the less harsh discipline could fulfill the disciplinary goals. Alternatively, the police officers should be taught why such misconduct should be disciplined with the discipline they perceived originally as too harsh.

We have also explored how respondents’ evaluations of expected discipline as too lenient are linked with their willingness to stick to the code of silence. Prior research has mostly indicated that there is no strong relationship between assessments of discipline as too lenient and the respondents’ willingness to report misconduct (e.g., Datzer et al., 2019; Kutnjak Ivković & Klockars, 1998; Kutnjak Ivković et al., 2016; Kutnjak Ivković & Sauerman, 2013; Kutnjak Ivković & Shelley, 2005, 2007, 2010). In about one-half of the scenarios we also found that the data fit the model of discipline indifference, but in the rest of the scenarios in our full models, we found that, compared to the respondents who evaluated discipline as fair, the respondents who evaluated discipline as too lenient were less likely to say that they would stick to the code of silence. The officers in this subgroup advocated for more severe discipline than they expected their agency to mete out and were also more likely to say that they would report. Assuming that their answers were not a consequence of their self-serving bias, future research could explore their views in more detail.

We also expanded the results to look at the effects of organizational justice on the decision to adhere to the code of silence. There is a consistent relationship between organizational justice and workplace deviance (Colquitt et al., 2005), and failing to report the misconduct of others—especially in policing—is a form of misconduct. Thus, we would have expected to see a consistent negative relationship between organizational justice and adherence to the code of silence. However, the effect of organizational justice on the respondents’ decision whether to report misconduct could be best described as inconsistent across models. Even in the more parsimonious models presented in Chap. 4, we only saw a significant negative effect in 3 of the 12 scenarios and a positive effect in 2 of the 12 scenarios, while in the majority of them—8 out of 12 scenarios—there was no relationship between organizational justice and the expressed willingness to adhere to the code of silence. After including all of the additional variables in this chapter, significant negative effects of organizational justice were present in six scenarios, significant positive effects in four, and there were none in the remaining two. For all of the corruption scenarios, the effect of organizational justice on adherence to the code of silence is negative, but varied between the other scenarios. While organizational justice increases adherence to the code of silence in certain scenarios, this should not be construed to mean that police executives do not need to enhance organizational justice in their organizations. Again, enhancing organizational justice may be a proverbial mixed bag, but the deleterious effects of enhancing organizational justice may be outweighed by the positive effects. Further research is needed to unpack the potentially complicated relationship of organizational justice which has been suggested by other scholars in this area (Nagin & Telep, 2017).

The conflicting results here suggest that the true effect of organizational justice may be more complicated than prior research leads on. These differences could be a result of the fact that we use organizational justice to predict adherence to the code of silence (i.e., failing to do something), whereas most prior research has looked at decisions to affirmatively engage in misconduct (see generally Culquitt et al., 2005). This subtle difference may be important and may not be as contrary to prior research as it initially seems. Recently, the importance of strengthening perceptions of organizational justice in police agencies has become increasingly ubiquitous because of the positive effects that come from this elevated sense of organizational justice (Wolfe & Lawson, 2020). An agency could still see the benefits of organizational justice in the behavior of officers consistent with prior research, but that organizational justice may inconsistently influence officers’ willingness to report the misconduct of others. In essence, the bolstered organizational justice may change how officers behave, but does not necessarily overcome fundamental elements of the police culture (i.e., adherence to the code of silence). Additional research is needed in this area to determine the veracity of this conjecture.

Finally, we turn to the effects of self-legitimacy. There is a growing body of literature which suggests that self-legitimacy can positively affect the attitudes and behaviors of police officers (Concepcion, 2021). Our work tried to extend the study of self-legitimacy into the study of police integrity, specifically the code of silence. The results here suggest that self-legitimacy yields an inconsistent effect across scenarios—regardless of whether it is only included in the model with other police integrity variables (i.e., Chap. 5) or along with the inclusion of additional relevant factors (i.e., this chapter). At first blush, the results from the full models presented in this chapter suggest that self-legitimacy may not be a desirable trait to instill in police officers, at least if the goal is to maximize willingness to report the misconduct of others. After all, in 7 of the 12 scenarios self-legitimacy significantly increased adherence to the code of silence, while exerting a significant negative effect in four of the scenarios.

Self-legitimacy increased adherence to the code of silence in the corruption scenarios, but exerted differential effects across other types of scenarios. The results for the excessive force scenarios are particularly diverse; self-legitimacy increases adherence to the code of silence in the scenario depicting verbal abuse of a citizen, yet it decreases the adherence to the code in the scenario depicting a failure to report a beating. These results are both supportive and contrary to prior research which consistently found that officers’ self-legitimacy is associated with lower willingness to use force and a greater desire to build positive partnerships with the community (Bradford & Quinton, 2014). However, these differences may not be inconsistent with prior research. Self-legitimacy talks about the confidence that officers have in the authority vested in them, which is created in dialogue with multiple constituencies (Nix & Wolfe, 2017). This may mean that—all else being equal—self-legitimacy could be the highest in those officers who have internalized the police subculture the strongest. For instance, self-legitimacy may exert a positive effect in the case of the verbal abuse of the motorist, because in this scenario the motorist’s actions could make him/her an “asshole” that needs to be treated differently (Van Maanen, 1978). Thus, the response from the officer depicted could be consistent with the expectations of an officer with self-legitimacy, whereas the actions of the officer who fails to report the beating of a suspect may be seen as something the officer with self-legitimacy would not do. The same potential explanation can be used to justify the inconsistent effects seen across other scenarios, apart from the corruption scenarios. Clearly, additional work on self-legitimacy is needed to address these issues and to determine whether the effects of self-legitimacy are generalizable from agency to agency, given the differences in organizational culture and climate.

The goal of our book has been to study factors that could be related to the police officers’ code of silence that exists in every police agency and to provide evidence-based suggestions for police administrators regarding different ways to curtail the code of silence. Our results vividly demonstrate that what police agency does or does not do—from the establishment of the official rules and their enforcement to the way supervisors treat their subordinates—directly affects police officers’ willingness to remain a part of the code of silence. Although the traditional police integrity factors, such as the evaluations of misconduct seriousness, severity of expected discipline, and the peers’ willingness to report misconduct, perform critical roles in shaping the code of silence, our results demonstrate that the expansion of the traditional approach through the organizational justice elements, discipline fairness, and self-legitimacy provide further valuable information about what makes police officers decide to protect misconduct in silence.