Abstract
How can police officers be encouraged to commit to changing organizational and personal practice? In this paper we test organizational justice theories that suggest that fair processes and procedures enhance rule compliance and commitment to the organization and its goals. We pay particular attention to (a) tensions between the role of group identity in organizational justice models and classic concerns about ‘cop culture’; and (b) the danger of over-identification with the organization and the counterproductive types of compliance this may engender. Results suggest that organizational justice enhances identification with the police organization, encourages officers to take on new roles, increases positive views of community policing, and is associated with greater self-reported compliance. Identification with the organization has generally positive implications; however, there is some danger that process fairness may encourage unthinking compliance with orders and instructions.
Developing and maintaining the commitment of police officers to changing organizational and personal practice is a perennial problem for police managers and is a mainstay of policing research. In England and Wales, the last two decades have seen a range of policing policies – neighbourhood policing, problem-oriented policing, the ‘confidence agenda’ (Stanko et al., 2012) – come and sometimes go. Although these policies have often involved major organizational change, the extent to which they have resulted in a wholesale shift in practice on the ground is often moot (Foster and Jones, 2010; Skogan, 2008). Yet the changing dynamics of police/public relations and the advent of new policies and technologies, on the one hand, and organizational attempts to adapt to and influence these dynamics, on the other, are often debated and discussed in ways that pay scant attention to the relationship between the practice of individual officers and the behaviour of the police organization more widely (for an exception see Chan et al., 2003). How can police officers be encouraged to alter their behaviour, and what institutional configurations might promote, or inhibit, change?
The unique nature of ‘police culture’ is often cited as a key barrier to implementing change programmes, refocusing officers’ activities and enhancing ‘professionalization’ (Foster, 2003; Skogan, 2008; Stanko et al., 2012). On these accounts, police cultures are marked by suspicion, internal solidarity, pragmatism and conservatism. Police officers have an action-oriented sense of mission coupled with a widespread cynicism and pessimism that may clash with some of the more service-oriented goals of modern policing (Loftus, 2010; Reiner, 2010). Furthermore, officer discretion increases as one moves from the top to the bottom of the organization, and ‘street-level cops’ are generally operating away from supervisory control (Goldstein, 1960). Together, these factors present a significant problem for those seeking to alter operational practice (Mastrofski, 2004). Moves towards service-based policing models, in particular, may encounter resistance, as those officers who see themselves primarily as thief-takers (Reiner, 2010) subvert or simply ignore efforts to refocus their activities and develop more sustainable relationships with the communities they serve (Loftus, 2010).
A considerable literature has built up around the issue of how and why police officers treat members of the public the way they do, the extent to which they ‘buy into’ new policies and practices, and the problems inherent in generating officer compliance in the face of the extent of their discretion. Studies have found, for example, that demographic (both officer and citizen) variables affect the ways in which officers react to members of the public (Mastrofski et al., 2000; Rabe-Hemp, 2008; Sun et al., 2008), and that a proclivity towards community policing increases the chance that officers will react to service-related requests (Rossler and Terrill, 2012; Myhill and Bradford, 2013). Yet the relationship between individual officers and the police organization is, as noted, often absent from such studies, as is consideration of the ways in which organizational structures or policies might encourage behaviour change. Given the centrality of the police organization within the wider policing literature, this lack of attention is somewhat surprising, and there are, of course, some exceptions. Gau and Gaines (2012), for example, found that variables related to attachment to supervisors and the quality of organizational decision-making and communication were associated with attitudes towards non-traditional practices such as order-maintenance policing. Similarly, using data from a UK police force, Myhill and Bradford (2013) reported that perceptions of organizational justice were an important predictor of positive attitudes towards serving the public. Finally, Tankebe (2010) found that the ‘organizational commitment’ of police officers in Accra, Ghana, was strongly predicted by the quality of their treatment by supervisors and their relationships with their peers.
In this paper we build on these recent studies to integrate an understanding of officer attitudes towards organizational goals with a body of literature that addresses the relationship between organizations and individuals. Work on organizational justice (Greenberg, 2011) has stressed that employees’ justice perceptions – particularly the perceived fairness of procedures and decision-making, and the quality of interaction and communication between managers and staff – are strongly linked to their compliance with organizational goals and to ‘citizenship behaviours’ that might benefit the organization and, in the case of policing, the public at large (Colquitt et al., 2001; Tyler, 2011; Tyler and Blader, 2000). We draw in this paper on data from a survey of officers serving in Durham Constabulary, 1 a police force in the north of England that stresses a service model of policing. We find that organizational justice – and particularly procedural fairness – is associated with compliance, commitment to community policing, and a willingness to engage in discretionary and extra-role behaviour. As predicted by prior work on justice within organizations, a large part of this effect is mediated by organizational identification; despite the classic concern with the negative implications of police culture, identification with a service-oriented police organization may, in as much as it is based on procedural fairness, be a positive force for change (see Myhill and Bradford, 2013).
Fairness, cooperation and compliance within police organizations
It is well established that the fairness of legal authorities – particularly the police – is linked to public propensities to defer to and cooperate with those authorities; procedural justice, and the legitimacy it generates, has also been linked to compliance with the law (Tyler, 2006, 2011; Tyler and Huo, 2002; Jackson et al., 2012). Based on the idea that fairness promotes normative commitment to authorities and the values they represent, the procedural justice model of policing contrasts with theories that stress a rational choice understanding of human behaviour – the idea that people comply with rules and norms as a result of the promise of reward or the threat of sanction (Nagin, 1998). Procedural justice theories stress instead the role of social connections between individuals and authorities; fair treatment by important group authorities such as the police activates people’s sense of belonging, group pride and legitimacy, and, concomitantly, their sense of duty in relation to group norms and values.
This model of the relationship between citizens and authorities such as the police resonates with, and is indeed partly based on, research that has looked at why people comply and cooperate within organizations and other more closely defined groups (Blader and Tyler, 2009; Lind and Tyler, 1988; Tyler and Blader, 2000). In this work, the importance of group membership and identity often takes centre stage. As Smith et al. (1998: 471) put it: people attend to the process by which authorities make decisions, rules, and policies . . . People who feel that authorities have treated them in a unbiased, dignified, and trustworthy manner are . . . more likely to comply with group rules. Benevolent treatment by authorities also shapes commitment to organizations, views about the legitimacy of organizations, and the willingness to engage in activities that benefit the group.
The reason people care about their treatment at the hands of group authorities is that they seek, and obtain, identity-relevant information from the behaviour of representatives of groups to which they belong. When they feel they are treated in a fair, open and honest fashion this encourages a sense of inclusion, status and pride within the group that, in turn, activates their responsibilities to it and encourages them to follow group rules and procedures willingly (Blader and Tyler, 2009; Smith et al., 1998; Tyler and Blader, 2000; Tyler and Degoey, 1995; Tyler et al., 1996). In essence, people are motivated to support organizations of which they feel valued members. Inclusion and value within groups have also been linked to feelings of self-worth (De Cremer and Tyler, 2005), and the experience of procedural fairness indicates to members of an organizations that they have its support, allowing for better management of uncertainty and enabling decision-making in difficult situations (a lack of fairness, by contrast, may close down options by suggesting that employees do not have the organization’s support).
Naturally, fairness within organizations extends beyond procedural justice, and people may come to identify with and feel proud of their organization based on a number of other relational criteria. Organizational justice is an umbrella term that encompasses research in social and organizational psychology and organizational behaviour (Colquitt, 2008; Greenberg, 2011). Most theorists recognize four elements of organizational justice:
Distributive – the perceived fairness with which inputs and rewards are distributed in the workplace.
Procedural – the perceived fairness of decision-making processes in the workplace (for example, accuracy, consistency, absence of bias, having a voice in the process).
Informational – the perceived fairness of information about decision-making processes and their outcome (for example, clarity, truthfulness and explanatory value).
Interpersonal – the perceived fairness of interpersonal interactions (for example, respectful and dignified treatment).
Nevertheless, there are debates in the literature about how these elements are defined and measured, and whether each is conceptually and empirically distinct (see Colquitt, 2001). Interpersonal and informational justice are often referred to jointly as ‘interactional justice’ because they are frequently highly correlated. Tyler and colleagues (Blader and Tyler, 2009; Tyler and Blader, 2003) also see little conceptual difference in the elements of organizational justice that relate to perceptions of fair process and decision-making and good interpersonal treatment and communication. They use the term ‘procedural justice’ to represent both procedural and interactional justice (the most obviously normative aspects of organizational justice). They note furthermore that procedural justice may have both formal and informal sources, suggesting the potential for people to draw a distinction between their immediate supervisor and senior leaders and the wider organization.
In this paper we follow the approach developed by Tyler and Blader and distinguish between just two elements of organizational justice: procedural (at supervisory and leadership levels) and distributive. Indeed, despite being an important element of organizational justice models, distributive justice arguably stands in contrast to procedural justice because it is more instrumental in nature (referring to costs and benefits at work) and draws primarily on equity theory (the idea that people compare their relative inputs and outcomes and act on the basis of whether they find the balance fair or not). Yet distributive justice may also have an important relational component, for example when it demonstrates to group members that they are equally valued and respected within the group.
The literature on organizational justice provides considerable empirical evidence that: (a) fairness within organizations is linked with group identification and pride; (b) fairness is linked to compliance with group rules and cooperative action; and (c) part of the link between fairness and compliance and cooperation is mediated by group identification (Blader and Tyler, 2009; Greenberg, 2011). Additionally, procedural justice often seems to be favoured over distributive justice (Greenberg, 2011; Tyler, 2011). According to this model, then, when police organizations behave in a procedurally fair manner towards their staff, this should encourage group identification, compliance with organizational norms and rules, a greater commitment to implementing new procedures and practices, and a greater willingness and confidence among staff to use their own initiative and professional judgement (because they feel the organization supports and values them).
Other motivations to cooperate and comply
Despite the evidence supporting the idea of organizational justice, rational choice models of human behaviour often appear to dominate in the human resource departments of both private and public sector organizations (Kohn, 1999; Tyler, 2011). Rewards (such as performance-related pay) and punishments (such as fixed limits on sickness absence) are frequently used to motivate staff. This instrumental approach suggests, first, that people respond to the risk of sanction, and comply with work rules and organizational policies when they believe they will be punished in some way if they do not. Second, it assumes that people respond to the promise of reward, comply and cooperate when they feel they will gain from doing so, and take on extra responsibilities when bonuses and other inducements are offered.
Another aspect of reward systems is the equality of allocation and, as suggested by organizational justice models, the distributive fairness of rewards could also be an important factor in securing cooperation, compliance and extra-role activity (Greenberg, 2011). Perceptions of distributive unfairness have been shown in a number of studies to result in counterproductive behaviours intended to redress the perceived imbalance, such as withdrawal of effort and stealing from the organization (Greenberg, 2011).
Tyler (2011: 27) characterizes these concerns as extrinsic, contrasting them with intrinsic motivations to work on behalf of the group that are encouraged by procedural justice. Whereas the efficacy of extrinsic motivations within organizations is contested, especially in relation to performance-related pay (Frey and Osterloh, 2012; Perry et al., 2009), their potential relevance in a hierarchical and quasi-military organization such as the police cannot be doubted, and they may be important factors shaping officer behaviour. The legal nature of much police work, in particular, places constraints on behaviour that cannot, indeed should not, be subsumed by intrinsic motivations generated by procedural fairness and group identification. Commitment to the organization should not be the only basis on which police act; they should also be constrained by formal rules and orders backed by the threat of sanction. Yet, notwithstanding this imperative, it has been suggested that extrinsic motivations can also influence identification with groups (Blader and Tyler, 2009). People may identify more with organizations when they feel fairly rewarded for the work they do; they may also identify with organizations they perceive to be efficacious in asserting and maintaining systems of rewards and punishments.
Identity, legitimacy and blind obedience
Three issues arise from the above. The first concerns the importance of identity-relevant information to individuals within social groups. People’s treatment by group authorities may be more or less meaningful to them depending on the nature and strength of their pre-existing identities. Fairness may mean little to people with uncomplicated, strong group affiliations, for example, since their status in, and commitment to, the group is more certain; conversely, fairness may mean more to those whose status or identity is in doubt (De Cremer, 2002; Sunshine and Tyler, 2003). Similarly, fairness may matter only in situations where group status is salient. Greater salience increases receptiveness to identity-relevant messages (Van Prooijen et al., 2002), whereas lower salience decreases receptiveness. On these accounts, social identity is largely prior to the experience of organizational justice (or injustice), and group identification moderates the links between fairness, cooperation and compliance. The strength of any link between fairness and cooperation (for example) will vary according to the strength of group identification.
Yet, important as these points are, in this study we concentrate on the role of fairness in shaping group identities within the police organization: the idea that group identity mediates the links between organizational justice, motivations and compliance. Police organizations have strong social identities, upon which officers draw to form their own self-image (Reiner, 2010; Waddington, 1999), suggesting that the nature and affective content of internal procedures will be important for most of them. Furthermore, individuals’ identities as police officers must to some extent be shaped by their treatment by the organization – as well as the links they develop with their co-workers in other ways – since, by definition, it has an existence clearly and definitively prior to, and separate from, their membership of it. It may thus make little sense to say that the effect of procedural fairness will vary according to the strength of group affiliation because procedural fairness, as an aspect of the way an organization treats its employees, is itself a cause of group affiliation. Naturally, we do not claim this argument to be definitive, and it would be a fruitful area for future research to examine whether justice effects vary across officers with different orientations towards the organization. However, concentration on the identity-forming aspects of organizational justice within police organizations seems, for the time being, to be justified.
The second issue is the relationship between procedural fairness, group identity and legitimacy. The concept of legitimacy plays a key role in the wider procedural justice literature, providing, for example, a bridge between the fairness of criminal justice agencies and people’s propensities to obey the law (for example, Jackson et al., 2012). There is no reason to suggest that similar processes are not at work within organizations, especially ones such as the police, whose very existence relies on internal and external processes of legitimation (Bottoms and Tankebe, 2012). However, we concentrate here on the link between procedural justice, group identity and compliance and cooperative behaviours, again leaving consideration of the role of legitimacy within police organizations to future research.
The third issue is the concern that procedural fairness, group identification and indeed legitimacy buy ‘blind obedience’ to rules and procedures. Kelman and Hamilton (1989) describe how strong identification with an institution or social group, and concomitant legitimation of its authorities, may serve to deaden individuals’ critical moral faculties and lead them to undertake actions that they might, in other circumstances, view as wrong and even morally repugnant. This is naturally a significant risk when the group involved is a police organization. If officers ‘over-identify’ with their organization, and merge personal and group goals to too great a degree, they may be more likely to follow instructions or policies that are normatively undesirable and detrimental to the public they are meant to serve. The danger that procedural fairness is a ‘double-edged sword’ (MacCoun, 2005) is particularly pertinent in the context of policing, and in this paper we examine the links between fairness judgements, group identification and two types of compliance. The first, soft compliance, refers to simple adherence to instructions and procedures. The second, hard compliance, refers to following instructions and orders irrespective of their substantive content (see below).
Summary and hypotheses
The commitment of police officers to organizational goals and practices may therefore be influenced by two sets of criteria: organizational justice and instrumental incentives. Both criteria may influence officers’ attitudes directly or via the extent to which they affect identification with the organization. These relationships are summarized in Figure 1, and the core hypotheses to be tested are:
Perceptions of organizational justice are associated with identification with the group (that is, the police organization).
Identification with the group shapes and informs officers’ motivations in relation to organizational citizenship (extra-role activity), empowerment (willingness to take initiatives and risks), adherence to organizational goals, and compliance with instructions and procedures.
Although much of any association between organizational justice and officer motivation will be mediated by group identification, some direct associations are expected (group identity will not mediate all the association between justice judgments and outcomes).
Procedural justice will be more important than distributive justice.
Instrumental factors will also shape and inform officer motivation, with, again, part of the association mediated by group identification. In line with existing literature, we expect these associations to be much weaker than those involving fairness judgements.
Conceptual map.
Research methods and data
The study reported here is based on a survey of police officers carried out in collaboration between Durham Constabulary and the College of Policing. 2 Durham is a small, non-metropolitan police force in the north of England, covering County Durham and Darlington. Durham’s organizational priorities were articulated to staff in 2011 via its ‘plan on a page’. Overall, the force aimed to ‘inspire confidence in victims and communities’, which it sought to do by ‘protecting neighbourhoods’, ‘tackling criminals’ and ‘solving problems’. Underpinning these goals there was a commitment to working together, motivating and developing staff, communicating effectively and reinforcing a culture of ‘excellence’.
Durham therefore had a strong interest in understanding what might motivate its staff to embrace community policing and develop service delivery. At the time of the study, however, and like many other forces in England and Wales, the force was going through a period of significant organizational change – resulting from budget cuts – involving a major structural reorganization and an impending reduction in staff numbers (most, although not all, of which was scheduled to occur among civilian staff rather than sworn officers). An important element of this process was a shift towards a non-traditional matrix management structure that maintained the traditional rank structure but reduced the overall numbers of Inspectors and Chief Inspectors. This change programme was one of the main reasons why Durham approached the College to help it develop an organizational survey. There was concern over how the programme was perceived by officers and staff, and how it was affecting people within the organization, but also an interest in developing a focus on issues of staff motivation and wellbeing, and to think about how improvements could be made in the future.
Organizational survey
The survey was carried out over a three-week period in June/July 2011 and administered online by Durham Constabulary. Every officer and member of staff was sent an email containing a hyperlink to the questionnaire, which had been programmed in Snap. A total of 1043 responses were received (a 40.2 percent response rate), consisting of 522 attributable to police staff (a response rate of 50.5 percent) and 479 to police officers (response rate 30.7 percent). 3 The analysis presented in this paper was based only on the police officer responses, because this was the more tightly defined group likely to be influenced by ‘cop culture’; police officers are also more likely to have contact with the public and are therefore best able to enact organizational priorities in relation to service delivery, as well, of course, as undermine them. 4
The profile of officer respondents appeared generally representative of the force as a whole. Some 25 percent of the sample was female, compared with 28 percent of serving officers in the force. Similarly, 24 percent of the sample was aged 18–35, compared with 27 percent of the force; for 36–45 year olds these figures were 46 percent and 47 percent, respectively, and for 45–55 year olds they were 25 percent and 26 percent. By contrast, constables were slightly under-represented in the sample compared with the force as a whole (64 percent compared with 73 percent), although the proportion of the sample who were sergeants and more senior ranks was comparable (for example, 17 percent of the sample were sergeants, compared with 16 percent of all officers). This disparity appears to have arisen because 8 percent of survey respondents did not report their rank. Like the Durham workforce, the sample was overwhelmingly white.
The questionnaire consisted of a total of 77 items (not including socio-demographic questions). Although some of the items had been included to measure issues of interest to the force, the majority were specifically designed to enable the analysis presented in this paper. Initially, a much larger number of questions had been designed or adapted from previous studies (principally, Tyler et al., 2007; and Tyler and Blader, 2003). These were reduced in number, and their wording refined, through discussion with the force; a small number of cognitive interviews with officers and a pilot questionnaire were also carried out in advance of the main survey.
Measurement issues
As suggested by Figure 1, three sets of variables were needed for analysis. The first were measures of what might be termed the ‘drivers’ of group identification and officer motivation – that is, measures of instrumental motivations and organizational justice. The second comprised measures of group identification. The third group included measures of attitudes concerning ‘organizational citizenship’ behaviours, orientation towards organizational goals and stated willingness to comply with organizational rules.
The survey fielded a number of questions relating to each of these subjects, all on Likert-type ordinal scales using seven-point numbered responses with two anchoring words (for example, ‘strongly agree’ and ‘strongly disagree’). We used latent variable techniques (in Mplus 6.2) to estimate, validate and model the key constructs required. Latent variables and observed indicators are shown in Tables 3 and 4 in the Appendix. Table 3 displays the input variables: threat of sanction, distributive justice, procedural justice and group identification. A five-factor Confirmatory Factor Analysis (CFA) model (with no cross-loadings) fitted the data well, according to the approximate fit statistics. The strongest correlation between latent variables in this model was .62, suggesting adequate discriminant validity. The five latent variables represented were:
Instrumental incentives
1. Threat of sanction
This variable represented respondents’ sense that they would be caught and punished if they did not meet work expectations. It was measured by three items, including ‘I would be held to account if I underperformed in my role’.
Elements of organizational justice
2. Distributive justice
This variable represented respondents’ assessment of the way pay and outcomes were distributed within the organization. It was measured by items such as: ‘Compared with other people in the force, I am satisfied with how well I am paid’; and ‘I am rewarded fairly for the work I do’. The variable therefore combines aspects of fairness and relative material reward.
3. Supervisory procedural justice
The perceived fairness of supervisors was measured by items such as ‘My supervisor gives me the chance to voice my opinion about decisions that affect me’.
4. Leadership procedural justice
Questions such as ‘Senior managers in this force don’t listen to the views of their staff’ (reversed) were used to measure perceptions of the procedural fairness of leaders and the wider organization.
Mediating variable
5. Organizational identification
Responses to two questions were used to derive a measure of identification with the force: ‘I have no emotional attachment to the force’ (reversed) and ‘I feel a sense of loyalty to Durham Constabulary’.
A similar process was used to derive measures of the response variables required, and question wording and model fit statistics are again shown in Table 4 in the Appendix. A total of five variables were needed:
Officer motivations
1. Extra-role activity
This variable represented respondents’ self-assessed willingness to do extra work and help others when required (but not necessarily for personal gain). It was measured by items such as ‘I am happy to take on extra work to help other people’.
2. Empowerment
Officers’ sense that they were empowered to use their initiative was measured by three items, including ‘I am confident about using my own judgement at work’.
3. Citizen focus
Attitudes towards the public and a service model of policing were measured by indicators such as ‘It is a waste of time trying to help some members of the public’.
4. Soft compliance
Officers’ self-reported soft compliance was measured by four items, including ‘I do what my supervisor asks me to do’ and ‘I follow the correct processes at work’. It represents stated willingness to follow the instructions of line managers and self-reported propensity to follow wider organizational rules and policies.
5. Hard compliance
Respondents’ sense that they simply followed all rules, orders and policies regardless of their content was assessed via three items, including ‘I follow my supervisor’s instructions even when I think they are wrong’. The aim was for these items to be qualitatively different from those relating to soft compliance because they each contained a qualifying statement that emphasized the contingency upon which compliance was to be based, thus providing a ‘harder’ test of a respondent’s reported willingness to follow orders. This variable was very strongly correlated with soft compliance; we show below, however, that its associations with other variables were somewhat different.
Results
A total of five Structural Equation Models were estimated. Each predicted one of the response variables outlined above, and each contained the four explanatory variables as well as organizational identification as a mediator. The general form of the models is shown in Figure 2, and the results are shown in Tables 1 and 2.
General form of Structural Equation Models estimated.
Models predicting extra-role activity, organizational identification, and citizen focus (standardized regression coefficients).
Notes: All variables are coded such that high equals more.
p < .0005, ** p < .01, * p < .05; . indicates path not applicable.
Models predicting compliance (Standardized regression coefficients).
Notes: All variables coded such that high equals more.
p < .0005, ** p < .01, * p < .05; . indicates path not applicable.
Looking first to Model 1 (Table 1), we find that, as predicted, both procedural and distributive justice factors were associated with identification with the organization. Leadership procedural justice had the strongest link with organizational identification, followed by distributive justice and supervisory procedural justice. Threat of sanction, by contrast, had no significant association with identification. Turning to the ultimate response variable in this model, self-assessed propensity to take on extra work was significantly, and positively, predicted by organizational identification and supervisory procedural justice, with group identification having the largest unique statistical effect by some margin. Procedural justice at the supervisory and leadership levels and distributive justice also had significant indirect effects via group identification.
Models 2 and 3 in Table 1 repeat the above findings in relation to the factors shaping identification with the organization. They differ, however, in relation to the factors influencing the ultimate response variables. In Model 2, empowerment was directly predicted by all the independent variables bar risk of sanction; distributive justice and procedural justice at both supervisory and leadership levels also have indirect effects via organizational identification. Model 3, by contrast, shows that only procedural justice at the leadership level and organizational identification were directly associated with holding positive attitudes about the public, although, again, distributive justice and supervisory procedural justice had significant indirect associations via organizational identification.
Identification with the police organization, then, is enhanced when officers feel fairly treated by supervisors and the organization as a whole, and when they feel rewards within the organization are fairly distributed. Identification with the organization is, in turn, linked to organizational citizenship behaviours (for example, extra-role activity), a sense of empowerment, and support for organizational goals (for example, community policing). Importantly, group identification did not mediate all statistical effects in all the models and there were a number of direct associations between the fairness of supervisors and/or the organization and the various outcome variables. This suggests that the experience of procedural justice has an influence on officers’ attitudes above and beyond the extent to which it shapes their identification with the police as an organization. By contrast, threat of sanction had no significant association with either group identification or officer motivation in any of the three models. Finally, note that the R2 values for the three response variables were high (and above .5 in two cases).
Table 2 turns to the issue of compliance. Results from two models are shown, predicting soft compliance (Model 4) and hard compliance (Model 5), respectively. In each, the component predicting organizational identification was the same as before. Yet predictors of the outcome variables varied significantly from the previous models. Strikingly, but perhaps not surprisingly, threat of sanction was a significant predictor of compliance. Equally striking, and more surprising, was that organizational identification was not a significant predictor in either model; neither was leadership procedural justice or distributive justice. Supervisory procedural justice was, however, associated with both forms of compliance, although it is interesting to note that the regression weight in Model 4 was almost twice that in Model 5 (.37 compared with .20); the perceived fairness of supervisors appears more strongly associated with soft than with hard compliance. Note also that the R2 for the compliance variables – particularly hard compliance – was much lower than in the other models.
Discussion
The models presented above provide evidence that police officers’ motivations across a range of ‘positive’ behaviours are associated with their perceptions of the fairness of both their immediate supervisors and the organization as a whole. Returning to the hypotheses laid out above, we found, first, that perceptions of organizational justice were associated with group identification. Second, identification with the group was associated with officer motivations across a range of behaviours. On average, those who felt more emotionally attached and loyal to the force were more likely to say they were ready to take on discretionary activities, and felt more confident and empowered to use their own initiative. They were also apparently more in tune with community policing.
Third, much, but not all, of the statistical effect of all three elements of organizational justice on motivations was mediated by identification with the organization. The exception was compliance: those who felt their supervisors treated them fairly also reported being more ready to comply with instructions and procedures, and this relationship was not mediated by identity judgements. Fourth, procedural justice was in general the most important component of organizational justice.
Finally, the threat of sanction predicted only compliance, and there was no link between sanction and organizational identification. As such the emphasis traditionally placed on instrumental incentives in performance management regimes seems misplaced. Positive orientations towards organizational goals in Durham were seemingly not fostered by threatening officers with sanctions for non-compliance.
Our findings cast new light on the links between police occupational identity and officers’ behaviours. There are two schools of thought in relation to the potential effects of police identities on behaviour. For some commentators, occupational culture operates as a psychological defence mechanism to help officers deal with the sometimes brutal realities of imposing authority and order in what can be chaotic and dangerous situations. Although officers may express undesirable views and opinions in the metaphorical arena of the canteen, this shared identity does not necessarily influence their actions when they do not have an audience of their peers (see Waddington, 1999). For others, culture is ‘figurative action’, and the shared identity built through the hearing and telling of stories can influence the decisions that officers make in the field (Shearing and Ericson, 1991). In this latter case, strong group identities within police organizations may have problematic implications for the ways officers interact with members of the public.
Studies of police culture have tended to emphasize this second, more negative, account. However the data presented here suggest it is at least possible that the experience of organizational justice within the police will foster positive identities based on a culture of cooperation and fair process that, in turn, will influence the way officers view and interact with the public. Far from creating a barrier to positive attitudes, identification with the organization was consistently associated with stronger motivations or self-assessed propensities across a range of desirable behaviours, including towards community policing. We would note, however, that Durham was already committed to implementing a service model of policing. It may be that a pre-existing commitment to such goals is necessary for these relationships to develop – that is, for identification with the organization to encourage positive attitudes towards citizen-focused policing, the organization may itself already need to be associated with such policing models.
Yet organizational identification did not entirely mediate the statistical effects of the other explanatory variables on officer motivations. We again note that we omitted consideration of organizational legitimacy from this study. It may well be that feeling fairly treated by the organization encouraged a sense among officers that its structures and processes were legitimate, and in turn promoted a feeling that, say, a community or service model of policing should be followed because this was the model mandated by the force. Future work might profitably attempt to examine legitimacy within police organizations to see if this is indeed the case.
Finally, it seems that procedural fairness can indeed foster a certain level of unthinking compliance with instructions. However, this was the case only in relation to supervisory procedural fairness and, despite our fears to the contrary, it does not seem that feeling a strong sense of identification with the organization in and of itself encourages officers to feel they should comply with instructions even when they think they are wrong. Procedural justice may indeed be somewhat ‘double-edged’ (MacCoun, 2005), for example because it encourages a level of blind loyalty within work teams. Yet it seems, on the basis of the evidence presented here, that any threat to the development of officers’ ability to make critical judgements about the orders given to them may be rather small, and, arguably, outweighed by the benefits relating to organizational citizenship.
Limits of the study
We acknowledge that this study has a number of limitations. First and most importantly, we measured attitudes not behaviours. The extent to which positive attitudes or motivations in relation to, for example, community policing translate into behaviours ‘on the street’ is an open and empirical question, and would be a fruitful area for future research. 5 Relatedly, there is the possibility that the survey participants ‘gamed’ their responses, providing answers they knew to be the ‘right’ ones in a given case. It is worth noting, in this regard, that there was significant variation in responses to many of the survey questions. For example, 45 percent of respondents agreed to some extent with the statement ‘some victims of crime are more deserving of a good service than others’, while 25 percent failed to agree that they felt ‘a sense of loyalty to Durham Constabulary’.
A further limitation is that our data were cross-sectional, meaning that we could assess only correlations between constructs rather than causal links. This problem is insurmountable, given the data available. Yet the fact that the survey was specifically designed to answer the type of questions posed in this analysis may provide some reassurance that we are at least accessing possible causal links that future research designs might build on.
Durham is a small, relatively rural force serving an area with few major urban centres. Whether the results presented here apply in more metropolitan areas is another open, empirical question that again warrants further investigation; we also note again the relatively low response rate, although, as described above, the sample was broadly reflective of the age, gender and rank structure of the force. Perhaps more importantly the survey was conducted at a time when the organization was confronting considerable, stressful, change and Durham, like all other forces in England and Wales, was awaiting news of significant staff reductions. Although the employment status of sworn police officers means the respondents in the officer sample would not have felt at personal risk of redundancy, 6 minimizing the possibility of bias to our findings arising from some unidentified portion of the sample directly fearing for their jobs, the prospect of staff cuts formed an important backdrop to the survey and may have influenced our results in more indirect ways. Future work might, for example, profitably examine whether officers are more or less sensitive to issues of organizational justice during periods of change or when they feel their organization is under threat in some way.
Finally, we have paid no attention here to the organizational structure of the force; advisedly so, since our interest was to establish the basic contours of a model of organizational justice as this might play out in a police organization. It remains an open question, therefore, whether individuals’ positions within the hierarchy, or their relationships within and across work units, influenced the answers they gave. Again, it would be fruitful for future research to assess whether organizational justice is more or less important to junior or senior officers, or whether those working in particular parts of the organization place greater emphasis on justice issues than their counterparts working elsewhere.
Conclusion
Since carrying out the survey, and based in large part on the evidence presented here, Durham Constabulary has made a number of changes to its staff management processes, invested significantly in staff development and begun to think more directly about issues of fairness. Force managers aimed to use organizational justice approaches to consider, first, how a difficult period of change could be best navigated and, second, how officers and police staff could be better motivated in relation to process and community-based policing.
Results from the survey were used to identify three work streams: staff wellbeing, communication, and recognition and reward. A series of internal focus groups were carried out to explore whether officers and staff felt valued and empowered by the organization and to understand the effects of organizational change. Along with the organizational justice research, this process clearly identified the importance of values for the relationship between management and staff, and led to greater organizational investment in values-based leadership and consideration of what this means for the force. Understanding of the importance of supervisor behaviour for relations within the organization, and between police and community, fed into training at the middle management level (Chief Inspector and Inspector levels, and police staff equivalent) that was specifically designed to encourage supervisors to think differently, to underline that they had the organization’s support to make the right decisions, and, at a time of reductions in staff numbers, to empower them to try new ideas and ways of working. Senior leaders have also been encouraged to meet regularly and reflect on practical issues faced by the force and their own role in solving them. Importantly, routine performance meetings – which in the past largely focused on crime, detections and public perceptions – have been revised to include discussion of the nature and quality of the relationship between managers, supervisors and staff.
The findings of this study and the work conducted by Durham suggest that organizational justice may be a fruitful avenue of exploration for police managers seeking to encourage in staff the extra commitment, initiative and citizen-focused approach that is crucial to the implementation of service-based policing models. Although the threat of sanction will always remain a viable and necessary motivational tool in an organization such as the police, this study suggests sanctions should be used with care, not least because the perceived fairness of such processes may have a bearing on the way officers identify with the organization and, consequently, their support for its policies in the long run. Moreover, it is interesting that, in respect to the aspects of workplace relations covered here, the police do not appear so different from other organizations. Our findings are very much in line with those from studies conducted among employees from a wide range of workplaces (Colquitt et al., 2001). On this basis, at least, claims about the uniqueness of police organizational culture may be somewhat exaggerated. Furthermore, the wider literature on organizational justice suggests that processes and interpersonal treatment perceived to be unfair are a significant organizational risk. In the context of cuts to both budgets and personnel, police organizations might do well to give their officers and staff a voice in organizational change programmes and ensure that managers and supervisors are seen to make fair decisions and to communicate those decisions openly and honestly.
Footnotes
Appendix
Outcome variables: Factor loadings from five-factor model with no cross-loadings.
| Extra-role activity | ||||||
| I am happy to take on extra work to help other people | 0.76 | |||||
| I’d go the extra mile at work if it helps Durham Constabulary | 0.78 | |||||
| I’m reluctant to take on more work than my share unless I benefit directly (reversed) | 0.66 | |||||
| Empowerment | ||||||
| I feel trusted to take important decisions on my own | 0.76 | |||||
| I am confident about using my own judgement at work | 0.75 | |||||
| I have the flexibility in my job to try out new ways of working | 0.76 | |||||
| Citizen focus | ||||||
| Some victims of crime are more deserving of a good service than others | 0.85 | |||||
| It is a waste of time trying to help some members of the public | 0.93 | |||||
| Some people do little to earn the respect of the police | 0.73 | |||||
| Soft compliance | ||||||
| I do what my supervisor asks me to do | 0.81 | |||||
| I ignore my supervisor’s instructions (reversed) | 0.69 | |||||
| I stick to the approved way of doing things when I am at work | 0.80 | |||||
| I follow the correct processes at work | 0.85 | |||||
| Hard compliance | ||||||
| I follow my supervisor’s instructions even when I think they are wrong | 0.66 | |||||
| I do what I am told by my supervisor without question | 0.60 | |||||
| I comply with force policies even when I think they are wrong | 0.88 | |||||
| Chi-square | 328.011 | |||||
| Degrees of freedom | 94 | |||||
| p-value | <.00005 | |||||
| RMSEA | 0.07 | |||||
| CFI | 0.97 | |||||
| TLI | 0.96 | |||||
| Correlation matrix of latent variables | ||||||
| 1 | 2 | 3 | 4 | 5 | ||
| Extra-role activity (1) | 1 | |||||
| Empowerment (2) | 0.71 | 1 | ||||
| Citizen focus (3) | 0.55 | 0.45 | 1 | |||
| Soft compliance (4) | 0.39 | 0.20 | 0.23 | 1 | ||
| Hard compliance (5) | 0.17 | 0.10 | 0.14 | 0.88 | 1 | |
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.