Keywords

Introduction

From the start of European settlement in the eighteenth century, law enforcement in Australia was beset by the same complex problems of unethical conduct as occurred in many locations. Expanding colonial policing was haphazard and frequently ­corrupt (Bryett et al. 1997). The introduction of colonial self-government in the nineteenth century involved attempts to make police more professional and accountable. However, officers were generally held in low regard, discipline was erratic and discretion was difficult to manage in the highly dispersed policing environment (Bryett et al. 1997; Prenzler 2010). Victorian and post-Victorian era prohibitions or restrictions on alcohol, gambling, and prostitution set the conditions for organized protection rackets that continued well into the twentieth century.

Independence from Britain in 1901 resulted in a federal system based on the former colonial boundaries. With no local police, law enforcement responsibilities were located within the six states: New South Wales, Queensland, South Australia , Tasmania, Victoria, and Western Australia. A small commonwealth force was established in 1917, which later became the Australian Federal Police (AFP) . The AFP is responsible for policing crimes against the commonwealth, such as drug trafficking, and also carries out regular police duties in the Australian Capital Territory. The eighth policing jurisdiction is the Northern Territory .

Numbers of sworn officers, as reported in police annual reports for 2012–2013, reveal the New South Wales Police to be the largest of the eight police departments, with approximately 16,000 sworn officers (around 20,000 total employees), followed by Victoria Police (13,000 sworn officers; 15,000 total staff) and Queensland (11,000 sworn; 15,000 total). Midsize departments are the AFP, Western Australia and South Australia, with between 5000 and 7000 sworn officers. Tasmania and the Northern Territory are the smallest departments with 1100 and 1500 sworn officers, respectively.

Police Integrity in Australia

Until the late 1980s, Australia’s police enjoyed high levels of discretion with little oversight. After widespread allegations of police abuses, a Royal Commission into the Queensland Police Service began a period of extensive reform, not just for Queensland, but paving the way for integrity research and future inquiries (e.g., The Wood Royal Commission in New South Wales 1997; The Kennedy Commission into the Western Australia Police in 2004) . In Queensland, the Criminal Justice Commission was formed in 1989 not only with responsibility for overseeing the Queensland Police, but also with an active research agenda. All Australian police departments are now subject to oversight agencies. Most agencies not only respond to allegations of police misconduct and corruption in a traditional reactive model of oversight but also proactively engage in prevention, research, and education about wrongdoing to improve the ethical health of police organizations. This has included attempts to measure ethical climate through surveying the ethical attitudes of police officers (Crime and Misconduct Commission 2010; Huon et al. 1995).

The premise of this “ethical climate” approach was the move away from the “rotten apple, theory of police misconduct and acknowledgement that integrity is the responsibility and product of agencies’’ own organizational practices. In other words, organizational culture does not simply reflect the sum of individual morality. As Huon et al. (1995, p. 1) state, ‘‘the system supports corruption via (i) an opportunity structure; (ii) on-the-job socialisation; and (iii) peer group reinforcement and encouragement of certain rule violations (O’Brien 1991).” In order to build a culture of integrity, agencies must set clear expectations for behavior, with clear rules and clear consequences when rules are violated. Agencies must also combat a possible “code of silence” that can protect officers from the consequences of violations by allowing behavior to remain hidden (Fitzgerald 1989).

In the USA, the work of Klockars et al. (2000) highlighted these aspects in the form of an ethical scenario survey that has since been replicated in multiple police departments over the world (Klockars et al. 2004). The survey tapped officers’ ethical attitudes not only in relation to a variety of behaviors but also in relation to their understanding of official rules, support for the code of silence (willingness to report others) , attitudes towards control mechanisms (support for disciplinary responses), and expectations of others (police culture). The following sections discuss each of these aspects in relation to the Australian policing climate .

Organizational Rules

The normative frameworks of police misconduct and associated responses are primarily dependent upon two mechanisms: formal rules (societal laws, organizational policies, etc.) and informal extra-legal perspectives (societal beliefs, public opinion, police culture) . The former can be enforced through formal processes, while the latter is more difficult to circumscribe. Formal rules govern behavior through describing expected standards and prescribing expected consequences for breaching those standards. Thus, rules lay down the foundations of any integrity system .

In Australia , police departments are governed by legislation at the commonwealth level and at the state (or territory) level. Each department, therefore, has a separate system of rules and processes for rule enforcement and their own codes of conduct and formal disciplinary processes. Broadly speaking, however, the evolution of police governing legislation and codes of conduct has been characterized by convergence in terms of more detailed prescriptions, consistent with the standards set out in international police codes of conduct, such as the United Nations (1979) Code of Conduct for Law Enforcement Officials and the International Association of Chiefs of Police (2002) Law Enforcement Code of Conduct.

The major inquiries outlined above, and ongoing scandals, have driven the refinement of laws and codes designed to clarify unacceptable conduct in areas such as conflicts of interest, information security, use of force, and off-duty behavior (Porter and Prenzler 2012a). This tightening of rules has included mandatory reporting of misconduct. The main substantive differences lie in enforcement practices. As one example, in 2011–2012, a scandal forced the Queensland Police Service to adopt a more explicit policy, and an internal communications strategy, to try to reduce police acceptance of gifts and benefits (Prenzler et al. 2013). Several years earlier, the police response to a similar scandal in Victoria became lost in bureaucratic inaction.

In 1991, as part of a larger program on police integrity issues, the National Police Research Unit (NPRU) launched an ethical climate survey that tested officers’ understanding of the inappropriateness of a variety of police actions (Huon et al. 1995, p. 26). A questionnaire was developed with 20 scenarios involving a hierarchy of types of misconduct—including assault, bribery, theft from a crime scene , modifying a statement, sleeping on duty, gratuities , cheating on assessment and personal identification checks. Each scenario required the respondent to rate the seriousness of the breach, their perceptions of colleagues’ and the department’s view, and their willingness to report incidents and to whom. Huon et al. (1995) distributed the questionnaires to police officers and recruits in the six Australian states and to senior officers attending a national training facility. A total of 683 responses were received to the survey. The overall mean for officers’ personal views was 6.7 (with 10 being most serious), with the department’s view put at 8.6 and the “typical officer’s” view put at 5.6. Thus, on average, respondents saw a divide between the police culture and the formal organizational position, with the former taking the infractions considerably less seriously than the latter. For personal views, the highest means were 8.2 for reckless driving, 8.0 for altering a rapist’s statement, and 7.9 for minor theft (cigarettes) from a crime scene. The lowest means were 5.3 for checking the ID of an attractive woman, 5.3 for a speeding officer attempting to avoid a fine, and 5.0 for an officer speaking rudely to a young person. Recruits viewed the scenarios most seriously, low- and middle-ranking officers least seriously (constables, senior constables and sergeants), and senior officers were midway. There was a similar decline across length of service, but rose back up from the 10–20 years’ service mark. Female officers viewed the scenarios more seriously than males.

The other main application of ethical climate surveys in Australia has been in Queensland, where the oversight agency surveyed recruits and first-year constables each year from 1995 to 2008 in order to monitor the impact of reform. Eight scenarios based on the NPRU survey were used in 1995, with 12 scenarios in 2008. A major review covering 14 surveys to 2008 was published in 2010 (Crime and Misconduct Commission 2010). The results were similar to the NPRU study, although with slightly higher overall means. For personal views, across all surveys, on a scale of 1–10, the highest (most serious) means were 9.8 for selling confiscated drugs, 8.7 for minor theft (cigarettes) from a crime scene and 8.6 for altering a rapist’s statement. The lowest means were 5.2 for accepting cartons of beer at Christmas from a publican who requested extra patrols, 5.8 for a personal job while on duty, and 6.3 for an officer speaking rudely to a young person. Indeed, the review noted an excessively tolerant attitude amongst first-year constables towards gratuities and personal jobs on police time and considerable reluctance to report misconduct—especially intermediate- and lower-level types. Across the years, recruits had fairly stable views. There was a slight trend upwards in some seriousness ratings by first-year constables. First-year constables had similar views to recruits for the four most seriously rated scenarios and lower ratings for most of the remainder—around one point on average on the 10-point scale. Females generally viewed scenarios more seriously.

Control Mechanisms

The existence of formal organizational rules, while setting expected standards, requires the establishment of enforcement mechanisms. Historically, Australian policing was characterized by recurring allegations of misconduct and a series of hamstrung, largely unsuccessful, judicial inquiries (Prenzler 2010). From the 1970s, the Victorian police ombudsman produced a series of investigative reports revealing widespread abuses, including kickbacks for emergency security notifications, unjustified shootings of mentally ill persons, violent interrogations, excessive political surveillance, sexual harassment and sex discrimination, harassment of police whistleblowers , abuse of strip searching, theft and on-selling of drugs, and leaking of information to criminals (Office of Police Integrity 2007a, 2009).

A major breakthrough in Australian police reform occurred with the Fitzgerald Commission of Inquiry in Queensland (1989), which revealed a set of police abuses centred on legal process of corruption and protection of vice. Major reforms included professional recruitment, ethics training, and civilian oversight . In New South Wales, the Wood Commission of Inquiry , which ran from 1994 to 1997, exposed diverse forms of corruption including routine assaults and excessive force, protection of organized crime, evidence tampering, opportunistic thefts, and extortionate gratuities (Wood 1997). More recently in Western Australia , the Kennedy Inquiry (2004) revealed diverse criminal conduct by police, including assaults, stealing, perjury, drug dealing, and disclosures of confidential information.

Over this period, police reform and improved accountability followed a stop–start pattern, with limited civilian oversight introduced from the 1970s. Police managers and police union leaders were generally resistant to external investigations and auditing of complaints processes (Prenzler 2010). Nonetheless, recurring scandals led to all police departments becoming subject to independent oversight. Australia has also been the site of a number of innovative integrity management strategies, including officer profiling and early intervention, covert operations and drug and alcohol testing (Porter and Prenzler 2012a). There were improvements in the reporting of complaints, and, in a number of departments, complaints were systematically analyzed to inform modified training and procedures (Porter and Prenzler 2012a).

More recently, attention has turned to the internal disciplinary systems of police departments, with increased awareness of the impact of perceived fairness and transparency on officer behavior. While traditionally punitive, for the purposes of deterrence, disciplinary systems now incorporate a number of mechanisms to improve the experience of officers, particularly for minor behavioral issues. For example, mediation and local resolution of complaints have been trialled in Queensland and Victoria in order to avoid protracted investigations with unsubstantiated findings (Porter and Prenzler 2012a). Early intervention systems have also been developed in many jurisdictions to highlight problematic behavior and deal with issues in a remedial rather than punitive manner. The acknowledgement of procedural fairness has also led to refining disciplinary processes in Queensland, New South Wales, and Victoria, providing officers with more information and opportunities to respond to allegations against them (Porter and Prenzler 2012a).

Code of Silence

Despite formal rules and control mechanisms, police culture has been raised as a powerful obstacle to rule adherence, with informal norms taking precedence over formally prescribed standards (Porter and Prenzler 2012a). The most widely cited of these is the “code of silence” that encourages officers to ignore wrongdoing they are witness to, breeding a culture of tolerance and even active workplace harassment. For example, the Queensland Crime and Misconduct Commission review (CMC 2010) of 14 ethical climate surveys identified reluctance to report misconduct as an ongoing problem, along with the view that discipline was harsh but with a low likelihood of misconduct being detected.

These types of findings support the theory that the perceived fairness of the system for responding to violations is a key factor in increased willingness to report colleagues’ wrongdoing. That is, it is expected that if members view the organizational response to be too harsh, they will be less likely to report wrongdoing than if the response is understood to be legitimate and proportionate to the behavior (Klockars et al. 2004). Kutnjak Ivković and Shelley (2010) found that officers self-reported less willingness to report misconduct where the expected discipline was viewed to be harsh compared with when discipline was viewed to be fair. Similarly, “organizational justice” research shows that feelings of unfair treatment can negatively affect job performance and rule adherence (Tyler et al. 2007). In contrast, perceptions of organizational justice have been positively linked to police officers’ “attitudes towards serving the public” (Myhill and Bradford 2013, p. 339) as well as a reduced likelihood of police officers supporting the code of silence and noble cause corruption (Wolfe and Piquero 2011).

Huon et al.’s (1995) questionnaires revealed a rough correlation between officer rank and willingness to report . For recruits and low- and middle-ranking officers, the preferred response was “no action” or “raise informally with a senior officer.” Senior officers tended to support reporting the incident to Internal Affairs. The report remarked on the decline in standards in the early and middle years of officers’ careers and concluded that training in ethics needs to be reinforced.

Public Expectations and Influences

In addition to informal influences within police organizations, the external environment has also been a powerful driver of police attitudes and behavior. In theory, negative public attitudes towards police can increase officer expectations of opposition and resistance, causing police to be less tolerant and more likely to escalate the amount of force used to do their duty (Smith and Hawkins 1973).

In Australia, there has been recurring friction between police and different groups, especially indigenous Australians and some ethnic minorities. On the whole, however, community satisfaction with police is very high (Prenzler and Sarre 2012). Surveys of the public have consistently shown general satisfaction levels around 75 %, with similar levels of support for specific questions about fair and equal treatment. For those survey respondents who had contact with police in the previous 12 months, satisfaction goes up to around 85 %. At the same time, there is clearly a very dissatisfied minority, with complaint numbers consistently high in the tens of thousands each year, including a large component concerned with excessive force.

The remainder of this chapter explores these integrity constructs in Australian policing through utilisation of a similar self-report survey to those used in the studies outlined above. Questions in response to a variety of scenarios measure both the attitudes of officers towards different infractions and their perception of the culture around these infractions (the majority of attitudes and the organizational response).

Method

Materials

The survey used in this study was based on the questionnaire developed by Klockars et al. (2000) in their U.S. National Institute of Justice study. The survey had been updated by Klockars et al. and was then reviewed by the two Australian police agencies who agreed to participate. The questionnaire contains two sections: ethical scenarios and background questions. There were 11 ethical scenarios (short descriptions of incidents) each followed by the same set of questions to measure constructs relating to perceptions of seriousness, rules, discipline, and code of silence. The 11 scenarios are provided in the introductory chapter. Minor modifications were made to suit Australian spelling and terminology. Scenario two was modified as follows (additions in italics):

A police officer is aware that there is an arrest warrant for a long time friend of his. Although he sees his friend frequently over a period of more than a week and warns his friend of its existence, he does not arrest him or pass on information about his friend’s whereabouts to other police.

The scenarios cover a range of behavior including rudeness, theft, receiving gifts and benefits, excessive force and failure to perform duties. Motivations for personal gain and organizational gain are represented, as well as the behavior of supervisors. Responses to the scenarios measure the following aspects of officers’ views.

Seriousness

Officers’ perceptions of the seriousness of each scenario were measured by a single item with a 5-point scale from 1 = “not at all serious” to 5 = “very serious.” Officers were also asked to rate how seriously they believe most officers would view the scenario, using the same scale. The officers’ own perception of the seriousness of the scenarios sheds light on their own integrity attitudes, while their perceptions of “most officers” indicate the integrity attitudes ascribed to by the broader police culture.

Violation

Officers’ understanding of the scenarios in relation to agency policy was measured by a single item that asked, “Would this behavior be regarded as a violation of official policy in your agency”? Responses were measured on a 5-point scale from 1 = “definitely not” to 5 = “definitely yes.” This item is an important measure of officers’ understanding of agency expectations regarding behavioral standards.

Discipline

Officers’ views on discipline were measured by two items. For each scenario, officers were asked to indicate the level of discipline they think should occur in response to the behavior, and which they think would occur in response. For each item, officers were presented with a range of six discipline options: 1 = “none”; 2 = “verbal warning/counselling”; 3 = “written warning ”; 4 = “suspension/disciplinary transfer”; 5 = ‘‘reduction in rank’’; 6 = “dismissal.” The first item, regarding the discipline that is believed should follow, is an important indicator of officers’ own integrity perceptions, while the second item that measures the perception of what discipline would follow is an important indicator of officers’ expectations of the agency response.

Further, officers’ perceptions of what discipline should and would follow can be used to calculate an index of fairness (the difference between these). Fairness is an important dimension of a disciplinary system, which has been linked to employee behavior. Disciplinary fairness was calculated by subtracting the level of fairness the officers believe should follow from the level of fairness they believe would follow (would–should). If this yields a positive number, the discipline that would follow is expected to be higher than the level that should occur (harsh discipline). A negative number indicates that the level that would occur is expected to be lower than the level that should occur (lenient discipline). No difference indicates that discipline is viewed to be at the level it should be (perceived fairness).

Reporting

The officers’ willingness to report the behavior described in the scenarios was measured by a single item that asked, “Do you think you would report a fellow police officer who engaged in this behavior”? Responses were measured on a 5-point scale from 1 = “definitely not” to 5 = “definitely yes.” One further item asked officers, “Do you think most police officers in your agency would report a fellow police officer who engaged in this behavior,” using the same rating scale. These two items are important measures of the reporting culture, or the ‘‘code of silence,” for the agency. While the first item indicates the officers’ own comfort in reporting, the latter item refers to the officers’ perception of broader agency culture (the norm).

Background

Following the scenarios, questions were asked regarding the respondents’ years of experience in the police, their rank, current assignment, and supervisory status.

Honesty

To gauge honesty, respondents were asked if they thought most officers would give their honest opinion when completing the survey, and whether they themselves were honest in their completion. The analyses that follow did not include responses from officers who said that they were not honest in their answers.

Procedure

The survey was hosted online by Qualtrics and a weblink distributed to all police officers internally by personnel at two Australian police agencies. The agencies required confidentiality as a condition of participation (thus, they are not named specifically here) and respondents were not required to provide identifying information. The survey remained open for 3 months in 2013. At the end of this period, the survey was closed and data downloaded into statistical package for the social sciences (SPSS) for cleaning and analysis.

Response

Table 3.1 shows the characteristics of the sample of 856 officers who completed the questionnaire. Unfortunately, this represents only around 5 % of the total numbers of sworn officers employed by the agencies at the time of the survey. At the same time, the representativeness of the sample was assessed by comparing it to figures published in police agency annual reports for the 2012–2013 period. Males made up 62 % and females made up 25 % of the sample (13 % did not disclose): a figure similar to that for all Australian police agencies. Senior officers were somewhat over-represented in the sample with general duties assignments under-represented.

Table 3.1 Characteristics of the sample (n = 856)
Full size table

The modal length of service for the sample was more than 20 years. In fact, around 60 % of the sample had served more than 10 years, showing an experienced sample. Less than 5 % had served under 1 year. The distribution of experience (time in service) reflects the rank distribution of the sample, with a skew to higher ranks that would be reached after at least 5 years in service. A variety of roles were evident in the sample, including specialist assignments.

Findings

Table 3.2 provides a summary of how the 11 scenarios were scored according to the four main survey constructs: seriousness, violation knowledge, discipline, and willingness to report. The table is ordered by the officers’ views of seriousness, from lowest to highest. It is evident that there is considerable consistency in this rank order across the remaining constructs, with some small deviations (i.e., the order of seriousness is similar to the order in which respondents would place the scenarios for severity of discipline and willingness to report, etc.). This would suggest that the constructs are highly related and part of a more general construct of integrity, as suggested in the literature (Klockars et al. 2000, 2004).

Table 3.2 Officers’ views about seriousness of misconduct, discipline it should and would receive and officers’ willingness to report
Full size table

The results pertaining to each construct are reported in turn below, followed by a more detailed analysis of how the constructs relate to one another and, particularly, officers’ willingness to report. Finally, differences according to sample characteristics are explored, including gender, length of service, rank, and supervisory status. Where differences are explored for significance with t-tests, Cohen’s d is also provided to signify the magnitude of the difference (the effect size). An effect size of 0.2 is considered small, 0.5 medium, and above 0.8 accepted as a large effect size (Cohen 1988).

Seriousness

Table 3.3 shows that, on average, all scenarios were viewed as being at least somewhat serious (the mean being at the scale midpoint or higher). There appear to be three groups of scenarios according to the officers’ own perceptions of seriousness. First are those that constitute lower levels of seriousness, which for this sample means a score of less than 4. These are scenario 7 (verbal abuse of motorist), scenario 1 (receiving gifts from small businesses), and scenario 5 (being granted leave in exchange for running errands). These constitute demeanour and personal gain infractions.

Table 3.3 Officers’ views about the seriousness of misconduct
Full size table

The second group represents more serious behavior, scoring between 4 and 4.6. These are scenario 6 (punching an offender in custody), scenario 8 (failure to report officer’s driving under the influence, DUI), scenario 2 (failure to arrest friend with warrant), and scenario 11 (Sgt. failing to intervene in assault of suspect). These cases represent failures to perform duties, particularly due to internal loyalties, as well as excessive force towards offenders and suspects.

The final group was judged to be the most serious, with scores between 4.8 and 5 (the top of the scale, representing “very serious”). These are four scenarios: scenario 9 (receiving discounts for referrals), scenario 10 (falsely reporting finding evidence), scenario 4 (shooting an unarmed person), and scenario 3 (theft of knife from crime scene). These constitute a variety of examples of misconduct, including an abuse of position for personal gain as well as organizational gain (padding evidence to gain a conviction), and excessive use of lethal force.

Table 3.3 and Fig. 3.1 show that, in terms of the relative seriousness between each scenario, respondents believed their views to be similar to “most officers” (the rank order of seriousness is almost the same for officers’ own view and their view of most officers). However, the degree of seriousness attached to each scenario was consistently viewed to be different: Respondents on average believed they view each scenario as being more serious than do most officers (p < 0.001), with medium effect sizes.

Fig. 3.1
figure 1

Mean officers’ ratings of seriousness and their perception of the views of most officers

Full size image

Violation of Official Policy

Table 3.4 shows that, on average, respondents believed all the scenarios to be a violation of official agency policy. Across the scenarios, only small percentages of respondents (typically less than 5 %) believed the behavior was not a violation (scoring 1 or 2 on the scale). A proportion of respondents was unsure (represented by the midpoint of the scale). In particular, nearly 15 % were unsure that verbal abuse of motorist (scenario 7) was a violation of policy, and 10.5 % were unsure that granting leave in exchange for errands was in violation of policy. These scenarios are amongst the less serious, as judged by the sample. In contrast, more than 90 % of the sample were sure that scenarios 11, 9, 10, and 3 were definitely violations. These represent some of the most serious scenarios (as rated by the sample): the supervisor turning a blind eye to violent behavior, discounts for referral of business, falsifying evidence and theft.

Table 3.4 Would this be regarded as a violation of official policy in your agency? (rank ordered by seriousness—least serious to most serious)
Full size table

Discipline

Table 3.2 also shows that the discipline level that respondents’ believe should be applied to each scenario largely followed the same pattern as the seriousness attributed to the scenario. The three least serious scenarios attracted mean discipline levels between 2 and 3, representing verbal and written warnings. The four scenarios rated on average as somewhat serious are prescribed sanctions amounting to suspension , transfer, or reduction in rank . Interestingly, the scenario for which respondents selected a reduction in rank as the penalty was scenario 11, which involved rating the behavior of the supervisor who did not step in when officer’s assaulted a suspect. Respondents clearly saw the action as serious, but not serious enough to warrant dismissal; rather, the person is deemed to need moving down from the supervisory role as they are not exhibiting the level of responsibility appropriate to this position. This is in contrast to the supervisor in scenario 5 who grants leave in exchange for errands; that supervisor is rated as necessitating only a written warning . The remaining scenarios were the top four most serious, and all most commonly rated as warranting dismissal. These were, receiving discounts in exchange for business referrals, falsifying evidence, shooting an unarmed person and theft. These actions were, therefore, not tolerated, with those involved deemed unsuitable to continue in the service.

Table 3.5 and Fig. 3.2 show that the rank order of discipline across the scenarios was almost identical for the discipline respondents believed both should and would occur. However, there were differences between the levels of discipline respondents believed should and would occur at the case level. With the exception of scenario 10 (falsifies evidence), respondents believed, on average, that the level of discipline that would occur was significantly different to what they believed should occur. Scenarios 2, 3, 5, 9, and 10 were all considered to attract more lenient discipline than they should, while the remaining scenarios were considered to attract slightly harsher discipline than they should. However, while these differences were significant, the effect sizes were small to medium. It is interesting again to note scenarios 5 and 11, where the supervisor’s behavior is the subject of judgement. In scenario 5, where the supervisor grants leave in exchange for errands, while the modal choice is the same, on average the sample thought the disciplinary response would be more lenient than it should be (perhaps suggesting that supervisors are treated more leniently overall). However, for scenario 11, where the supervisor turns a blind eye to an assault of a suspect, respondents thought the disciplinary response would be harsher than necessary; equating to a difference between a demotion (should occur) and dismissal (would occur). This suggests that, while generally not tolerated, the sample is willing to remove the individual from the position of responsibility rather than remove them from the service altogether.

Fig. 3.2
figure 2

Average score for the discipline that should and would be received for each scenario. Scenarios are ordered according to the respondents’ own perception of seriousness (low seriousness to high seriousness)

Full size image
Table 3.5 Officers’ views about the discipline the misconduct should and would receive
Full size table

Table 3.6 shows the results of this fairness calculation—as described in the method section—for each scenario. Overall, as noted above, the mean differences were small (between − 1 and + 1 out of a possible scale of − 5 to + 5). This would suggest that, on average, officers believed that fair discipline would be received. Across the scenarios, differences can be observed in the means—with five of the scenarios yielding negative means, indicating on average lenient discipline; and six scenarios yielding positive means, indicating on average harsh discipline. The scenarios found most lenient were the granting of leave in exchange for errands (scenario 5) and taking goods from a crime scene (scenario 3). The scenarios seen as most harsh were failing to report an officer’s DUI (scenario 8) and punching an offender (scenario 6).

Table 3.6 Difference in officers’ views about the discipline the misconduct should and would receive (rank ordered from most lenient to harshest)
Full size table

Willingness to Report

Table 3.7 and Fig. 3.3 describe the participants’ willingness to report the scenarios, and their belief about the willingness of most officers to report. On average, respondents were willing to report the majority of scenarios. Only one scenario, (scenario 7: verbal abuse of motorist) received an average score less than the midpoint of the scale—indicating unwillingness to report. Once again, this construct is related to the overall perception of seriousness of the scenarios, with the rank order of willingness to report mirroring the rank order of seriousness ratings. Those least likely to be tolerated are again, therefore, receiving discounts for referrals of business, falsifying evidence, theft and shooting an unarmed person.

Fig. 3.3
figure 3

Average responses for officers’ own willingness to report and perception of most officers’ willingness to report each scenario. Scenarios are ordered by the officers’ perception of seriousness from least to most serious

Full size image
Table 3.7 Officers’ views about the willingness to report misconduct
Full size table

For all scenarios, respondents believed they were significantly more likely to report the incident than most officers would be (p < 0.001), with medium to large effect sizes. This shows that, on average, the sample saw their own behavior as more ethical than the majority, suggesting that, while others might subscribe to a “code of silence,” they would be much less likely to do so. This discrepancy is interesting in that the perception of the code of silence may be stronger than the evidence that it actually exists. However, given the small sample, it could be the case that the respondents are indeed different from the majority of officers. Perhaps those that self-selected to complete the survey are also those more likely to report wrongdoing. The largest differences were seen for the willingness to report the theft scenario, followed by the discounts for referrals, followed by granting leave in exchange for errands and falsifying evidence. Interestingly, the smallest differences in perception of own and others’ willingness to report were seen in the two scenarios they were least and most willing to report. This indicates that, at the extreme ends of the spectrum, they see their own behavior as slightly more aligned with their perception of the majority.

Correlates and Predictors of Officers’ Willingness to Report

For each scenario, a percentage of officers was unwilling to report the incident (indicated by a score of 1 or 2 on the 5-point scale). Table 3.8 explores the possibility that this unwillingness may be explained by a belief that the behavior does not constitute a violation. While this may explain some of the unwillingness, it does not explain the majority. Most notably, nearly a third of respondents believed scenario 7 (verbal abuse of motorist) to be a violation of policy but were unwilling to report it, and just over a fifth of respondents believed scenario 1 (receiving gifts from small businesses) to be a violation but would not report it. Instead, reporting behavior would seem to be linked more to the perception of the seriousness of the activity, with unwillingness to report decreasing as the perceived seriousness of the incident increases.

Table 3.8 Percentage of respondents who are unwilling to report a scenario even when they believe it to be a violation of policy
Full size table

The fairness of the discipline predicted to result from the behavior may also affect an officer’s decision to report an incident. Officers who believe the result will be too lenient may feel reporting is pointless, while officers who feel the result will be too harsh may feel reporting would be unfair. Figure 3.4 shows how scores of fairness relate to willingness to report for each scenario. The trend line shows a slight negative relationship, whereby scenarios judged to have lenient-to-fair discipline will generally be more likely to be reported than those thought to incur harsh discipline.

Fig. 3.4
figure 4

Scenarios according to officers’ perception of fairness (horizontal axis) and willingness to report (vertical axis). Dotted line represents the trend line

Full size image

Moving away from the individual scenarios, the survey constructs can be averaged across all scenarios to measure the relationships between them at the aggregate level (Table 3.9). Spearman’s correlations reveal significant relationships between officers’ willingness to report and all other measured constructs. Specifically, officers show significantly greater willingness to report when seriousness is high (in their own view and for most officers), the behavior is considered to be a violation, the discipline that should follow is higher, the discipline that would follow is higher, it is believed most officers would report and discipline is unlikely to be unfairly harsh. The strongest relationship with willingness to report is for officers’ own seriousness judgments, followed by the discipline that should occur. Also high are the relationships of willingness to report with the two constructs that measure the “culture” of most officers: most officers’ seriousness judgments and how likely most officers would be to report. The smallest (albeit still significant) relationships are with the discipline that would follow, the unfairness of the discipline, and the belief that it is a violation.

Table 3.9 Correlations between the survey constructs averaged across all scenarios
Full size table

A step-wise linear regression confirmed these results. When all seven constructs were entered as predictors of officers’ willingness to report, five were retained as significant predictors (“violation” was excluded as a non-significant predictor and “discipline (would)” was excluded due to multicolinearity), explaining 89 % of the variance (R 2 = 0.89, F(5, 731) = 554.54, p  < 0.001). Table 3.10 shows the results of the regression. All retained predictors provide a unique contribution to predicting willingness to report. Officers’ perception of seriousness remains the strongest predictor, accounting for 65 % of the variance.

Table 3.10 Regression results for predictors of officers’ willingness to report
Full size table

Demographic/Background Differences

The constructs were averaged across all 11 scenarios before comparing for differences by years of experience, rank , gender, and supervisory status. Table 3.11 shows that, on average, years of service were significantly positively correlated with most of the integrity constructs, with the exception of fairness perceptions. However, the correlations were small to moderate. Similarly, rank was significantly positively correlated with most of the integrity dimensions, with the exception of the seriousness perception of most officers and the discipline it would attract (uncorrelated), and a significant negative correlation between rank and unfairness, showing that, as rank increases, perceptions of unfairness decrease.

Table 3.11 Average construct scores across 11 scenarios; relationships with experience and rank, and differences by gender and supervisory status
Full size table

Years of service predicted willingness to report (F(1, 723) = 21.09, p < 0.001), although R 2 was only small (0.028). Rank also predicted willingness to report (F(1, 610) = 52.66, p < 0.001), but again R 2 was small (0.079). A multiple regression with both variables entered as predictors found that only rank remained a significant predictor when both were taken into account (F(2, 607) = 25.83, p < 0.001, R 2 = 0.078). This means that, beyond years of service (or when controlling for years of service), the rank of an officer predicts the willingness to report; such that the higher the rank, the greater the willingness.

Using a stringent alpha of 0.0045 to correct for the multiple number of comparisons (Bonferonni correction), some differences were evident by gender and supervisory status (Table 3.11). Female officers had a significantly less favourable view of ‘most officers’, judging most officers’ seriousness perceptions and willingness to report lower than did male officers. Female officers also showed a significantly more favourable view of the discipline system, with a more lenient view on what discipline would result. However, effect sizes were small to medium. The comparison by supervision status revealed fewer differences. Supervisors were, on average, significantly more willing to report the scenarios and have a harsher view on the level of discipline that should occur in response. However, again effect sizes were small, with only the difference in willingness to report showing a medium effect size.

Discussion

Policing in Australia , like many other countries, had undergone periods of scandal and reform over the past few decades, with a changing integrity landscape. The survey reported here provides a useful comparison to previous, similar ethics surveys conducted in Australia on different samples of police officers—the NPRU survey (Huon et al. 1995) and the series of surveys conducted by Queensland’s Crime and Misconduct Commission from 1995 to 2008 (CMC 2010).

In the present study, it is encouraging that mostly officers recognized the scenarios as depicting violations and viewed them as serious. For each of the scenarios, at least 80 % of respondents answered that it would be a violation of agency policy. For some scenarios, this increased to around 99 %. All 11 scenarios were rated as at least somewhat serious on average (all means were at or above the scale midpoint). This is similar to the findings from the NPRU survey and the CMC ethics surveys. Similar too are the types of infractions that, on average, drew the highest and lowest ratings of seriousness. The NPRU and CMC reports both highlighted that theft from a crime scene and altering a suspect’s statement rated amongst the most serious of the scenarios presented. Similarly, of the current scenarios, theft from a crime scene and planting evidence were amongst the top three most serious, lower only than excessive use of lethal force. Similarly, the prior studies found gratuities and rudeness to be amongst the least serious infractions, which the current study also supports.

Most officers expressed willingness to report the scenarios. On average, respondents were willing to report 10 of the 11 scenarios, the exception being the rudeness scenario. Less than 40 % of officers were unwilling to report each scenario, although this still represents a substantial minority for some scenarios. The two scenarios least likely to be reported were the case of rudeness and also accepting gratuities . These were often seen to be violations but were not seen to be particularly serious in contrast to the other scenarios. Rudeness is often the topic of complaints against police by members of the Australian public. Indeed, some police forces have implemented strategies to particularly identify and reduce customer service complaints in contrast to other issues that demand more serious investigation. For example, triage processes of complaint handling, local resolution of complaints as well as “mystery shopper” initiatives to test front counter service at police stations (Porter and Prenzler 2012a).

The proportion of the sample unwilling to report decreased to 10 % or less for some scenarios, particularly serious behavior such as theft and shooting an unarmed suspect. This is somewhat encouraging given the CMC’s finding of greater reluctance to report misconduct in Queensland over their survey period (CMC 2010). However, a proportion of officers in the current study recognised the violations, but was still unwilling to report them. This is despite the fact that most jurisdictions in Australia have mandatory reporting of misconduct, with some supporting this requirement through provision of confidential reporting channels (e.g., dedicated phone lines and support staff; Porter and Prenzler 2012a). For some scenarios, as many as 20–30 % of respondents were aware that the behavior would be considered a violation of policy, but stated they would be unwilling to report it. This was the case for the rudeness scenario and the gratuities scenario, which were also the two scenarios viewed least serious. Indeed, the primary factor that predicted officers’ willingness to report the scenarios was how they perceived the seriousness of the behavior. While all integrity dimensions measured were positively associated with officers’ willingness to report, the officers’ own view of the seriousness of the behavior made the largest contribution to willingness to report. Further exploration of the likelihood of engaging different modes of reporting, such as informal or formal (as in the NPRU study), or even anonymous reporting, would be an interesting addition for future surveys.

Similar to previous studies, integrity was positively related to years of experience and rank . A number of relationships were observed across the dimensions of integrity with rank and years of experience as a police officer. All relationships were in the same positive direction, showing that understanding of, and support for, integrity increased with experience and rank. This is similar to the finding of the NPRU study; however, that study was also able to compare recruits to find an early decrease in ethical standards during those first few years on the job. The present study was limited only to those officers already sworn in and so was unable to explore this issue. This might be a useful focus for further exploration.

Interestingly, rank was positively related to the willingness to report, even when controlling for years in service. While rank and years of experience are highly correlated, supporting the view that officers must gain experience to be promoted, rank has a unique contribution to predicting officers’ willingness to report. This suggests that attributes associated with rank affect integrity beyond the amount of experience. Exploration of the roles, attributes and training at different ranks, and their influence on ethical attitudes, would be a useful addition to this area.

In contrast, both prior Australian studies found that females viewed the scenarios as more serious than males. This was not found in the current study, with no gender difference in the overall seriousness ratings. Female officers did, however, have a dimmer view of police culture and a greater willingness to report, on average, than male officers. Further research could attempt to unpack the relationships between the integrity constructs for males and females to understand whether there are gender differences in the relative importance of, for example, police culture versus personal views in the willingness to report an infraction.

Implications and Conclusion

According to the current survey, officers’ understanding of the seriousness of behavior would appear to be the most important factor in breaking the code of silence. Experience, rank, and supervisory status were positively related to respondent’s own views of the seriousness of the scenarios, showing that this experience and role status may be associated with how aware officers are of the consequences of infractions. However, these individual factors do not show a large effect on officers’ views. Thus, while those officers who reach positions of responsibility may reflect more on the seriousness of infractions, it is ideal to proactively engage all officers with this mindset through raising awareness of the consequences and changing the normative attitude. Incorporating ethics messages into training throughout the organization is a standard recommendation, while some agencies have also provided case-specific materials in the forms of “learning the lessons” bulletins or articles in internal publications (Porter and Prenzler 2012b) and have implemented reforms oriented towards enhancing customer service and reducing service-related complaints (Porter and Prenzler 2012a).

Indeed, the fact that the rudeness scenario rates, on average, only just above the midpoint of the scale is concerning. Annual reports of those agencies that take complaints from members of the public about police (including police agencies themselves), both in Australia and overseas, show that rudeness or incivility tend to be amongst the largest categories of complaints/allegations received (Crime and Misconduct Commission 2013; Independent Police Complaints Commission 2013; Porter and Prenzler 2012a, p. 222). Studies of police legitimacy show that such treatment can undermine public confidence in police and willingness to cooperate with police (Bayley 2002; Decker 1981; Murphy 2009). Public cooperation is essential for police effectiveness in safeguarding the public from crime (Murphy et al. 2008).

Fairness, as measured by the concordance between appropriate and likely discipline, was also important to the willingness to report and, generally, officers saw the discipline as fair (differences were small). In Australia, there have been recent attempts by some agencies at improving the procedural fairness of internal discipline systems and complaint-handling procedures (Office of Police Integrity 2007b; NSW Police Force 2012, p. 10; Queensland Government 2011).

Police culture was also important, with significant relationships between perceptions of the views and behavior of the majority and respondents’ own views and likely behavior. The respondents’ own views were, however, significantly different from their views of the culture (most officers), with respondents believing themselves to be more ethical than the majority. Further, respondents’ own views showed higher correlation with, and greater prediction of, willingness to report than did their views of the culture.

In conclusion, on average, the respondents to the survey exhibited positive trends towards understanding the seriousness of violations and being willing to take action in reporting colleagues. In comparison to the previous Australian surveys conducted nationally by the NPRU and in Queensland by the CMC, the present survey highlighted some similar findings, despite the different samples and time periods. There seems to be consistency in views of relative seriousness of different types of infractions, as well as some consistent differences by rank and experience in the police service. It is encouraging that perceptions of fairness of discipline and willingness to report are more positive in the current sample. It is evident that respondents are somewhat cynical about police culture, viewing themselves as being more ethical than most officers in terms of both perceptions of seriousness of misconduct and willingness to take action in response. Proactive attempts to instil both personal values and cultural normative beliefs may, therefore, serve to improve police integrity systems.