Introduction

Police legitimacy is a well-studied phenomenon within the literature. It has been examined in a wide range of criminological fields including road policing (which is a stream of policing dedicated to traffic matters; MacQueen & Bradford, 2015; Sahin et al., 2017; Van Damme & Pauwels, 2016), terrorism (Cherney & Murphy, 2013), and social media (Grimmelikhuijsen & Meijer, 2015). Police legitimacy has been linked with overall satisfaction and compliance with police (Demir et al., 2020; Jackson et al., 2012; MacQueen & Bradford, 2015; Sahin et al., 2017; Seo & Lee, 2018). The role of uniforms worn by police is another aspect considered in previous research that has examined views of police legitimacy (Bell, 1982; Jenkins et al., 2021; Johnson et al., 2015; Simpson, 2017). Despite this extensive literature looking at police legitimacy and the impact of police uniforms, we still do not understand the relationship between uniforms, legitimacy, and intentions to offend. To address the gap in knowledge regarding the relationship between police uniforms, views of police legitimacy, and intentions to offend, this study examines the impact of police officers delivering a road safety program to young people in both uniform and plain clothes (sometimes referred to as civilian clothes while on duty).

Background literature

Police legitimacy

Sunshine & Tyler, (2003, p. 2) defined legitimacy as “a property of an authority that leads people to feel that the authority or institution is entitled to be deferred to and obeyed”. A growing body of literature has begun to consider the impact of views of police legitimacy within a road safety or traffic enforcement context (MacQueen & Bradford, 2015; Mastrofski et al., 2016; Mazerolle et al., 2013a; Sahin et al., 2017; Van Damme & Pauwels, 2016). For many people, the majority of their interactions with police will be within a traffic context (Johnson, 2004). However, there is limited research that examines the impact of perceptions of police legitimacy on an individual’s intention to offend on the road. Previous research has discussed the views of police legitimacy as a mediator for cooperation and compliance with police officers from members of the general public (Mazerolle et al., 2013b). Yet this research examining a citizen’s view of police legitimacy typically occurs as a product of an official or enforcement interaction with police, often through a procedural justice lens (Madon et al., 2017; Murphy & Cherney, 2011; Sunshine & Tyler, 2003; Tyler, 1990). For example, research by Mazerolle et al., (2013a, b, c) modified the language and messaging during a roadside checkpoint interaction with police to improve perceptions of procedural justice, including views of police legitimacy, and, as a result, drivers reported increases in satisfaction and cooperation. Perceptions of police performance have also been linked to the view of police legitimacy reported by members of the public (Jonathan-Zamir & Weisburd, 2013). However, research is yet to examine the impact of views of police legitimacy on the intentions of citizens to offend. This study further seeks to determine the impact of police uniforms in achieving positive views of police legitimacy, which may subsequently impact intentions to offend.

The impact of police appearance and uniforms

Recent research has demonstrated that previously overlooked factors such as the way police present themselves in public may impact the views of police legitimacy held by the public. These factors include their facial expressions (Simpson, 2021), their use of personal protective equipment such as masks during COVID-19 (Sandrin & Simpson, 2022), minor alterations to the uniform (Jenkins et al., 2021), and the donning of what is publicly viewed as militarised attire (Blaskovits et al., 2022). The analysis of the role of uniforms on civilian perceptions of ability, skill, compliance, or trustworthiness for professional services is not a new concept (Bickman, 1974; Bushman, 1988; Tenzel & Cizanckas, 1973; Volpp & Lennon, 1988), nor is it limited to the policing profession (Hribar et al., 2023). The impact of uniforms is not even limited to the external impact, with recent research by Simpson & Sargeant, (2022) linking uniforms with an officer’s own perception of their accountability, aggression, approachability, friendliness, and professionalism. Police officers in uniform are generally perceived as more legitimate than those in plain clothes as uniforms serve as a visible symbol of legal power and authority (Bell, 1982; Johnson et al., 2015). A recent study in Canada found that small manipulations to the operational police uniform resulted in significant changes in perceptions of officers regarding community engagement, professionalism, and officer safety (Jenkins et al., 2021). The study by Jenkins et al., (2021) even controlled for views of police legitimacy prior to examining the impact of uniform manipulations. In another study, Simpson, (2017) found that police were viewed more favourably by members of the public when presented in a fully operational police uniform, as opposed to wearing civilian clothing (such as the attire a detective would wear). However, most studies on this topic used still images of police and presented them to participants to comment on. Our contention is that there is no literature available that has examined the impact of uniformed or plain clothes police officers in the field and subsequently reported the impact of their uniform on views of police legitimacy. These impacts that uniforms have on views of police legitimacy could have a subsequent impact of legitimacy on citizens’ compliance with police.

Compliance

Compliance, in a road safety context, refers to drivers following both the directions of police and the road rules. A significant body of research highlights that in most fatal and serious injury crashes, there are contributing factors resulting from traffic offences (Davey et al., 2022a, b; Davey et al., 2022a, b). Road trauma (injuries resulting from land transport related crashes) is the second leading cause of death for young people (Australian Institute of Health & Welfare, 2022), and as such, there is a significant amount of research that has examined various means and methods to improve driver compliance with road rules including deterrence-based research (Bates & Anderson, 2019; Bates et al., 2017, 2020, 2023; Truelove et al., 2019, 2022), public messaging, and procedural justice or police legitimacy research (Demir et al., 2020; Mazerolle et al., 2013a; Mazerolle et al., 2013b; Sahin et al., 2017). The design and appearance of police uniforms can greatly influence the perceptions of legitimacy (Simpson, 2017) and thus the level of compliance among citizens, as they serve as a visual representation of the authority and professionalism of the police force.

Young driver crash risk and driver education

While young people aged between 16 and 25 years only account for 13% of the licensed population, they account for approximately 29.2% of the fatalities on Queensland roads in 2021 (Queensland Department of Transport & Main Roads, 2022). As a result of this crash risk and the over-representation in fatalities and serious injury crashes, significant resources are dedicated to the prevention and education of young drivers to reduce their involvement in crashes. Within the literature, a majority of evaluation research on the impact of driver education as an intervention aimed at reducing crash risk suggests that it leads to limited or no reduction in crash risk for young drivers (Christie, 2001a; Elvik et al., 2009; Glendon et al., 2014; Lonero & Mayhew, 2010). In one literature review, Christie, (2001a) concluded that driver education cannot be considered an effective crash countermeasure. However, the results of recent research that examined a police-led driver education program indicated that there was a significant change in young driver intentions following the program (Anderson et al., 2021). This change was present even for the most behaviour-change-resistant group of young drivers: males with high levels of sensation seeking. However, that research did not identify why this program resulted in a change in young driver intentions.

This study

While existing research demonstrates that the way police present themselves influences the public’s views on police legitimacy, there are limitations in the current literature that this study aims to address. Specifically, there is a scarcity of field experimental research examining the impact of uniformed and plain clothed police officers on views of police legitimacy and how these views affect intentions to offend. Additionally, previous literature on police-led road safety education has shown some efficacy in the outcomes (Anderson et al., 2020), but it remains unclear whether the presence of uniformed police played a role in those outcomes. To address these gaps, the current study employs an experimental design, collaborating with frontline police officers facilitating a road safety program for students nearing high school graduation. This approach allows us to examine the real-world impact of uniformed and plain-clothed police on views of police legitimacy and intentions to offend.

Method

This study used a randomised field experiment to investigate whether the delivery of a driver education program by uniformed or plain clothes officers affected the views of police legitimacy among young people. A survey was administered to a sample of high school students receiving a police road safety program. The survey consisted of a battery of scales designed to measure views of police legitimacy and their future driving intentions on the road.

Study context

The Life Awareness Workshop (LAW) presentation is a police-led, school-based, road safety program delivered to schools located in south-east Queensland, Australia. This program was designed to address the high-risk driving behaviour exhibited by young drivers while providing them with the knowledge to make safer decisions behind the wheel. The program runs for approximately two hours and is interactive, with students encouraged to ask questions and engage with the police facilitators. The program focuses on the FATAL5, which is a Queensland road safety campaign that highlights the five leading causes of serious injury and fatal crashes: speeding, seatbelts, fatigue, impaired driving, and distractions (such as mobile phones (Harbeck et al., 2018).

Participants

Participants in this study were grade 12 students (n = 380) from private (fee-paying) high schools located in Queensland, Australia. All participants were receiving the LAW presentation at their school. The participants were aged between 16 and 18 years of age based on their school year and licence status. Within the control condition (program delivered by police officers in uniform), 38.9% (n = 103) of the participants were male. Within the experimental condition (program delivered by police officers in plain clothes), 40.9% (n = 47) of the participants were male. A total of six schools participated in the study, and two schools received the experimental plain clothes condition (n = 115). The average number of students present at each school was 63.33 (SD = 27.20; R = 16–89).

Procedure

This study used a randomised field experiment to examine the effect of an educational and informal interaction with police officers on a young person’s views of police legitimacy, views of police performance, and their intentions to offend on the road. In the control condition, police facilitators wore their standard operational uniform, and in the experimental condition, police officers were in plain clothes with their standard accoutrements (such as a belt, firearm, handcuffs, and oleoresin capsicum (OC) spray). The same officers delivered the program across both conditions. The officers that presented the program are both male, middle-aged, and have over a decade of experience in traffic-focused policing. The decision regarding the delivery of the program and allocation to conditions were made by the Queensland Police Service taking into account any other policing duties that required further interactions with the public requiring officers to be uniformed during that shift. Participants completed a survey at the conclusion of the LAW program. To control for differences in program delivery, the same officers facilitated both the control and experimental condition sessions of the LAW program. Ethical clearance for the involvement of young people, as well as the collection of data in cooperation with a Queensland Police Service program, was obtained from both the University ethics board and police research review committee.

Measures

Participants were provided with the survey in a paper and pencil format. The survey was anticipated to take 10 min to complete. The measures relevant to the current study are described below.

Police legitimacy and performance items

The survey completed by participants included several prevalidated scales. A scale measuring police legitimacy and performance was developed by condensing previously utilised scales (Sunshine & Tyler, 2003; Tyler, 1990; see Appendix 1 for the complete police legitimacy and performance scale). The final police legitimacy scale consisted of 19 items which are mean scored across 4 subscales. The subscales include Obey and Confidence in Police, Police Performance, Cooperation with Police, and Police Efficiency. Responses were provided across a five-point Likert scale spread from “Never” to “All the time”. Examples of the questions include the following: “I feel a moral obligation to obey police”, “I have confidence in police”, “I question the legitimacy of the police’s authority over me”, and “On the whole, police officers are honest”. There is a significant body of procedural justice and police legitimacy research available that has used these questions to measure police legitimacy and performance (Antrobus et al., 2015; Murphy et al., 2010, 2014; Oliveira & Murphy, 2015; Seo & Lee, 2018). Of the four subscales, Cronbach’s alpha scores were used as a secondary method of ensuring reliability. This information has been reported below in Tables 1 and 2.

Table 1 Independent t-tests examining police legitimacy across the conditions
Full size table
Table 2 Independent t-tests examining police legitimacy subscales across experimental conditions
Full size table

Behaviour of Young and Novice Drivers Scale (BYNDS)

Self-reported driving intentions were measured using the BYNDS. Originally, the BYNDS consisted of a 44-item scale developed to specifically assess risky behaviour, driving violations and misjudgements of young drivers (Scott-Parker et al., 2010), and has been previously used in a number of studies (Bates et al., 2017; Scott-Parker & Oviedo-Trespalacios, 2017; Scott-Parker & Proffitt, 2015; Scott-Parker et al., 2010, 2012). Two modified subscales of the BYNDS were used in our survey. The modified version of the BYNDS has been previously used in young driver research (Anderson et al., 2020) and includes only the fixed and transient offending subscales as well as additional items focusing on mobile phone use while driving. Overall, the modified BYNDS scale consisted of 23 items which were mean-scored across two individual subscales: fixed offending (offences that are not occurring for the entire journey, such as speeding or running a red light) and transient offending (offences that occur through the journey, such as drink driving). Of the 23 items examined, 8 were fixed offending items, and 15 were transient offending items. Cronbach’s alpha scores of both subscales were consistent with previous uses of the BYNDS.

Data analysis

Initial data screening identified that there were incomplete surveys within the sample. As the survey contained several pre-validated scales, in cases where participants fully completed an individual scale within the survey, their data was retained. Incomplete scale data was discarded. In total, 38 surveys were discarded as they were partially completed no individual subscales were fully completed. Scale reliability was examined to ensure that identified scales had Cronbach’s alphas of > 0.70 (Hinkin et al., 1997). Means and standard deviations were calculated for continuous-level data. As the scales were developed using Likert-scale data, parametric methods were used to analyse these data. This allowed for inferential analyses such as the use of a t-test to compare mean scores. There is research that indicates that the analysis of Likert data as parametric data for analysis is acceptable (Harpe, 2015). The significance level for all analyses was set at 0.05.

Following the t-tests, a bivariate correlation analysis was conducted to examine the relationship between views of police legitimacy and intentions to offend. Subsequently, a mediation analysis was performed to assess the overall impact of uniforms on intentions to offend, with views of police legitimacy serving as the mediator. The mediation analysis was conducted in SPSS using the PROCESS macro (Hayes, 2017). The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available for public distribution.

Results

Uniforms and views of police legitimacy

One aim of this study was to investigate the impact of police uniforms on views of police legitimacy among a school-age young driver sample. A key result of this study indicates that having the program delivered by police officers in uniform resulted in more positive views of police legitimacy when compared with the group where police officers facilitated the program in plain clothes (see Table 1). That is, views of police legitimacy were more positive for the group that received the program by uniformed police officers and the difference between the groups was statistically significant (t(261) = 2.274, p = 0.012). The Cronbach’s Alpha score (α = 0.951) demonstrates excellent reliability of the scale (Tavakol & Dennick, 2011). The computed effect sizes suggest that the difference reported could be important, with a small to medium effect size (d = 0.30, 95% CI [0.04, 0.55]).

As there was a significant difference in the overall police legitimacy scale, we further explored the individual subscales for a more in-depth understanding of which subscales had the greater impact. With equal variances tested and assumed, results in Table 2 show that three of the four subscales are significantly different across the experimental conditions. The Cronbach’s Alpha scores (α = 0.922, 0.954, 0.893, and 0.898) demonstrate excellent reliability for each subscale (Tavakol & Dennick, 2011) and are consistent, if not higher, with previous uses of the subscales (Murphy et al., 2010). In the subscales of Obey and Confidence (t(279) = 2.994, p = 0.002), Police Performance (t(269 = 0.954, p = 0.024), and Police Efficiency (t(268) = 0.898, p < 0.001), presentations delivered by uniformed police resulted in statistically significant increases in positive views of the police. An examination of effect sizes suggested that these differences could be important, with small (d = 0.2) to medium (d = 0.5) effect sizes (Cohen, 2013) in the Obey and Confidence (d = 0.37, 95% CI [0.12, 0.62]) scale, as well as the Police Efficiency scale (d = 0.38, 95% CI [0.13, 0.64]). Police Performance had a smaller effect size (d = 0.26, 95% CI [0.001, 0.51]). The Cooperation with Police subscale results did not show significant differences between the control and experimental conditions (t(276) = 0.893, p = 0.383). Further, the analysis of the effect size (d = 0.038, 95% CI [− 0.21, 0.29]) reveals a less than small effect size, indicating that the difference between uniform and plain clothes was non-significant in this case.

Uniforms and intentions of young drivers

Another aim of this study was to investigate the impact of police uniforms on the intentions to offend of participants. The intentions of young drivers to offend were examined following the LAW presentation across the experimental and control conditions. The results of two t-tests are displayed in Table 3. The results demonstrated that both transient and fixed offending benefitted significantly from the presence of uniformed officers, compared to the experimental condition.

Table 3 Independent t-tests examining offending intentions across the conditions
Full size table

An analysis of both the transient and fixed offending scales shows a statistically significant difference between the uniformed and plain clothes’ conditions in the study. The impact of the uniform on the means was negative, with the likelihood or intentions to offend decreasing within the uniformed group. The Cronbach’s Alpha scores (α = 0.950 and 0.955) demonstrate an excellent reliability of each of the subscales (Tavakol & Dennick, 2011) and are higher than previous uses of the BYNDS (Scott-Parker et al., 2012). Examination of effect sizes suggested that these results may have important real-world implications, with small (d = 0.2) to medium (d = 0.5) effect sizes (Cohen, 2013) in both the transient offending scale (d =  − 0.269, 95% CI [− 0.541, 0.003]) as well as the fixed offending scale (d =  − 0.277, 95% CI [− 0.549, − 0.005]).

Views of police legitimacy and intentions of offend

To examine if any correlation exists between views of police legitimacy and intentions for young drivers to offend, a correlation analysis was conducted. The results of the analysis demonstrate that a moderate negative correlation exists between the two variables. That is, as the views of police legitimacy increase, the reported intentions to offend decrease for both fixed and transient offending. Results of the Spearman Rho correlation analysis show that there is a moderate negative correlation for transient offending (rs =  − 0.471, p < 0.001) indicating that as views of police legitimacy increase, there is a decrease in the reported intentions to offend (Table 4). Likewise, there was a slightly weaker negative correlation for fixed offending (rs =  − 0.314, p < 0.001) indicating a similar outcome.

Table 4 Correlation analysis between views of police legitimacy and intentions to offend
Full size table

Uniforms, views of police legitimacy, and intentions of offend

To understand the interactions between uniforms, views of police legitimacy, and intentions to offend, a series of mediation analyses were conducted using PROCESS. The two intentions to offend subscales were analysed separately for clarity.

Transient offending

The first mediation analysis focused on the intentions to commit transient offending. Results indicated a significant direct effect of the uniform condition (β =  − 0.168, SE = 0.079, t =  − 2.131, p = 0.034, 95% CI [0.323, 0.013]) on the views of police legitimacy (see Fig. 1). In addition, there was a significant total effect of the uniform condition on intent to commit transient offending (β = 0.133, SE = 0.049, t = 2.706, p = 0.007, 95% CI [0.036, 0.231]). The direct effect between uniform and intent to commit transient offences accounted for a portion of the total effect (β = 0.094, SE = 0.046, t = 2.033, p = 0.043, 95% CI [0.003, 0.186]), while the indirect effect, mediated by views of police legitimacy, was also significant (β = 0.039, SE = 0.024, 95% CI [0.003, 0.095]). The mediation model accounted for approximately 16.8% of the variance in intent to commit transient offending (R2 = 0.168, F(2,207) = 20.932, p < 0.001). The uniform condition accounted for 3.4% of the variance in intent to commit transient offending (R2 change = 0.034, F(1,208) = 4.539, p = 0.034). Views of police legitimacy accounted for 13.5% of the variance in the outcome variable (R2 change = 0.135, F(1,207) = 37.946, p < 0.001).

Fig. 1
figure 1

Mediation model for uniform, views of police legitimacy, and intent to commit transient offences

Full size image

Fixed offending

The second mediation analysis focuses on the intentions of drivers to commit fixed offences. Results indicated a significant direct effect of the uniform condition (β =  − 0.163, SE = 0.079, t =  − 2.069, p = 0.034, 95% CI [− 0.318, − 0.008]) on the views of police legitimacy (see Fig. 2). In addition, there was a significant total effect of the uniform condition on intent to commit fixed offending (β = 0.105, SE = 0.032, t = 3.238, p = 0.0014, 95% CI [0.041, 0.168]). The direct effect between uniform and intentions to commit fixed offences accounted for a portion of the total effect (β = 0.085, SE = 0.031, t = 2.719, p = 0.007, 95% CI [0.023, 0.147]), while the indirect effect, mediated by views of police legitimacy, was also significant (β = 0.0196, SE = 0.031, 95% CI [0.001, 0.061]). The mediation model accounted for approximately 12.9% of the variance in intent to commit fixed offending (R2 = 0.129, F(2,208) = 15.455, p < 0.001). The uniform condition accounted for 4.78% of the variance in intent to commit fixed offending (R2 = 0.048, F(1,209) = 10.485, p = 0.0014). Views of police legitimacy accounted for 2.0% of the variance in the outcome variable (R2 = 0.02, F(1,209) = 4.280, p = 0.0398).

Fig. 2
figure 2

Mediation model for uniform, views of police legitimacy, and intent to commit fixed offences

Full size image

Discussion

This study examined the impact of uniformed and plain clothes on police views of police legitimacy and intentions to offend following a police-led road safety program. The results have demonstrated that when uniformed police officers were facilitating a road safety education program for young drivers, participants that received the program from officers in their full uniform reported more positive views of police legitimacy and intentions to offend on the road. This demonstrates that uniformed police, proactively engaging with the community, can improve the views of police legitimacy and the reported intentions of participants to offend in the future.

Views of police legitimacy

This study found that when the program is facilitated by police officers in their uniform, significantly more positive views of police legitimacy were reported by the participants. While this study focused on road safety, the views of police legitimacy, performance, and willingness to cooperate with police may have positive impacts beyond the proposed road safety benefits. The link between views of police legitimacy and compliance or cooperation with police is well-documented (Jackson et al., 2012; Mazerolle et al., 2013a; Muratbegovic et al., 2014; Seo & Lee, 2018); however, the understanding of the impact on offending is limited. Farren et al., (2018) identified that procedural justice and police legitimacy may be key to reducing the intentions of young people to offend. Within that context of the current study, although the LAW program aims to reduce traffic offending, the impacts of the program on views of police legitimacy held by the young drivers could flow on to reducing the likelihood of other offending behaviour. Thus, the impact of these types of programs and interactions with young people could be instrumental from a community policing perspective, beyond the topic of the presentation.

An examination of individual subscales within the police legitimacy scale, obey and confidence, police performance, and police efficiency was also significantly more positive in the control condition as opposed to the experimental. This indicates that the impact of police uniforms may not just be limited to the perceptions of authority, trustworthiness, or confidence in the police which have been previously discussed in the literature (Bickman, 1974; Bushman, 1988; Jenkins et al., 2021; Simpson, 2017), but also impact the perceptions of police performance and efficiency as demonstrated in this study. It is clear from this study that the uniform worn by police officers has a significant impact on the views of police legitimacy held by members of the public beyond authority and trustworthiness and should be considered further in future police legitimacy research.

Another important aspect of this study to consider is the pre-existing Australian culture of respect and cooperation with police, unlike that of the USA where trust and confidence in law enforcement are much lower (Murphy et al., 2014). This pre-existing Australian culture of respect and cooperation with police could explain the lack of significant findings in the cooperation with police subscale within the police legitimacy scale. While there was a more positive result reported in the cooperation with police subscale by participants that received the program from uniformed police, the difference between the groups was not significant. Due to the existing cultural willingness of Australians to comply with police, there may be little impact from police uniforms on this aspect of police legitimacy. Future research should examine this phenomenon outside of the Australian context.

While there is limited recent research available on the impact of uniforms on views of police legitimacy, seminal research supports the findings of this study which identified positive impacts of police uniforms (Joseph & Alex, 1972). More recent research has further reinforced the idea that uniformed police are perceived by the public more favourably when compared with plain-clothed officers (Simpson, 2017). Similarly, the concept of high visibility or identifiable policing extends to the vehicles they drive, with more recent research by Simpson’s, (2019) finding that members of the public view police in marked vehicles as more favourable compared with those in unmarked vehicles. Even the accoutrements worn by officers both when uniformed and plain clothed have impacts on the perceptions of legitimacy, performance, and accountability by members of the public (Simpson, 2018).

The results of this study are promising to support the involvement of uniformed police in educational and informal interactions with young people as a method of improving views of police legitimacy and performance. While these educational duties are often left to community organisations and other NGOs, this research suggests that there may be a benefit to the relationship between police and young people with the involvement of police in the education space.

Young driver intentions to offend

When examining the intentions to offend by young drivers, there was a significant difference in the reported intentions to offend by participants that received the program from uniformed police officers when compared with plain clothes police officers. A possible explanation of this phenomenon is grounded in previous research by Simpson, (2017) which found that uniformed police, when compared with plain clothed officers, were viewed as much more approachable, respectful, and accountable. This could be leading to the improvements in views of police legitimacy discussed above, which have a cascading impact on the intentions of young drivers to offend on the road.

The impact of the police program on reported intentions to offend by young drivers was expected with similar research being conducted previously on the same program, although the study was unable to identify if the improvements in young driver intentions were a result of the program or the delivery by a police officer (Anderson et al., 2020). This study builds upon that research by examining the impact of the involvement of uniformed police within the presentation. This suggests that a possible reason for the successful nature of this road safety intervention, when there is a significant body of research that suggests that classroom-based interventions are ineffective (Christie, 2001b), could be a result of the involvement of police officers.

Views of police legitimacy and intentions to offend

A possible explanation for the differences in the intentions of young drivers across the plain clothes and uniformed conditions could be a result of the differences in views of police legitimacy between the two conditions. A correlation between views of police legitimacy and intentions to offend identifies that there is a significant negative moderate correlation indicating that as views of police legitimacy increase, the intentions of participants to offend decrease. This finding is further supported by a systematic review of interventions involving police procedural justice and police legitimacy concluding that police legitimacy may be linked to improvements in citizen compliance and cooperation (Mazerolle et al., 2013b). Within the road safety context, there is a significant body of research that examines the police-directed compliance of drivers who have positive views of police legitimacy (Jackson et al., 2012; Mazerolle et al., 2013a; Muratbegovic et al., 2014; Van Damme & Pauwels, 2016). However, this study highlights the existence of a more self-directed willingness to comply resulting from more positive views of police legitimacy which warrants further examination.

Uniforms, views of police legitimacy, and intentions to offend

Mediation analysis of uniforms, views of police legitimacy, and intentions to offend demonstrate that intentions to offend, both fixed and transient, are impacted by uniforms worn by police both directly and mediated through views of police legitimacy. The mediation results also highlight a difference between fixed and transient offending. For transient offending, a portion (16.8%) of the intentions to offend variance was explained by the mediated model (uniform > views of police legitimacy > intentions to offend), whereas only 3.4% of the variance could be attributed to the direct impact of police uniforms. This demonstrated that the impact is not just from the presence of police uniforms, but the impact those uniforms have on the views of police legitimacy held by participants. For fixed offending, the results were less noticeable with the mediation model only accounting for 12.9% of the variance, and the direct impact of uniforms accounting for 4.78%. The reduced impact on fixed offending is also demonstrated in the correlation analysis, where the ability to impact intentions for fixed offending is limited. A possible explanation for this finding comes from previous literature that reported most individuals are typically more compliant with fixed offences (Bates et al., 2016). This leaves very little room for views of police legitimacy or uniforms to impact behaviour, possibly explaining lower correlation and mediation values.

Limitations and future research

This study relied on self-report data from senior school-age students. While self-report data is well used within social sciences (Chan, 2010), it is not without its limitations. Furthermore, we acknowledge the possibility that a power differential may exist leading to bias in the answers when students were asked to anonymously disclose views of police legitimacy and intentions to offend immediately following a presentation from police officers. The use of fee-paying schools may have impacted the diversity of the sample able to be recruited into the research; however, it should be noted that fee-paying schools within Australia are financially diverse and cater to a range of sociodemographic sectors.Footnote 1

While statistically significant and meaningful differences were observed between both the control and experimental conditions and within views of police legitimacy, performance, and willingness to cooperate with police and young driver intentions on the road, how this translates into actual behaviour change is unknown. Furthermore, while the samples included students from a range of schools, the limited demographic data collected to protect the anonymity of participants limits the generalisability of these findings. Future research should seek to identify if there is a similar relationship between views of police legitimacy and other types of offending. Furthermore, the exploration of other methods of improving views of police legitimacy and the impact on reported willingness to offend should also be examined. These methods could include community engagement activities or other informal educational programs facilitated by police to target any number of issues within the community. Finally, the reported reduction in willingness to offend in this paper should be examined longitudinally to examine if young drivers translate their reported intentions to offend into a reduction in offending behaviour.

Implications

Theoretically, this study highlights an important factor, the presence of police uniforms, that should be considered when examining citizen and police interaction and the outcomes for police legitimacy. While a majority of research to date has considered concepts such as procedural justice when measuring impacts on police legitimacy and focused on the direct interaction between police and a member of the community, this study demonstrates that environmental factors such as the uniform of officers during an interaction should also be considered.

Practically, this study highlights an important implication for police proactively engaging within the community. While programs such as the LAW workshop could be delivered by civilian facilitators to save money and increase the availability of police resources, this paper demonstrates the benefits of unformed police facilitation. This paper highlights that facilitation by uniformed police improves views of police legitimacy and reduces the reported intentions to offend by young drivers. Also of note is the breadth of applicability of these findings; while this study focused on the road safety or road policing context, these positive results for police legitimacy and intentions to offend could impact a multitude of other aspects of policing.

Conclusion

This study builds upon previous research and examines the impact of a police uniform on views of police legitimacy when compared with plain clothed officers. When this issue has been examined previously, studies used photos to assess the impact of varying uniforms or outfits worn by police (Jenkins et al., 2021; Johnson et al., 2015; Simpson, 2017). This study used an experimental approach to examine the effects of an education program delivered to schools by police officers. The experiment used the same police officers to deliver the intervention. In the control condition, the officers wore their uniform, while in the experimental condition, the officers wore plain clothes. These findings suggest that among a group of young people, the presentation of the program by uniformed police officers led to significantly more positive views of police legitimacy. A mediation analysis demonstrated that while uniforms have a direct impact on the reported intentions to offend when mediated by views of police legitimacy, significantly more of the variance in reported intentions to offend is explained. It should be noted, however, that in both experimental conditions, the reported views of police legitimacy were positive, suggesting that views of police legitimacy are not only based on uniforms but a myriad of factors (such as perceived fairness and trustworthiness) that each requires individual consideration.

The implications of this study have the potential to extend beyond the topic of the intervention program, in this case, road safety. Specifically, this study highlights the need for research examining police legitimacy to consider external aspects of the interaction, such as the presentation of police officers, beyond direct communication and standard procedural justice factors. This consideration will support future police legitimacy research, which has been shown to be linked with the likelihood to report crimes, seeking help, and cooperating with the police (Hinds & Murphy, 2007; Mazerolle et al., 2013a; Mazerolle et al., 2013b). Overall, this study demonstrates that while uniforms have a direct impact on reported intentions to offend and views of police legitimacy, considerably more variance in the intentions to offend is explained when views of police legitimacy are considered in a mediation model with intentions to offend.