Understanding the decline: a procedural justice approach to the key factors behind the downward shift in opinions of police

ABSTRACT

Recently, there has been a marked decrease in public satisfaction with police. Reduced trust and confidence in police can lead to strained police-citizen relationships and increased conflicts. This study aimed to understand this decline in an Australian context. First the media’s influence of public opinions was examined by analysing a total of 855 newspaper headlines from two distinct periods: the first marked by a year of high public satisfaction in police and the second with a low public satisfaction rating. Secondly, to examine the public’s opinion specifically, a content analysis was conducted using a procedural justice framework to analyse 198 public comments posted to news articles. While the headlines showed consistent media sentiment between the two time-periods, there was a marked increase in police-related articles in the low-satisfaction period. Additionally, the public commentary analysis revealed a shift from positive to more neutral and negative sentiments with trust, neutrality, and respect as focal points. This research highlights that while media portrayal remained consistent, the amplified coverage and shifting public opinions emphasise the need for police to enhance trust-building efforts and prioritise procedural justice. Understanding and addressing these perceptions are crucial for fostering harmonious police-citizen interactions.

KEYWORDS:

• Police
• law enforcement
• procedural justice
• public opinion

Introduction

The public rely on the police to maintain order, protect lives and property, prevent and detect crime, and ensure public safety (Hine & Davenport-Klunder, 2022). The police’s ability to perform this role relies heavily on the public’s trust and confidence (ANZPAA, 2021). When the police are viewed as legitimate, people are more likely to comply with the law and cooperate with the police (Van Damme, Pauwels, & Svensson, 2015). In contrast, if police-citizen relationships break down, it can result in a lack of confidence and trust in police and, ultimately, a lack of compliance which is essential to maintaining peace and order (ANZPAA, 2021). Reduced cooperation may result in underreporting of crimes and a lack of intelligence for police operations (Boateng, 2018). It may also lead to community members taking on judicial functions, injustices, and opposition to police orders in high-stress situations like protests (ANZPAA, 2021).

Recently, there has been a sharp decline in public satisfaction with the police in general. For example, public trust in police in England and Wales has fallen from 62% in 2018 to 55% in 2020 (Office for National Statistics, 2020). In the US, trust in law enforcement fell from 53% in 2019 to 48% in 2020 (Brenan, 2020). Furthermore, from 2011 to 2019, there has been a significant increase in civil unrest such as the number of riots, general strikes, and anti-government demonstrations worldwide (Institute for Economics & Peace, 2020). Australia, historically known for high levels of satisfaction with the police, has also experienced a decline in public confidence (Morgan, 2017, 2021; Productivity Commission, 2023). The public rating of police ethics and honesty was at a high level in 2017, with 76% of the respondents considering it very high or high (Morgan, 2017). However, there was a significant decline in 2021, with only 51% of the public rating police ethics and honesty at a very high or high level, reaching a record low (Morgan, 2021). This decline in public confidence in police is a major source of concern for police-citizen relationships.

There are several possible explanations for this notable decline in public trust and confidence in police. Firstly, there is a growing societal awareness and scrutiny regarding a perceived lack of police accountability. High-profile controversial cases in the US involving police such as incidents involving George Floyd, Breonna Taylor, and others, have heightened public awareness of potential police misconduct (Sierra-Arévalo, Nix, & Mourtgos, 2023). The public’s demand for accountability in policing is reflected in large-scale movements like ‘Defund the Police’ and ‘Black Lives Matter’ which have been pivotal in bringing these critical issues into the public arena. These movements and demands of accountability also resonated outside of the US including countries such as Australia. The issue of police accountability in Australia found a parallel in the context of Aboriginal deaths in custody sparking protests and demonstrations across Australia. Hence, the raised awareness of potential police misconduct may have led to a decline in public trust and confidence in police.

Additionally, the rise of alternative ideologies, fuelled by conspiracy-driven extremism and ‘alternative facts’ media, has likely intensified misunderstandings and hostility towards the police (for example see Hardy, 2023). This is particularly evident in the growth of the ‘sovereign citizen’ movement, which rejects government and police authority, adhering instead to personal interpretations of the law. Such ideologies may have contributed to the declining public opinion of police, fostering a climate of distrust and scepticism towards law enforcement in an increasingly polarised society.

Lastly, the implementation of unpopular government policies during the pandemic likely played a significant role in escalating public dissatisfaction. Notably, Australia’s adoption of an aggressive zero-COVID suppression strategy, characterised by frequent and prolonged lockdowns and strict mask mandates, contributed to this discontent (for example see BBC News, 2021; Francis, 2020). As the primary enforcers of these measures, police officers often found themselves on the front lines, bearing the brunt of public frustration and backlash. This direct association with the enforcement of stringent and often controversial restrictions may have significantly impacted the public’s perception of the police, contributing to a decline in their overall approval and trust. Furthermore, there has been a growing distrust in government with police often seen as the frontline representatives of government (ANZPAA, 2021). Indeed, Prenzler and Ransley (2023) identified a decline in public trust towards Australian politicians and public servants over the past decades which is aligned with numerous scandals (see Prenzler & Ransley, 2023 for an overview).

While the aforementioned factors are plausible contributors to the recent decline in public trust and confidence in the police, it is important to recognise that the extent of their impact remains uncertain. It is not definitively clear if these specific factors were the primary drivers of the notable drop in public opinion, or if other, unidentified factors also played a significant role. The complexity of public sentiment means that attributing the decrease in trust and confidence to these issues alone may oversimplify a multifaceted issue. Therefore, a comprehensive analysis that considers a broader range of influences is necessary to fully understand the dynamics at play in shaping public attitudes towards police.

How public perceptions of police are formed

To address public perceptions of police, it is necessary to first understand how these perceptions are formed which can be a complex and multifaceted process. Socialisation and cultural beliefs can influence how someone perceives police such as values and beliefs transmitted by family as well as wider society attitudes towards criminal justice and law enforcement (Cheon et al., 2022). Further, these factors can be influenced by systemic racism, inequality, and police misconduct (Baranauskas, 2023; Coventry, Dawes, Moston, & Palmer, 2015; Lai & Zhao, 2010; McKernan & Weber, 2016). Personal experiences with police officers, either positive or negative, can also influence an individual’s perceptions (Murphy & Tyler, 2017). These experiences include being a victim or a witness of crime and interactions with police during traffic stops (Van Damme, Pauwels, & Svensson, 2015; Weitzer & Tuch, 2005).

However, most citizens have limited interactions with police, instead, the media portrayal of the police plays a substantial and crucial role in shaping public perceptions (Baranauskas, 2023; Engel, 2005). Indeed, the media is the primary source of information for most people (Prenzler, 2020). While Australians rely on a diverse range of media sources, with approximately 73% using online news, 42% accessing news through television, and 25% through social media and/or blogs (Hughes, 2022b, 2022d). Traditional media remains the preferred and trusted source for most Australians, while the adoption of digital news media is increasing (Hughes, 2022a). Despite traditional news outlets’ commitment to responsible journalism (Prenzler, 2020), they often have a selective focus on ‘newsworthy’ stories and spin articles to a more newsworthy perspective. Researchers have identified criteria used in the selection and editing of news stories, including ‘immediacy, dramatization, personalisation, simplification, titillation, conventionalism, and novelty’ (Chibnall, 1977, p. 23; Prenzler, 2020, pp. 37–38). When considering how this might influence public perceptions of police, positive media coverage can lead to more positive public attitudes, while negative coverage, particularly concerning police brutality incidents, tends to result in more negative opinions (Choi & Lee, 2021; Kesic, Ducat, & Thomas, 2012; Mazerolle, Bennett, Davis, Sargeant, & Manning, 2013; Tucker, 2019). Media coverage also influences the interpretation and response to police incidents and policies (Walker & Henderson, 2020). Overall, public perceptions of the police have significant implications for law enforcement practices, public safety, and the relationship between the police and the communities they serve. Understanding the factors that shape these perceptions is crucial for improving police-citizen relationships and designing effective strategies to build trust and confidence in policing. One way to understand the public’s main concerns of the police is through a procedural justice theory lens.

Procedural justice theory emphasises the importance of fairness in decision- making processes, involving components such as ‘voice, neutrality, respect, and trustworthiness’ (Tyler, 2003). Applying procedural justice principles in policing includes providing individuals with the opportunity to express their concerns, making unbiased and transparent decisions, treating individuals with respect, and being perceived as competent and reliable (Murphy & Tyler, 2017). When police officers adhere to these principles, individuals perceive the process as fair, leading to increased trust and legitimacy in the police and positive attitudes towards law enforcement (Hinds & Murphy, 2007; Murphy & Tyler, 2017; Murphy, Hinds, & Fleming, 2008; Tyler, Fagan, & Geller, 2014). Conversely, the absence of procedural justice principles can result in perceptions of unfairness, reduced trust, and negative attitudes towards the police (Murphy & Tyler, 2017; Tyler, Fagan, & Geller, 2014). Adhering to procedural justice principles can enhance trust, legitimacy, and the effectiveness of policing efforts (Murphy & Tyler, 2017; Tyler, 2003). Hence, to understand the shift in public opinion and the decline in sentiment regarding police, we can investigate which aspects of the four pillars of procedural justice framework are of particular concern to the public, shedding light on the influencing factors contributing to this noteworthy drop.

Overall, the recent decline in public trust and confidence in the police is a complex phenomenon. While there have been some recent societal shifts such as heightened awareness of police accountability issues, the emergence of alternative ideologies, and the enforcement of contentious government policies, which may have influenced public opinions about police, the actual factors and their precise impact and scope remain unknown. The current study is the first to take a broad exploratory approach to explore the factors behind the recent decline. Essentially, this study explores the declining public opinions of the police in Australia to identify the reasons behind the decline using a procedural justice framework. Specifically, it explores both the role of news media in shaping public perceptions and the public’s opinion of police directly. These research findings provide insights to policing agencies and policymakers on factors contributing to the decline in public opinions, enabling them to implement strategies and reforms to improve policing practices, community relations, and public trust in the police. The study aims to guide evidence-based policies that promote police accountability, transparency, and community engagement.

Method

Research aim

This study explored public opinions and sentiment towards police in an Australian context. Specifically, it aimed to identify factors that may have led to the recent decline in public opinions of police in Australia using a procedural justice framework. To do this, a multifaceted approach was employed (see Figure 1). First, as the media are a key influencer of public opinions, a media analysis was conducted to establish if the message being conveyed by the media had changed from 2017 when public opinions were notably high to 2021, a year marked by a reported decline in public opinions. Second, to explore the public’s opinions directly, an analysis of public commentary on media posts was conducted. To achieve the aims, four main questions were asked:

Figure 1. Overview of methods.

1. How were the media portraying police in 2017 and 2021?

2. Is the media portrayal of police in 2021 different from 2017?

3. What were the public opinions of police in 2017 and 2021?

4. Were the public opinions of police in 2021 different from 2017?

The Australian context

Prenzler and Ransley (2023, p. 1) outline that Australia is recognised as a prosperous and established democracy that operates under a parliamentary government system. This structure is distinct from the American model, as the executive branch is led by the majority faction in the parliament. The federal government is headed by a Prime Minister, while at the state level, Premiers lead, accompanied by ministers responsible for various government departments. Essentially, the two primary political parties in Australia exhibit similar policy positions, especially when compared to the broader ideological spectrum seen in the US between the Democratic and Republican parties. The Australian political landscape is characterised by a narrower range of policy difference, with both major parties often finding common ground on many key issues. This contrasts with the American political system where the Democratic and Republicans frequently display more distinct and polarised stances on various policies. Prenzler and Ransley (2023, p. 1) highlight that the primary responsibilities of government functions, particularly in sectors like criminal justice, are primarily managed by its six states and two territories (p. 1). They also note that Australia has generally rated highly on public sector integrity scales.

Policing in Australia is characterised by its uniformity across various jurisdictions. The country is divided into eight policing jurisdictions, each corresponding to each of the states/territories¹ and the federal jurisdiction. Despite this division, policing practices across these jurisdictions are remarkably similar (Hine & Carey, 2020). The consistency in procedures, regulations, and approaches to law enforcement ensures a cohesive policing strategy nationwide. This uniformity is a notable feature of the Australian policing system, differing from countries with more decentralised or varied policing practices.

Australia’s media ownership is smaller and more concentrated compared to other countries. For example, while the US has a diverse media ownership resulting in greater diversity in political and ideological perspectives across different newspapers, Australia has more of a high concentration of media ownership which leads to less diversity in viewpoints and editorial stances (Hughes, 2023). In Australia, media is regulated by the Australian Communications and Media Authority (ACMA) with government influence generally low.

Phase 1 – media analysis

Phase 1 – data source

Police-related headlines were drawn from leading Australian newspapers based on readership numbers and representation of each state/territory (see Appendix). Overall, the analysis encompassed 10 newspapers with a combined readership of 25.387 million readers, which is significant considering the population of Australia is only 26.437 million (Australian Bureau of Statistics, 2023). Therefore, these newspapers reach a broad and diverse audience, providing a comprehensive portrayal of the media’s representation of police throughout the continent.

Phase 1 – data collection

Headlines were extracted from the 10 newspapers using ProQuest’s Australian and New Zealand Newsstream database. Time periods were selected based on empirical research that revealed high and low points of public satisfaction in police (for example, see Gallup, 2017; Productivity Commission, 2017; Productivity Commission, 2023). Data were extracted for the one-year period preceding the public opinion polls in 2017 and 2021. Specifically, data were extracted for time-period 1: 1 May 2016, to 30 April 2017, and time-period 2: 1 May 2020, to 30 April 2021.

To locate relevant newspaper headlines, document titles were searched using the terms ‘police,’ ‘cops,’ or ‘law enforcement’. The search was restricted to newspaper sources, and the document types selected were articles and news pieces. Headlines were then sourced from the 10 newspapers. The study specifically focused on headlines rather than the entire articles because headlines often encapsulate the most salient aspects of news stories, capturing the primary message and angles that the media wishes to convey to its audiences (Gabielkov, Ramachandran, Chaintreau, & Legout, 2016). Indeed, Dor (2003) found that most people’s opinons are formed by only scanning the headlines rather than going through the complete story. Furthermore, Gabielkov, Ramachandran, Chaintreau, and Legout (2016) found that 59% of readers do not progress beyond headlines suggesting that headlines tend to be most influential and memorable in forming opinions rather than the article itself. Rameshbhai and Paulose (2019) outline the benefits of using headlines to research opinions in their sentiment analysis that mined opinions on newspaper headlines.

The initial search yielded 362 headlines from time-period 1 and 1,009 headlines from time-period 2. The headlines were then downloaded into an Excel spreadsheet. Data cleaning involved removing duplicate headlines to ensure that each article was unique and did not skew the data. Headlines were then checked for relevance, with headlines excluded if they pertained to alternative meanings of the terms such as ‘language police’ or ‘boxer cops a knock to the head’. Given that the current study was particularly interested in an Australian context, headlines related to international policing events were excluded from the dataset. Overall, after the data cleaning and exclusion process, there were 219 headlines from time-period 1 and 636 headlines from time-period 2 for analysis.

Phase 1 – data coding and analysis

This phase of the study aimed to specifically address the first two research questions. Therefore, headlines were coded for variables relating to the headline itself including; date published, newspaper publisher, newspaper ownership. The headline’s sentiment was also coded. Positive sentiment was coded when police were portrayed in a favourable or commendable lights, spotlighting their achievements, actions, or initiatives. These headlines use affirmative words or expressions that evoke sentiment of support, approval, or satisfaction concerning the police. Conversely, negative sentiments were coded if they conveyed criticism, challenges, or disapproval aimed at the police. Such headlines feature words or phrases that evoke sentiments of dissatisfaction, disappointment, frustration, or even anger towards the police. Neutral headlines were also coded which were those that conveyed in a straightforward, objective manner, without overtly praising or criticising the police.

To identify the main policing factors discussed by the media during the two time-periods, the headlines were also thematically analysed for ‘Topical Themes’. In doing so, we aimed to establish the message being sent by the media to the public about police which may, in turn, influence public opinions. These themes were not predetermined but emerged organically from an in-depth reading and analysis of the headlines. The process allowed for the capture of diverse representations present in the media coverage, providing a comprehensive understanding of the prevailing themes that may have influenced public perceptions of police.

Phase 2 – public commentary analysis

Phase 2 – data source

To investigate the public’s opinions directly, phase 2 analysed public commentary from a carefully selected sample (outlined below) of publicly available online newspaper posts on Facebook. The benefits of using social media comments in research to understand public perceptions are well-established within the policing literature (for example see Reddick et al., 2015; Bragias, Hine, & Fleet, 2021; Davenport-Klunder & Hine, 2023; Fleet & Hine, 2022) with benefits including the diverse range of opinions that social media can capture, its time-efficient and cost-effective approach, and the unobtrusive nature of online data collection. Specifically, Facebook is the most popular social media platform in Australia, with an overwhelming 89% of respondents indicating its use (Yellow Social Media Report Survey, 2020), and it is also the most popular platform across various demographics, thereby enabling a diverse and representative sample. Importantly, a substantial majority of Australians (70%) place greater trust in news disseminated through traditional media channels such as credible news outlets’ pages (Yellow Social Media Report Survey, 2020).

Phase 2 – data collection

The second phase specifically addresses research questions 3 and 4. To identify differences in public opinions, a systematic process of matching newspaper articles from the two distinct timeframes was conducted. Specifically, headlines for the two time- periods were matched based on their sentiment, topical theme, and subject matter (see Table 1). To do this, all newspaper articles posted to Facebook by the same publishers featured in the first phase were searched. This process resulted in seven newspaper articles for commentary extraction – the odd number arose from the pairing to account for a representative sample size of the original comments.

Table 1. Matched newspaper articles for public commentary extraction.

Newspaper article time-period 1	Date published	Number of comments	Newspaper article time-period 2	Date published	Number of comments	Matching details
Queensland Police release body-worn camera footage of speeding motorists	Courier Mail, 22 December 2016	23	Speeding blitz: 22 drivers, 3 hours, 1 road	Courier Mail, 9 July 2020	7	Traffic and road safety Mutual focus on speeding issues Common point of police-public interaction
Front line female police officers hurt in attacks	Daily Telegraph, 19 November 2016	52	Police face more ‘life and death’ situations than ever	Herald Sun, 9 November 2020	18	Police safety and occupational risks Focus on the risks, challenges, and occupational hazards faced by police officers
			Police leave in larger numbers despite recruitment drive	Sydney Morning Herald, 20 July 2020	49
Man shot at Westfield Hornsby was given day release from mental health facility	Sydney Morning Herald, 10 June 2016	57	‘Systemic failures’: Family performs courtroom haka to remember Jack Kokaua,	Sydney Morning Herald, 8 November 2020	29	Police use of force Police-involved fatalities by use of force Specifically concerning individuals with mental health issues

From the matched newspaper articles, the public comments posted to each article were extracted. The comments were manually exported to an Excel spreadsheet. To ensure accuracy and relevance to the research topic, irrelevant comments were excluded (for example tagging or comments about policing outside of Australia). Overall, the analysis consisted of 132 comments from time-period 1 and 102 comments from time-period 2. Lastly, it is worth noting that some comments may fall under multiple thematic categories. This overlap reflects the complexity of public opinion and the multi-dimensional nature of the sentiments expressed.

Phase 2 – data coding and analysis

The public comments were content analysed. Similar to the first phase, data were coded for sentiments of positive, neutral, and negative. To provide meaning to the public comments, a procedural justice framework was employed coding for themes of ‘voice’ (statements in relation to the opportunity to express opinions and concerns during an interaction with the police), ‘neutrality’ (statements in relation to impartiality of the police and/or the extent to which they treat everyone fairly and without bias), ‘respect’ (statements in relation to the behaviour and degree to which the police treat individuals with dignity and show an understanding of their needs and concerns), and ‘trust’ (statements in relation to confidence and trust in police and/or whether the police have the public’s best interest at heart).

Similar to the first phase, ‘Topical Themes’ were also thematically analysed to identify the primary themes prevalent in public discourse that might be influencing public opinion about the police. These themes were not pre-established; instead, they emerged naturally through the thematic analysis of the public commentary. This method facilitated the encapsulation of a broad spectrum of viewpoints expressed in public statements. Consequently, it provided a holistic understanding of the dominant themes that could have shaped the public’s perception of the police.

To ensure the quality and reliability of this study, a codebook was created for both phases to provide a consistent framework for coding and analysing (Nowell, Norris, White, & Moules, 2017). Double-coding was employed on a subset by an independent analyst to ensure consistency and reliability in the coding process (Cornish, Gillespie, & Zittoun, 2014; Elliott, 2018). Steps were taken to maintain rigor throughout the analysis. Coders engaged in both immersion and periodic distancing from the data, allowing for more critical, objective interpretations (Vaismoradi, Turunen, & Bondas, 2013). These steps combined ensure the study’s robustness and reliability.

Results

Study 1 – Media Analysis of Headlines

Descriptive analyses of the headlines revealed that there were almost three times more police-related articles published in time-period 2 (2020–2021) (n=636) compared to time-period 1 (2016–2017) (n=219).

Sentiment

The examination of the sentiments conveyed by the headlines to the public revealed that, while the quantity of newspaper articles increased from time-period 1 to time-period 2, the overarching sentiments remained somewhat consistent across both time-periods (see Table 2). Neutral headlines remained dominant and were exemplified by phrases like ‘police investigate’ or ‘police present evidence,’ and accounted for 69.46% of all headlines in the initial time-period and 76.26% of all headlines in the subsequent time-period. The next most frequently occurring sentiment portrayed by the media were the negative headlines. Such headlines often contain language that questions police competency or emphasises their failures, such as ‘ex-cop charged after drug and gun raids,’ or ‘violent dispossession at hands of police.’ In time-period 1, these sentiments made up 15.53% of the headlines, which increased to 18.71% in time-period 2. Like the positive sentiments, these proportions were relatively similar across the two time-periods. The most notable disparity, however, arose within the positive headlines. Positive headlines were characterised by phrases like ‘cops nab $101M ice haul’ or ‘speaking up for those who no longer can: police homicide boss awarded Australian Police Medal.’ Though positive sentiments were relatively infrequent in both time spans, the discernible discrepancy lay in their distribution. In time-period 2, a mere 4.87% of all headlines were of a positive sentiment, markedly fewer than the 14.16% observed in time-period 1. Although this difference was ultimately still less than 10%.

Table 2. Sentiment of headlines by time-period.

	Time-period 1		Time-period 2		Difference between time-periods
Sentiment	n	%	n	%	n	%
Positive	31	14.16	31	4.87	–	-9.29
Neutral	152	69.46	485	76.26	333	6.80
Negative	34	15.53	119	18.71	85	3.18
Unclear	2	0.91	1	0.16	-1	-0.75
Total	219	100.00	636	100.00	417	0.00

Topical themes

To identify the main topics discussed during both time-periods, the thematic analysis revealed an array of topical themes as outlined in Table 3. The number of articles in each theme experienced an increase from the 2016–17 period to 2020–21, however, the general sentiment across articles remained fairly stable. Neutral articles continued to be the most prevalent, while topics associated with controversies consistently garnered negative sentiment during both periods. Although positive stories were present, they were outnumbered by their neutral and negative counterparts.

Table 3. Topical themes of headlines by sentiment per time-period.

	Police investigations				Controversies and misconduct				Crime				Public safety and community				Police welfare
	Time-period 1		Time-period 2		Time-period 1		Time-period 2		Time-period 1		Time-period 2		Time-period 1		Time-period 2		Time-period 1		Time-period 2
Sentiment	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%
Negative	2	0.91	7	1.10	20	9.13	61	9.59	1	0.46	9	1.42	5	2.28	12	1.89	–	0.00	3	0.47
Neutral	50	22.83	130	20.44	7	3.20	19	2.99	27	12.33	81	12.74	14	6.39	63	9.91	4	1.88	24	3.77
Positive	5	2.28	5	0.79	–	0.00	–	0.00	13	5.94	3	0.47	4	1.83	7	1.10	3	1.37	7	1.10
Unclear	1	0.46	–	0.00	–	0.00	–	0.00	–	0.00	–	0.00	–	0.00	–	0.00	–	0.00	–	0.00
Total	58	26.48	142	22.33	27	12.33	80	12.58	41	18.72	93	14.62	23	10.50	82	12.89	7	3.20	34	5.35
Negative	4	1.83	13	2.04	1	0.46	4	0.63	1	0.46	5	0.79	-	0.00	5	0.79	34	15.53	119	18.71
Neutral	17	7.76	27	4.25	10	4.57	37	5.82	23	10.50	37	5.82	–	0.00	67	10.53	152	69.46	485	76.26
Positive	3	1.37	4	0.63	1	0.46	–	0.00	2	0.91	3	0.47	–	0.00	2	0.31	31	14.16	31	4.87
Unclear	–	0.00	1	0.16	–	0.00	–	0.00	1	0.46	–	0.00	–	0.00	–	0.00	2	0.91	1	0.16
Total	24	10.96	45	7.08	12	5.48	41	6.45	27	12.33	45	7.08	–	0.00	74	11.64	219		636

A notable distinction emerged between the two time-periods, primarily attributed to the global pandemic. During the second time-period, a distinct ‘COVID-19’ theme surfaced, absent during the initial period. This theme encompassed headlines pertaining to police activities related to COVID-19 rules and protocols such as mandatory vaccinations, quarantine breaches, restrictions, border controls, and mask enforcements. In total, there were 74 COVID-19 related headlines in the second time-period constituting 11.64% of all headlines for that period. These 74 headlines accounted for only a fraction of the additional 417 articles published in the later time- period compared to the first. Specifically, they represented approximately only one-sixth (17.75%) of the extra articles published in the later time-period. Interestingly, these COVID-19 articles predominately portrayed the police in a neutral light with 67 headlines conveying neutrality, along with only five negative headlines and two positive headlines. To explore the potential influence of the COVID-19 theme and its impact on the results, a subsequent analysis was conducted excluding the COVID-19 headlines. This revised analysis revealed that the majority of proportions between the two time-periods aligned even more closely.

Other themes to emerge were headlines related to police investigations and operations. This theme represented various actions and aspects directly related to police operations, investigations, tasks, and responsibilities. Essentially, this theme focused on the core functions and day-to-day operations of the police service and tended to be mostly neutral in sentiment. Similarly, a group of headlines emerged that focused on crime and criminal activities, emphasising the criminal elements the police must combat and investigate – these too tended to be neutral. Another theme revolved around traffic and road safety, encompassing headlines related to road safety campaigns, police response to accidents, and reporting of road tolls.

Another category to emerge related to public safety and community relations. More specifically, this theme focused on public safety, missing persons, and the role of the police in ensuring public safety and was also often portrayed in a neutral sentiment. Conversely, another set of headlines revolved around the welfare of police officers, encompassing their physical, mental, and emotional well-being and may involve stress, fatigue, police deaths, or other health-related issues tied to police work.

In contrast, a distinct set of headlines concentrated on controversies, misconduct, and challenges that were mostly negative in sentiment. These headlines typically stemmed from political issues, disagreements, limitations in police operations, corruption, misconduct, and disciplinary concerns. Lastly, a theme emerged that was more related to organisational levels of policing; that is, policy and governance and included aspects of governance issues, reforms, and changes to administrative structures.

Study 2 – Public Commentary

Overall sentiment

Analysis of the public’s comments revealed a notable shift in sentiment. There was a drop in positive sentiment from 41.82% of all comments in the first time-period to 14.77% in the second time-period (see Table 4). Additionally, neutral and negative sentiments increased. Neutral sentiments increased from 7.27% in time-period 1 to 25% in time-period 2, while negative sentiments increased from 49.10% of all comments in the first time-period to 60.23% in the second time-period.

Table 4. Sentiment of public commentary per time-period.

	Time-period 1		Time-period 2		Difference between time-periods
Sentiment	n	%	n	%	n	%
Positive	46	41.82	13	14.77	-33	-27.05
Neutral	8	7.27	22	25.00	14	17.73
Negative	54	49.10	53	60.23	-1	11.13
Unclear	–	0.00	–	0.00	–	0.00
Total	110	100.00	88	100.00	-22	0.00

Procedural justice themes

Themes of procedural justice were extracted from 158 of the total 198 comments, consisting of 76 comments from the first time-period and 82 from the second time-period. Table 5 shows there was a shift in public opinions about the police between the two time-periods.

Table 5. Procedural justice framing of public comments by sentiment per time-period.

	Trust				Neutrality				Respect				Voice				Total
	Time-period 1		Time-period 2		Time-period 1		Time-period 2		Time-period 1		Time-period 2		Time-period 1		Time-period 2		Time-period 1		Time-period 2
Sentiment	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%	n	%
Positive	29	38.67	10	21.28	–	0.00	–	0.00	1	100.00	–	0.00	–	0.00	–	0.00	30	39.47	10	12.20
Neutral	4	5.33	5	10.64	–	0.00	4	17.39	–	0.00	2	18.18	–	0.00	1	100.00	4	5.26	12	14.63
Negative	42	56.00	32	68.09	–	0.00	19	82.61	–	0.00	9	81.82	–	0.00	–	0.00	42	55.26	60	73.17
Total	75	100.00	47	100.00	–	0.00	23	100.00	1	100.00	11	100.00	–	0.00	1	100.00	76	100.00	82	100.00

Trust

For both time-periods, trust dominated the comments. However, the results indicate a shift from positive sentiments about trust to neutral and negative sentiments of trust from time-period 1 to time-period 2. Specifically, positive comments about trust declined from 38.67% in time-period 1 to 21.28% in time-period 2, while neutral comments showed a slight increase from 5.33% to 10.64%. Furthermore, negative trust comments also increased from 56.00% to 68.09%. These comments offer a nuanced perspective of public sentiment towards the police. Many voice concerns, especially regarding police behaviour during traffic enforcement, while others praise the officers’ unwavering dedication. Throughout these discussions, ethical considerations within the service remain a recurring theme.

Neutrality

Interestingly, there were no comments about neutrality in the first time- period but emerged as an area of public concern in time-period 2 with 82.61% of these comments being negative and 17.39% were neutral. These comments primarily highlighted concerns regarding the impartiality of traffic law enforcement, suggesting revenue-driven motives, particularly post-COVID-19 spending. Moreover, a significant portion of these comments underscored the inadequate training police receive in handling individuals with mental health issues, emphasising the need for proper education to ensure fair treatment. At the same time, some comments acknowledged the challenges officers face in situations they are not equipped for.

Respect

Compared to the first two categories, respect was relatively underrepresented with only one comment in the first time-period, although, this rose to 11 comments in the second time-period. When respect was mentioned, it was usually in a negative light (81.82%) with some (18.18%) neutral commentary. A dominant opinion within this theme was the demand for dignified treatment of individuals grappling with mental illness, asserting that mental illness should not equate to a death sentence. There were broader concerns about the impact of societal shifts towards political correctness on police authority. The emphasis was also placed on the police needing to earn respect by treating citizens respectfully and concerns raised over ethically questionable practices, such as the strip-searching of minors.

Voice

Voice was discussed the least of all the four pillars of procedural justice. It was only mentioned once in the second time-period. This singular comment emphasised the importance of the police listening to the public’s perspectives and the necessity for officers to exercise greater discretion, favouring cautioning over merely fulfilling quotas.

Discussion

This research sought to understand the factors that may have led to the recent decline in public opinions toward the police in Australia. To examine the media’s influence of public opinions, a thematic analysis was conducted of news headlines. Additionally, to examine the public’s opinions of police directly, a content analysis using a procedural justice framework was conducted of public comments on Facebook news posts as well as a thematic analysis to find the main topics discussed. The findings revealed that despite a fairly consistent and neutral representation of the police during the two time-periods by the media, there was a notable shift in public opinions from positive to more neutral and negative sentiments. In particular, the public expressed procedural justice concerns of trust, neutrality, and respect.

Media exposure

The findings revealed that the media portrayed the police fairly consistently across the two time-periods with the majority of headlines being neutral in both time-periods. This is, perhaps, not surprising and the neutral sentiment finding may be more a reflection of professional and objective journalism (given the data selected of popular and credible news outlets). However, it was notable that there were three times as many articles published in the second time-period (where public perceptions of police were reported as being low). Such large exposure of policing headlines may have played a pivotal role in influencing public opinions. Hence, while the proportion of negative articles in the media did not substantially increase, the sheer volume of both negative and general police-related headlines may have exerted an influence on public opinions. Indeed, Choi and Lee (2021) found a positive correlation indicating that the media can shape public perceptions of the police, with those perceptions aligning with the specific media content viewed by individuals. Furthermore, Roberston et al., (2023) found that with every negative word in the headline, the click-through rate increased. Overall, these findings suggest that when it comes to the sentiment being portrayed by the media about police, there were no substantial changes, however, the media was exposing the public to more policing matters. Hence, while the public may not be exposed to a higher proportion of negative media headlines, they may be more focused on these headlines – combined with the sheer volume of this exposure, this may be a contributing factor to explaining the decline in public opinion.

Newsworthiness

Aside from the sentiment, the findings revealed the topical themes being discussed about the police – that is, the policing topics that the media considered newsworthy. In both time-periods, the topics discussed by the media remained relatively stable. Though, there was the addition of a new topic in the second time-period of COVID-19 which was not present in the initial time-period. However, the introduction of COVID-19 into the headlines is unlikely to have had any substantial influence on public opinions. Firstly, headlines that were related to policing practices during COVID-19 were almost always presented with a neutral sentiment. Secondly, the number of headlines related to COVID-19 only represented a small proportion (less than one-sixth of the additional articles within the second time-period).

Additionally, the analysis suggests that the recent surge in public awareness and the movement to hold police accountable for potential misconduct incidents might not be a significant factor influencing public opinions by the media. While instances of police controversies and misconduct were identified as a recurring theme in the media, its presences were not exclusive to the second time-period, nor was it disproportionately highlighted. Hence, findings from the analysis of the topics discussed in the media do not appear to indicate any substantial difference between the two time- periods.

Ultimately, these findings combined suggest that despite previous research suggesting that the media plays a significant role in shaping public opinions (Choi & Lee, 2021; Donovan & Klahm, 2015; Dowler & Zawilski, 2007; Graziano, 2019; Johnson, Maguire, & Kuhns, 2014), the media in this data set may not have had a substantial impact in general. However, the sheer volume of articles published in the second time-period exposing the public to policing matters may have some influence.

Shifts in public opinions

Despite a fairly consistent representation of the police by the media across the two time- periods, the study revealed that there was a notable shift away from positive sentiments towards the police to more neutral and negative sentiments. This trend aligns with Roy Morgan’s image of professions surveys, which indicated a significant decrease in Australians’ satisfaction with police (Morgan, 2017, 2021) which is also a trend reflected globally (Brenan, 2020; Institute for Economics & Peace, 2020; Office for National Statistics, 2020).

More specifically, the analysis of the procedural justice elements revealed that the public had shifted from positive sentiments of trust to more negative sentiments of trust. This shift indicates more apprehension and scepticism among the public regarding their trust in police. Prenzler and Ransley (2023) identified a decline in public trust towards Australian politicians and public servants over the past decade. This decline is aligned with numerous scandals, which may have influenced public expectations about the ethics of public officials. As a result, Australians appear to be setting higher standards for their public representatives, becoming less tolerant and more vigilant about misconduct. This broader landscape offers context to the current study’s findings regarding the police. It is possible that the police are experiencing a similar erosion of public trust, consistent with this overarching trend.

These findings align with previous research that emphasises the importance of trust in shaping public opinions of the police (Tyler & Huo, 2002; Weitzer & Tuch, 2005). The percentage rise in negative-trust related comments may signal a decline in the perceived legitimacy of police actions and decisions, suggesting that public trust in police has diminished in the recent time- period. This finding highlights the need for police to prioritise rebuilding public trust through respectful interactions with the public (Tyler & Huo, 2002). Indeed, Murphy, Mazerolle, and Bennett (2014) found a direct link between trust and citizens’ willingness to obey and cooperate with the police. In particular, they suggest that trust is deeply influenced by personal interactions that accumulate over time through consistent positive experiences. For instance, Murphy, Mazerolle, and Bennett (2014) found that even when officers embraced procedural justice during common interactions such as roadside breath tests, it provided opportunities for drivers to voice concerns, ensuring neutrality by explaining the randomness of stops, and ultimately establishing trustworthiness by expressing concerns about community safety, and ending the interaction with gestures of respect towards the driver. This method offers an efficient and direct way for the police to cultivate positive community relations and therefore increase public confidence and trust (Murphy, Mazerolle, & Bennett, 2014).

While trust had always been a topic of interest across both time-periods, concerns about neutrality and respect were new concerns that were raised in time-period 2. The prominence of neutrality emerged in the latter time-period. Studies like Lum et al. (2011) and Cauffman et al. (2021) have previously shown that when the police are viewed as biased, their perceived legitimacy plummets. Tyler & Wakslak (2004) also emphasised that non-neutral police actions, especially in certain communities, contribute to dwindling trust. These findings suggest that the public is increasingly concerned about the neutrality of police actions and decisions which may impact their trust and perceptions of police. To address this concern, police should strive to maintain a neutral stance in their interactions and decision-making processes. For example, the media analysis revealed concerns of non-neutral policing and excessive force. Training and education programs for police that raise awareness about racial profiling and de-escalation techniques can help to improve neutrality (Weitzer & Tuch, 2005).

Finally, the results also indicated a recent decline in respect. The role of respect in policing is pivotal. Individuals, no matter their role in an interaction, expect respectful treatment from the police (Van Damme, Pauwels, & Svensson, 2015). Previous studies, like those by Porter and Prenzler (2017), show that perceived rudeness from officers leads to a significant number of complaints. Further, when police interactions are respectful, it strengthens public trust (Sunshine & Tyler, 2003; Mazerolle, Bennett, Davis, Sargeant, & Manning, 2013). When officers engage respectfully with individuals during encounters, it fosters fairness, justice, and procedural legitimacy, ultimately bolstering the public’s positive perceptions of police.

To strengthen the relationship between police and citizens, several established successful strategies could be applied. Specifically, to address these public concerns, policing authorities should focus on trust, neutrality, and respect. Weitzer and Tuch (2005) stress the importance of police officers consistently acting in the public’s best interest with sincerity, which can counteract negative sentiments stemming from misconduct allegations. Their research also underscores the benefits of community policing, as those who perceive its practice in their neighbourhoods generally have a more favourable opinion of the police and increased trust (Weitzer & Tuch, 2005).

To enhance police-citizen relations, policing agencies should prioritise strategies that facilitate procedurally just dialogue (Mazerolle, Bennett, Davis, Sargeant, & Manning, 2013). By ensuring citizens have opportunities to voice their opinions during interactions, it significantly boosts their perception of police legitimacy, trustworthiness, and their willingness to cooperate. Mazerolle, Bennett, Davis, Sargeant, and Manning (2013) underscored the importance of empowering individuals to express their views, which positively impacts their opinions about the police. Conversely, marginalising individuals or suppressing their voices can erode trust, leading to a widening disconnect between the police and the community (Mazerolle, Bennett, Davis, Sargeant, & Manning, 2013).

Furthermore, the New York Police Department’s Courtesy, Professionalism, and Respect (CPR) program, established in 1997, successfully engaged in respectful policing to decrease crime and citizen complaints (Davis, Mateu-Gelabert, & Miller, 2005). This initiative, through comprehensive training, emphasised values that resonated with the community, such as courtesy, professionalism and respect. A key feature was the active involvement of community leaders, an effort that promoted mutual understanding between the police force and the communities they serve. Notably, the program included verbal training to help officers de-escalate tense situations. While CPR was implemented citywide, its success in reducing public complaints was especially evident in precincts where leaders actively adapted and reinforced its principles, highlighting the importance of local leadership in community policing (Davis, Mateu-Gelabert, & Miller, 2005).

Limitations of the present study

These findings need to be considered within the scope and limitations of the study. The study focused on a specific sample of news outlets which may lead to sampling bias. The omission of smaller regional and local newspapers might have excluded localised sentiments. However, by integrating national papers with wide reach, the research aimed to minimise this impact, striving for a well-rounded representation inclusive of diverse demographic groups. Additionally, the focus on reputable news sources does not account for the influence of more sensationalist media outlets which could potentially instil more negative sentiments. Future research should examine the influence of sensationalist outlets on public opinions to comprehend the full spectrum of media influences on public opinions. Furthermore, the focus on headlines for the media analysis has the potential to miss the depth and context that might be found within the body of the articles. Nevertheless, it is worth noting that headlines often carry significant weight in framing a story’s message for the audience (Gabielkov, Ramachandran, Chaintreau, & Legout, 2016), often playing a more influential role in shaping public opinion than the extended article content.

While the use of Facebook comments as one of the data sources provides a valuable resource for gauging public opinion and behaviour, the data collected is subject to certain limitations regarding demographics. Due to the constraints of a secondary data source and ethical considerations, the demographics of the Facebook data remains unknown. Nevertheless, given Facebook’s widespread popularity in Australia (89% usage reported in the Yellow Social Media Report Survey, 2020) across diverse demographics, it still facilitated a broad and representative sample. Although it is not possible to definitively identify the commenters and their nationalities, including those from Australia, by centering our analysis on Australian newspapers and issues related to Australian policing, we could reasonably infer that the opinions expressed were primarily from individuals with a heightened interest in Australian law enforcement matters. Another concern of using Facebook comments is self-selection bias in that people who choose to comment on Facebook may not be representative of all users. Those who are more vocal and opinionated are likely to comment, potentially leading to an overrepresentation of extreme views. On the other hand, users might restrain their true opinions due to privacy concerns or the public nature of Facebook comments. However, it was beyond the scope of this study to focus on the extent of these views but, instead, to identify the main concerns about policing. Furthermore, unlike interviews and surveys where responses might be influenced by the questions asked, Facebook comments are unsolicited, providing a more natural and spontaneous insight into public opinion. In addition, Facebook comments often include detailed personal stories and experiences, offering depth that might not be captured through traditional surveys and polls (Burles & Bally, 2018). Future research may consider the use of a variety of data sources including traditional interviews that allows for more nuanced reflections.

Conclusion

This study explored how media portrayal and public perceptions of police evolved in an Australian context. To do this, it examined both the media headlines about police as well as public commentary posted to news articles comparing time-periods marked by high and low public satisfaction. Findings revealed a clear negative shift in public sentiment, despite the media maintaining a consistently neutral portrayal of police. However, while the tone of media coverage remained fairly consistent, the number of police-related articles surged in the second low-satisfaction time-period. Hence, the consistent exposure of police matters by the media (both negative and positive) may be influencing public opinions. This raises the possibility that the frequency of police-related media coverage, rather than the sentiment, could be contributing to public perceptions. An analysis of the public commentary revealed that negative concerns of trust, neutrality, and respect were driving the negative shift.

One potential explanation emerges: Australians may have rising expectations about police conduct. This shift emphasises the evolving nature of public opinion and underscores the importance of the pillars of procedural justice in shaping positive police-citizen relations. Police must embody these principles to potentially increase public satisfaction and foster a more cohesive police-citizen relationship.

Specifically, policing agencies should consider updating training programs to emphasise the importance of trust, neutrality, and respect in everyday police work. Police departments should implement robust feedback mechanisms to help understand community sentiments in real-time, allowing them to address concerns promptly. Moreover, these findings reiterate the need for the principles of procedural justice to be at the forefront of policing. As policing agencies integrate these findings and adjust their practices, it opens avenues for rebuilding trust which is vital for community safety and cohesion.

By consistently integrating trust, neutrality, and respect in their daily interactions and broader practices, police have the potential to rebuild and sustain a positive relationship with the public. Effective policing relies on public cooperation, and a trusting relationship facilitates not just cooperation but also compliance and mutual respect. In today’s changing society, the way police engage and communicate with the public can either strengthen or strain their relationship with the community.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. ACT Policing is under the jurisdiction of the Australian Federal Police.

2. No Western Australia newspapers were included due to database restrictions and paywalls on relevant archives. However, ‘The Australian,’ (a national paper) was included, to ensure a wide reach.

Appendix

Online Readership Figures of Major Australian Newspapers as of March 2021.

Table

State/Territory^2	Online newspapers (readership in millions)	Publisher
Australian Capital Territory (ACT)	The Canberra Times (0.74)	Australian Community Media
New South Wales (NSW)	Sydney Morning Herald (7.68)	Nine Entertainment Co.
	The Daily Telegraph (3.03)	NewsCorp Australia
Northern Territory (NT)	The Northern Territory News (0.047)	News Corp Australia
Queensland (Qld)	The Courier-Mail (1.77)	News Corp Australia
South Australia (SA)	The Advertiser (1.16)	News Corp Australia
Tasmania (Tas)	The Mercury (0.3)	News Corp Australia
Victoria (Vic)	The Age (5.19)	Fairfax Media
	Herald Sun (2.84)	News Limited
National	The Australian (2.63)	News Corp Australia

(Hughes, 2022c; Hughes, 2023).