Australian Institute of Criminology

Skip to content

Confidence in the criminal justice system

Trends & issues in crime and criminal justice no. 387

David Indermaur and Lynne Roberts
ISSN 1836-2206
Canberra: Australian Institute of Criminology, November 2009

Foreword | Confidence in the criminal justice system has emerged as a critical issue at the interface of the administration of justice and political pressures in western democracies. For more than a decade, governments in the West have felt acute pressure to make the criminal justice system more relevant, more transparent and more accountable. The 'crisis of confidence', particularly in judges and sentencing, has led to a range of high profile policy announcements seeking to 'modernise' the criminal justice system. This trend was most pronounced in the United Kingdom in the period from around 1998 to 2004 under the Blair government and led to an outpouring of analyses, investigations and reforms aimed at improving confidence in the criminal justice system. This paper reviews some of these developments but also takes a closer look at the nature of public confidence in the criminal justice system in Australia. Using the results of the latest Australian Survey of Social Attitudes, aspects of public confidence are examined with a particular focus on the evaporation of confidence in the criminal justice system from police, through courts to corrections. This 'evaporation effect' has been observed in all countries where confidence in various aspects of the criminal justice system has been studied.

Adam Tomison

Public confidence is fundamental to the operation of the criminal justice system. The system depends on the participation of victims and members of the public who participate as witnesses and jurors. Low levels of public confidence also leads to disrespect and dissatisfaction with those responsible for administering the system. The political debate surrounding dissatisfaction has become well established over the past decade. Given the centrality of public confidence, it is not surprising there has been intense interest in measuring, understanding and addressing this phenomenon.

This paper examines how confidence in the criminal justice system needs to be understood as a multidimensional construct with distinct differences in levels of confidence between the three major components of the system—police, courts and corrections. Public confidence declines from the police, to courts to prisons, suggesting the public views each component individually, rather than the criminal justice system as a whole. It is argued that the best way to improve confidence in these criminal justice institutions is to enhance the perception that the institution is acting on behalf of citizens and representing their interests.

International and Australian research

Roberts (2007) provided an international comparison of confidence levels in the criminal justice system with Australia recording a rating of 35 percent of citizens indicating a great deal or quite a lot of confidence in the system, comparing unfavourably with similar western countries. For example, the figure was 57 percent in Canada and 49 percent in the United Kingdom. Overall, Australia was ranked 27 out of the 36 countries included, with levels of confidence ranging from a high of 79 percent in Denmark to a low of 19 percent in Lithuania.

Recent Australian research (Jones et al. 2008) examined public confidence in the criminal justice system in New South Wales. This telephone survey asked 2,002 members of the public about their level of confidence in six aspects of the criminal justice system. The proportion who were fairly or very confident ranged from 30 percent (confidence in criminal justice system to deal with cases promptly) to 75 percent (confidence in the criminal justice system to respect the rights of the accused). However, the survey did not differentiate between different institutions within the criminal justice system.

Research in Canada, the United Kingdom and Australia has also examined levels of confidence in components of the criminal justice system. Research conducted by the Canadian government (Ipsos-Reid 2002 cited in Roberts 2007) depicted a systematic decline in levels of confidence, from the national police (88%), state police (82%), supreme court (78%), local court (62%) to the prison system (49%) and finally the parole system (36%). This decline or 'evaporation effect' has also been documented in the MORI 2003 survey in the UK (Hough & Roberts 2004). While three-quarters of respondents expressed confidence in the police, less than half the sample had confidence in the prison system.

Within Australia, findings from the 2003 Australian Survey of Social Attitudes (AuSSA) indicated that the majority (72%) of the public have a great deal or quite a lot of confidence in the police, while less than one-third (29%) have confidence in the courts and the legal system. While the confidence levels for both police and courts have gradually declined over time (in 1995, 76% reported confidence in the police and 35% in courts), public confidence levels in the police have consistently been at least double those in courts (Bean 2005).

Roberts (2007) hypothesised that confidence in the criminal justice system is typically lower than confidence in other public institutions as the criminal justice system has to cater for the competing interests of a wider range of parties (eg including victims and offenders) than many other public institutions. Other factors hypothesised to account for the low levels of confidence include the limited public knowledge of the criminal justice system beyond the police (Hough & Roberts 2004), limited contact with the criminal justice system beyond the police (Roberts 2007; Smith 2007) and negative media coverage of the criminal justice system (Hough & Roberts 2004).

In understanding confidence in the elements of the criminal justice system, it is important to note how the construct confidence interacts with, or is affected by, some other key measures. These include the accuracy of the individual's perception of crime, and their reliance on the news media as a source of information on it. These matters have been explored in research on attitudes to crime and justice (eg Hough & Roberts 2004; Indermaur & Roberts 2005; Jones et al. 2008; Roberts 2007; Roberts & Indermaur 2009). From these studies a consistent picture emerges. First, the news media provide the main source of information on crime and justice for most people. However the type of media that the public relies on varies (eg from broadsheet to tabloid, from analysis to infotainment). Indermaur and Roberts (2005) found that the accuracy of crime perception is associated with the main source of information on which respondents rely. Where talkback radio and commercial television are the mainstay, perceptions are least accurate. A similar finding is reported by Jones et al. (2008) in their analysis in New South Wales. However, as Indermaur and Roberts (2005) observed, associations here do not imply causality. Pfeiffer and colleagues (2005), analysing data from Germany, found that, while crime was decreasing in Germany, most people believed it was increasing and this belief was associated with the desire for tougher sentences. These researchers found that a belief in increasing crime was associated with a pattern of television viewing that focused on crime stories. Further, Roberts and Indermaur (2009) reported that the desire for harsher penalties was significantly and positively associated with the respondents' self-rated importance of television.

The evaporation of confidence is mirrored in the evaporation of trust and respect for different professionals within different agencies of the criminal justice system. The police consistently attract more trust and respect than judges. Lawyers, especially defence lawyers, attract some of the lowest ratings. For example, Canadian research reported 72 percent of the public expressed high levels of trust in police officers, compared with 59 percent for judges, 52 percent for prosecutors and 34 percent for defence lawyers (Ekos Research Associates 2000 cited in Roberts 2007).

Measuring confidence in the criminal justice system

Confidence in the criminal justice system can be measured in a number of ways. Ratings of trust and confidence are often used interchangeably and correlate highly (Roberts 2007). Alternative methods include asking the public to rate the performance of criminal justice staff or comparing levels of confidence in the criminal justice system with other public institutions (Roberts 2007).

Much of the research examining public confidence in criminal justice proceeds by asking respondents in general about their level of confidence in the criminal justice system or how well they think the criminal justice system is doing. However, by asking about the criminal justice system researchers may be creating grounds for further misunderstanding by masking different levels of confidence in its individual components. Examination of confidence in individual components shows that the public thinks about the police very differently than they think about courts and corrections (Roberts 2007). Further, individuals may vary in their understanding of the institutions that comprise the criminal justice system. For example, based on focus group research followed by a national survey in the United Kingdom, Smith (2007) reported that the criminal justice system is generally perceived by the public to consist of police and courts, with prisons and community correction agencies seldom included.

The Australian Survey of Social Attitudes

AuSSA is a biennial mail-out survey that collects ongoing data on Australians' social attitudes and behaviours over time (Gibson et al. 2005). Since its inception in 2003, AuSSA has included a range of crime and justice items. In 2007 the Australian Institute of Criminology commissioned new items, including measures of confidence in police, criminal courts and prisons.

AuSSA 2007 consisted of a survey completed by 8,133 adults from all states and territories in Australia. Three versions of the survey were fielded with final response rates ranging from 39 to 42 percent. The data have been weighted by education level to produce national estimates. The dataset (Phillips et al. 2008) for the current analysis was provided by the Australian National University. To assess levels of public confidence, the survey included four items about each level of the criminal justice system (police, courts, prisons). These items were used to form scales of confidence in each. Full details of the psychometric properties of the scales and statistical tests for the results reported below are available in Roberts and Indermaur (2009).

Confidence in police

A large minority of Australians have direct contact with the police every year. About four in 10 AuSSA 2007 respondents had some contact with police in the 12 months prior to the survey. Respondents (n=5,187 to 5,222) were asked how much confidence they had in the police for a range of situations. The results (Figure 1) indicate that the majority of respondents have quite a lot/a great deal of confidence in the police to solve crime (74%), to act fairly (74%) and to respond quickly to crime (54%). However, less than half (48%) had quite a lot/a great deal of confidence in the police to prevent crime.

Figure 1 Confidence in police (%)

Figure 1 Confidence in police

Survey respondents were asked their level of agreement with the statement 'There is a lot of corruption in the police force in my state or territory'. Almost half (49%) neither agreed nor disagreed, or were unable to decide. Approximately one-quarter each agreed (26%) and disagreed (25%) with the statement. Not surprisingly, those who agreed with the statement had significantly lower confidence in the police than those who could not decide or neither agreed nor disagreed, who, in turn, had lower confidence than those who disagreed with the statement. The level of confidence therefore appears to be linked with perceptions of the level of corruption in the police force. Those who perceive more corruption have lower confidence.

Confidence in criminal courts

The majority of Australians have little direct contact with the criminal courts. Only one in 20 (5%) AuSSA survey respondents reported any contact with a criminal court (magistrates, district or higher court) in the past year. AuSSA 2007 respondents were asked:

How much confidence do you have in the criminal courts...

  • to have regard for defendants' rights?
  • to have regard for victims' rights?
  • to deal with matters quickly?
  • to deal with matters fairly?

The responses, shown in Figure 2, indicate that Australians have more confidence in the criminal courts to have regard for defendants' rights (70%) than victims' rights (47%) or to deal with matters fairly (52%). Of concern, less than a quarter (22%) has quite a lot/a great deal of confidence in criminal courts to deal with matters quickly. The desire for harsher sentences was significantly negatively correlated with confidence in the courts (rho=-.25, p<.001). A higher degree of confidence in the courts was associated with a reduced desire for harsher sentences. Interestingly, the results from the AuSSA 2007 survey indicate that those who had contact with the courts over the previous 12 months had higher levels of confidence in the courts and were less likely to be in favour of tougher sentencing.

Figure 2 Confidence in criminal courts (%)

Figure 2 Confidence in criminal courts

Confidence in prisons

Respondents rated how much confidence they had in the prison system to perform a number of roles. As shown in Figure 3, the majority of survey respondents had very little or no confidence in the prison system in terms of rehabilitating prisoners (88%), in deterring future offending (85%), in teaching prisoners skills (68%) or as a form of punishment (59%).

Figure 3 Confidence in the prison system (%)

Figure 3 Confidence in the prison system 

Confidence across the criminal justice system

Responses from the four items about police, courts and prisons were used to construct scales of confidence in these institutions. Possible scores range from four to 16, with higher scores reflecting greater confidence. Confidence levels in the three sectors of the criminal justice system are presented for the five largest states in Figure 4. In each state, the public has greatest confidence in the police, followed by courts and the least confidence in prisons.

Figure 4 Confidence in police, courts and prisons by state (scale score)

Figure 4 Confidence in police, courts and prisons by state 

a: Five largest states (by population) shown only.

The scales measuring confidence in the police, prisons and courts are moderately positively correlated (police and courts r=.38, p<.001; courts and prisons r=.43, p<.001; police and prisons r=.29, p<.001). This means that while individuals differentiate between criminal justice institutions in terms of confidence, those who have more (or less) confidence in one institution tend to have more (or less) confidence in other institutions.

The evaporation effect

Of central interest in regard to public confidence in the criminal justice system is the distinct evaporation of confidence moving through the criminal justice system from police through courts to corrections. Understanding this effect requires an investigation of three distinct domains affecting public perceptions: public knowledge of the three parts of the criminal justice system; the media treatment of them; and what the three aspects of the criminal justice system represent at a psychosocial level.

Public knowledge and direct experience with the three parts of the criminal justice system decline with progress through it. Police are highly visible and the first point of contact with the criminal justice system. Both high visibility and increased contact are factors that have been linked with higher levels of confidence (Roberts 2007; Smith 2007). The public has less direct knowledge of prisons than other parts of the criminal justice system, with UK estimates that four out of five have had no direct contact with prisons (Roberts & Hough 2005).

Where members of the public do not have direct experience or knowledge of these institutions, they rely on media portrayals. Knowledge of prisons is largely based on media representations, with many sharing incorrect perceptions of prison life. Low levels of public confidence in prison may be associated with incorrect perceptions and underestimations of the severity of prison life (Roberts & Hough 2005). Research in the United Kingdom suggests that confidence in prisons is affected by the perception that prisons have become a soft option (Smith 2007).

Analysis of movement through the criminal justice system from a psychosocial perspective also aids understanding of the evaporation of public confidence. The crime control mandate of police is closely aligned with public sentiment (Jackson & Sunshine 2007; Roberts 2007) In his analysis of the 'universal hierarchy of confidence', Roberts (2007: 171) focuses on how members of the public appear to align themselves with the crime control model rather than the due process model. In essence, Roberts argues that the public is much more concerned about the effectiveness of the system in controlling crime and less with legal process, especially the rights of the accused. The guiding principle of the courts, that it is better to acquit 10 guilty people than convict one innocent one, is not enthusiastically shared by the public and Roberts cites evidence from the United Kingdom and Canada that indicates the public are less than satisfied with this legal balance. The reason that the police attract more confidence than the courts, Roberts argues, is that they are more closely aligned with the crime control mandate. The courts and the judiciary are seen as aligned with the less popular due process model.

The journey from the police to courts thus represents the movement from the rather appealing and entertaining focus on crime fighting through the psychologically ambiguous process of sentencing, mixing as it does, a concern for both punishment and treatment. The public consistently perceives the courts as too lenient (Indermaur & Roberts 2005). The crisis of confidence in the courts can be characterised by rifts between practitioners, governments, the media and the public. Although many commentators interpret the driving force to be for increased punishment, some (eg Freiberg 2005) have noted that most of the recent big sentencing reforms have mainly been for more transparency, accountability and certainty. This is reflected in calls for mandatory sentencing, truth in sentencing, determinate sentencing and similar mechanisms.

The transformation of posture is complete by the time attention is turned to corrections. By this stage the offender has become the focus of concern. The focus is now on what works, what doesn't, their needs and what can be done to assist them, ensure effective resettlement in the community and ensure that they don't reoffend. The industry acting on behalf of the offender shares some of the opprobrium attached to these popular social enemies.

Measurement limitations

The construct confidence may not be as simple is sometimes assumed. At an immediate level it can be read as synonymous with trust (Roberts 2007). However, the results discussed here suggest that what is being measured is more of an attitudinal variable than any reflection of judgment or assessment of performance. This may present problems where confidence is used as a type of performance measure. The notion of using an attitudinal measure as a form of performance indicator is flawed as it cannot be assumed to be a public assessment of performance. Political or bureaucratic imperatives to build trust in the criminal justice system may mask this vital point.

The status of confidence as an attitudinal variable can be seen in the comparisons between the social institutions. The continuum of confidence really reflects what they mean to the individual respondent. They can not reasonably be taken as a genuine attempt to assess the performance of the agency. Apart from anything else, this would be a task requiring information that the vast majority of respondents simply do not have. It is reasonable to assume that in answering this question about confidence the respondent will primarily draw on what they imagine the institution means to them, that is, how it sits in relation to their interests.

People are likely to have a positive attitude to those elements that they can perceive as acting on their behalf or in their interests. Thus defence and police will always receive higher confidence ratings than banks, journalists and politicians and this should not be seen as reflecting on their actual performance but rather as a general posture towards what they imagine the institution represents for them. This helps explain why the police can sustain high levels of confidence while at the same time only a minority of respondents indicate that they believe there is little corruption in the police force. Evidence for this view can be found in the results of focus groups conducted in the United Kingdom (Smith 2007). In a qualitative investigation of understanding of and responses to the standard questions used in the British Crime Survey to measure confidence in the criminal justice system, one of the significant findings was just how malleable these responses are. Smith (2007) found that the level of confidence can be altered by placing the questions at different points in the questionnaire. This suggests that 'top of the head' rather than deeply considered responses are given.

In understanding the lack of confidence that the public has in the criminal justice system, we need to look not only at differences between the various institutions of criminal justice, but to see them as reflections of government policy. It is now well documented that the current era is characterised by declining levels of trust in government, coupled with rising expectations of service.

Americans' trust in government has declined sharply in the last quarter century. A similar loss of trust has been found in 18 other democracies. Citizens now expect higher levels of recognition, respect, and status from the government. Criminal justice serves as a flashpoint for this change in citizen attitudes because so many Americans have contact with the criminal justice system and because the hierarchical design of criminal justice institutions juxtaposes so starkly with the egalitarian demands of the public (Sherman 2002: 25).

To sum up, measures of confidence should not be viewed uncritically. Tempting as it is to simply take measures of public confidence as performance indicators, as Roberts (2007: 176) rightly points out, ultimately these measures '...reveal a lot about the public and nothing about the system'.

Key implications

What are the implications of these findings for increasing public confidence in the criminal justice system? Smith (2007) reports that a focus on improving the service component in criminal justice is indicated. Activities to improve public confidence in the criminal justice system should focus on the police and courts, as these agencies have the highest public profile. Messages should address common misperceptions using clear, simple statements that explain how the system works. Public information should focus on the key areas of sentencing, improvements in services to victims, and citizen-focused policing. Criminal justice system activity should seek to improve community engagement by addressing local concerns and highlighting successes (Smith 2007).

Several major initiatives to increase public confidence in the courts have been undertaken in Australia. The South Australian Courts Administration Authority has developed a strategic plan and conducts regular surveys, reviews practices and uses the internet to provide public information. Targets in the plan included measurable increases in public confidence so that the effectiveness of the strategy can be evaluated. The Victorian Sentencing Advisory Council has been active in providing the public with information and seeking feedback on the work of the courts and sentencing as a way to build confidence (Freiberg 2008). Similarly in New South Wales, the Sentencing Advisory Council commissioned a survey into public confidence (Jones et al. 2008) in an attempt to better understand the nature and dimensions of public confidence.


The results reported here clarify three major points important to understanding and addressing confidence in criminal justice as a concept and as a measure. First criminal justice or the criminal justice system should not be the subject of investigations as it is a highly contaminated construct representing a crass amalgam of distinctly different social institutions. The level of confidence in each of these criminal justice institutions varies widely. Second, understanding the evaporation effect whereby confidence dries up moving from the police, to courts to prisons relies on an interpretation of what each of these institutions means to the individual respondent. Third, the best way to improve the confidence and regard of normal citizens to institutions within the criminal justice system is to enhance and optimise the perception that the institution is acting on behalf of citizens and representing their interests.


  • Bean C 2005. Is there a crisis of trust in Australia? In Wilson S et al. (eds), Australian social attitudes: the first report. Sydney: UNSW Press: 122–140
  • Freiberg A 2008. The Victorian Sentencing Advisory Council: incorporating community views into the sentencing process, in Freiberg A & Gelb K (eds), Penal populism, sentencing councils and sentencing policy. Leichhardt: Hawkins Press: 148–164
  • Freiberg A 2005. Sentencing, in Chappell D & Wilson P (eds), Issues in Australian crime and criminal justice. Chatswood: LexisNexis Butterworths
  • Gibson R et al. 2005. Introduction, in Wilson S et al. (eds), Australian social attitudes: the first report. Sydney: University of New South Wales Press: 1–11
  • Hough M & Roberts J 2004. Confidence in justice: an international review. London: Home Office
  • Indermaur D & Roberts L 2005. Perceptions of crime and justice, in Wilson S et al. (eds), Australian social attitudes: the first report. Sydney: UNSW Press: 141–160
  • Jackson J & Sunshine J 2007. Public confidence in policing. British journal of criminology 47: 214–233
  • Jones C, Weatherburn D & McFarlane K 2008. Public confidence in the New South Wales criminal justice system. Contemporary issues in crime and justice no. 118
  • Pfeiffer C, Windzio M & Kleinmann M 2005. Media use and its impacts on crime perception, sentencing attitudes and crime policy. European journal of criminology 2(3): 259–285
  • Phillips T et al. 2008. The Australian Survey of Social Attitudes, 2007. Canberra: Australian Social Science Data Archive
  • Roberts J 2007. Public confidence in criminal justice in Canada: a comparative and contextual analysis. Canadian journal of criminology and criminal justice 49: 153–184
  • Roberts J & Hough M 2005. The state of the prisons: exploring public knowledge and opinion. Howard journal 44: 286–306
  • Roberts LD & Indermaur D 2009. What Australians think about crime and justice: results from the 2007 Survey of Social Attitudes. Research and public policy series no. 101. Canberra: Australian Institute of Criminology. 
  • Sherman L 2002. Trust and confidence in criminal justice. National Institute of Justice journal no. 248: 22–31
  • Smith D 2007. Confidence in the criminal justice system: what lies beneath? Ministry of Justice research series 7/07. London: Ministry of Justice