Introduction

The Rule of Law holds that the actions of all individuals and entities in a society, including government institutions, are governed ‘by rules fixed and announced beforehand – rules which make it possible to foresee with fair certainty how the authority will use its coercive powers in given circumstances, and to plan one’s individual affairs on the basis of this knowledge’.1 Effective operation of the Rule of Law requires transparency and legitimacy, that is, that the authority ‘demonstrates to itself and to citizens its rightful possession of power. For institutions to flourish, they should demonstrate to citizens that they are trustworthy and that they possess legitimate authority.’2

In a society governed by the Rule of Law, the courts and judiciary are accorded institutional responsibility to interpret and apply the law.3 Unlike the legislative and executive branches of government, the judiciary has no independent force to compel individuals to comply with its application of legal principle, so perceptions of its legitimacy are even more important because ‘its authority… depends upon public acceptance of its role. That acceptance requires a certain level of faith.’4 An individual’s level of trust or confidence in courts will not only affect their willingness to turn to them for help, but also the likelihood that they will comply with court decisions.5

Public trust or faith in courts used to be taken for granted. However, declining trust in public institutions, including courts, has been a feature of democratic societies for some time,6 and has arguably been given further impetus recently as part of a deliberate political strategy to undermine the notion of objective truth which includes placing less value on expertise of all kinds.7

The rise of consumerism from the 1960s onwards drew greater attention to the quality of the individual experience of government, including the judiciary, so much so that by the late 1990s a major Australian research report on the relationship between the courts and the public was able to confidently assert that:

Public confidence in the work of the courts is no doubt won in a number of ways, in particular through transparently impartial adjudication and the skilful application of the law. But it is also won by an obvious commitment to public service in the activities of every facet of a court.8

At the same time, the task of winning that confidence has become more difficult. A changing discourse between courts and the media has resulted in increased criticism of criminal courts, focussed largely on frequent suggestions that they are ‘out of touch’ with community values in their sentencing of offenders.9 More broadly, the impartiality of the judiciary has also been questioned, with allegations of perceived gender bias,10 and concerns about the over-representation of Indigenous and minority groups in the justice system.11

In dialogues about the relationship between the courts and the public in the context of confidence and legitimacy, there is often an unstated assumption that such things are measurable. As a former Chief Justice of Australia pointed out:

We talk about public confidence in such things as the courts, the democratic process, the institutions of government, or other aspects of public life as though we are referring to an observable state of mind of a sufficiently large group to represent public opinion. We assume that such confidence is as measurable as, say, the approval of a political leader, or the popularity of a celebrity.12

Certainly, there are frequent attempts to measure trust and confidence in courts among the wider population and, more specifically, among court users and participants. The findings, in turn, are used to identify ways of improving public perceptions and the justice experience of individual members of the public, particular types of court users, as well as members of particular groups. However a diversity of approaches, some inherent limitations of survey methodology, and a lack of a clear conceptual framework, can make it difficult to make meaningful interpretations, or useful comparisons, from such survey data.

In this paper, we explore understandings of ‘trust’, how it is defined, and how it might be measured in the court context. Focussing on survey methodology and the use of a procedural fairness framework as the most expedient method available to courts, we then consider how its components may be more fully conceptualised and better applied to efforts to explore the factors that promote public confidence in, and acceptance of, the legitimacy of courts. We then identify some approaches to the practical application of this research in court management, particularly when standardised measures are used in surveys that are undertaken both regularly and frequently.

What is ‘trust’?

It is axiomatic that that any attempt to measure or assess the extent of a particular phenomenon, requires an agreed definition of that phenomenon. The definition needs to be sufficiently robust to delineate it from other, allied, or associated phenomena or concepts. A clear definition also allows meaningful comparisons to be made between different attempts to measure the same phenomenon, and exclude those that are ineffective.

The concept of ‘trust’ in the justice process has been researched from a variety of perspectives that explore different types of trust, e.g., individual trust; interpersonal trust; trust in statements (the credibility of a source of information); or trust in organizations or institutions, such as the police, the jury system, tax authorities and the courts. As Van Djik points out, institutional trust can be further distinguished between its ‘diffuse form (trust in an institution in general) and specific form (trust in an institution to do something specific such as uphold the law)’.13 So, for example, lines of research, for example, have explored the factors that promote trust and disclosure in police interviews,14 or disengagement from terrorism.15 Research has also explored cultures of mistrust of adult and child victim complainants in sexual assault cases,16 as well as courtroom factors promoting trust and credibility, such as cognitive fluency17 and the impact of various physical features of the court environment.18 Ongoing research has explored trust in the jury system;19 in particular the long-standing mistrust of juries,20 trust in witnesses from minority communities21 and trust in evidence given with the assistance of interpreters.22

However, as has often been observed, ‘trust’ is a term that ‘attracts frequent and disparate use’.23 The terms ‘trust’ and ‘confidence’ are often used without definition and frequently used interchangeably.24 Trust and confidence are also often conflated with legitimacy.25 How best, then to distinguish these concepts?

A common way to make a clear distinction between ‘trust and confidence’ is to conceptualise the former as ‘a confident expectation regarding another’s behaviour.’26 This definition has three core elements. Firstly, that trust is given prospectively, without actual evidence as to the way the other will behave. Giving trust, or choosing to trust, carries with it an element of risk.27 Secondly, that the individual who chooses to trust another has confidence in that other’s future actions and in their own ability to judge or predict those actions,28 so that confidence is an outcome of trust.29 Thirdly, ‘trust’ in this sense also carries with it an understanding that the future behaviour of the trusted other is viewed as preferable by the individual who trusts. This is made explicit in Braithwaite’s definition of trust as ‘a relationship between actors or groups in which one party adopts the position, expressed either verbally or behaviourally, that the other will pursue a course of action that is preferable to alternative courses of action.’30

Procedural justice theory views ‘trust’ as one of four components necessary to achieve public legitimacy, which might be equated with ‘confidence.’31 It views an individual’s decision to trust as based primarily on an assessment of the character of the other (whether individual or institution).

However, the distinction between ‘trust’ and ‘confidence’ can also be drawn in other ways. ‘Trust’ is often conceptualised as a characteristic of individuals, whereas confidence, in the sense of ‘public confidence’, is seen as related to the assessment or perception of the ability of an individual or institution to perform its functions e.g. does a given population have confidence in the police or government.32 Others argue that ‘confidence’ is a passive quality of an individual’s beliefs or attitudes, whereas ‘trust’ refers to the way individuals act.33 Research has established three theoretically and statistically validated empirical components or factors to be used in measuring institutional trust: Ability, Benevolence and Integrity,34 that appear to capture the competence, action and future behaviour components described above.

Regardless of how it is defined, ‘trust’ is also a concept that is not necessarily absolute, but variable. The degree or extent of trust is tied to the predictability of the behaviour of the ‘other’. So individuals may have greater or lesser degrees of trust in particular individuals or institutions.

Measuring Trust

A clear definition or understanding of what is being investigated is, therefore, the first step in investigating trust and confidence. The next step involves the selection of an appropriate method of measurement.

Survey questionnaires are often the principal method used for investigating trust and confidence. These take various forms, but the most common format asks respondents to indicate their agreement or disagreement with a series of statements — usually along a 5-point (‘Likert’-type) scale, where 1 = strongly disagree, 2 = disagree, 3 = neither disagree nor agree, 4 = agree, 5 = strongly agree).35

Large-scale general opinion surveys typically contain one or two questions about trust generally in courts/legal process, often in comparison with other institutions, as part of a much larger survey instrument. These questions are also often drafted in very general terms. The European Commission’s extensive Eurobarometer survey, for example, contains one question that asks respondents in its constituent countries whether they tend to trust or not to trust their country’s justice/legal system.36 The Australian Survey of Social Attitudes (AuSSA)37 takes a similar approach. For example, the 2017 AuSSA contained 111 questions of which only one related to trust in courts and simply asked respondents to rate their level of trust in each of courts and private companies on scale of 1 to 10 (between ‘No trust at all’ and ‘Complete trust’).38 Court-specific surveys, such as the US National Center for State Courts’ Annual Survey of Public Attitudes to State Courts39 and one conducted for the Californian Courts in 2005,40 investigate the general public’s attitudes to the courts in more detail. The Californian survey, for example, consisted of 98 questions which explored respondents’ attitudes to public institutions, their knowledge of courts, sources of information on courts, court experience, their expectations of and views about the performance of courts, their perceptions of procedural fairness and distributive justice, and collected detailed demographic information.41

Other surveys investigate the views and experience of specific justice system stakeholders, such as police, judges, prosecutors, criminal defence lawyers and witness assistance professionals.42 The European Commission for Efficiency in Justice has published a handbook to assist courts in EU member states design and conduct court user satisfaction surveys43 and courts that have adopted the International Framework on Court Excellence will conduct such surveys as a matter of routine.44

Researchers have also used survey questionnaires to explore jurors’ views and experiences of the court system,45 and to measure perceived community trust in the jury system.46 Most recently, researchers in the JUST: Jury System study developed a new 7-factor measure of the trust in the jury system. Use of those 7 items (ability, benevolence, integrity, identification, impartiality, fairness and respect), uniquely predicted jurors’ willingness to serve on a jury and their own preference for a jury trial.47

Survey methodology has limitations. The first point to note is that when individuals are asked simply to self-report their levels of trust or confidence, what their answers reveal is actually the respondent’s perceptions, their subjective assessment. So, the individual’s response indicates their state of belief about their level of trust. However, it is well-documented in social psychology that behaviour is a more accurate indicator of individual attitudes and beliefs than self-assessment; there is a wide gap between what people say they do and what they do in practice.48

Furthermore, questions that simply ask respondents to self-assess their overall level of trust of confidence in a particular institution or class of individuals, such as those used in the AuSSA survey, reveal little about the basis upon which the surveyed populations make that self-assessment, that is, the criteria they are using in determining their responses. There may also be a difference between the criteria that those populations actually apply in making their assessments and the criteria that those who work in the courts think that they ought to apply. As Gleeson pointed out, while judges might assume that: ‘That faith, or confidence, depends on or requires, … a satisfaction that the justice system is based upon values of independence, impartiality, integrity, and professionalism, and that, within the limits of ordinary human frailty, the system pursues those values faithfully,’49 … in fact

There may be a difference between what members of the public actually think about the courts, when they think about them at all, and what a judge believes would or should cause a hypothetical, fair-minded, well-informed observer to be concerned about judicial competence, independence or impartiality. …. Any professional group that seeks to assess the esteem in which it is held by outsiders is undertaking a risky exercise. They need to be sure they are listening to voices from outside, and that they are not working in an echo chamber.50

The difficulties associated with accurately measuring attitudes and beliefs using self-reported survey data means that researchers often prefer indirect methods, such as those typically used in experimental studies, in which participants are blind to the manipulated variables and are randomly assigned to manipulated groups, as a way of teasing out the effects of those variations unconsciously on participants’ behaviour and, hence, on their attitudes or beliefs.

Behaviour is also very context-dependent. Global or abstract beliefs and attitudes often differ from actions in response to a specific set of case facts or a particular context.51 For example, a notable paradox has been observed between the generally positive attitudes and high levels of citizens’ trust in juries, especially among citizens who serve on juries, and the widely held and persistent view in the community that jury duty is an onerous burden best avoided.52 Lack of information is one common cause of the observed value-action gap.53 Whatever the causes of the gap, the consequence is that different results are obtained using a survey questionnaire versus observations in a naturalistic setting.54

A further point to note about survey research is the need to ensure that the survey questions (‘measures’) are designed and tested so that they are valid and reliable. Social scientists distinguish three types of measurement validity: construct, content and criterion validity. ‘Construct validity’ refers to whether the test actually measures the concept it is intended to assess. ‘Content validity’ refers to whether the test is representative of what it purports to measure, i.e., does it address all relevant parts of the construct? ‘Criterion validity’ refers to whether the outcomes are similar to those obtained by other measures of the construct using a different test.55 So, for example, if the survey contains a number of questions directed to measuring the same thing, do those questions result in consistent responses?

Selecting a theoretical framework to underpin the design of a survey helps to address these issues. A survey that does not have a consistent explanation or theory for the way it is designed is likely to contain implicit biases or assumptions; these, in turn, will influence the validity of the measures, or survey questions.

Trust Research and theoretical frameworks

Approaches to researching trust and confidence in the justice system have been informed from a variety of theoretical viewpoints, some of which draw on research that has been done in the in other fields into the way that individuals form and develop trust. For example, recent research directed to individual and interpersonal trust in police-suspect interviews has applied social influence tenets to elicit cooperation (such as reciprocity, liking, social proofs, social authorities, commitment)56 derived from Cialdini’s marketing research.57 Research by Braithwaite and others on the way that individuals develop and apply criteria called ‘trust norms’ to identify others as trustworthy58 has also been applied in legal contexts, as discussed elsewhere in this edition.59

One of the most influential theoretical frameworks in the justice sector has been provided by Procedural Justice Theory. This approach shifts the measurement focus to the behaviours of an authority figure in terms of what they communicate regarding the quality of the social relationship between the individual citizen and the authority. This theory has been applied in measuring individuals’ views of police,60 courts61 and juries.62

Using procedural justice theory to investigate public perceptions of courts has a number of advantages. Firstly, it draws on an existing research base which demonstrates that perceived fairness of process is a key component in creating acceptance of the legitimacy of Western justice institutions, to the extent that the way that individuals are treated in the process appears to impact on their levels of confidence to a greater extent than the outcome of the case in which they are involved.63 Secondly, procedural justice research has tended to rely largely on survey data, which is likely to be a more cost-effective method for courts than more resource-intensive forms of qualitative and quantitative research.

In the remainder of this paper, we examine the procedural justice approach to investigating trust and confidence and discuss how best to construct survey measures that can be used to apply it to courts. We note that some procedural justice research using survey data has been criticised for lack of construct in the measures used,64 and there have also been concerns expressed about inconsistency between the measures used in different studies making it difficult to generalise and compare findings (external validity).65 It is also important that where a study attempts to measure more than one of the four components of procedural fairness, that measures used in relation to each are distinct.

A set of accepted standardised measures has now been widely used in procedural justice research into police in Australia.66 There is not yet any such set of standard measures in relation to courts, although the brief set of measures proposed in ICCE Measure 2 (discussed below) provides a useful starting point. We hope that this paper will contribute to the development of a more comprehensive suite of procedural justice measures for use in courts that may, in time, lead to the development of a more standardised approach.

We suggest that conceptualising the components of procedural justice more fully in the court context can assist courts to develop measures that investigate more deeply the factors that impact on public perceptions, rather than relying on court users’ self-assessments of ‘trust’ or ‘confidence’. This then results in a more robust set of findings, which have potential for wider application.

We conclude by expanding on the benefits to courts of doing this measurement work; of the ways in which more accurate and detailed data that sheds light on the way they are perceived by the public can be used by the courts. We suggest that more reliable insights into the factors that contribute to building trust and confidence in courts can be used not only to improve their service to their existing users, but to inform their engagement with the broader public, the business community and government.

Conceptualising Procedural Justice Measures for Courts

The ‘group values relational model’ that underpins procedural justice theory posits that our ‘sense of belonging and status in a social group or community affects our perceptions of justice.’67 An assessment that a procedure is fair results in the individual feeling respected and valued, which, in turn, promotes co-operation and compliance with the law. In this way, perceived fairness serves as a proxy for the level of confidence that citizens have in their courts, that is, the extent to which citizens perceive the courts to be acting legitimately and are inclined to respect and comply with court decisions.

Procedural fairness has emerged as the predominant approach to surveying perceptions of courts in recent years. It underpins (in part) the court user satisfaction component of the Global Measures of Court Performance68 proposed by the International Consortium for Court Excellence (‘Measure 2’), which includes a short sample survey as a template that can be adapted by individual courts ‘as necessary to suit local requirements and conditions.’69 This has been used by courts in a number of jurisdictions,70 including the Family Court and Federal Circuit of Australia71 and the County Court of Victoria.72 A procedural justice approach has also been influential in attempts to measure public perceptions of the United States State Courts73 and more detailed investigations into attitudes to courts in particular States.74

However, we have yet to see a consistent approach to the design of survey measures to investigate procedural fairness in courts. We suggest that this is, in part, because the four principal components of procedural fairness – ‘trust’, ‘voice’, ‘neutrality’ ‘respect’ have not yet been fully conceptualised for use in the court context. In the following section, we offer some thoughts on how this task might be approached, drawing on a number of these court surveys to identify how the investigation of these components might be mapped to specific survey measures. We follow this with a discussion of some of the more general factors for courts to consider in designing surveys to measure trust and confidence.

‘Trust’

Trust’ in the procedural justice sense refers to an assessment of the motivation or character of the decision-maker, that is, their degree of sincerity or the extent to which they are being honest, open, caring and trying to do what is right.75 It requires that individuals feel that both they and their problems are taken seriously by the legal system and requires demonstration of protection and concern for individual rights.76 In the case of courts, we suggest that it includes views about the trustworthiness of the institution as a whole, as well as of its constituent components, so that, for example, in a jury trial, it could encompass ‘the integrity and credibility of judges …, but could also refer to how the public think about the jury system itself and particular verdicts.’77

Survey measures might explore courts’ respect for rights in a general sense.78 At a more granular level, measures might explore behaviours that diminish trust — such as incompetence, indifference, neglect, venality—as well as those that serve to enhance it.79 So, for example, questions that explore perceptions of the extent to which judges or other court personnel are competent,80 are honest and fair,81 make timely decisions,82 that court proceedings are conducted efficiently,83 are understood by the public,84 and the extent to which juries are representative of the public,85 might all serve to explore different attributes or aspects of trust.

Research findings that demonstrate an association between citizens’ levels of trust and their participation in social and political activities,86 suggest that activities that promote citizens’ engagement with courts also help to engender trust.87 This illustrates the difference between measuring ‘abstract’ attitudes and perceptions as opposed to perceptions borne of concrete experiences. It also suggests that surveys of court users should include questions that ask participants about the frequency and extent to which they have had dealings with the courts and, if so, in what capacity.88

‘Voice’

‘Voice’ as it is conceptualised in procedural justice also refers to the capacity for an individual to tell their side of the story in their own words before a decision is made.89 The complex nature of legal proceedings, and the use of ‘legalese’, may result in individuals feeling ‘voiceless’ in a practical sense, even where they have legal representation. The increasing cost of access to justice and a consequent rise in the number of people accessing court without legal representation, also impact on individuals’ ability to ‘voice’. This is why some survey measures in courts with high numbers of unrepresented parties have explored in more detail aspects of the experience of court users who do not have legal representation such as their ability to navigate court processes, question witnesses, present their argument before a judge.90 Moves by courts to simplify procedures and provide greater support for unrepresented parties are all issues that might be explored as relevant to ‘voice.’91

The ability to voice also includes ‘being heard’ which signifies that the individual is listened to. It is important that this aspect is also explored in survey measures relating to users’ perceptions of their ability to express their voice.92

‘Neutrality’

The notion that judges are independent decision-makers, that they are ‘neutral, principled decision makers who make decisions based upon rules and not personal opinions’93 is a cornerstone of the Rule of Law. In procedural justice terms, neutrality must be demonstrated; transparency and openness, consistency in decision-making, reasoned explanations for decisions, clear and accessible processes, are all methods by which the public can be assured about judicial even-handedness and the absence of bias.94

Neutrality also encompasses the notion that all citizens are equal before the law. Concerns about political bias, as emerged in one recent US study,95 and concerns about racial and gender bias96 are aspects of neutrality that require attention in any investigation of public perceptions of the courts. Survey measures that explore respondents’ perceptions in this regard,97 for example, the extent to which they believe the court is unbiased and treats them or others fairly,98 together with the collection of demographic data that enables survey responses to be analysed, for example, on the basis of gender, group, or racial identification, can also enable courts to investigate in more detail the way that they are perceived by different sectors of the community.

‘Respect’

In procedural justice, ‘respect’ applies at all stages of the court process and encompasses the behaviour of all court personnel. It includes treating court participants professionally – that is, with dignity and courtesy.99 Recent US research confirms a public expectation of professional behaviour by judges, including decorum, courtesy and timeliness.100 Survey measures that explore perceptions of respect include explicit questions relating to the behaviour of judges and other court personnel,101 while others simply asked for impressions of the general level of respect and dignity accorded to citizens by courts.102

Measures might explore the extent to which the court environment is respectful, for example, the extent to which courts are accessible (easy to locate and find their way around)103 safe104 and provide a supportive physical environment with sufficient amenities for court users.105 They might also explore the extent to which court users are supplied with information about the court process (which can be confusing) and sources of help.106 Appropriate survey measures may explore court users’ experience of obtaining information from the court about court processes, whether in person, or via a court website,107 as well as their understanding of the process to be followed during the hearing of their case, the outcome and any next steps.108

Applying Procedural Fairness to Measure Trust and Confidence in Courts

In suggesting how the four components of procedural fairness might be more fully developed for application in the court context, we do not mean to suggest that this should lead to a ‘one size fits all’ approach to exploring confidence in courts. The first step for any court in determining how to explore public perceptions of their legitimacy will be to determine what they wish to measure and why.

What an individual court chooses to measure might be influenced by what is of particular importance to that court. A court might choose to focus on equality of treatment, for example, by attempting to measure the perceptions and experiences of different categories of courts users. The findings might also be intended to better help it to reach particular population in the community. A court with a high percentage of self-represented litigants might choose to focus in more detail on customer service, support and information. A new court might be more concerned about community knowledge or understanding of its role. A criminal court might choose to focus on the role of jurors or community perceptions of the sentencing process.

Data from these surveys can be used in a wide variety of ways. A court that identifies a need to build trust among a particular group of court users, for example, victim witnesses, might use that data to support an application to government for funding to build specialised support services and to communicate the availability of those support services to those individuals. Data might reveal a need for training of court staff and judges, the need for community education or outreach, the need to improve direct communication with court users, or to use different, more accessible, methods to reach particular groups of individuals (for example, the use of ‘Apps’ to provide information on mobile phones or other devices).

Regular, frequent, attempts to measure public perceptions can provide ‘real-time’ feedback to court staff and judges, enabling early identification of particular issues and timely responses. This might avoid problems escalating to the point where they require more resource-intensive interventions at a later stage. Regular surveys also enable a court to measure its performance over time, which can further assist in identifying methods and tools which work to improve public trust and confidence.

Surveys of trust and confidence using standardised measures can enable more meaningful comparisons – between individual courts or between individual court locations — which can in turn inform appropriate, targeted, responses. It can also facilitate courts learning from each other, and sharing resources, tips and tools that help build community trust and confidence.

However, while procedural fairness has been a useful lens for measuring individual experience, there is a need for caution in using court user surveys as a proxy for general public trust and confidence in justice. As we have noted, research demonstrates that concrete engagement with an institution generally has the effect of increasing an individual’s level of trust in it. However, abstract perceptions of courts can be very influential in public and political discourse, as criticisms of courts over sentencing demonstrates.109 Surveys that target a broader population group in the community are likely to yield more valuable data that can enable courts to tackle negative perceptions among individuals who have no direct experience of the courts themselves.

Finally, we suggest that there is value for courts in supplementing surveys with other methods of data collection, for example, debriefing juries, forming and holding regular meetings with court user groups and implementing a court complaints system. Indeed surveying court users about their experience without providing a direct avenue for complaints might appear to be something of a tokenistic exercise. It also provides a method of demonstrating a concrete response to feedback, which in turn may help further trust and confidence.

Conclusion

The task of measuring ‘trust’ is not a simple one and calls to mind the old adage ‘Not all that counts can be measured, and not all that is measured counts.’ In this article we have examined the way the concept of trust is defined and how it might best be measured in relation to courts, focussing on survey methodology as a practical and economic option.

Large scale quantitative surveys that simply ask a given population to report their global levels of ‘trust’ or ‘confidence’ yield very little in the way of meaningful data without some indication of how the individuals who respond to the survey understand those terms, and what factors, experience, or knowledge, influence their responses.

The value of comparative approaches is also questionable, for example, measuring trust in courts against trust in other institutions, for example, private corporations, unless it tells us how individuals make those comparisons. A finding that an individual trusts a court more than a private corporation is meaningless unless the individual is assessing both institutions on the same basis. Comparing levels of trust or confidence between different courts, or court systems, may be a more useful exercise, but is also one that needs to be approached carefully and, we would suggest, in a way that is sufficiently detailed and well-grounded to identify with specificity the contextual factors that may impact those differences.

Although not the only possible theoretical approach, procedural fairness has proved a useful framework for courts seeking to measure public perceptions of their work, and has been applied most often in the form of quantitative surveys, either of court users or of particular communities. In this article, we have discussed examples of survey measures from actual court user surveys that, either explicitly or implicitly adopt a procedural fairness approach.

As an approach to investigating courts it is at an early stage of development and there is little scholarship that has explored its application. We have attempted to address that gap by offering some suggestions as to how the components of procedural fairness might be more fully conceptualised to enable to courts to undertake more detailed investigations of public perceptions of their work. Further implementation of methods that permit inferential statistics to assess cause-and-effect relationships is recommended.

We have also identified some of the very real advantages for courts in attempting to measure public trust and confidence using this approach, particularly where standardised measures are adopted. Rather than viewing this as an irregular, ad hoc, activity, we suggest that it should be both a regular and frequent component of court management activities.

We have also noted some limitations and qualifications that suggest an ongoing need to consider carefully how best to measure public support for the role of courts in sustaining the rule of law. Court user surveys can be an important tool, but they do not provide a ‘magic bullet’ to ensuring a critical component of the Rule of Law – public trust and confidence in the courts.