The Public and the Justice System: Attitudes, Drivers and Behaviour - A Literature Review
This literature review examines evidence on what public attitudes to the justice system are, what drives these attitudes, what effect these attitudes have on behaviour, and what works to improve such attitudes.
3 Drivers of Attitudes
Key Findings
- While some studies suggest that demographics or experience of victimisation influence people's attitudes to police, courts, and the justice system as a whole, evidence suggests that these are secondary to the influence of people's experience of the justice system, and the key issue is how experiences vary across different groups in society. More Scottish research into the relationship between different social identities, experiences with the system, and people's evaluation of those experiences would be useful.
- The majority of Scottish people have had some direct contact with the justice system, most commonly the police. Some groups are more likely to have contact, such as young people and victims of crime.
- Contact with the justice system has a clear relationship with attitudes to the justice system. Evidence shows that evaluation of the system's procedural justice is crucial to the nature of the impact contact has on attitudes. That is, people's evaluation of whether the system and its representatives behaved fairly, neutrally, respectfully, allowed everyone to have their say, and had trustworthy motives.
- Those feeling a contact was procedurally just are more likely to be satisfied with the contact and its outcomes, even when outcomes are not favourable (though more research is needed on the link between process and outcome evaluation).
- Negative experiences with the justice system have a bigger impact on attitudes than positive ones.
- Perceptions of anti social behaviour and social cohesion in the local area, and general satisfaction with the local area have been found to be related to attitudes to local police, suggesting that people may hold police responsible for the state of the community. One US study has suggested the relationship could be the other way round, with confidence in police reassuring people about their local area.
- There is consensus in the literature that visibility of police does improve attitudes to the local police, especially where this includes increased availability and accessibility of police officers.
- Police engagement and communication with the public using newsletters have been shown to have a small positive impact on attitudes to local police.
- Perceptions of local crime and national crime are different and driven by different factors. Neither, however, seem to be major drivers of attitudes towards the justice system.
- It appears that having more knowledge about the justice system may have a small positive impact on attitudes, but this may be mediated by the source of knowledge, level of engagement with justice information (both in terms of openness to it, and opportunity to ask questions about it), and the moral context of information.
- The public are not universally punitive, and have different sentencing preferences across different kinds of crimes, offenders and circumstances. The relationship between sentencing preferences, or perceptions of sentencing practice, and attitudes towards the courts and the justice system, have not been fully explored, and more research is required to determine whether such a relationship exists, which direction it goes in, and the size of its influence.
- The media can be an important source of information for people, and can influence, but does not directly drive, people's attitudes to the justice system, both because the media itself is diverse and communicates a wide range of information and messages, and because people choose which media sources to consume and are happy to disregard messages that do not chime with their existing beliefs and experience.
- Public attitudes to the justice system are correlated with trust in other public institutions, and with levels of concern about social change. Attitudes will also depend in part on individual country or social group's history and their political and social context.
- The strongest drivers of public attitudes to the justice system seem to be those closest to the individual - personal experience of the justice system, and of the local neighbourhood. Other drivers, such as the media, may come into play when people do not have such direct experience.
Introduction
3.1 This section of the report identifies all the potential drivers of attitudes to the justice system examined in the literature review, and explores the extent to which each potential driver does in fact influence people's attitudes, and where possible the mechanism by which this happens.
Demographics
3.2 In social research, often the first port of call when trying to understand attitudes or experiences is to examine differences in such by demographics, i.e. variance across gender, age, ethnicity, and various measures of class and wealth. The topic in hand is no exception to this. Many surveys of attitudes to police, and courts, have looked at how more or less positive different groups are about the different parts of the justice system, and some have gone further and explored whether these differences are signs of a direct relationship between demographics and attitudes, or whether there are other mediating factors.
Gender
3.2.1 Starting, then, with gender, there are mixed findings both as to whether someone's gender influences their attitudes to the justice system and as to which gender is more positive about the system. The only Scottish study to consider this to date9 found that women were more likely to be satisfied with police than men (Hale & Uglow 2000), and similarly the European Social Survey has found that women were more likely to say they trust the police than men (Kääriäinen, 2007). Evidence from the BCS is more mixed in terms of confidence in the criminal justice system as a whole - a slightly lower proportion of younger women than younger men (16-24) were confident that the CJS meets the needs of victims, but 7% more women than men aged 75 and over were confident in this (Walker 2004, page 5).
3.2.2 Further afield, a review of US evidence concluded that there is no consensus about the effects of gender on attitudes towards the police (Brown & Benedict 2002), and this can be seen across individual studies. For example, Gainey and Payne in the US found that gender was not related to perceptions of police performance (2009: 314); a Canadian study found that men had more negative views of the police than women (O'Conner 2008); and Brick et al's youth survey found that once other factors were controlled, there was no gender difference in attitudes to the police (2009). This point is crucial - we will see below that multivariate analyses often show that the apparent relationship between demographics and attitudes to police (or the courts/criminal justice system) is due in fact to mediating factors.
3.2.3 Similarly, Moorhead et al's review of evidence on attitudes to the courts (civil and criminal), found that some studies found gender to be related to satisfaction with courts, and others found it not to be (2008: 16), and a US study found the relationship to be complex, with white females more negative about courts than white males, while there was no gender difference amongst Black or Latino people (Sun & Wu 2006).
3.2.4 On balance, then, it is hard to determine whether an overall gender effect exists in public attitudes to the justice system, police, or courts. Further research using existing Scottish data, or generating new qualitative or quantitative data, would be useful to explore whether such a gender effect exists, and where it does, why such an effect exists.
Age
3.2.5 Turning now to age, Scottish data from the 1980s and 90s showed that satisfaction with the police increased with age (Hale & Uglow 2000), and this is echoed by findings of an international survey, findings in Canada, and a review of US research (see Jang et al 2010; O'Conner 2008; and Brown & Benedict 2002 respectively). Studies in both Canada and the USA have found that age has the strongest relationship with attitudes to the police of all the demographic variables (see O'Conner 2008 and Wu & Sun 2009).
3.2.6 In an effort to understand this finding, in 1982 a large number of interviews were undertaken with Scots between the ages of 16 and 21 in order to ascertain their experience with and views of the police and the wider justice system. It was found that 79% of interviewees had had at least one contact with police, the majority of which were 'of a trivial nature'. This high level of contact was due to the large amount of time this age group spent in public spaces (Dobash et al 1990: 313).
3.2.7 It was found that these contacts had a negative impact on young people's perceptions of the police, no matter whether they were victims, witnesses or accused. So, for example, of those with contact 43% described police behaviour as 'bad' and 11% said they were 'frightened' by their contact. Overall, 49% of those with contact were dissatisfied with their contact and only 26% expressed satisfaction. Reasons for responses centred on perceived attitudes of and treatment by the police, e.g. level of politeness, rather than on the outcome of the contact (Ibid: 317). Flint et al have also found in interviews with young people in Scotland that they felt that the police could not protect them, but could harass them in public places10 (see Flint 2007). This links in with the concept of procedural justice which will be explored in-depth later in this report. To date, similar research into older people's experiences and views of the police, or indeed any part of the justice system, has not been published.
3.2.8 Perhaps unexpectedly, then, the BCS finds the opposite trend in regards to attitudes to the criminal justice system as a whole, with younger people tending to be more confident than older people (Walker 2004, see also Butler and McFarlane's 2009 findings in Australia). And as with gender, Moorhead et al's review of attitudes to courts found mixed findings as to the relevance of age to satisfaction with courts.
3.2.9 We might have expected common trends in the effect of age across aspects of the justice system, but this does not appear to be the case. It appears, rather, that that the influence of age might depend on the part of the justice system asked about and/or the framing of the question. No study to date has explored the reasons for this.
Ethnicity
3.2.10 The third demographic category explored in many studies is ethnicity. This is one area where there are clear differences across countries, because of the different ethnic groups making up, for example, the USA compared to Scotland, and the different historical relationships between different ethnic groups within countries. So, there is a large body of evidence in the North America showing that ethnic minorities (generally Black, Latino, and in one study, White people) are generally markedly less positive about the justice system (usually police) than the ethnic majority population (see O'Conner 2008; Brown & Benedict 2002; McCluskey et al 2008).
3.2.11 This does not seem to be the case in Scotland, or in England and Wales. A Scottish survey in 1996 found that ethnic minority respondents were generally more positive about the police than white respondents (although they were less likely to say they would be willing to involve themselves after witnessing criminal acts) (Ditton 1999).
3.2.12 This trend has been reflected throughout the years in the findings of the BCS, which routinely show that people from ethnic minority groups to be more confident in all aspects of the criminal justice system than white people, with the exception of confidence that the system respects the rights of the accused and treats them fairly (Allen et al 2007; Walker 2004). Similarly, ethnic minority respondents tend to be more confident in the police than white respondents, with the exception of the mixed ethnic group (Jansson 2006; Jansson et al 2007). Clancy et al's 2000 analysis of BCS questions on confidence in various strands of the justice system found that the 'net satisfaction' (positive responses minus negative responses) with the police for ethnic minorities was almost half that of white respondents, though both were still in the positive. For all other strands of the system, except prisons, net satisfaction was higher among ethnic minorities than white respondents (2001: 83). It should be noted however that in England and Wales it is only in recent years that people from ethnic minority groups have tended to have more favourable views of the police. Previously, opinions in many ethnic minority groups tended to be less favourable (see Bradford 2011).
3.2.13 Again, evidence in relation to courts is less clear, with six studies finding ethnicity to be not independently associated with satisfaction and two finding it to be linked to judgements of procedural justice (see below) (Moorhead et al 2008: 16, see also Higgins et al 2009).
3.2.14 There are two main limitations to the analyses outlined above. Firstly, by grouping all ethnic minorities together, these analyses mask quite large differences in the views of different ethnic groups. For example, analyses of 2004/05 BCS findings by ethnic group found that levels of confidence that witnesses were treated well were higher among people from Asian and Chinese or Other minority ethnic groups than among White people, while Black people were less confident than White people on this issue (Allen et al 2007). Other demographic factors may be relevant here, such as wealth, as average wealth differs across ethnic groups.
3.2.15 Secondly, these studies only give some insight into whether, and not into how ethnicity is related to attitudes to the justice system. As with age and gender, some suggest that it is not ethnicity in itself, but the way it interacts with other factors which influences attitudes, such as experience, with levels and type of experiences with the justice system varying across ethnic groups. We will return to this after considering one final demographic factor.
Class
3.2.16 A handful of studies have considered the impact of factors that could be said to relate to class, such as education, financial status and deprivation. For example, the SCJS has found that those living in the 15% most deprived areas in Scotland were significantly more negative about their local police across several indicators than those living in the rest of Scotland (Page et al 2010; Scottish Government Social Research 2011 - though this could be due to these areas being generally more exposed to the police than other areas, see below for evidence on the impact of direct contact). Other studies, from outside the UK, have found negative relationships between wealth and attitudes (Butler & McFarlane 2009; Kääriäinen 2007), though findings in regards to level of education are mixed (see Jang et al 2010; Brown & Benedict 2002), and findings in regards to courts were also mixed (Moorhead 2008).
Summary
3.2.17 We have seen throughout this discussion of the relationship between demographic factors and attitudes to the justice system that it is difficult to disentangle demographics from other factors. Many studies, and especially those utilising more sophisticated statistical techniques, question the relevance of demographics in the light of analyses where demographics cease to be of statistical significance once other factors are included in analysis, especially those around personal experience of the police (for example, experience of stop and search), and perceptions of things like anti social behaviour and personal safety (Myhill & Beak 2008; Flatley et al 2010; Gainey & Payne 2009; Clancy et al 2001; Jansson et al 2007; Jansson 2006).
3.2.18 In other words, it appears that it is not being a woman, or a young person per se that influences a person's attitudes to police or courts, but rather the experiences that a person has, and the kinds of experiences people go through vary across different demographic groups.
3.2.19 It should be noted that there is a lack of complex analysis of the relationship between belonging to different social groups in Scotland and attitudes towards, and experiences with, the justice system. A small number of surveys have ascertained some patterns, but there is a need for more complex statistical, or more in-depth qualitative, research to explore the ways in which the relationships between various social groups and the Scottish justice system operate.
3.2.20 So, in summary, while some studies suggest that people's gender, age, ethnicity and levels of deprivation influence their attitudes to police, courts, and the justice system as a whole, evidence suggests that these are secondary to the influence of people's experience of the justice system. This has not been tested in Scotland. The following sections will turn to these more experiential and perceptual factors, beginning with an exploration of how and to what extent experience of crime may influence attitudes to the justice system.
Experience of crime
3.3 Some research studies have examined the impact that experience of crime (i.e. being a victim of crime) has on attitudes to the justice system. As with perceptions of crime, simple comparison of victim and non-victim attitudes show some (small) differences, but more complex analysis suggests that other factors are at play. So, in Scotland, surveys in the 1980s and 1990s found that victims of crime tended to be less satisfied with local policing than non-victims (Hale & Uglow 2000), and more recently the SCJS has found some small differences between victim and non-victims in confidence in different aspects of the criminal justice system, with victims less confident that the criminal justice system is effective in bringing people who commit crimes to justice (victims 47%; non-victims 58%) and that the system deals with cases promptly and efficiently (victims 38%; non-victims 43%) (Scottish Government Social Research 2011: 104), and there have been similar findings in England and Wales and the US (see Jansson et al 2007 and Gainey & Payne 2009 respectively).
3.3.1 However, more complex statistical analysis shows that it is not the fact of victimhood itself which influences attitudes to the justice system, but the nature of the resulting contact with the system when the crime is reported. So, for example, Myhill and Beak have found that victimhood was not independently associated with confidence in the police, but perceptions of the standard of service received from the police were (2008: 9). Other studies do find a unique effect of victimisation, though it does tend to be small. So we find that by and large it is a mediating factor - contact with the justice system - which influences people's attitudes to the justice system, rather than the experience of being a victim per se. Clearly it is not just victims of crime who have contact with the justice system, and the next section of this report will examine the evidence on how and the extent to which contact with the system - whether as a victim, accused, witness, person stopped in the street, juror, and so on - influences attitudes to the justice system.
Prevalence of contact
3.4 Before we turn to evidence on direct experience of the justice system, It is worth considering how prevalent contact with the justice system is. The 2010/11 SCJS found that 70% of respondents who had heard of at least one criminal justice organisation had ever had contact with the police, 26% had had contact with the Scottish Court Service, 20% with the Crown Office and Procurator Fiscal Service; 19% with the judiciary, 10% with the Scottish Prison Service, and 8% with criminal justice social work. In contrast, 23% said they had never had any contact with the Scottish criminal justice system, (Scottish Government Social Research 2011: 101), a similar figure as in 2009/10 (25%, see Page et al 2010).
3.4.1 Of course rates of contact may vary across groups, and as we have seen, a 1990 study of young people's experience with police in Scotland found that 79% of Scottish youth had had contact with the police, partly due to the large amount of time they spent in public places. Very few of their respondents, however, had had contact with the courts, and even fewer with the prison service (Dobash et al 1990). And as we have seen above, victims of crime are more likely to have contact with the justice system, and certain groups are more likely to be victims, or repeat victims, than the public at large (such as young men).
3.4.2 Some have suggested that contact with the police has declined alongside crime and victimisation rates (Jokinen et al 2009, Bradford et al 2008), though as noted below the downward crime trend is not as clear cut in Scotland as it is in England and Wales.
Experience of the justice system
3.5 The evidence shows a clear relationship between direct experience of the justice system and attitudes towards it (Clancy et al 2001; Gau 2010), and many studies suggest that those who have had contact of some kind with the police or courts are generally less likely to be confident than those with no contact (Bradford, Jackson & Stanko 2009; Bradford et al 2008; Skogan 2009; Walker 2004; Allen et al 2007; Sun & Wu 2006; Dobash et al 1990; Brown & Benedict 2002; Orr & West 2007).
3.5.1 There have also been differences identified within 'those with contact' as to what impact that contact is likely to have on attitudes such as confidence (Wu & Sun 2009). So, for example, Genn & Paterson found that those whose civil 'justiciable problems' had led to legal proceedings or adjudication were consequently less positive about the courts than those whose problems had not. It has also been found in the USA that litigants tend to be less positive about fair procedures, outcomes and respectful treatment in the courts (Sun & Wu 2006). On the criminal side:
- a Scottish study looking at young people's attitudes to the police found that all those who had contact with the police evaluated that contact negatively, but this was especially the case with those who had been suspects (Dobash et al 1990);
- evaluations of different aspects of the criminal justice system have been found to vary across different user groups (van de Walle 2009);
- people who have been stopped by the police while on foot are particularly negative about the police (Allen et al 2007);
- analyses of the BCS and other data in England and Wales has concluded that public initiated contact has led to more confidence than police initiated contact (Bradford et al 2008, see also Myhill & Bradford 2012 and Skogan 2006); and
- in the USA, it has been found that evaluations of police effectiveness are influenced by public initiated contacts, while evaluations of prevalence of police misconduct are influenced more by police initiated contacts (Miller & Davis 2008 - though we will see below that views on police fairness are more important to subsequent legitimacy judgements than views on police effectiveness, see for example Tyler & Fagan 2008).
3.5.2 Evidence shows that contact in itself does not reduce confidence, but rather that the key factor is how satisfied people are with that contact, with people who are satisfied more likely to have confidence in the police or courts, and those not satisfied with the contact less likely to be confident (Moorhead et al 2008; Myhill & Beak 2007; Bradford, Jackson & Stanko 2009). Thus it is quality of contact which is key (see Gau 2010).
3.5.3 So what factors influence judgements about the quality of contact, and thus shape satisfaction with individual instances of contact with the justice system? Table 1 on page 26 outlines the different aspects of contact with the justice system which have been found to influence people's evaluations of that contact (these factors are statistically significant in quantitative research, or the key messages from qualitative research).
Procedural Justice
3.6 It is notable that most of the factors related to satisfaction and confidence in Table 1 are to do with the behaviour and conduct of police and court personnel, and the messages that this sends to those in contact with them, rather than to do with the outcome of the contact, as may be expected. This reflects a large and growing body of evidence which suggests that the most influential aspect of an experience with the justice system in terms of subsequent attitudes to the system is a person's perceptions of the 'procedural justice' of that experience, in terms of four elements of fairness:
- Having your say - research shows that it is important that people feel that they have an opportunity to state their case during a legal process, whether this is to the police at the time of an incident, or in a criminal or civil court (see Tyler et al 2007; Tyler 2004; Bradford, Jackson & Stanko 2009; Moorhead et al 2008). This clearly links to the 'being taken seriously' theme in Table 1.
- Neutrality - there is also strong evidence that people are more likely to be satisfied with contact, and confident generally, when they perceive police and courts to be neutral in their procedures, their conduct and their decision making. In other words, when police and courts are seen to be fair in the sense of being unbiased, using consistent procedures and facts and not personal opinions. This in itself requires some degree of openness and transparency in procedures and decisions if it is to be 'seen' (see Tyler et al 2007; Tyler & Huo 2002; Tyler 2004; Bradford, Jackson & Stanko 2009; Moorhead et al 2008; Buckler et al 2007). In a similar vein, Buhlmann and Kunz found that demonstrated judicial independence had a positive impact on the development of public trust in the justice system (Buhlmann & Kunz 2011). This links to the 'fair and appropriate conduct' section of Table 1.
- Treatment - Thirdly, it is crucial that people feel like they have been treated with respect (see the 'respect' and also 'inconvenience' sections of Table 1), and this includes being treated politely and feeling that their rights have been respected, for example the right to complain (see Tyler et al 2007; Duffy et al 2008; Bradford et al 2008; Tyler & Huo 2002; Tyler 2009; Moorhead et al 2008; Tyler 2004). This is especially important in that treatment by police officers or other justice professionals sends messages to members of the public about their standing in the community, i.e. whether they are worthy of the respect of the community (see Bradford, Jackson & Stanko 2009; Bradford et al 2008; Tyler 2004)
- Motive-based trust - Finally, the perceived motives of police and courts is important, in that they are felt to have the interests of members of the public - and the community at large - at heart (Tyler et al 2007; Tyler & Huo 2002; Moorhead et al 2008; Bradford, Jackson & Stanko 2009).
Table 1: Aspects of experiences relevant to evaluation of contact with the justice system
| Aspect of contact | Positive examples | Negative examples |
|---|---|---|
| Respect |
|
|
| Fair and appropriate conduct |
|
|
| Being taken seriously |
|
|
| Inconvenience |
|
|
| Outcome |
|
|
3.6.1 In terms of the means by which these four elements come to bear on actual experiences, Moorhead et al have suggested that the following 'process oriented factors' shape the evaluation of an individual court experience:
Factors Shaping Procedural Justice Judgements of a Court Experience
1. the expectations of, and information provided to, participants;
2. the quality of participation granted to participants (by which we mean in particular the extent to which, and process through which, participants are able to get their story out in a way that they view as accurate and fair);
3. the quality of treatment and, in particular, respect shown to the participant during their time at court;
4. issues of convenience and comfort (which may be themselves related to respect if inconvenience and discomfort is perceived as disrespectful treatment); and
5. judgments about court and tribunal personnel (what Tyler and Huo refer to as motive-based trust).
Moorhead et al 2008: 39-40
3.6.2 We would expect many of these to also hold true for contacts with the police, with the fourth point relating more to initial response time, than the convenience of court times.
3.6.3 The extent of the effect of these procedural justice judgements has been demonstrated by many and varying research studies. So, for example, Jackson and Sunshine found in their English study that of all the factors studied, procedural justice had the most impact on dissatisfaction with police effectiveness, in that it explained 35% of the variance in such dissatisfaction across the sample (2007: 227). Likewise Tyler's 1990 Chicago study found that a full 59% of variance in confidence in legal authorities was explained by judgments about the quality of police service and judgments about how fairly the police treat community residents (Tyler & Huo 2002: 197). All such studies find that procedural justice judgements influence attitudes to the legal authority more than outcomes or performance, though some find process to be far and away more influential (see the work of Tyler, Myhill & Bradford 2012), while others find performance to be nearly as important (See Hinds & Murphy 2007). For more on outcomes see below.
3.6.4 In terms of the exact nature of the influence of procedural justice on attitudes, a growing number of studies have found that perceiving procedural justice is associated with increased satisfaction with the legal authority in question, which in turn is associated with increased confidence, and increased perceived legitimacy of the authority. This chain of relationships has been supported by studies in the UK, US, and Australia, and by panel and longitudinal studies11 (see Murphy et al 2008; Gau 2010; Gau 2011; Hinds & Murphy 2007; Myhill & Bradford 2012; Tyler 2004; Tyler & Huo 2002).
3.6.5 This chain does not end with legitimacy, and chapter four of this report will look at the consequent impact of legitimacy on co operation and compliance with legal authorities.
3.6.6 While there is strong evidence that procedural justice impacts upon people's attitudes to the justice system, studies have found variance in its impact across different situations or social groups.
3.6.7 We have already seen that there are different patterns of attitudes across different kinds of justice user, which may well reflect the different expectations and relationships brought about by different situations, both on the part of the user and justice professionals. So, for example, Tyler (2010) found that procedural justice theory holds in prisons, but points out that 'The exact manner in which the elements of fair treatment are enacted depends on the setting and will vary depending whether that setting involves the courts, the police, or corrections' (page 130).
3.6.8 Similarly, studies which have examined the interaction between demographics, procedural justice, and willingness to cooperate with police, have found that ethnicity influences the strength of the link between procedural justice judgements and willingness to cooperate (see Murphy & Cherney 2011; Cherney & Murphy 2011). More on this in Chapter 4, but for now it is important to note that membership of particular social groups, especially minority groups, may well influence people's reactions to justice system experiences.
Asymmetry
3.7 Given this evidence that judgements about procedural justice are key to people's evaluations of their contact with the justice system, and the findings outlined at the start of this experience section that those who have had contact are often found to be less confident that those who have not, does this mean that the justice system is generally found by its users to not be procedurally fair? Once again, the evidence shows that the reality is more complex than this. In line with wider psychological literature (see Skogan 2006), research shows that negatively evaluated contacts with the justice system have a much larger impact on confidence than positively evaluated contacts (Bradford, Jackson & Stanko 2009; Bradford et al 2008; Tyler & Huo 2002; Myhill & Beak 2008; Skogan 2006). Negative contact are therefore not necessarily more common, but do have a bigger impact than positive ones. This is known as asymmetry.
3.7.1 There is some debate as to whether positively rated contacts have any impact at all on confidence in the justice system. Some studies have found that positively evaluated contacts have no impact on confidence levels (Skogan 2006), while others have found that they do have an impact, albeit a small one, especially in comparison to negatively evaluated contact (Bradford, Jackson & Stanko 2009; Bradford et al 2008, Tyler & Huo 2002).
3.7.2 Some studies also give a more nuanced picture. Myhill and Beak found in England and Wales that positively evaluated public initiated contact with police led to higher odds of being confident, but positively evaluated police-initiated contact made no difference to the odds of being confident (2008: 7). Similarly, Myhill and Bradford found that the effect of contact with the police as a victim was asymmetrical, as was police initiated contact (to the extent that satisfactory contacts had no statistical association with confidence), while non-victim self-initiated contacts were more symmetrical in the relative size of influence of positive and negative experiences on confidence (2012: 11-13). Finally, Bradford, Jackson & Stanko 2009 found in London that positively evaluated contacts can improve confidence in police fairness and community engagement, but not in police effectiveness (Bradford, Jackson & Stanko 2009, see also Bradford et al 2008).
Influence of attitudes on contact
3.8 Some also point to evidence that the relationship between attitudes and contact also goes the other way, with people's prior attitudes to the police or courts influencing how people interpret any contact they have with them (van de Walle 2009; Skogan 2006; Bradford, Jackson & Stanko 2009), and Skogan argues that the impact of prior attitudes on interpretation of direct experience is stronger and therefore more influential than the impact of experience on attitudes to the justice system.
3.8.1 Myhill and Bradford have recently examined this in England and Wales with panel data12 , and in line with previous US studies (see Tyler & Fagan 2008) they found an asymmetrical relationship between pre-existing confidence and satisfaction with contact with police, in that negative pre-existing views predicted dissatisfaction with an encounter with the police, but positive pre-existing views were not found to be related to positive evaluations of contact experiences. It should be noted, however, that despite this relationship between negative pre-existing views and dissatisfaction with contact, almost two thirds of those with pre-existing low confidence in their study were satisfied with their contact with the police as a victim (Myhill & Bradford 2012: 19-20).
3.8.2 Evidence also shows that direct contact with the justice system not only influences how positive or negative people are about the justice system, but actually changes the criteria by which people evaluate the justice system. So, for example, van de Walle cites research into attitudes to public services from the 1970s, which showed that evaluations based on recent experience are more pragmatic than the general, ideological or stereotypical evaluations of those without recent experience (2009: 386). Likewise, Moorhead et al refer to research into views on courts which found that those with personal experience based their views more on 'process evaluations' - i.e. the detail of what happened, when and how in their specific experience - while those without direct experience based their views more on evaluations of perceived relative fair treatment of people by the courts across different groups (i.e. how far the courts are perceived to discriminate) (2008: 27).
Outcomes
3.9 The final aspect of experience with the justice system examined by the literature is the outcome of such experience. We have seen above that there is strong evidence suggesting that the most important aspect of contact with the justice system in terms of subsequent attitudes to the system, is judgements about the fairness of the process undertaken. This does not mean, however, that the outcome of contact is irrelevant to attitudes to the justice system, and indeed evaluations of outcome are generally found to be significantly related to attitudes to the justice system, though not as strongly as process (See Tyler & Huo 2002; Moorhead et al 2008; Tyler & Fagan 2008; Myhill & Bradford 2012). Myhill and Bradford explain it thus:
It would be disingenuous to suggest that victims, particularly those who experience violent or other serious offences, place no value on seeing an offender brought to justice. A criminal justice outcome alone, though, appears less likely to result in overall satisfaction than good interpersonal treatment and a tailored response.
Myhill & Bradford 2012: 20
3.9.1 Some argue that the procedural justice literature does not sufficiently explore the relationship between outcomes and process in peoples evaluations of their experience (see Tankebe 2009 and Moorhead et al 2008). Tankebe points to studies which have shown that the relative importance of fair procedures and favourability of outcome varies according to factors such as the level of hostility in a civil dispute (2009: 14-15). Moorhead et al identify two levels of complexity here. First, it can be difficult to disentangle how far judgements of procedural justice are shaped by the favourability of the outcome, and vice versa. Second, it can be difficult to define the 'favourability' of an outcome - even if a person has ostensibly 'won' a court case, for example, they may not necessarily be content with the outcome for reasons such as the levels of compensation, whether there has been an apology, and so on (2008: 19). On the other hand, where studies ask participants themselves to assess outcome favourability (as in Tyler & Fagan 2008), we again run into problems around whether process factors influenced this assessment.
3.9.2 There is a large social psychological literature on this topic13 , which was too large and complex to include in the scope of this review. A separate review of this literature would be required to fully explore the relationship between process and outcomes assessments.
Vicarious Contact
3.10 Of course, direct personal experience is not the only source of information people have about individual contacts between members of the public and the justice system - people share their experiences with their friends and family, and read about the experiences of others in the media (which we will return to below). Such vicarious contact with the justice system is rarely explored in the literature, with the following exceptions:
- Brown and Benedict refer to research which showed that witnessing or having knowledge of police misconduct has significant negative impact on attitudes to the police (2002:551)
- A study in the USA showed that vicarious contacts significantly affected public attitudes to the police, though not as strongly as direct personal contact (Miller & Davis 2008)
- Also in the USA, it has been found that: "Being a crime victim or having a family member as a crime victim reduces the odds of being satisfied with police by approximately 34 percent. Even more significant, being harassed or having a family member being harassed by police substantially decreases the odds of satisfaction with police by 67 percent, controlling for all other individual indicators" (Wu & Sun 2009: 143, emphases added).
3.10.1 It stands to reason that vicarious contact will vary from person to person depending on their social network, and whose experiences they identify with. Vicarious experiences therefore may well vary across different social groups, for example across different ethnic groups.
3.10.2 This is once again an area which requires more research to determine the impact of vicarious contacts on attitudes towards the justice system. In particular, it would be useful to explore whether and to what extent vicarious contact influences attitudes to the justice system in Scotland, and how this varies across different social groups.
Perceptions of neighbourhood and Anti-Social Behaviour
3.11 The sections above have examined the influence of direct experience of crime, and of the justice system on confidence in police and courts. The remainder of this report will examine more indirect factors which influence attitudes to the justice system, beginning with the influence of people's perceptions of their own neighbourhood on their confidence in police14 . Three kinds of neighbourhood factors have been found to be related to confidence in the local police:
1. Perceptions or concern about anti social behaviour or disorder - several studies have found that people who perceive there to be high levels of ASB in their area, or are concerned about ASB in their area, are less likely to be confident in the police (Myhill & Beak 2008; Jackson et al 2009; Jackson & Bradford 2009; McCluskey et al 2008; Sprott & Doob 2008; Gainey & Payne 2009). Jackson et al have pointed out that the trend in confidence more closely matches perceptions of ASB than trends in actual crime rates in England and Wales (Jackson et al 2009:5), which echoes Sprott and Doob's assertion that while people may see ASB as a signal of crime rates more generally, crime and ASB rates are not in fact related (2008: 341-2).
2. Perceptions of social cohesion in the local area - Likewise, studies which have examined social cohesion and trust, that is, whether the people in the local area have shared values, morals, and identity, have found this to be significantly related to confidence in the police (Jackson et al 2009; Jackson & Bradford 2009; Jackson & Sunshine 2007; Maxon et al 2003)
3. General satisfaction with the local area - with those more satisfied with their area being more confident in the police (Mawby 2004; Sprott & Doob 2008; Wu & Sun 2008).
3.11.1 Taken together, these findings have been described as supporting the 'expressive model', defined thus:
The expressive perspective proposes that the police are viewed as representatives and guardians of the community. Individuals look to the police to strengthen moral structures. It follows that when signs of social breakdown are evident, the police will be judged accordingly, beyond and despite what is happening to crime…According to this perspective, more day-to-day concerns over anti-social behaviour, disorder and incivilities, signs of low community cohesion, and declines in moral authority move toward the foreground of public confidence in policing. In part this is because these things loom larger in most people's lives than do more serious crimes. People look on the police less as super-cops roaring past in patrol cars to the scene of a bank robbery and more as old-fashioned representatives of community values and norms - symbols of moral authority - there to address more everyday problems.
Jackson et al 2009: 5-6
3.11.2 This view of police as representatives and upholders of 'community' who are evaluated on the basis of people's perceptions of their local neighbourhood, is supported by three recent studies in England and Wales (Jackson et al 2009; Jackson & Bradford 2009; Jackson & Sunshine 2007), which also found that factors associated with the more 'instrumental model' - whereby police are judged according to crime rates, fear of crime and perceptions of risk - were much less influential on confidence in the police.
3.11.3 It is thus important that people can identify with the police, and this is where the influence of neighbourhood factors such as ASB and social cohesion aligns with procedural justice theory. It was seen above that through utilising procedural justice, officers communicate that people are their equals, and part of the same community. At this more general level police are judged according to how well they deal with ASB. This was demonstrated in a Scottish study examining responses to ASB, which found that when people were not happy with the response to complaints about ASB, they saw this as a sign both that the authorities 'did not care', and that the authorities saw their area as a 'problem area' (Flint 2007: 66). These sentiments reflect both the motive based trust strand of procedural justice and the importance of the role of police as guardians of the community.
3.11.4 In fact, analysis of a 2007 survey in the North East of England found that the three factors most influential on respondent's confidence in the police, in descending order, were:
- perceptions of whether police treat people with fairness and dignity;
- whether people identified with the police; and
- whether there were problems of social cohesion and trust in the community.
3.11.5 Together, these explanatory variables accounted for 39.7 and 34.9 per cent of the variance of the respective measures of public confidence in policing (Jackson & Sunshine 2007: 227).
3.11.6 This relationship between perceptions of the neighbourhood and confidence in the police may, however, not be as clear cut as it appears. Skogan has pointed out that many studies identify a relationship between these two factors, but some claim that perceptions of the neighbourhood shape confidence in the police, which others claim that confidence in the police shapes neighbourhood perceptions (and theoretically the relationship could go both ways). Skogan used American longitudinal data to examine the direction of the relationship statistically, and he found that, in fact, people did not hold police accountable for local crime, but confidence in police did reassure people about their neighbourhood conditions (Skogan 2009 (data from 1983)).
3.11.7 This study has not been replicated since, or in the UK, but if it were, and the findings held true, this would mean that the expressive model has the relationship the wrong way round, and in fact perceptions of people's neighbourhood, including perceptions of ASB, of social cohesion, and even general satisfaction, are shaped in part by confidence in the police, and not the other way around. Longitudinal research is required to ascertain whether this is the case.
Visibility of police
3.12 It is often held that the visibility of police officers in a community is key to both public confidence in the police, and public reassurance. The importance of this does seem to be borne out by the evidence, in terms both of research participants directly stating the importance of visibility, and also in terms of statistical examination of the relationship between having seen police patrols and attitudes to local police and crime.
3.12.1 So, for example, the past two years of the SCJS have found that 89% of Scottish people say that it is important to have a community police officer who knows and patrols the area (Page et al 2010: 112; Scottish Government Social Research 2011: 112). Similarly, Flint et al's research into the impact of ASB strategies in Scotland found that an increased police presence in people's neighbourhoods was "important in providing reassurance and in symbolising local agencies' commitment to the neighbourhoods and their willingness to tackle antisocial behaviour" (Flint et al 2007: 64-65) The researchers also noted that respondents were particularly positive about community police officers who they knew personally, and that not all residents saw increased police presence as reassuring.
3.12.2 Research in England and Wales has similarly found that the public identify increased foot patrols as the most likely thing to reassure them about crime (Charlton 2010; Mawby 2004).
3.12.3 Other research has examined this issue by looking at the difference in attitudes between people who had seen police patrols and people who had not. These studies have found that people who have seen police patrols are more likely to be satisfied with local police services (Hale & Uglow 2000, Scotland), or have more confidence in the police and local council in dealing with ASB and crime issues in the area (Smithsom & Flint 2006, England & Wales). Similarly, there has been a positive link found between awareness of local neighbourhood policing teams in England and Wales and evaluations of various aspects of local police performance (see Flatley et al 2010: 120); between living within 5 miles of a police station in a rural area and perceptions of the police (see Mawby 2004), and between having informal contact with police officers and attitudes towards the police (see Hough & Roberts 2004).
3.12.4 A handful of studies have examined the relative importance of visibility compared to other factors by including it in more complex multi-variable statistical analysis. Studies in London have found that perceptions of increased visibility were linked to more positive views about police effectiveness, fairness and community engagement (Bradford et al 2008), and that improvements in police visibility combined with police communication with the public, and the number and quality of personal contacts, are linked to increases in trust and confidence in police; decreases in fear of crime, and improvements in people's quality of life (Bradford, Jackson & Stanko 2009: 4). This link was found to be independent of the other potential drivers outlined in this paper, such as demographics, direct contact with police, and concerns about ASB.
3.12.5 Finally, a US panel study has found that the degree of visible policing perceived between the two waves of the survey significantly impacted both concern about crime and, to a greater extent, confidence in the police, and in fact recent police visibility was found to have the largest positive relationship with confidence in the police of all the factors examined by the study (Skogan 2009).
3.12.6 Some research has suggested that there is a preference for foot patrols and cycle patrols over patrols by car (Mawby 2004, Hough & Roberts 2004), and this combined with the importance of local officers who know the area outlined above, suggests that the issue is not only visibility, but the availability and accessibility of police officers, and the messages this sends about their level of engagement in the community.
3.12.7 Given the apparent consensus in the literature that police visibility does influence confidence in the police, it is worth taking a moment to consider Scottish findings in regards to perceived visibility of the police. Table 2 shows the findings of the most recent wave of the SCJS on the visibility of police in the local area.
Table 2: 2010/11 Findings on Visibility of Police in the Local Area
| Awareness of police patrols in your area | Views on levels of police patrols in your area | ||
|---|---|---|---|
| Police patrol regularly | 52% | Not enough | 56% |
| Police do not patrol regularly | 39% | About right | 38% |
| Do not know if police patrol regularly | 9% | Too much | 1% |
3.12.8 The table shows that while a small majority reported that police patrolled their area regularly, most people also felt that there were not enough patrols in their area. This was even more pronounced in the responses from the people living in the 15% most deprived areas, where 60% were aware of police patrols, but 65% felt that there were not enough patrols (Scottish Government Social Research 2011: 110-112).
Communication from police
3.13 A final potential driver which is related specifically to local police is communication from the police. This can take the form of newsletters, emails, local meetings and so on, which are designed to inform people about issues such as crime in their area, and local police operations. When this is examined in the literature, there tends to be a focus on newsletters.
3.13.1 As with police visibility, such direct communication between police and the community is something that research respondents themselves have suggested would increase their confidence or reassurance (Charlton 2010, Flint: 2007), and it has also been found to be statistically significantly linked to general confidence and judgements of police effectiveness (Bradford, Jackson & Stanko 2009; Bradford et al 2008).
3.13.2 The influence of police communication on confidence has been tested by two experiments in England. Hohl et al undertook a quasi-natural experiment in London whereby a leaflet containing information on what local police were doing was distributed in some areas between two sweeps of the Metropolitan Police's Public Attitudes Survey (the survey also covered control areas where leaflets were not dropped). The contents of the leaflet were designed using qualitative data on what Londoners wanted from such a leaflet, namely that it was clear, concise, locally relevant, from the police and containing information about neighbourhood policing. The researchers highlighted that communication that did not explicitly cater to what the public wanted risked having a negative impact. The study found that the leaflet had a statistically significant positive impact on confidence, specifically perceptions of police engagement and belief in police effectiveness, (though there was no effect on perception of police fairness) (Hohl et al 2010).
3.13.3 Undertaking a similar experiment, Quinton compared attitudes of those who had been given crime maps and/or local crime information, to those who had not. Those who had received the crime maps and information were found to be slightly more likely to be positive overall about neighbourhood policing, specifically in regards to thinking that the police understood their concerns and dealt with the things that mattered. Additionally, those who had been given the crime maps were slightly less likely to think that crime was going up locally (Quinton 2011).
3.13.4 Both of these pieces of research note that the effect of the communication was small (and indeed, in the case of Quinton's study, conflicting), particularly in comparison to the effect of personal encounters. Both did, however, emphasise that the effect was statistically significant, and suggested that if left as a one off exercise the effect would fade, but if such communication was sustained, the effect could potentially grow.
3.13.5 The fact that the influence of both of these communication experiments was primarily on people's ratings of police engagement, suggests that it could be as much the act of communication as the content of such which influenced attitudes. As such, tailoring communication to what people wanted to know was crucial, and the Metropolitan Police, in a separate exercise, have used surveys to identify four different groups of Londoner with different attitudes to crime and police, for whom they have developed different communication strategies (see Bradford et al 2008).
3.13.6 Some have suggested that there is potential for such communication to have a second type of effect, by improving public knowledge about the system and crime. The next section will examine the role of people's factual knowledge about the justice system, and about crime levels, in shaping their attitudes towards the justice system.
Knowledge about crime
3.14 It is widely asserted that a key driver of public attitudes to the justice system is people's level of factual knowledge about crime, and about the system itself. There are two parts to this theory - first that people hold (negative) misperceptions about crime and the justice system, and second that this has a negative impact on attitudes.
3.14.1 Taking crime first of all, people's perceptions of crime, its frequency and severity, could theoretically have an impact on attitudes to the justice system, if people hold the justice system directly responsible for such crime rates. We will now look at trends in perceptions of crime, both local and national, before considering what drives perceptions of crime, and what impact these perceptions have on attitudes to the justice system.
3.14.2 The SCJS has found that the proportion of Scottish adults believing that particular crimes were common in their local area generally decreased from the early 1990s to 2010/11 (Page et al 2010; Scottish Government Social Research 2011). There has been a similar trend in perceptions of overall crime rates in local areas (see Table 3), with a decrease in the proportion of respondents saying that there has been more crime over the past two years, and an increase in the proportion saying that there has been less crime or it is about the same.15 It is worth noting that women, victims of crime, and those living in the 15% most deprived areas in Scotland were more likely say there was more crime.
Table 3: SCJS Perceptions of local area crime over past 2 years
| 2006 | 2009/10 | 2010/11 | |
|---|---|---|---|
| A lot/little more crime | 32% | 25% | 23% |
| About the same | 57% | 61% | 64% |
| A lot/little less crime | 8% | 10% | 9% |
3.14.3 Interestingly, responses were very different when people were asked about crime in Scotland as a whole, with nearly twice as many respondents in 2009/10 thinking the crime rate had increased in Scotland as a whole than in their local area (45% and 23% respectively, Scottish Government Social Research 2011: 75 see also Hutton 2005 and Anderson et al 2002 for more Scottish findings).
3.14.4 This gap between perceptions of local and national crime rates is a recurring finding in many western countries (see Hough & Roberts 2007, Windzio et al 2007, Paulin et al 2003), including England and Wales where this gap has been widening since 2004/05 (see Flatley et al 2010). In England and Wales the consistently high proportion of people thinking that crime has increased nationally contradicts the BCS's own victimisation rate, which shows that crime has been decreasing in England and Wales since 1995 (Ibid). This suggests that perceptions of national crime rates do not have a direct relationship with actual crime rates in England and Wales. In contrast, perceptions of local crime rates did correlate to some extent with relative crime rates in the local area (Ibid: 112), and similarly Paulin et al found that local crime rate estimates were more accurate than estimates of national crime rates (2003: 12).
3.14.5 The actual crime trend is not so clear in Scotland, although recorded crime in 2010/11 was the lowest since 1976 (Scottish Government 2011a), and while there has been no clear trend in victimisation rates as measured in Scottish crime surveys since 1993, there was a statistically significant reduction in victimisation between 2008/09 and 2010/11 (see Scottish Government Social Research 2011: 23). Nonetheless, a higher proportion of adults believe that local, and national, crime is increasing, than those who think it is decreasing.
3.14.6 Similarly, studies in various countries have shown that people overestimate the proportion of crime that involves violence, and believe that crime is increasingly violent (see Hutton 2005; Paulin et al 2003; Anderson et al 2002; Hough & Roberts 2007; Butler & Mcfarlane 2009). So, for example, when a Scottish survey asked in 2005 what proportion of police recorded crime involved violence, the average response was 54%, compared to the actual proportion of 6% (Hutton 2005: 247), and comparison of results across jurisdictions and over time consistently find that people overestimate the proportion of crime which involves violence (with Scottish results the most extreme with 2% accurate estimate, 15% small overestimate, 83% large overestimate - see Hough & Roberts 2007: 208).
3.14.7 It seems clear, then, that many people have misperceptions about both the frequency of crime, especially national crime, and the severity of crime. As suggested above, there do seem to be different drivers of perceptions of crime rates at the local and the national level (see Jansson et al 2007). Jansson et al's analysis of the BCS suggests that perceptions of local crime is driven more by personal experience, factors to do with the local area, and risk of victimisation, all of which chimes with SCJS findings outlined above. On the other hand, they suggest that national crime rate perceptions are driven more by 'general characteristics' like age and newspaper readership (ibid: 55).
3.14.8 Duffy et al have also put forward some potential explanations from US evidence as to why views about the local area tend to be more positive. These centre on the idea of 'hometown favouritism' whereby people tend to see their own area as better than the others that make up the rest of the country, partly because they feel their values are matched by others in their area, and partly because of a belief that they would not choose to live in a worse than average place. They also point to perceptual biases around the scale of crime numbers at a national level (2008: 25).
3.14.9 In terms of the effect of these perceptions of crime on attitudes to the justice system, analyses of the BCS have found that while perceptions about crime are significantly related to confidence in the police, it is a small effect, and this is not a major driver of attitudes to the police (see Jackson & Sunshine 2007). Jackson et al (2009) argue that both fear of crime (as measured by perceived crime rates) and confidence in the police are driven by the same factors - namely perceptions of local disorder, cohesion and community efficacy - rather than being strongly related to one another.
3.14.10 To conclude then, we have seen that perceptions of local crime and national crime are different and driven by different factors, with local perceptions driven by personal experience of the local area and generally more positive than perceptions of national crime, which are more likely to be driven by perceptions of other areas, possibly gleaned from the media. Neither, however, seem to be major drivers of attitudes towards the justice system, though there is some suggestion that common drivers shape both confidence in the police and perceptions of crime, namely perceptions of the local area, which we explored above.
Knowledge about the justice system
3.15 Turning now to knowledge about the justice system itself, Table 4 shows results of Scottish surveys which have shown that a large proportion of Scottish people feel that they know very little about the Scottish criminal justice system:
Table 4: Self reported knowledge levels
| Not very much | Nothing at all | |
|---|---|---|
| The Scottish Criminal Justice System* | 64% | 17% |
| The Crown Office and Procurator Fiscal Service* | 57% | 23% |
| What happens to offenders in court^ | 61% | 9% |
| What happens to offenders in prison^ | 60% | 23% |
*Source - Scottish Crime and Justice Survey 2010/11 (Scottish Government Social Research 2011)
^Source - Hutton 2005: 247
3.15.1 Scottish studies have also found widespread lack of awareness about the range of sentencing options available to Scottish courts, particularly community sentences (Scottish Government Social Research 2011; TNS System Three; Anderson et al 2002, see also below).
3.15.2 Much of the literature outside of Scotland in regards to public knowledge about the justice system focuses on trends in sentencing. And so, studies have tested participants by asking them about their perceptions of justice system trends, and compared the responses to statistics to test the accuracy of these perceptions. These studies have found that people consistently underestimate the percentage of those found guilty for particular crimes who are given prison sentences (Hough & Moxon 1985; Roberts & Hough 2005c; Hough 2003; Roberts et al 2008; Allen et al 2007; Hough & Roberts 1999; Feilzer 2007).
3.15.3 It is established, then, that people's factual knowledge of the justice system, particularly sentencing practice, is low. The important question is whether and to what extent this has an effect on public attitudes to the system. There are broadly two ways that research has explored whether there is a relationship between knowledge and attitudes - by analysing the findings of surveys which have asked about both, and by conducting experiments where participants are provided with information in various ways, and their knowledge and attitudes both before and after are measured16 . These experiments are different from the police communication experiments outlined above in that they were designed to convey facts and figures rather than act as reflexive engagement between police and their local community.
3.15.4 Taking the survey analysis first, some analyses of the BCS and similar survey in other countries have identified correlations between knowledge levels and attitudes to the justice system, especially views of sentencing and judges (Butler & Mcfarlane 2009, Hough & Roberts 1998, 1999, 2004, 2007). So, for example, Hough and Roberts (1999) found misperceptions about trends to be correlated with the attitude that sentencing is too soft, as shown in the box below.
3.15.5 Most of these analyses, however, do not measure the size of the impact of knowledge levels on attitudes; explore whether the relationship is statistically significant; or whether there are other factors responsible for the apparent relationship, in other words, whether there is some other factor which influences both knowledge and attitudes (such as source of information - see section below - or trust in different information sources).
Correlations between Knowledge of Trends and Attitudes to Sentencing
1 Changes in national crime rate. (Those saying `a lot more crime' were most likely to think sentences far too soft.)
2 Changes in use of imprisonment. (Those saying prison use `the same/down' were most likely to say sentences far too soft.)
3 Estimated number of convicted muggers who were sent to prison. (Under-estimators were most likely to say sentences were far too soft.)
4 The proportion of recorded crime involving violence. (Over-estimators were most likely to say sentences were far too soft.)
5 Estimated number of convicted burglars who were sent to prison. (Under-estimators were most likely to say sentences were far too soft.)
16 Estimates of the clear-up rate. (Under-estimators were most likely to say sentences were far too soft.)
from Hough & Roberts 1999
3.15.6 This means that while the evidence suggests that there is a link between knowledge levels and attitudes to the system and sentencing, from this evidence we cannot definitively conclude that a direct relationship exists, or indeed which direction it goes in - i.e. whether it is low knowledge which leads to low trust and confidence or whether low trust and confidence results in a lack of interest or trust in any information which could improve knowledge.
3.15.7 Moving on to information experiments then, we will look at studies which involve the communication of (mainly statistical) information about the justice system and its trends. This has been done in a range of different ways, such as booklets, seminars, videos, and deliberative workshops17 , and some experiments have compared the effects of different communication modes. These experiments find that provision of such information leads to some improvement in knowledge about the system (as measured by before and after tests), and some small improvements in attitudes to the justice system (Mirrlees-Black 2002, Singer & Cooper 2009, Salisbury 2004, Chapman et al 2002, Hough & Park 2002).
3.15.8 These findings, however, are not as straightforward as they may appear. Mirrlees-Black (2002) and Salisbury et al (2004) have found that, despite the fact that participants themselves attributed increased satisfaction to their increased knowledge, there was not a statistical link between the two. Both studies put forward a different potential explanation for this.
3.15.9 Mirrlees-Black (2002) found that those who had not improved their scores in the knowledge test were just as likely to have improved confidence as those who had improved their scores, and concluded that improved attitudes are retained for longer than improved knowledge.
3.15.10 Salisbury et al found that while knowledge levels had only improved for the groups who had received the booklet, there were similar increases in confidence among all participants, i.e. both those who had, and had not received the booklet, i.e. those who had only been surveyed at two points in time. This suggests that it was the fact of being engaged with on the subject which increased confidence. This chimes with the findings outlined above on communication with police, that the fact of communication increases confidence more than the factual content of communication.
3.15.11 There is also some complexity in the findings around which respondents change their attitudes and in which direction. The studies where changes in attitudes were analysed further found that some people's attitudes became more negative rather than more positive, and some people were more likely than others to have become more positive in their attitudes.
3.15.12 So, for example, those who were less positive after receiving information in Mirrlees-Black's study explained this by referring to "the length of a custodial sentence served, greater awareness of crime generally and a concern that they had been fed propaganda" (Mirrlees-Black 2002: 188). This suggests a mistrust of the source of information, which we will return to below. Chapman et al also found that those with lowest confidence were more likely to increase in confidence, and those who said they were 'very interested' in law and order issues were less likely to change their attitudes (2002: xi). This mirrors Feilzer's point that the extent to which improved knowledge can increase positive attitudes depends on the levels of knowledge people already have, the interest they have in the issues, and the extent to which information can reflect a system which 'deserves' positive attitudes (see Feilzer 2007).
3.15.13 All in all, the evidence on the relationship between knowledge levels and attitudes to the justice system shows a complex picture, in which it is difficult to isolate the effect of people's knowledge from other factors to do with the source of such knowledge. This is both in terms of where or who it comes from, and how it is communicated.
3.15.14 Taking these one at a time, there is some evidence that some people do not trust information they perceive as coming from 'the government' about justice, or indeed generally, especially information in the form of statistics. It has been shown that the same crime information is more likely to be trusted, and thus taken on board, if it is believed to be from an independent source than if it is seen as 'government information' (Duffy et al 2002).
3.15.15 Similarly, a study into trust in statistics generally in England and Wales has shown that more people disagreed than agreed that official statistics are accurate (40% and 32% respectively - Bailey et al 2010). The majority of respondents (60% and 61%) felt that government officials, and the media, do not present statistics honestly, reflecting concern about 'spin' amongst justice research participants (see also Stead et al 2002; Maruna & King 2008; Duffy et al 2008). And when respondents were asked about trust in specific statistics, where those statistics were not trusted, the most common reason given tended to be that they did not match personal experience (Duffy et al 2008). This chimes with the wider findings of this review that personal experience is more influential to attitudes than system knowledge.
3.15.16 On the other hand, Duffy et al found that police were amongst the most trusted sources of information, but this trust was not being capitalised upon, as they also found that police were amongst the least often cited sources of information (2008: 41).
3.15.17 In terms of how information is communicated, there is some evidence that the more interactive the method of communication, the more impact it may have - whether it be a leaflet handed out by police rather than posted (Singer & Cooper 2009), or a format which allows people to ask questions and discuss the issues, such as a deliberative workshop (see Green 2006 and Hough & Park 2002). This seems to suggest that two way communication may be more effective than one way 'education' of the public, in which they are provided with information without the opportunity to discuss it or ask questions.
3.15.18 Finally, in regards to the tone and content of communication, some research has pointed to the importance of context, and of any explicit or underlying moral and emotional messages in the communication of information to the public. So, for example, Stead et al's focus groups with people who had been provided information found that, on the one hand, information about the size of the prison population did not mean much to their participants, as they were not clear on how big it should be. On the other hand, points about the values underlying community sentences (such as 'paying back' and 'making good') resonated much more with participants than points about their effectiveness (Stead et al 2002: 4).
3.15.19 This demonstrates that there is an important moral dimension to the way that people understand what the justice system is, does, and should be doing. This cannot be separated from 'information' and 'knowledge', which can never be morally neutral, in that what information is communicated will always be selective and carry assumptions about what is important.
3.15.20 We have seen, then, that the evidence around the effect of knowledge about the justice system on attitudes towards it is complex. It appears that having more knowledge may have a small positive effect on attitudes, but this may be mediated by the source of knowledge, level of engagement with justice information (both in terms of openness to it, and opportunity to ask questions about it), and the moral context of information. We will return later the role of the main source of indirect information about the justice system - the media. First we will examine people's perceptions of sentencing.
Sentencing perceptions and preferences
3.16 In the second chapter of this report we saw that despite the fact that most survey respondents tend to say the courts are too lenient, people's preferences in specific cases tend to be very similar to what the courts would do. We also saw in the previous section that there is a correlation between knowledge about sentencing and attitudes to the justice system. This section will examine the evidence on sentencing attitudes and preferences in more depth, concluding with a consideration of whether and to what extent such attitudes and preferences influence attitudes to the justice system itself.
3.16.1 So, as we have seen, when asked in polls or surveys whether the courts are 'too tough', 'too lenient' or 'about right', the vast majority say that the courts are 'a little', or 'much' too lenient. This has been found in Scotland (Anderson et al 2002), England and Wales (Allen et al 2007; Hough & Roberts 2004; Smith 2010), and internationally (Hough & Roberts 1999; Stalans 2009; Roberts & Stalans 1997; Butler & McFarlane 2009). For example, Anderson et al found in a Scottish survey that 3% of respondents thought that current sentencing was too tough, while 32% thought it was `a little too lenient' and 38% that it was `much too lenient' (Anderson et al 2002: Section 3.5). This suggests that generally people are not supportive of sentencing practice in Scotland.
3.16.2 Further research, however, suggests that this finding is not a direct commentary on what courts are actually doing. This can be demonstrated by studies which ask participants to decide on sentences for particular case studies. These studies tend to find that the sentences people choose are on average about the same, or more lenient, than the sentence which would be imposed in a real world court. This has been found in Scotland (Anderson et al 2002; Hutton 2005; TNS System Three 2007), and further afield, including England and Wales, Poland, Canada and Australia (Hough & Roberts 1998; Krajewski 2009; Roberts & Hough 2005c; Chapman et al 2002; St Amand & Zamble 2001).
3.16.3 The exceptions to this, where sentences chosen were found to be markedly more punitive than the sentence that would be imposed by the courts, are in three kinds of circumstances. First, when studies take place in the Netherlands (Keijser & Elffers 2009), or Switzerland (Kuhn 2002 - though it should be noted that researchers found that results were skewed by a small number of very punitive respondents, and actually most people were more lenient than the courts - page 123-124). It is possible that sentences are more lenient in these countries than countries like Scotland. Second, in scenarios where the most serious kinds of cases are used (see Roberts & Hough 2005c) people have been found to be more punitive than actual courts. And third, people have been found to advocate longer prison sentences than the courts impose, where they have chosen a prison sentence for the scenario (see Roberts et al 2008).
3.16.4 So what is the reason for the apparent difference in attitudes to sentencing across these two kinds of question? Does it simply demonstrate misperceptions amongst the public as to what sentences courts are imposing? Or is it more to do with the ways that people answer different kinds of questions?
3.16.5 The key difference between the two questions (about the leniency of courts and about sentencing preferences for specific case studies) is the generality of the former compared to the specificity of the latter. Research has shown that when people answer general questions about sentencing, they use mental shortcuts to formulate answers, and use the most easily accessible information to do so. This results in people thinking of the worst kinds of offences, and offenders - i.e. very serious crimes committed by repeat offenders, which are not representative of crime in general. In this situation, people are also inclined to think of the most memorable sentences, and assume that a small number of these are representative of sentencing as a whole i.e. atypically lenient ones (see Stalans 2002; Roberts & Stalans 1997; Stalans 2009; Hutton 2005; Hough & Roberts 1999; Keijser & Elffers 2009). In these ways, short general questions tend to lead to apparently punitive responses.
3.16.6 A Canadian study tested this by asking separate questions about violent and non-violent offenders. They found that 80% of people asked about violent offenders felt that courts were too lenient, compared to less than half for non-violent offenders (Roberts & Stalans 1997). Interestingly, it has been found that those people with strongly held views are less susceptible to this shortcut thinking than those with mixed or less strong views (see Stalans 2009; Roberts & Stalans 1997). This perhaps suggests that it is more difficult to influence strongly held attitudes with the supply of information.
3.16.7 On the other hand, it has been shown that where questions are more specific, contain more information, and are more interactive (such as in an interview or deliberative workshop rather than a structured survey), the views elicited tend to be less punitive (Hutton 2005; Keijser & Elffers 2009; Kury et al 2009; Indermaur 2009). This is because more information and more opportunity for consideration and discussion allows people to think through their responses rather than using mental shortcuts. For example, one German study used a survey to determine punitivity levels, and then interviewed a sample of those who had been found to be most punitive in the survey. It was found that levels of punitivity as measured by the interviews were half of those measured by the survey (Kury et al 2009: 67).
3.16.8 Several studies have found that the more information given about a case, whether it be about the offender, the available sentencing options, or the case circumstances, the less severe the sentence suggested (see Roberts & Hough 2005b; Roberts & Hough 2005c; Kury et al 2009; Roberts & Stalans 1997; Hough & Roberts 1999; Keijser & Elffers 2009). This is not an unanimous finding however, as Paulin et al found that in their seven case studies more information resulted in more severe sentences in some cases, and less severe sentences being suggested by participants in others (Paulin et al 2003). Clearly it would depend at least in part on the nature of the additional information which was provided, though the general point from this research is that specific, realistic case studies give people more of a chance to think the case through, without relying on stereotypes and assumptions to provide top of the head responses.
3.16.9 And so it seems that short, general questions cannot fully capture the complex attitudes people have about sentencing (see also Coulsfield 2004; Sprott 1999), which can depend very much on context and careful thought. We will now turn to the responses to these more complex and specific questions, and consider what public preferences are in terms of sentencing practice.
3.16.10 Without going into the specific details of sentencing preferences in particular cases, numerous case study research studies have found that the four key factors influencing sentencing preferences are:
- Crime seriousness - the most important consideration. It seems crucial to the public that sentences are proportionate to the seriousness of the offence. Within this, clear lines are drawn between serious, violent offences, where offenders can be seen as dangerous, and for which prison is seen as appropriate, with a focus on public protection, and less serious, non violent offences such as property offences, which are more often assigned community sentences, with a focus on rehabilitation (see Hutton 2005; Anderson et al 2002; Stalans 2009; Roberts & Hough 2005c; Roberts & Stalans 2004; Stalans 2002; Krajewski 2009).
- Attitude - Participants in these studies also made a distinction between those who did and did not express remorse, and attempt to make amends for their actions (Roberts et al 2008; Roberts & Stalans 2004; Kury et al 2009; Roberts & Hough 2005b)
- Previous Record - a clear distinction is also made between repeat and first time offenders in the severity of sentence which is deemed appropriate (Anderson et al 2002; Roberts et al 2009)
- Offender circumstances/characteristics - some, though not all, studies found that some offender characteristics or circumstances influenced sentencing severity or purpose, such as whether they were young offenders, or had a drug addiction (both tending to lead to less severe and more rehabilitative sentences) (see Anderson et al 2002; Roberts & Stalans 2004).
3.16.11 The key message here is that people's sentencing preferences are complex and dependant on many factors, with different types and severity of sentence and sentencing purposes coming to the fore in different circumstances (see Roberts et al 2009; Paulin et al 2003; Stalans 2009). And so people have been found to support sentencing based upon the specific details of individual cases rather than standard penalties for particular offences (see Roberts et al 2009; Maruna & King 2008; Hutton 2005), and to support a range of apparently contradictory sentencing purposes such as rehabilitation, restoration, and punishment both across and within cases (see Maruna & King 2008; Coulsfield 2004; Roberts & Stalans 2004).
3.16.12 Finally, it is important to note that the sentences people choose in these exercises are not necessarily the only ones they would deem acceptable. Roberts and Hough (2005) asked those who had suggested a prison sentence for a specific scenario if they would find an alternative (community) sentence (one year of supervision, 200 hours work, and victim compensation) acceptable. They found that 79% of those who had originally opted for prison said they would find this alternative acceptable (Roberts & Hough 2005b: 223). They refer to three similar studies where community sentences are suggested in the place of custody, and in line with the findings outlined above, they found that the substitute was more likely to be seen as acceptable when the offence was non-violent and the alternative community service, and less likely to be seen as acceptable when the office was violent and the alternative a fine (Ibid: 225).
3.16.13 It should be noted that these sentencing preferences may well be influenced by perceptions of sentence types, i.e. of prison, community sentences, and fines, which were outlined in the previous chapter.
3.16.14 We have seen, then, that in general the public are not universally punitive, and have different sentencing preferences across different kinds of crimes, offenders and circumstances. Of course, not every person in these studies suggested the same sentences, and some were found to be generally more punitive than others in their suggestions. We will now look at what causes some people to be more punitive than others.
3.16.15 Studies examining drivers of punitivity have consistently found some demographics to influence punitivity, though the influence is relatively small (Kesteren 2009; Maruna & King 2009). Men, older people, the less educated, and tabloid readers have been found to be more punitive (Kesteren 2009; Maruna & King 2009; Hough & Moxon 1985; Jansson et al 2007; Kuhn 2002). Victims are generally found to be no more punitive than non-victims (Kesteren 2009; Kury et al 2009), though in New Zealand the one exception of female victims of male assault has been identified (Paulin et al 2003). One international study has found a link between punitivity and lower confidence in the police, and living in a country with higher inequality (i.e. the USA or the UK) (Kesteren 2009).
3.16.16 Maruna and King have demonstrated a link between personal beliefs about offenders and punitivity, in that the most punitive people are those who believe that crime is a choice and that criminals cannot change their ways and the least punitive people believe that crime has 'social origins' and people can change, with everyone else having moderate punitivity (Maruna & King 2009a: 18-19).
3.16.17 The most commonly cited drivers of punitivity are, however, perceptions about crime and society. Those with high fear of crime, who think crime is rising, or have concerns about social change are found to be most punitive (Jansson et al 2007; Kury et al 2009; Sprott 1999; Hutton 2005; Windzio et al 2007; Hough & Moxon 1985; King & Maruna 2009). Punitive attitudes, for those who have them, seem to fit into a wider narrative of rising crime and societal breakdown, though as we have seen, in the context of specific cases, the public are by and large not universally punitive.
3.16.18 This section has so far examined the evidence in regards to what people's perceptions of sentences are, what drives them, and what makes some people more punitive than others. The most important question in regards to the topic at hand, however, is the relationship between attitudes to sentencing and attitudes to other aspects of the criminal justice system, such as attitudes to the system as a whole, or in regards to attitudes to courts and judges.
3.16.19 Unfortunately, for our purposes, the relationship between either sentencing preferences, or perceptions of sentencing practice, and attitudes towards courts, judges, and the justice system more broadly (such as confidence and perceptions of legitimacy), have rarely been explicitly or robustly explored, beyond the finding above that punitivity is linked to confidence in police.
3.16.20 In a relevant vein, Smith's (2007) survey in England and Wales asked people explicitly what they thought about when answering a question about their confidence in the justice system, and what they thought would improve their confidence. The most common answer to the former question was 'consistency in sentencing', and the second most common response to the latter was 'tougher sentencing' (after 'more police on the streets (Smith 2007). This suggests that the public themselves in England and Wales see sentencing practice as key to their own confidence.
3.16.21 However, two caveats should go with this. First, the exact wording of the confidence question was 'How confident are you that the criminal justice system is effective in bringing people who commit crime to justice?'. This is a very specific question, focusing directly on justice outcomes (i.e. sentencing), rather than other factors which may be relevant to confidence more generally, such as fairness of process (see above). Second, the responses to these questions may well have been brought about by similar mental shortcuts as the leniency question outline at the start of this section. As with the leniency question, respondents were asked for quick, top of the head responses to simple questions about complex issues. Indeed, the author of the report does caution that people's perceptions of what would increase their confidence may differ from what would in reality increase their confidence (Smith 2007).
3.16.22 Some correlation has been found in England and Wales between perceptions of leniency in sentencing and seeing judges as 'out of touch' or having low confidence in the courts (Hough & Roberts 1999; Hough & Roberts 2004). Leniency of sentencing has also been found to be the reason cited by research respondents for their low level of confidence in the courts (Hough & Roberts 2004). The size and direction of this relationship have yet to be explored.
3.16.23 Conversely, a Dutch study has found that dissatisfaction with sentencing does not lead to dissatisfaction with the courts and judges themselves, who are seen to be independent, impartial, and fair (Keijser & Elffers 2009). This echoes the procedural justice findings above, which found process to be more important for confidence than outcomes. Clearly, however, these Dutch findings cannot be assumed to be transferable to a Scottish context. This is therefore a major gap in the evidence and more research into the link between sentencing perceptions and preferences, and attitudes to the justice system, are required to determine whether a direct or indirect relationship exists, which direction it goes in, and the size of its influence.
Media influence
3.17 There is a widespread assumption that a major or even the main driver of people's attitudes to the justice system is the media, but in fact the relationship is very rarely examined analytically. The research that does exist, as ever, shows a more complex picture. This section will examine the evidence in an attempt to ascertain the nature and size of the relationship between the media and people's attitudes to the justice system.
3.17.1 First, let us consider what we mean by 'the media'. As with 'the public', the media is not a homogeneous group, and includes a wide range of formats, styles, and messages. Importantly, there is also variance in the reach of different media sources.
3.17.2 For example, the BCS has asked its respondents where they get the most information about their local police and police in the country as a whole. Table 5 shows the top twelve responses, and while the top two sources of information for both local police and police generally are media sources (local newspapers, TV and radio news programmes), the next four for local police are about personal and vicarious experience, with documentaries and newspapers appearing towards the bottom of the table. The picture is more mixed for police generally, where newspapers are more often cited as an information source. It should be noted, however, that the question is about where information comes from, and it does not explore the relative influence of information from the different sources (see below).
Table 5: Where people get the most information about the CJS
| Information Source | Local Police (%) | Police in General (%) |
|---|---|---|
| Local Newspaper | 55 | 38 |
| News programmes on TV/Radio | 35 | 59 |
| Word of mouth/information from others | 32 | 21 |
| Personal experience | 23 | 20 |
| Relatives' and friends' experiences | 21 | 19 |
| What see been going on in their neighbourhood | 17 | 10 |
| TV documentaries | 12 | 23 |
| Tabloid newspapers | 11 | 27 |
| Broadsheet newspapers | 9 | 22 |
| Radio programmes | 8 | 15 |
| Local authority/local government/local council | 5 | 4 |
| Local police (including police's own website) | 3 | 6 |
Adapted from BCS 2004/05 Allen et al 2007
3.17.3 Individual media sources are not, then, universally consumed by everyone, and used as a source of information about crime and the justice system. And as Table 5 shows, people turn to different sources when asked about different dimensions of the justice system. It is suggested by the research literature that the extent to which people rely on the media as a source of information depends on how much they can glean from more direct sources, i.e. their own experience and that of those around them.
3.17.4 For example, research finds that when asked about local justice factors (e.g. ASB, police) or about things which they have personal or vicarious experience of (crime victims and suspects and their families etc) people are less likely to use the media as a source of information than when people are asked about national factors, or things they have no direct or vicarious experience of (Anderson et al 2002, Scottish Government & Cosla 2009, Ormiston 2010, Stalans 2009, Carrabine 2008, Dowler 2002, Callanan & Rosenberger 2011, Boda et al 2011). Personal and vicarious experience, when people have it, is thus a more important driver of attitudes to the justice system than information from the media.
3.17.5 Of course, some aspects of the justice system cannot be experienced directly, or are rarely so, such as trends - especially in crime and sentencing. Some studies have examined the content of news media and found that it has devoted an increasing amount of coverage to crime stories (Stalans 2009) and focuses disproportionately on serious violent crime, and stranger on stranger crime (Hough & Roberts 2004, Duffy et al 2008, Windzio et al 2007, Weitzer 2002). Some suggest that this explains the high proportion of people who think crime is increasing, seemingly independent of actual crime statistics, though detailed studies have found mediating factors between fear of crime and media consumption (see Carrabine 2008). Tracking studies have, however, found that peaks in fear of crime can be measured following media coverage of violent crime, followed by reduction in fear following stories about crime crackdowns (Duffy et al 2008).
3.17.6 Studies focused more directly on the influence of the media on attitudes to the system itself have found that coverage of the police is generally positive, and this may in fact lead to less confidence when unrealistic expectations are created that cannot be met in real life encounters (see Carrabine 2008, Leishman & Mason 2003). Some US studies have tracked public confidence in police alongside reporting of police misconduct, and have found that such reporting does have an impact on people's attitudes to the police (Callanan & Rosenberger 2011), but not necessarily a lasting impact (Weitzer 2002), or on all attitudes to police, with diffuse support (Brown & Benedict 2002) or views on police effectiveness (Miller & Davis 2008) not affected.
3.17.7 Boda et al's survey in Italy, Bulgaria and Lithuania found a link between higher media consumption and higher trust in the police, and more mixed findings for trust in courts, with reading newspapers leading to higher trust, and watching TV to lower trust. They also found, however, a correlation between trust in the media itself and trust in police and the courts and concluded that there was no substantial evidence of a direct strong link between media consumption and levels of fear of crime and trust in police and courts (Boda et al 2011). Roberts & Stalans (1997) also found this, which suggests that it is a wider trust in institutions that is relevant here (see next section).
3.17.8 One review of the evidence has suggested that the issue here is that the media are not effective at telling people what to think, but rather have an agenda setting function. In other words, the media is effective at telling people what to think about, and perhaps even how to think about it (Green 2006).
3.17.9 So, we have seen that the relationship between media content and people's attitudes is not direct and straightforward. The field of media studies itself has increasingly focused not on the content of the media, but the different ways that people consume it. It is now widely acknowledged that people do not unquestioningly absorb all media messages. People choose both what media sources they consume, and how to interpret what the media says (see Carrabine 2008).
3.17.10 A growing body of research, then, demonstrates that people choose which media sources to consume based on their existing attitudes and preferences (Boda et al 2011, Duffy et al 2008, Miller & Davis 2008, Windzio et al 2007, Dowler 2002, Feilzer 2009), and interpret different sources and messages in different ways according to their trust in each source, and the extent to which information and messages fit with their experience (as we have seen above) and existing attitudes (Duffy et al 2008, Hough & Roberts 2004, Carrabine 2008, Callanan & Rosenberger 2011, Leishman & Mason 2003, Roberts & Stalans 1997).
3.17.11 For example, a 2003 MORI poll in England and Wales asked a similar question to that outlined in Table 5, but added a secondary question about trust in sources. Table 6 shows the responses, and it can be seen that levels of trust in sources of information differ, and just because a person uses a particular source, does not mean they believe its contents. This is especially striking for tabloids, where 60% of respondents cited them as a source, but only 22% said they trusted them to tell the truth.
Table 6: Sources of information and trust
| Source of the most information (%) | Trust that the source tells the truth about how crime is being dealt with (%) | |
|---|---|---|
| TV news/documentaries | 95 | 87 |
| Local newspaper | 85 | 77 |
| Broadsheet newspaper | 68 | 60 |
| Relative/friend's experience | 67 | 89 |
| Tabloid newspapers | 60 | 22 |
| Internet | 24 | 30 |
| Police | 23 | 68 |
| What you learned in school/class | 19 | 57 |
| Newsletter from the police sent to your home | 18 | 49 |
| Telephone helpline | 5 | 36 |
(Adapted from Duffy et al 2008)
3.17.12 Similarly, Roberts and Stalans describe an experiment where respondents were shown one of two news reports on a violent confrontation between students and police, one which blamed police and one which blamed students. It was found that:
'Respondents who viewed a news report that conflicted with their prior attitudes (e.g., pro-police respondents watched report attributing blame to the police) perceived the newscast as less objective, and perceived the reporter as less credible and as being partisan to the political position consistent with the newscast. Respondents who were exposed to information inconsistent with their attitudes discounted the information by concluding that the media were biased and hostile toward the view of the other side.'
Roberts & Stalans 1997: 94
3.17.13 People therefore not only make judgements both about the trustworthiness and reliability of particular media (and other) sources, they also make judgements about the authenticity of particular stories and pieces of information (see also Leishman & Mason 2003).
3.17.14 The fact that people can, and do, choose to disagree with media messages, lays bare one last important point - that what the media says does not necessarily reflect what they public think. Just as the media have been shown to not directly reflect the reality of crime trends, it also cannot be said to reflect the reality of the views and preferences of the public themselves (see Indermaur & Hough 2002). This is an important point for those in government and the justice system, who may be tempted to take media coverage as a direct representation of public opinion.
3.17.15 These findings demonstrate that the media can be an important source of information for people, and can influence, but does not directly drive people's attitudes to the justice system, both because the media itself is diverse and communicates a wide range of information and messages, and because people choose what media to consume and how they react to it.
Wider social and political context
3.18 Having examined the potential drivers of attitudes to the justice system which are directly related to the justice system, it remains only to look at the evidence regarding the wider social and political context - factors which lie outside the realm of crime and the justice system, but nonetheless are related to public attitudes.
3.18.1 First, it is widely established that public trust in the justice system or its constituent parts is correlated to trust in other institutions, in government and in other people (Slomczynski & Janicka 2009, Ivkovic 2008, Jokinen et al 2009, Van De Walle 2009, Jang et al 2010, Sun & Wu 2006, Duffy et al 2008, Tyler & Huo 2002, Jokinen et al 2009). The nature and direction of this relationship is unclear, with some ascribing declining trust across the board to a general loss of deference to authority (van de Walle 2009), some suggesting that trust in all institutions has a common source (Slomczynski & Janicka 2009, Ivkovic 2008), and some suggesting that trust in government influences trust in justice institutions (Jang et al 2010, Jokinen et al 2009). Whatever the nature of this relationship, it is clear that trust in the justice system cannot be isolated from wider trust in government and authority, and so may be influenced by factors which have little or nothing to do with the operation of the justice system itself.
3.18.2 In a similar vein, some international surveys have found links between the quality of governance, level of government corruption, or equality in a country, and attitudes to police or courts (Ivkovic 2008, Sun & Wu 2006, Kesteren 2009).
3.18.3 Second, several studies have found a relationship between attitudes to the justice system, and feelings about social change and cohesion. These studies suggest that attitudes to the justice system, particularly police, are shaped in part by concerns about change in society, and the levels of social cohesion in the local area, that is, how far there is a feeling of community and togetherness (Tyler & Huo 2002, Hutton 2005, Jackson & Bradford 2009, Jackson & Sunshine 2007). According to this theory, attitudes to the police are shaped in part by factors not within the control of the police, who are seen as guardians of the community (although as we have seen above, police can have a powerful role in communicating through their behaviour who belongs as part of the 'community' and who does not).
3.18.4 It is important to recognise that public attitudes to the justice system, as a whole, or its constituent parts, will always in part be influenced by the particular history and culture of an individual society or geographical area (such as a country). International surveys, and studies in countries quite different from Scotland show that people's approach to crime and justice issues, and their attitudes to the justice system, often vary substantially across cultures (and indeed across time) (see Fortete & Cesano 2009, Ghassemi 2009). Indeed, we have seen that some of the research conducted in the USA is not comparable to Scotland due to the different ethnic make up of the two countries, and the different relationship between ethnic groups.
3.18.5 Of course, some of this variance will be due to the different experiences of the justice system in different places, but some will also be down to the different ways that issues and experiences are understood, processed and contextualised across cultures.
3.18.6 Finally, and related to the previous point, within individual societies there are groups with different levels of power, and dominance in the culture (these groups may, or may not have a demographic basis). So, for example, Buhlmann and Kunz compare confidence in the justice system amongst political and economic 'winners and losers' (the former defined by comparing political preferences with current governments, and the latter by stratifying income) and found that 'winners' were more likely to be confident in the justice system than 'losers' (Buhlmann & Kunz 2011: 332). We will return the role of social group membership when we discuss the mediating impact of ethnicity on the relationship between legitimacy and compliance below.
Summary of significant/relevant drivers
3.19 We have now reviewed all the possible drivers of attitudes to the justice system directly examined in the research literature. Table 7 summarises the findings in the approximate order of the size of impact on attitudes to the justice system.
Table 7: Summary of relationship between drivers and attitudes to the justice system
| Driver | Strength of Evidence | Nature of relationship | Description of relationship | Part of justice system |
|---|---|---|---|---|
| Contact and vicarious contact with police and/or courts (procedural justice) | Moderate (no Scottish studies and few courts studies) | Direct and large | Of all the drivers, contact with the justice system has the clearest and biggest impact on attitudes. Procedural justice evidence shows that perceived fairness, neutrality, respect and honest motives lead to more confidence in the system | All (much more evidence on police, but some on courts). |
| Perceptions of neighbourhood and ASB | Moderate (direction of relationship contested) | Direct and large (though direction contested) | Evidence shows that perceptions of local ASB, and of local neighbourhood cohesion are related to attitudes to police, though attitudes to police may in fact reassure people about neighbourhood conditions, not the other way around. | Police |
| Visibility/ accessibility of police | Strong | Direct and large | Evidence shows that visibility and accessibility of police in the local area increases confidence in the local police | Police |
| Communication from police (other parts of system not studied) | Strong | Direct and medium | Communication from the police that meets people's desires from such communication can increase positive attitudes, as it communicates police engagement with the community (though smaller impact than direct contact) | Police |
| Wider social and political context | Moderate | Direct, possibly large | The correlation between trust and confidence in justice agencies, and trust and confidence in government and other public institutions suggests there is a common source of attitudes, which is beyond the realm of justice. The different historical relationships between populations and their public institutions, both as a whole, and for specific groups within societies, is also relevant. | All |
| Media use | Moderate | Direct and small | The media is a source of information, especially important where people do not have personal or vicarious experience to draw on, but people choose what media sources to consume, and contest messages that do not match their existing attitudes and experience | All (more crime and sentencing, where there is less personal experience) |
| Experience of crime (victimhood) | Moderate | Direct and small | Being a victim has a small direct effect on attitudes to the justice system, but having contact with the system as a result of victimhood has a bigger effect (see above). | Mostly police |
| Demographics (age, gender, ethnicity, class etc) | Moderate (mixed findings, little complex analysis in Scotland) | Indirect and medium (mediating factors) | Attitudes to the justice system (mostly police) do vary by demographics, but to a small extent, and this may be due in large part to the different experiences different groups have with the system. | All (most evidence on police) |
| Knowledge about the justice system and crime | Moderate (more complex analysis required) | Unclear, possibly indirect | While there are correlations between knowledge and attitudes, the evidence currently does not show a clear or large relationship. Possible mediating factors are the sources of information, trust in those sources, and levels of interest in information | All |
| Sentencing preferences and attitudes | Weak | Unclear - relationship exists but size and direction unclear | Evidence shows people have nuanced sentencing preferences based on severity and circumstances of crime. We also know people underestimate sentencing practice, and this correlates with negative views about judges. The size and strength of this relationship has to date not been tested | Courts and Judges |
3.19.1 We can see from the table that, according to current evidence, the most important drivers of people's confidence, trust, satisfaction and perceptions of the legitimacy of the justice system, especially the police, seem to be personal experience, whether that be of direct contact with the system, seeing or hearing from the police, experience and perceptions of the local neighbourhood, or even stories about the experiences of people you know or identify with. Within these experiences it is crucial that people feel that the justice system - as represented by people such as the police - values and cares about individuals and the community, behaves fairly, respectfully, neutrally and takes seriously the things that matter to people.
3.19.2 For parts of the justice system that are not as visible to the public as the police, such as sentencing, fewer people have direct or vicarious personal experience to inform their attitudes. In these circumstances inferences are made from available information, which is more likely to include media sources. The effect of such media use depends on which media sources are used, and the degree to which information aligns with existing attitudes and experiences.
3.19.3 The second and last columns in Table 7 show that evidence is stronger in some areas than others. There are notable gaps in the evidence on attitudes to particular parts of the system - while people's attitudes to the police have been examined in-depth, attitudes to the courts, other parts of the system such as prisons, and to the civil system have to date not been examined as thoroughly. Further research in this area would give a much clearer picture as to whether the same factors (e.g. perceived procedural justice, communication like newsletters) influence attitudes to these other parts of the system.
3.19.4 Similarly, which many studies examine people's attitudes to sentencing, to date none have fully explored the size and strength of the relationship between attitudes to sentencing and attitudes to the justice system more widely.
3.19.5 Finally, in many areas there is a distinct lack of Scottish evidence, for example police communication, and views of minority groups, and while studies from England and Wales may be comparable, this is not necessarily the case, and more research would be required to test findings from other jurisdictions in a Scottish context.
Contact
Email: Carole Wilson
There is a problem
Thanks for your feedback