Public judgement on sentencing: Final results from the Tasmanian Jury Sentencing Study

Foreword | This seminal study, which was funded by the Criminology Research Council, is the first reported study to use jurors in real trials to gauge public opinion about sentences and sentencing. Using jurors is a way of investigating the views of members of the public who are as fully informed of the facts of the case and the background of the offender as the judge. Based upon jurors’ responses from 138 trials, the study found that more than half of the jurors surveyed suggested a more lenient sentence than the trial judge imposed. Moreover, when informed of the sentence, 90 percent of jurors said that the judge’s sentence was (very or fairly) appropriate. In contrast, responses to abstract questions about sentencing levels mirrored the results of representative surveys. The results of the study also suggest that providing information to jurors about crime and sentencing may be helpful in addressing misconceptions in these areas.

Replication of this study may be of assistance to policymakers and judges who wish to know what informed members of the public think about sentencing. Portrayals of a punitive public are misleading and calls for harsher punishment largely uninformed.

Adam Tomison
Director

Public opinion surveys conducted worldwide over the last four decades have consistently found that between 70 and 80 percent of respondents believe that sentences are too lenient (Gelb 2006). Responding to research suggesting that judges are out of touch with public opinion, Chief Justice Gleeson of the High Court of Australia suggested that, instead of surveying uninformed members of the public, it might be more useful if jurors—as more informed representatives of the public—were asked about the sentences in the particular cases they have deliberated on (Gleeson 2005). The Tasmanian Jury Sentencing Study, which surveyed 698 jurors from 138 trials between September 2007 and October 2009, was inspired by this suggestion.

The aims

The central aims of the study were first, to develop an innovative method of gauging informed public opinion on sentencing by using jurors in criminal trials and second, to explore the use of jurors as a means of better informing the public about crime and sentencing. An earlier paper in this series, based on quantitative data from the first 51 trials, addressed the preliminary research question, namely the willingness and feasibility of using jurors as a source of public opinion (Warner et al. 2009). Some preliminary results describing jurors’ choice of sentence and their reactions to the sentences imposed were also outlined. This paper confirms the early results, which suggested that surveying jurors is a promising means of gauging public opinion. It shows that the jury survey method provides valuable insights into the relationship between information, reflection and first-hand experience, and the formation of public judgement on judicial sentencing that researchers, policymakers and judges can rely on.

The method

The three stage mixed method approach supplemented two surveys with face-to-face interviews. Jurors were recruited over a two year period from all criminal trials in Tasmania in the three cities in which the Supreme Court sits—Hobart, Launceston and Burnie. In the first stage of the study, each jury returning a guilty verdict was invited by the judge to participate in the study by remaining in court to listen to the sentencing submissions. Before the sentence was imposed, jurors completed Questionnaire 1 which asked them:

• to indicate the sentence that they thought the offender should receive;
• to answer questions about crime and sentencing trends; and
• to give their views on sentencing severity and whether judges were in touch with public opinion.

Those willing to participate further were sent a package containing the judge’s sentencing comments, an information booklet about crime and sentencing, and a second survey form. Questionnaire 2 repeated the questions in the first survey about judges, sentencing practices and crime trends and it asked extra questions about the sentence that the judge had imposed, the contents of the sentencing remarks and the usefulness of the information package.

In Stage 3, 50 jurors were interviewed. Initially, it was planned to select jurors on the basis of whether their opinions had remained unchanged or had become more lenient or more severe. However, the results revealed that respondents could not be unambiguously classified in this way and so jurors were selected from a representative spread of offence types and juror demographics. The semi-structured interviews discussed the juror’s reaction to the judge’s sentence and sentencing remarks and provided the opportunity for jurors to reflect, discuss and consider more deeply the views they had expressed in the two surveys and the reasons behind any changes in their opinions (Davis, Warner & Bradfield forthcoming).

The results

Response rate

Responses were received from 698 jurors from 138 trials out of a possible 162 trials that returned a guilty verdict in the two year period. This translated to a 36 percent response rate. The rate varied considerably between cities, with a response rate of 59 percent in Hobart and 14 percent in Burnie. Possible explanations for this include differences in court facilities impacting on the jury experience, proximity to the research team and juror reluctance to be associated with sentencing outcomes in smaller communities. There were some variations in response rates between judges, but little difference between offences types, length of trial or deliberation. As discussed earlier, the respondents were reasonably representative of the general population, refuting the claim that juries are not representative of the community (French 2007). While Australian born residents and the 45–64 year age group were slightly over-represented, jurors were less likely to be unemployed than the general population and more likely to be better educated and have a higher income.

Most Stage 1 respondents (88%) agreed to participate in Stage 2 and 64 percent (n=445) returned their forms. Almost half of the Stage 2 respondents were willing to be interviewed, providing a pool of 212 jurors from which 50 were selected for interview.

Figure 1: Judge and juror’s sentence compared by type of offence (%)

n=706. This figure includes all jurors’ responses for all offenders

Comparing juror’s sentencing choice with the judge’s sentence

Jurors selected a sentence from a menu of options that was designed to alert them to the range of possible alternatives and to avoid too great a focus on sentences of imprisonment (Hough & Roberts 1999). At Stage 1, the juror’s sentence was compared with the judge’s sentence using a constructed variable (Stage 1 Comparative Sentence Variable) that recorded whether the juror’s sentence was more severe, less severe, or the same as the judge’s sentence. This showed that 52 percent of jurors selected a more lenient sentence than the judge. Figure 1 cross-tabulates the responses by type of crime. For sex, violence and drug offences, the responses were quite evenly split between more severe and less severe sentences. Jurors were most likely to be more lenient than the judge for property offences and culpable driving cases, but for the latter the numbers were small (11 respondents only).

Jurors’ knowledge of crime trends and sentencing patterns

Four questions tested jurors’ knowledge of crime and sentencing. They were asked about recorded crime levels over the last five years. Recorded crime rates have been declining for about five to 10 years, nationally and in Tasmania. However, only seven percent of respondents thought that crime had decreased and 27 percent thought it had increased a lot. In response to the question about the proportion of crime that involves violence, only 17 percent of jurors correctly responded that a quarter or less involved violence and 41 percent thought that more than half was violent. Responses to the question about knowledge of the proportion of convicted offenders who were sent to prison for burglary and rape showed that 71 percent underestimated the imprisonment rate for rape and 80 percent did so for burglary. These findings of misconceptions about crime and sentencing trends are consistent with Australian research and international findings (Gelb 2006; Jones, Weatherburn & McFarlane 2008; Roberts & Indermaur 2009).

Jurors’ general opinion of current sentencing practices

To avoid the limitations of answering a single question about sentencing levels (Roberts & Stalans 1997), the study asked jurors to distinguish between four kinds of crimes—violence, property, drug and sex offences. The question asked whether jurors thought that current sentences were much too tough, a little too tough, about right, a little too lenient or much too lenient. Across all offence types, the majority responded that sentences were too lenient. This was most pronounced for sex and violence offences, with 80 percent and 76 percent of jurors saying that sentences were too lenient.

A punitiveness index was created using juror’s responses across the four offence categories. A comparison of the mean scores on this index (t-test) showed that more punitive respondents were more likely to:

• think crime had increased (p=.000);
• overestimate the proportion of crime involving violence (p=.000); and
• underestimate the proportion of convicted rape offenders who were imprisoned (p=.018).

This is consistent with previous research showing that public misperceptions about crime and sentences are associated with a belief that sentences are too lenient (Hough & Roberts 1999; Roberts & Indermaur 2009).

Jurors’ views of judges’ sentences

There was a high overall level of satisfaction with judicial sentencing among jurors. In the Stage 2 survey, jurors were asked to rate the appropriateness of the judge’s sentence on a four point Likert scale and 90 percent said that the sentence was appropriate, evenly split between very appropriate and fairly appropriate. There was some variation in satisfaction levels across different crime types. As Figure 2 shows, jurors were least satisfied with sentences for sex and drugs offences, for example they were less likely to say that sex and drugs offences were very appropriate and more likely to say that they were inappropriate.

As a follow-on from the question about the appropriateness of the sentence, jurors were asked (unless they thought the sentence was very appropriate) to indicate what the sentence should have been. A variable was constructed from the responses comparing the severity of the judge’s sentence with the juror’s view (Stage 2 Comparative Sentence Variable). The distribution of categories on this variable showed that a little more than a third (37.5%) thought that the judge should have imposed a more severe sentence. This was least likely for property offences (28%) and most likely for sex offences (46%) and drug offences (46%). This showed that the reason why jurors were less likely to say that sentences for these offence categories were appropriate was because they thought that the judges’ sentences were too lenient. The Stage 2 Comparative Sentence Variable was used to compare the acceptance of the judge’s sentence by those who had selected a more severe sentence at Stage 1 with those who had selected a more lenient sentence. This supported the results of the analysis of the responses to the appropriateness of the sentence, namely that those whose sentence choice at Stage 1 was more lenient were more likely to endorse the judge’s sentence than those who had selected a more severe sentence. Specifically, 57 percent of those who had selected a more severe sentence at Stage 1 still wanted a more severe sentence after they had received the sentencing remarks compared with just 18 percent who still wanted a more lenient sentence.

Figure 2: How appropriate was the sentence for each crime type? (%)

Changes in general attitudes to sentence at Stage 2

In Stage 2, after being informed of the judge’s sentence and receiving the information booklet, jurors were again asked whether sentences for the four offence categories were too tough, about right or too lenient. Table 1 shows that the much too lenient and a little too lenient responses decreased across all crime categories and the about right responses increased. However, after combining the responses to create a three point scale, the most common response across all offence types remained too lenient, except for property offences where about right was the most common response.

These general views about the leniency of sentences can be contrasted with jurors’ views about the judge’s sentence in their specific case. At Stage 2:

• 66 percent of respondents thought that sentences for violent offences were too lenient, even though only 35 percent wanted a more severe sentence in the particular case they deliberated on;
• for sex offences, 70 percent thought that sentences were generally too lenient even though only 46 percent wanted a more severe sentence at Stage 2;
• for property offences, 46 percent said that sentences were too lenient but only 28 percent wanted a more severe sentence in the case they deliberated on; and
• for drug offences, 49 percent said that sentences were too lenient but 46 percent wanted a more severe sentence in the case they deliberated on.

The data were analysed to determine if jurors’ general attitudes to sentence differed depending on the crime type of their particular trial. It could be hypothesised that knowledge of the sentence imposed and the sentencing information supplied about the offence may have a greater impact on attitudes to that particular type of crime. This association held for violent and property crimes but not for sex and drug offences. Jurors from violence and property cases were more likely than other jurors to say that sentences in general for violence and property offences were about right, but:

• 62 percent of jurors from violence cases still said that sentences for violent offences were too lenient even though only 49 percent had suggested a more severe sentence than the judge; and
• 36 percent of jurors from property cases still said that sentences for property offences were too lenient and 30 percent had suggested a more severe sentence than the judge.

This ‘perception gap’ or lack of consistency between jurors’ views about the particular offence they deliberated on and their general attitudes was therefore most pronounced in the categories of sex and violence cases. Clearly, there were some jurors from sex and violence cases who persisted in their general view that sentences for those general categories were too lenient, even though they were satisfied with the judge’s sentence in the particular case they deliberated on.

Use of the booklet and changes in jurors’ knowledge of crime and sentencing trends

After jurors had received the crime and sentencing booklet, their knowledge of crime and sentencing trends improved. However:

• 38 percent still said that recorded crime rates had increased;
• 37 percent still overestimated the proportion of crime that involved violence; and
• 49 percent still underestimated the imprisonment rate for convicted rapists.

Not all Stage 2 respondents read the crime and sentencing information booklet in full (62%) but those who did were significantly more likely to have given accurate answers to the questions about crime trends, the proportion of crime that involves violence and the burglary imprisonment rate (p=.000).

As well as being asked about how closely they read the booklet, jurors were asked a series of questions to measure its usefulness. Most (74%) planned to keep it and most rated it as informative and helpful. Many also thought that other jurors would be very interested in receiving such a booklet. One-third of jurors reported that they had discussed the information on crime trends with family or friends, 28 percent discussed the information on sentencing trends and 68 percent discussed the sentence.

Are judges in touch?

In both surveys, jurors were asked whether judges were in touch with public opinion about sentencing. In Stage 1, a total of 70 percent of jurors said that judges were either very in touch (13%) or somewhat in touch (57%). In Stage 2, the proportion of those who thought that judges were in touch with public opinion increased to 83 percent, with the very in touch responses doubling to 26 percent. This pattern was the same when the data were run using Stage 2 respondents only. It could be concluded, therefore, that the change in jurors’ perception of judges is associated with the knowledge of the sentence and the information received.

Further discussion of the results can be found in the full report (Warner et al. forthcoming).

a: Due to rounding, percentages may not total 100

Key implications

The myth of the punitive public

The fact that 52 percent of jurors chose a more lenient sentence than the judge and only 44 percent were more severe than the judge shows that informed members of the public are not as punitive as many representative surveys have suggested. This finding mirrors previous vignette studies that have also reported that when views of members of the public on a specific case are compared with those of judges, the judges’ sentences tend to be as severe or more severe than those of the public (Diamond & Stalans 1989; Lovegrove 2007). Moreover, when informed of the sentence at Stage 2, 90 percent of jurors thought that the sentence was very or fairly appropriate and only around a third thought that the judge should have imposed a more severe sentence.

Leniency, punitivity and malleability

One interesting finding was that those who had selected a more lenient sentence than the judge at Stage 1 were significantly more likely to agree with the judge’s sentence at Stage 2 and more likely to say it was very appropriate than those who had selected a more severe sentence than the judge at Stage 1. In other words, jurors who were more punitive were less tolerant of the judge’s sentence and less malleable in their views than the more lenient jurors, as measured by their Stage 1 sentence choice. This accords with Lovegrove’s (2007) findings.

Public opinion is multidimensional

Public opinion is not one dimensional; rather, it is multidimensional and contingent on particular circumstances. The jury survey methodology, which covers all trials over a lengthy period and therefore picks up a realistic assortment of sex, violence, drug and property cases, is better able to reveal broad differences in attitudes to particular offence types than the standard vignette methodology. The results showed a striking disparity in attitudes to different types of offences. For property offences, jurors were more than twice as likely to be less severe than the judge than more severe. For sex, violence and drug offences, the split between less and more severe was much more even. This difference in offence types was borne out in Stage 2. When asked how appropriate the judge’s sentence was, jurors were most satisfied with property offence sentences (57% very appropriate) and least satisfied with drug and sex offence sentences (around 35% very appropriate). Comparing the judge’s sentence with the juror’s preferred sentence at Stage 2 showed that jurors were least likely to have preferred a more severe sentence for property offences (28%) and most likely to have preferred a more severe sentence for sex and drug offences (46%).

The perception gap

The study showed that there was a distinct contrast between the jurors’ responses to the stimulus of a particular trial and their responses to an abstract question about sentencing levels. While the view that sentencing levels are too lenient moderated somewhat after jurors had received more information in Stage 2, a clear dichotomy remained between their responses to the sentence imposed on the offender in the trial they deliberated on and their responses to the question about general sentencing levels in sex, violence and property offences, but not in drug cases. This dichotomy persisted when the general views of respondents were separated so that general attitudes for offence types were limited to jurors who had deliberated in a case of that offence type. The analysis showed that the perception gap remained in the case of sex offences, diminished but remained in violent offence cases and all but disappeared in property cases.

This jury study is not the first to observe the differences between responses to abstract questions and the stimulus provided by an individual case (Diamond & Stalans 1989; Hutton 2005). However, it is the first to look at the impact of increased information on this dichotomy and the first to find that extra information and increased exposure to a real trial and a real sentence on an individual offender has a differential impact depending on offence type.

Attitudes towards judges

Just as 90 percent of jurors thought that the sentence imposed by the judge was appropriate, a substantial majority of 83 percent also thought that judges were in touch with public opinion. In contrast with representative surveys that have found that only 18–20 percent of respondents thought that judges were in touch with the public (Hough & Roberts 1998; Mirlees-Black 2001), jurors in this study who all had first-hand contact with judges were much less likely to say that judges were out of touch.

Impact of information

While jurors were shown to be as poorly informed about crime and sentencing trends as other members of the public, the results suggest that modest improvements in knowledge levels can be gained by providing better information directly to those who come into contact with the criminal justice system. Participants thought that other jurors would be interested in receiving such information and the results suggest that providing jurors in all trials with a crime and sentencing booklet and the reasons for the judge’s sentence has the potential to change attitudes. Moreover, because a majority of jurors discuss the sentencing outcome of their case with others, jurors also have the potential to act as conduits of information to the rest of the community. However, given that jury service touches only a minority in the community and that the provision of more information does not always lead to attitude changes, it seems that it is not a complete solution to the problem of misperceptions in the wider community. For some people, the belief that sentences are too lenient is difficult to shift.

Conclusion

The final results of the Tasmanian Jury Sentencing Study confirm the preliminary findings reported in an earlier paper (Warner et al. 2009), which suggested that representative surveys cannot be taken at face value. The results show that a substantial majority of jurors with firsthand experience of judges consider that sentences are appropriate and that judges are in touch with public opinion. By surveying members of the public who have engaged directly with the criminal justice system in a much more meaningful way than those who form their perceptions secondhand via the mass media, the study has shown that the jury survey methodology provides a better approach to finding a reliable source of informed public judgment of judicial sentencing.

The study has also shown that there is value in engaging jury members by giving them more information about sentencing patterns and crime trends and by informing them of the judges’ reasons for the sentences that they have imposed. The method has the potential to further explore differences in informed public opinion about the seriousness of different offence types and to investigate the contrast between punitiveness as measured by sentence choice in an individual case with the responses to an abstract question on general sentencing levels.

References

• Davis J, Warner K & Bradfield R forthcoming. Interviewing the jury: Three case studies from the Tasmanian Jury Sentencing Study, in Bartels L & Richards K (eds), Qualitative criminology: Stories from the field. Sydney: Federation Press
• Diamond S & Stalans L 1989. The myth of judicial leniency in sentencing. Behavioural Sciences and the Law 7(1): 73–89
• French V 2007. Juries—a central pillar or an obstacle to a fair and timely criminal justice system? Reform 90: 40–42
• Gelb K 2006. Myths and misconceptions: Public opinion versus public judgment about sentencing. Melbourne: Sentencing Advisory Council
• Gleeson AM 2005. Out of touch or out of reach? The Judicial Review 7: 241–253
• Hough M & Roberts JV 1999. Sentencing trends in Britain: Public knowledge and public opinion. Punishment and Society 1(1): 11–26
• Hutton N 2005. Beyond populist punitiveness? Punishment and Society 7(3): 243
• Jones C, Weatherburn D & McFarlane K 2008. Public confidence in the New South Wales criminal justice system. Crime and Justice Bulletin: Contemporary Issues in Crime and Justice no. 118. Sydney: NSW Bureau of Crime Statistics and Research
• Lovegrove A 2007. Public opinion, sentencing and lenience: An empirical study involving judges consulting the community. Criminal Law Review 2007: 769–781
• Mirrlees-Black 2001. Confidence in the criminal justice system: Findings from the 2000 British Crime Survey. Research findings no. 137. London: Home Office
• Roberts LD & Indermaur D 2009. What Australians think about crime and justice: Results from the 2007 Survey of Social Attitudes. Research and public policy series no. 101. Canberra: Australian Institute of Criminology. http://www.aic.gov.au/publications/current series/rpp/100-120/rpp101.aspx  
• Roberts JV & Stalans L 1997. Public opinion, crime and criminal justice. Boulder, CO: Westview Press
• Warner K, Davis J, Walter M, Bradfield R & Vermey R forthcoming. Jury sentencing survey: Final report. Criminology Research Council
• Warner K, Davis J, Walter M, Bradfield R & Vermey R 2009. Gauging public opinion on sentencing: can asking jurors help? Trends & Issues in Crime and Criminal Justice no. 371. Canberra: Australian Institute of Criminology. http://www.aic.gov.au/publications/current series/tandi/361-380/tandi371.aspx  

About the authors

Kate Warner is a Professor at the Law Faculty, UTas.

Dr Julia Davis is an Associate Professor at the School of Law, UniSA.

Dr Maggie Walter is a Senior Lecturer at the School of Sociology, UTas.

Dr Rebecca Bradfield is an Honorary Research Associate and Ms Rachel Vermey is a research assistant at the Faculty of Law, UTas.