1.1 Background to this study
The Post-Corroboration Safeguards Review was established by the Scottish Government to consider what additional safeguards and changes to law and practice might be required if the general requirement for corroboration in criminal cases was removed. Its final report recommended that "[t]he time is right to undertake research into jury reasoning and decision-making. Simultaneous changes to several unique aspects of the Scottish jury system should only be made on a fully informed basis."
There is a substantial body of jury research worldwide. However, the vast majority is concerned with the traditional common law jury found in most major English language jurisdictions, which typically: has 12 members, a choice of two verdicts (guilty or not guilty), and is required to reach a unanimous verdict on which the whole jury is agreed.
The Scottish jury, by contrast: has 15 members; chooses between three possible verdicts (guilty, not guilty and not proven); and is required to reach a 'simple majority' (eight out of 15 jurors) to return a verdict. Unlike the traditional common law jury, a Scottish jury cannot fail to reach a verdict ('hang'), because 15 jurors cannot split in a way that allows this.
There is very little existing evidence on what (if any) difference the distinctive features of the Scottish system make to the jury's operation in practice, in comparison with the traditional common law jury. In 2017, the Scottish Government commissioned Ipsos MORI Scotland, in collaboration with Professors James Chalmers and Fiona Leverick (University of Glasgow) and Professor Vanessa Munro (University of Warwick), to undertake research on jury decision making. The main element of the research was a mock jury study - the largest of its kind in the UK to date - involving 64 mock juries and 969 individual participants.
1.2 Research questions
The key research questions for this study were:
- What effects do the unique features of the Scottish jury system have on jury reasoning and jury decision-making?
- What are jurors' understandings of the not proven verdict and why might they choose this verdict over another verdict?
Specifically, the research explored whether there were any differences in jury reasoning and decision-making depending on:
- The number of verdicts: three verdicts (guilty, not guilty and not proven), or two verdicts (guilty and not guilty).
- Jury size: 15 members or 12 members.
- The majority required before jurors can return a verdict: a simple majority or unanimity among the whole jury.
In essence, the research was designed to test whether or not any of these features appears to incline juries in one direction or another in terms of their verdict and the process by which they arrive at this when other aspects (such as the trial content) are held constant.
Two further research questions focused on (i) effective methods for communicating with juries and (ii) the use of pre-recorded evidence. These questions were addressed by evidence reviews, published in 2018.
1.3 What do we know from previous research?
In this section, we summarise the findings from the small body of existing jury research which is directly relevant to the distinctive features of the Scottish system (jury size, the majority required to return a verdict, and the not proven verdict). In summarising findings, we also discuss some of the limitations associated with specific mock jury studies. Appendix B includes a more detailed discussion of the methods used in mock jury research in general, and the issues that should be considered in weighing findings from such studies.
1.3.1 Jury size
There is no existing research that specifically assesses the impact of having 15 jurors, as is the case in Scotland. However, researchers in the United States have examined the possible impact of reducing the traditional common law 12-person jury to a smaller size (usually six). This research indicates that 12-person juries are preferable to smaller juries, on the basis that a smaller jury is less likely to be properly representative of the community, more likely not to contain members of minority groups, more likely to deliberate for a shorter time (at the expense of better deliberation), and possibly less likely to recall evidence accurately. These findings do not, however, imply that increasing the size of the jury beyond 12 would necessarily have positive effects. Indeed, there is some evidence suggesting that the efficacy of group decision-making can be impeded as groups increase in size.
1.3.2 The majority required for a verdict: simple majority and unanimity
There is no prior research that has directly considered the impact of the Scottish requirement that jurors reach a 'simple majority' before returning a verdict. However, researchers have examined the effect of requiring different majorities for a verdict. For example, Hastie, Penrod and Pennington's extensive Inside the Jury research compared three different decision rules: one requiring all 12 jurors to agree unanimously, one requiring 10 votes out of 12 for a verdict, and one requiring eight votes out of 12. Other studies have generally compared a requirement of absolute unanimity with an alternative somewhat short of this (such as a rule requiring two-thirds or more of jurors to agree on a verdict).
Such research suggests, unsurprisingly, that juries are more likely to hang (fail to reach a verdict) when all jurors are required to agree unanimously compared with requiring a majority short of unanimity. There is also some evidence to suggest that requiring jurors to reach a unanimous verdict may be associated with better quality deliberation - "deliberations are likely to be evidence driven and more thorough than when a majority rule is in place".
1.3.3 The not proven verdict
There has been limited research to date on the effect of the not proven verdict, and that which exists has a number of limitations. First, previous studies of not proven have used trial summaries (either written or audio), which is less realistic than showing jurors audio-visual recreations of trials. Second, all but one study to date involved jurors returning individual verdicts, rather than deliberating as a group as they would in a real trial.
Bearing these limitations in mind, one of the four published studies found that having the not proven verdict available was associated with jurors being significantly less likely to convict. Differences in the other studies were not statistically significant (though they tended to point in the same direction).
There is also very limited evidence on jurors' understanding of the not proven verdict. Where this has been examined, the focus has often been on beliefs about the possibility of a retrial after a not proven verdict. According to the Jury Manual, it is "thought to be good practice" for judges in criminal trials to "inform the jury specifically that 'not proven' is a verdict of acquittal and that the accused cannot be tried again for the same offence". However, a 1993 opinion poll commissioned by the BBC found that 48% of the Scottish public wrongly believed that a not proven verdict meant the accused could be retried if new evidence became available. Similar views have also been observed in mock jury studies. For example, Hope and others found that around a third of participants believed the accused could be retried after a not proven verdict, despite having heard judicial instructions stating this was not possible. They also observed a strong perceived 'stigma' associated with the not proven verdict.
A small number of mock jury studies have also examined differences in whether jurors returning not proven verdicts were more likely to think the accused was probably guilty (but the evidence was insufficient for a conviction) than those returning not guilty verdicts. Curley and others found that ratings of 'likelihood of guilt' were considerably higher among those who favoured not proven compared to those who preferred not guilty (a mean of 58.0 compared to 39.2). In contrast, Hope and others did not find any significant difference in 'likelihood of guilt' ratings between jurors who favoured not proven and jurors who favoured not guilty.
In summary, in so far as existing jury research has engaged with the key research questions in the present study, it suggests that:
- 12-person juries generally have higher quality deliberations than smaller juries. However, it is unclear whether increasing jury size further (to 15) has any benefits.
- Requiring jurors to reach 'unanimity' may make deliberations more thorough, but may also result in more 'hung' juries. There is no existing research on the Scottish 'simple majority' requirement.
- Evidence on the not proven verdict's impact is equivocal but suggests that jurors may be less likely to convict when the not proven verdict is available.
- There are some misunderstandings about the possibility of a retrial after a not proven verdict. Some limited evidence also suggests that jurors who favour not proven may be more likely than jurors who favour not guilty to think the accused is probably guilty.
The research questions for this study clearly cannot be fully answered on the basis of existing evidence. It is in this context that the Scottish Government commissioned the large-scale mock jury study which is the focus of this report.
1.4 Report structure
The remainder of this report is structured as follows:
Chapter 2 describes the research methods for our mock jury study. It explains the steps taken to maximise realism and the limitations of the study.
Chapter 3 presents findings on the effects of the unique features of the Scottish jury system (three verdicts - including not proven, 15 jurors, and requiring a simple majority verdict) on verdict choice.
Chapter 4 discusses how the unique features of the Scottish jury system impact on the way in which juries reach their decisions.
Chapter 5 examines juror understandings of the not proven verdict.
Chapter 6 summarises the conclusions and discusses the potential implications of any changes to the unique features of the Scottish jury system.
Annexes A to K contain further detail on the study aims, background and methods, including: the full research questions, a summary of issues to consider when assessing mock jury research, a more detailed summary of previous research on the not proven verdict, summaries of the content of the trial videos used in the research, copies of the research materials, and details of the statistical tests used. They also include additional findings tables, and the remit and membership of the Research Advisory Group for the project.
The key terms section at the start of this report explains some of the key legal and other terms referred to in this report.