Citizens' Assembly of Scotland: research report

Chapter 7: Conclusions and Recommendations

The Assembly was the first citizens' assembly conducted at a national level in Scotland. Its successful completion was a notable achievement in the face of a number of challenges. Most significantly, the onset of the COVID-19 pandemic caused the Assembly to be paused for six months before resuming online until its completion.

In the following, we consider the internal dimension of the Assembly – its design and delivery and the members' experience – and the external dimension of the Assembly – its wider reception and impact. We draw conclusions about what worked well, what were the key challenges, and what learning can be drawn. We make recommendations to inform the design and delivery of future assemblies and other deliberative approaches in Scotland and internationally.

The internal dimension of the Assembly

Members' experience

For the majority of members, taking part in the Assembly was overall a rich and rewarding experience, both individually and collectively. This is common in assemblies and other deliberative approaches, which provide a level and quality of participation for members which they are unlikely to have experienced in other ways. Nonetheless, it was clear that the work of the organisers and facilitators was a key factor in members' experience. The care for the welfare of members and practical, technological and emotional support for their participation was notable.

Ensuring member inclusion and participation is a key design feature of citizens' assemblies. Overall, most members in the Assembly felt included in, and empowered by, the process. Over the course of the meetings, there was increased self-confidence and belief in their ability to participate. This personal growth was also associated with a greater sense of their ability and interest in participating in other political and civic activities: an 'activating' aspect to their experience.

There was some variation to this general picture of members' experience. Female members overall reported slightly lower levels of satisfaction; and lower levels of comfort and opportunity to express their views in small-group discussions. They were also observed to participate less in all-Assembly plenary sessions in the early weekends but engaged more equally with men in later weekends. Organisers sought to encourage and support this improvement. There was also a small minority of members that reported dissatisfaction, less positive experiences and a lack of inclusion.

Design and delivery of the Assembly

There are a number of common features associated with the design and delivery of citizens' assemblies. These include the provision of expert evidence, opportunities for members' to learn and reflect collectively, deliberation between members' about the topic, and a process that supports members to draw conclusions or make recommendations, depending on the defined task.

The breadth of the remit given to the Assembly is unusual when compared to other citizens' assemblies and deliberative approaches. Characteristically, they focus on a specific issue and are tasked to respond to a particular question. A range of design and delivery decisions follow from the remit. The scope of the Assembly was expressed as 'what kind of country are we seeking to build' and it was tasked to respond to three broad questions. This presented a range of challenges to its design and delivery and also had an impact on its wider reception. Still, in practice, that breadth was used as an (arguably necessary) opportunity for members to play a role in shaping the agenda of the Assembly and the topics for discussion.

A range of design and delivery plans generally flow from decisions about how to implement the remit of an assembly. In this case, the Assembly was planned rather more on a weekend by weekend basis, particularly between weekends one to four where the topics for discussion reflected the results of discussions from the prior weekend. This placed pressure on all the organisers to design the following weekend, and crucially to identify expert speakers, in a tight window of about four weeks. This timing placed a constraint on speaker selection and availability, and also on the time for preparation and scrutiny of speaker input. In practice, if there had been a longer interval between each weekend meeting, then it would have likely mitigated these constraints somewhat.

The move online for the second half of the Assembly meetings was a significant change in the delivery of the Assembly and had a broad impact. First, it created a new constraint on member participation and inclusion. Organisers invested time and practical and technological support in advance of restart to ensure all members' were able to connect online and use the virtual meeting platform to participate fully. There were teething problems in the first online weekend for some members, relating to connection issues and members' understanding of how to participate online. This improved significantly in the following weekends as a result of interventions from the organisers, and members' growing familiarity and confidence with the format. The length of weekend meetings was also reduced, reflecting concern about what might reasonably be expected from members in the online environment. It was also evident that most activities simply took longer than in person. Together, this compressed the time available for recommendation formation, scrutiny and agreement. There were additional challenges for facilitators who had online collaborative tools to use alongside completion of group discussion tasks and managing group dynamics. Some members found engagement easier online, others less so; but there was a common sense of regret that the opportunities for informal and social interaction valued in the in-person weekends were missing.

A key element of any citizens' assembly is to build members' knowledge and understanding of the issue they have been asked to consider to help inform their deliberations. Members of the Assembly heard from a range of experts and advocates across a broad spread of topics, particularly in weekends 1-4. The quantity and breadth of knowledge and evidence presented were sometimes experienced by members as difficult to absorb. The quality of delivery we observed from the selected experts varied, although the majority of members commonly expressed satisfaction. There were opportunities to hear from a greater diversity of perspectives on some topics, but which could not be taken due to the constrained time available between weekend meetings to identify and access diverse expert input. The six month pause in the middle of the Assembly made it more difficult for members to recall and engage with the evidence presented in the first four weekends, despite concerted efforts by organisers to remind members about what they had heard when the Assembly restarted. There is evidence members learnt from the evidence they were provided and from each other, which informed their views. A greater diversity in evidence provision formats, to meet the range of learning styles and needs of the assembly members, could have further enhanced this learning. Their knowledge was greater at the end of the Assembly when compared with members of the public we surveyed.

The majority of deliberative work undertaken by members in assemblies takes place in small group discussions. This places a premium on the quality of facilitation in small groups. The facilitators at the Assembly demonstrated a commitment to the Assembly process and to the care and support of members. Regular briefing and debriefing at and after every weekend encouraged a culture of reflective practice and improvement. There was evident variation in how well facilitators managed group dynamics to ensure inclusion, and encouraged members to engage in deliberative discussion. Some of this variation reflected different levels of skill, but also that facilitators were drawn from a range of different organisations and professional traditions. There was not a consistent understanding of deliberative facilitation practice. Equally, weekend design decisions were routinely finalised very close to Assembly meetings, which led to facilitators receiving detailed instructions with less opportunity for preparation than might be ideal.

Developing and fostering deliberation is a central design feature of citizens' assemblies. This is not a conventional mode of participation and so members require support through the process to encourage a range of deliberative values and practices. Some forms of deliberation improved over the Assembly weekends, a common finding in other assemblies. Some of this resulted from proactive facilitation, which might have been informed by feedback from the research team, from organisers and facilitators' reflective practice, and from members' growing in confidence and familiarity. The practice of scrutiny and challenge – of expert evidence and between members – was the least well developed. Our dataset suggests there are likely to have been a number of influences on this, including: cohesion and consensus were valued explicitly and repeatedly emphasised in the Assembly; there may have been insufficient time given to it in small group discussions, methods and techniques may not have systematically fostered constructive challenge, and its importance may not have been given enough emphasis.

Responsibility for the governance of the Assembly – which we take to comprise oversight, advice, organisation, design and delivery (including media communication and public engagement) of the Assembly – was distributed across a spread of different groups, namely: convener, secretariat, stewarding group, design team and facilitation team. This constitutes quite a lot of 'moving parts' to articulate the relationships between. It was evident that the tight timescale from the inception of the Assembly and the broad remit added to the challenge of working this through to common agreement. Together, there was a shared commitment to the members, to working together and to the successful delivery of the Assembly. There was a distinction in function between these groups but the boundaries were not clearly distinguished and there was a lack of collective and complete understanding of the respective roles that each would play. Arrangements for oversight of the evidence provided to the Assembly in particular were not optimal. The lack of collective clarity was experienced at different points as more or less frustrating, affected collaboration, and placed constraints on the overall effectiveness of the governance. In practice accommodations and adaptations were made that prevented any significant fissure to emerge.

There is value in reflecting on the role and importance of the research itself for the Assembly. The research was independent but deliberately embedded in the process. This meant that the Assembly organisers' commitment to continuous improvement was supported by regular feedback after each weekend from the research to the organisers and the stewarding group. This feedback was not uniformly applied but did inform changes and improvements in design and delivery, particularly in the early weekends. Unprecedented openness and access for the research, and a comprehensive mixed-methods design, enabled a unique level of granularity and robustness in the data obtained.

The external dimension of the Assembly – its wider impact and reception

There is less available evidence on the external impact of citizens' assemblies than on the internal dimensions. This research can contribute to learning on that, drawing on its analysis of media coverage, public awareness and the perceptions of journalists and politicians as key actors mediating the relationship between the Assembly and its public.

Wider public awareness and understanding of citizens' assemblies is important if there is to be support and legitimacy for their role in decision-making by governments and other public authorities. Media coverage can make an important contribution to public perceptions but citizens' assemblies have not always received significant media attention. The analysis of written media coverage of the Assembly indicates this was variable, with monthly reporting greatest across the in-person weekend meetings and at the time of the publication of the Assembly's report. The content was largely neutral or positive overall, focused on the process at the beginning of the Assembly and its outcomes and report at the end. But the pattern of which media covered the Assembly means that it would not have been seen by much of the Scottish population. The pause in the Assembly and the media focus on the pandemic clearly contributed to this. The organisers increased the communications capacity available initially and took a more strategic approach which had a positive effect, particularly with local media coverage which focused on particular members from the locality.

From our survey analysis, public awareness of and engagement with the Assembly were low. Still, support for the use of citizens' assemblies in principle increased over the Assembly period, and over two thirds of the population surveyed after the Assembly's completion indicated willingness to participate in a future assembly. The Assembly's recommendations were supported by a majority of the participants in the population survey, which is an important finding for policy-makers taking forward this work.

The Assembly was perceived positively overall by the journalists and politicians we interviewed. Although there were some more cynical initial perceptions about the purpose of the Assembly, by the end most were more positive and concluded that it had worked much better than they had initially anticipated. The general view was that the Assembly process had been well run, that members had clearly had a positive experience, and that it could have a positive influence as a prototype for future assemblies or other deliberative processes. This indicates importantly a recognition of legitimacy of the Assembly. Concerns expressed about the Assembly centred on the breadth of the remit, the quality of deliberation, the feasibility of Assembly's recommendations and the limited engagement with the wider public, and were unclear about how the Assembly could have impact. There were doubts expressed about the Assembly's influence; whether and how its recommendations will be taken up by the Scottish Government or Parliament, due to the lack of any formal legislative basis for the Assembly, and a perceived uncertainty about its mandate. Questions were raised about how the Assembly's recommendations sat alongside statutory requirements and common policy processes. There does not seem to have been clear advance plans about how the Assembly's recommendations would be dealt with. It is difficult to assess at this point what the medium to long term impact of the Assembly's report will be.

There are lessons to draw from this first experience of democratic innovation in Scotland in the form of a national-level citizens' assembly. This will help clarify how future assemblies would relate to, and connect with, the decision-making processes in the existing democratic institutions of Government (central and local) and Parliament and other public governance functions such as public audit and scrutiny. This is important for the wider legitimacy of assemblies and to meet the expectation given to members that their participation would influence decisions.

Recommendations for future assemblies

The following recommendations draw from the learning taken from this specific assembly, and we hope serve as a contribution to the growing body of evidence and experience internationally in democratic innovation broadly, and the particular use of citizens' assemblies.

• Assembly Remit: Recognise the importance of decisions about remit for design, delivery and governance and the benefits of a clearly articulated task and specific question. Decisions on remit and the time needed for an assembly are integral and interlocking: the broader a remit, the more time is likely to be needed for the assembly. The process that is used to determine an assembly remit can itself serve to enhance its mandate and legitimacy.
• Governance Framework: Collectively agree roles and responsibilities in a clear and distributed governance framework that distinguishes responsibilities for oversight, advice, design and delivery. This can contribute importantly to build in checks and balances on key aspects such as evidence provision and process design.
• Assembly Phases: Recognise the importance of three distinct phases to assemblies: inception, delivery and impact; and the need to plan for, resource, and give sufficient time to, each phase from the beginning of the process. At the inception phase, decisions about remit, governance and research are key; this will importantly shape the design and delivery phase. The impact phase of any assembly cannot be assumed but needs to be planned for and should include considerations about where accountability will lie for impact and how and when that is provided to assembly members and the wider public.
• Assembly Impact: Set out a clear mandate and identify how any assembly will interact with, and influence, the decision-making of existing democratic institutions.
• Public Engagement: Consider how to integrate engagement with the wider public into any assembly process to help build their awareness and understanding and enhance the legitimacy of assembly outcomes.
• Capacity Building: Build capacity in skills and resources for this kind of participatory work, focused on deliberative principles and practices. This includes training in specifically deliberative facilitation.
• Research: Concurrent research should be embedded and used to inform the Assembly's design and governance. The research should be fully funded and have a duration that enables an assessment of impact.