The Constitution Unit Blog

Why do citizens’ assemblies work? Evidence from the citizens’ assemblies on Brexit and Social Care.

       As the debate about the UK’s relationship with the EU continues to dominate the political agenda, citizens’ assemblies have been mooted by several high profile figures as a possible way to break the Brexit impasse. Here Sarah Allan and Rebecca McKee explain how and why citizens’ assemblies are able to assist and improve the policy-making process through engaging and informing ordinary members of the public.

Citizens’ assemblies have been gathering more attention amongst politicians, the public, and the media in recent weeks. For some this model of public engagement is entirely new. Yet, the history of citizens’ assemblies and methods like them extends back to the 1970s. Since then they have been used around the world to bring together representative groups of the public to deliberate on controversial and complex issues. Countries that have had citizens’ assemblies include Canada, the United States, Australia and Belgium. Most famously Ireland’s citizens’ assembly and constitutional convention played key roles in change on abortion and gay marriage.

The core purpose of a citizens’ assembly is to give decision-makers access to the informed and considered views of the public. A citizens’ assembly can be said to have worked when these three factors are delivered to a high standard. We use the examples of the Citizens’ Assembly on Brexit (CAB) and the Citizens’ Assembly on Social Care (CASC) to show that it is possible to deliver on these principles.

‘The views of those that took part in our citizens’ assembly have been vital in informing our thinking and the model also provides a possible route for further public engagement and building the support that any reforms will need.’ Clive Betts MP, Chair of the Housing, Communities and Local Government Select Committee.

The public

Is it possible to recruit a representative group of participants?

The primary goal of citizens’ assembly recruitment is to secure a broadly representative sample of the population as assembly members. The population of interest varies depending on the assembly topic. CASC was commissioned to look at the devolved issue of social care, so participants were only recruited from England. CAB dealt with the UK’s exit from the EU, so its membership was UK-wide. Both topics were issues of policy so participants were restricted to those eligible to vote in either general elections for CASC, or the 2016 European Referendum for CAB.

Citizens’ assemblies often use stratified random sampling to ensure that assembly members are representative of the public on key demographic and attitudinal criteria. For both CAB and CASC the stratification accounted for basic demographics; age, gender, ethnicity, region, age and social class. For CAB we also included stratification on referendum vote to ensure participants broadly reflected the way the UK voted in the EU referendum. CASC participants were dealing with the issue of social care funding, so we included stratification on attitudes towards the size of the state.

Despite stratification there are structural societal issues that could act as barriers, preventing some people from attending. For example single parents, carers and people with additional needs may face challenges in taking part. Well-planned citizens’ assemblies take steps to overcome these barriers. The evidence shows that providing compensation or incentives, such as an honorarium or travel expenses is important. For both CAB and CASC, participants were gifted money for attending per weekend. We also paid for all hotel and travel expenses, including in the case of CASC for participants’ carers, where this was required. Our budget catered for additional adjustments such as BSL and DSL interpreters, although in the end these were not needed. Venues and materials also need to be accessible.

Table 1 – Recruitment targets and assembly members by stratification criteria

At CAB the full 51 people we expected attended weekend one and only 1 person had to drop out for weekend two because of illness. For CASC 47 of the 50 we were expecting attended weekend one and all returned for weekend two. The table above details how these participants compared to the wider population. The recruitment was largely successful in securing a broadly representative sample. Feedback from participants was very positive, with 87.8% of CAB assembly members agreeing that the assembly was diverse enough to consider all perspectives.

We also checked to see if there were groups of people who might have felt unable to attend. 15% of the 5,000 people who answered the CAB survey and went into the recruitment pool had caring responsibilities for a household or family member. People with caring responsibilities made up 16% of the final participants. Although we didn’t stratify for this, it’s important to know that the model doesn’t exclude people with caring responsibilities. Further work is needed to look at the barriers faced by other groups, particularly those with learning disabilities.

Can people participate equally?

Another key consideration is whether participants are able to participate equally in assembly activities. Citizens’ assemblies are carefully designed to try and avoid any systematic imbalance.

The data we gathered at CAB meant we could look at whether there was variation in how long different groups of people spoke for in table discussions. We found that there was no statistically significant variation across any of the demographic or attitudinal criteria listed above.

Relatedly, it’s important for participants to feel that the assembly environment is respectful, and one in which they feel comfortable discussing their views and experiences. Feedback from both CAB and CASC was very positive. In total 100% of CASC, and 92% of CAB, members agreed or strongly agreed that ‘My fellow participants have respected what I had to say, even when they didn’t agree with me’.

‘[it’s surprising] how respectful and engaged such a diverse range of people have been towards one another on such an emotive subject’ CASC Assembly Member

Informed: Is it possible to provide fair and balanced information on contentious issues?

Citizens’ assemblies have three stages; learning, deliberation, and decision making. The learning stage is when participants receive the majority of their information about the topic. It is vital that the information is accessible, balanced, and comprehensive.

‘How else would you receive informed decisions/views from the general public? Not many avenues would allow people to receive 4 days of information on which to base their opinions.’ CASC Assembly Member.

Both assemblies contained multiple different sources of information including; briefing papers, speaker presentations and panels, Q&A sessions with the speakers, and answers to any unanswered questions from weekend one which were handed out at weekend two. Assembly members were very positive about the amount of information they received. 96% of CASC participants said they had had enough information to participate effectively and 98% said that they had learned a lot during the assembly about adult social care funding.

Both CAB and CASC also benefited from an independent advisory board or group. These advisors are experts on the topic who, between them, broadly cover the spectrum of opinions on the issue. Their role is to oversee all the materials handed to assembly members, the choice of speakers, and how the speakers are briefed. They check that the breadth of arguments and opinions on the topic have been covered, and ensure there is no bias in the language used. At CASC the select committee officials provided extra oversight of the materials, choice of speakers and overall assembly design.

90% of CAB assembly members agreed or strongly agreed that the information they had received was fair and balanced between different viewpoints. Additionally 94% agreed or strongly agreed that they had heard from a broad range of opinions during the assembly. 

Considered: Are people able to deliberate on the issues?

For assembly members to use the information from the learning stage there needs to be enough time and space for them to engage in considered debate. Participants at CASC took part in 28 hours of deliberation each over the two weekends, making a combined effort of 1,316 hours of discussion.

Participants also need to feel comfortable with the topic and the different arguments. We asked assembly members whether they had felt able to get to grips with the information: 96% of CASC members said they had understood almost everything that the speakers had to say; 92% of CAB members felt the same way. In terms of table discussions, 100% of CASC members agreed or strongly agreed that they understood everything the people in their group had to say, for CAB this was only slightly lower at 92%.

Table facilitators and expert speakers at CASC and CAB fed back very positively about the way participants engaged in discussions. They praised the quality of both the questions that assembly members asked, and the table discussions more generally.

‘This has been an interesting and thought-provoking experience. There was a lot I didn’t know or understand about social care and the assembly has afforded me the opportunity to not only learn more but also meet individuals with various opinions and backgrounds.’ CASC Assembly Member.

Conclusion

Citizens’ assemblies bring together a representative sample of the population to give their informed and considered input to policy-making. In so doing, they help build understanding between people with different views, and give a powerful voice to people from all walks of life. They also help build understanding between the public and decision-makers: the former gain an appreciation of the complexity of key issues and trade offs that need to be made; the latter understand what the public would do when confronted with the reality of choices facing decision-makers. This in turn encourages politicians to find a solution where they have reached an impasse or a situation is politically difficult – as Health and Social Care Committee Chair Sarah Wollaston said recently on this blog. In a world where politics is often characterised by alienation, disaffection and division, tools such as citizens’ assemblies have a vital role to play.

This article draws on data from both the Citizens’ Assembly on Social Care and the Citizens’ Assembly on Brexit. Where it quotes data from one assembly only, this is because relevant data was not collected for both examples.

About the authors

Sarah Allan is Head of Engagement at Involve. She was Design and Facilitation Lead for the Citizens’ Assembly on Brexit and overall lead for the Citizens’ Assembly on Social Care.

Rebecca McKee is a Research Fellow at the Constitution Unit. She was the Research Associate on the Citizens’ Assembly on Brexit and was a Senior Project Officer at Involve working on the Citizens’ Assembly on Social Care.