Over the weekend, The Democratic Society was involved in the European Citizens’ Panel, run by the European Commission, with the support of Kantar Public (a market research agency) and Missions Publiques (a specialist citizen deliberation organisation). Bertelsmann Stiftung acted as academic experts.

We helped with the design and facilitation of the event, which ran from Friday night to Sunday lunchtime, and a group of facilitators from the civil society network that we are creating on the European Citizens’ Consultation supported Missions Publiques with independent facilitation.

These are my first reflections – I’m sure there will more thorough research and reporting put out in due course.

What was the purpose?

The European Citizens Consultations (ECCs) are happening across 27 EU member states over the next few months. With original impulse from President Macron, they are designed to give a sense of what European citizens want for Europe’s future, in advance of the European elections next year, and the mandate of the new Commission.

The ECCs are not a single thing. Each member state has said they will undertake them in their own way, and certainly some will do more and some will do less.

The central common element is a digital consultation being run by the European Commission. The Citizens’ Panel that met over the weekend was intended to choose the themes and questions for that digital consultation. The French government had already said that they would use the themes that emerged as the core elements of their consultation approach, and others may do the same.

Who came?

Kantar Public, the market research agency who have the contract for undertaking the regular Eurobarometer survey, recruited a group of 96 participants from all 27 countries.

To ensure that each country could send a man and a woman, there wasn’t an even distribution between countries. Many countries had two participants, none had fewer than two, and none had more than six. This meant that France, say, was comparatively underrepresented compared to Malta.

The participant selection was also designed to create an audience representative on gender, age, employment status and economic status. The recruitment plan was designed to ensure that representation was spread across the countries, so that, for instance, in country X Kantar Public’s team would have to find a man under 30 who was employed and a woman over 65 who described herself as being under financial stress.

Participants did not have to speak English, or even understand it. In the final group, just over half said that they had some knowledge of English, but this varied widely between countries.

What was the setup?

We were hosted by the European Economic and Social Committee, who were unbelievably flexible, welcoming and supportive. Because the event was finalised at short notice, there were a lot of last-minute arrangements to sort out.

Because people had to express themselves in their native language, we used interpretation (the usual setup with interpreters in cabins and people wearing headsets that you’ve probably experienced, or seen on news reports if not).

The interpreters, who were giving up a bank holiday weekend, arranged the participants in groups so that everyone could speak in their native language and hear in a language that they understood (though not always their native language). With 22 languages spoken (the official languages minus Maltese and Irish) this was a multi-dimensional jigsaw puzzle but it worked – there were only a couple of people for a couple of sessions who needed someone with them to translate by whispering in their ear (chuchotage) rather than through the interpretation service.

The one inflexibility this introduced is that, because interpreters can’t work in all languages at once, the distribution of the participants had to be fixed for the whole two days, there was no option of mixing and combining groups. Still, there was a good mix in each group – I don’t think we had fewer than four nationalities in any of the rooms.

What happened?

We started with a blank sheet of paper. There was some talk beforehand of having a long list of topics that the EU worked on, or key themes that had come out of other engagement exercises like Eurobarometer, but it was thought best to start from zero.

After an introduction to the process and some icebreaking on Friday night, and a validation of the rules and confirmation that people understood the process the following morning, discussion started in groups.

Participants were asked to nominate topics that they would consider to be the main issues that should be asked to their fellow European citizens, in the context of a conversation on Europe’s future. Each group was facilitated by an independent person, with a note-taker recording the issues and argumentation for each.

After 1h45 minutes of discussion, the topic lists were brought together by the facilitators and the six most frequently raised issues were marked as the “top six”. These were:

  • Education and Youth
  • Equality, Fairness and Solidarity
  • Environment
  • Making rules and making decisions
  • Migration and Refugees
  • Security and Defence

These “top six” were then pinned – people could reopen them and continue the discussion on them, but they were noted as being significant topics that would be part of the twelve selected.

After lunch, participants were asked to continue their discussions with a prompt question, about what they think Europe will be like in 2040. The aim in this session was to enrich and deepen the list of topics from the morning.

Finally, before the plenary session at the end of the day, participants were asked to choose the two most important topics that they had discussed, excluding the “top six”, and present those back in the plenary.

Each group presented back in the plenary session, and there was then a vote to select the “second six” to go with the “top six” to form the twelve topics for the citizen consultation. From the fourteen topics presented, similar ones were merged so in the end ten were voted on.

The vote was positive only (you could vote for but not against), and each participant was asked to cast no more than six votes. The electronic voting system in the Committee was not ideally suited to this voting setup, but we made it work after a couple of initial hitches and the topics and voting results can be seen below:

  • Health/Quality of Life/Ageing Society (merged) – 86 votes
  • Social Protection – 74
  • Economic security – 67
  • Maintaining the Union in a future crisis situation – 61
  • Work/Technology/Employment and Technological Development (merged) – 55
  • Agriculture/Fisheries/Food Security – 54
  • Climate change – 47
  • Local vs EU decision making – 46
  • Size of the EU (states joining or leaving) – 45
  • More or less integration of the states of the Union – 39

I was a little surprised that Climate Change and institutional arrangements didn’t make the cut, but “Environment” and “How decisions are made” were already in the top six so presumably participants thought that the issues were sufficiently covered under that heading.

At that point the first day ended but each group had been asked to nominate one or two participants to take part in an evening session in which the group presented back to Kantar Public’s question design experts. Had we had longer with the participants, this would have been a focus of a daytime session, but with just a weekend to work with, this had to be an evening one. The disadvantage was that the session had to be conducted in English, because there was no interpretation available, but this was a report-back session rather than one where anything was going to be decided.

The participant volunteers joined their group facilitators and Kantar Public, and fed back the key points of the discussion in their groups relevant to each of the twelve selected topics. They were supported by the facilitators’ and note-takers’ notes. Kantar Public’s team (who had also been sitting in on the sessions) then asked questions for clarification, and explained how questions could be written to be open or closed, and how to avoid leading or biased questions.

The plan had been for Kantar Public’s team to spend an hour drafting a first set of questions on the twelve topics, before presenting them back to the group for initial feedback, but the group presentations took longer than expected and Kantar Public had a lot of material to work on. Rather than presenting the questions back to the volunteers at 2300, we brought them to the venue early and gave them first sight of them at 0830 before the 0900 start.

Kantar Public produced a long list of 39 questions, arranged under the twelve topics selected by participants. They merged “equality” and “social protection” to allow space for a set of cross-cutting or “transversal” questions that picked up on common issues arising.

In the original event plan, we had thought about giving each group one or two topics and asking them to choose one question from each to send to the plenary for approval, but on looking at the question list and thinking about the breadth of discussion the day before, we changed the plan on the Sunday morning and gave all the groups all the questions to consider.

This was the area where there was the most discussion in the facilitation team. Giving all the groups all the questions meant that every participant could express an opinion on everything and no-one would feel that they had been prevented from talking about, say, the environment. However, it was also a big workload for the Sunday morning, particularly since interpretation meant that the whole questionnaire draft had to be read to participants to ensure that they were able to hear it in their own language.

However, after considerable discussion, this approach was thought preferable to giving each group a subset of themes or issues, which would have increased their opportunity to go deeply into the questions, but (given that we couldn’t rearrange the groups because of interpretation) prevented some participants from having a say on issues that they cared about.

In each group, participants were given ten votes to distribute among the 39 questions (one vote per question maximum). They were told that the target was for there to be at least three open questions in the survey as a whole, and one question on each topic.

The participants expressed some concern that the workload of 39 questions was too heavy, but the discussions did get under way and with a half hour extension on the planned time in groups, each group voted successfully.

The vote in each group was added together and the top question in each section selected. In the end, only two of the open questions were selected and ten of the closed ones. On the basis that we needed one more open question, participants in plenary were given a vote between the two open cross-cutting questions, and chose the second of the two options as a thirteenth question.

Finally, participants voted to approve the list as a whole, and we had our twelve topics and our thirteen questions.

Thanks and reflections

It’s worth recording here what a great piece of teamwork this panel was. Arranged at short notice, reworked as we ran it, it demanded and received a huge level of commitment from the Commission, from Kantar and its partners. I’ve never run a fully multi-lingual citizen panel before and although we had the immense benefit of the EESC and the interpretation service, there was a lot more discussion between participants than I had thought there might be. On some tricky issues dividing Europe, you could see those from countries with different political attitudes listening to and reflecting on each other’s views.

Not everything worked. We needed more time than we had, for a start. Certainly another day would have allowed for deeper deliberation, or even running the event across two weekends – but the logistical upheaval of bringing people to Brussels for that would have been impossible.

The discussions were good but the slope of the decision making process (from blank sheet to final questions in a day and a half) was very steep. Participants were positive about the experience in the discussions at the end of the event, but I will not be surprised if the evaluation forms tell me that they found themselves rushed at certain points.

The questions, too, will need a little polishing, coming as they do from late night work by Kantar Public and being drawn from a very broad and diverse discussion. I don’t know – I’m writing this before the questionnaire is released – how faithful the final version will be to what emerged on Sunday afternoon, but I hope it will be a very close correlation. Perhaps not every multiple choice will be the same but I hope to see the same fundamental questions, and the twelve topics that emerged.

The final contribution, from Commission DG of Communications Timo Pesonen, reflected that this was a very new experience for the Commission, who had run citizen dialogues before but never an event like this, designing a participation approach participatively. The door was very clearly open to more, and although there are lots of lessons on what to do and what not to do that we can draw from it, for a first time event, pulled together over the course of no more than a month, it feels like a success.

What’s next

The real mark of success, of course, will be the impact of the European Citizens’ Consultations and the themes that the participants came up with. We’re talking about them, and this event, at the launch of the Open Government Network for Europe on 22 May, and you can sign up to join us there right now.

We’re also, with our colleagues at the European Policy Centre, running a European civil society network on the Citizens Consultations, and if you would like to find out more, or share what’s happening on the ECCs where you are, please let us know.

Thanks to Hannah Starman, Lena Morozova-Friha, Stephen Boucher and Marcin Gerwin for acting as facilitators, to Paul Butcher and Satguine Maison for taking notes, and to Corina Stratulat for her work in the core facilitation team.

Further Links


Published by Anthony Zacharzewski

Anthony Zacharzewski was one of the founders of Demsoc in 2006. Before starting work for Demsoc in 2010, he was a Whitehall civil servant and a local government officer.