7. Recommendations for the conduct of future assemblies

We stress that this was the first of a planned series of assemblies. As such, it was partly experimental, with the intention being to learn from it to improve the conduct and content of future meetings.

This first Assembly involved more than merely the day itself. The call for contributions provided a firm basis for preparing an agenda for the day, already demonstrating the value of user engagement. This approach is worth repeating for future assemblies, perhaps also allowing time for feedback from prospective attendees and reflection on issues and topics that do not appear in responses to the call. Moreover, in the days following the Assembly we received a number of email comments from delegates as they reflected more deeply on the discussions. There needs to be some formalisation of this source of more considered thoughts following future assemblies. GSS theme groups could play a major role in this engagement.

Inevitably, with a limited amount of time on the day itself, a choice had to be made on what topics to cover in the breakout sessions. While future meetings might return to all or some of these topics, other topics must also be considered, and future calls should invite suggestions. Other suggestions which have already been made are consumer behaviour or attitudinal data, cost of living, education, higher education, transport, food and agriculture, defence, the value of following international standards (especially in the context of Brexit), and a session on how to create an official statistical system with more porous boundaries so that expertise from outside can be utilised (such as via secondments). It might be appropriate to include a session where attendees look at suggestions and prioritise proposals (perhaps in a second day) since the current approach made it difficult for the Authority to get a view on what delegates themselves would prioritise.

It would be helpful, and would encourage future participation and engagement, if details were given of how the outputs from this Assembly will feed in to the Authority’s work over the next few years.

Further recommendations for future assemblies included:

1. Enable other ways for people to input in person on the day, as it was not always possible to feed in to sessions (for example if one wanted to participate in two parallel sessions or had clashing meetings).
2. Perhaps enable online breakout rooms for the online participants, to take place over the breaks.
3. Facilitate greater representation from policy professionals, whose perspective was missed.
4. Encourage greater participation from sectors which might benefit from greater representation, such as business and local government.
5. At future assemblies, all delegates should be able to see others’ responses, whether in the notes taken at the meeting or via Slido, and these should be generally available after the meeting.
6. It might be helpful to have a session focused on what will be removed from the Authority’s workplan. The Authority could lead this session, for example providing a list of questions (for example, should we do X, or dump Y, devote more or less resource to Z?). This would bring home the practicalities arising from resource limitations.
7. Some sort of cost-benefit analysis of the Assembly should be made. This might be in terms of changes which occurred as a consequence of the Assembly.
8. The Authority should identify which of the concerns raised by users are already a focus of attention, and feedback such a list to NSEUAC.
9. Perhaps an online survey about what delegates see as priorities could be conducted. It would be interesting to see a before- and after-Assembly comparison of the results.
10. The Assembly delivery group created a framework for the sessions, but not all sessions adhered to that framework, perhaps because alternative approaches best suited the topic. Nonetheless a standard framework at some level does facilitate capturing user concerns.
11. An important issue raised was how often the Statistical Assembly should be held. The Lievesley Review said triennially. However, there is some enthusiasm for more frequently, not least because of a feeling that people want the statistical system to be more nimble, agile and responsive. Annually seems too often, given the number of people involved and the resources and effort required, so perhaps biennial would be suitable.
12. An elaboration of the biennial suggestion was to run every two years, with a full Assembly like this first alternating with a more focused meeting (so that the full Assembly took place every four years). The focused meeting could be devoted to particular topics, or could be jointly organised with other bodies (for example, the RSS and the Confederation of British Industry).