• 10:30 – Introduction
• 10:35 – Action and sub-group update
• 10:45 – Census 2031 Planning
• 11:15 – EAP219 Statistical Design for Census 2031
• 12:00 – Future of Population and Migration Statistics update
• 12:15 – EAP220 2031 Census Topic Consultation evaluation criteria
• 13:00 – Any other business / Close

Panel members

• Sir Bernard Silverman (Chair)
• Ana Basiri
• Oliver Duke-Williams
• Carl Emmerson
• Nik Lomax

Office for National Statistics

• Owen Abbott (Presenter)
• Mary Gregory (Presenter)
• Cath Hollyhead (Presenter)
• Joy Preece (Presenter)
• Michelle Waters (Presenter)
• Charlie Wroth-Smith
• Tom Tarling (Secretariat)
• Susan Williams (Secretariat)
• Emily Winstone (Secretariat)

1. Introduction

1. Sir Bernard, the chair, welcomed attendees and panel members. Natalie Shlomo had given her apologies.
2. Following the resignation of the National Statistician, the chair noted the panel would continue to provide methods assurance to appropriate leaders in ONS.

2. Actions and sub-groups update (Tom Tarling)

Actions:

1. Action 100: Consultation Updates
   • The National Statistician’s recommendation, and MARP’s statement on this, were both published on the UKSA website.
   • MARP Statement link
   • UKSA Recommendation link
   • This action can be closed. Follow-ons from the recommendation will be brought to the panel in future meetings.
2. Action 114: ONS to produce and provide to the panel an explainer on papers covering the topic of Multiple System Estimation
   • This is currently being reviewed, and aimed to be shared onwards with the panel in July 2025.
3. Action 115: ONS to provide a worked example to the panel, following up on the Census Benefits update presented at MARP37
   • This has been provided to MARP members and their feedback is being considered by the ONS. This action can be closed.

MARP Subgroup updates:

1. MARP Sub-group activity since the previous meeting was discussed, with the Labour Market group having published advice in April 2025:
   • Comments on proposals for the design of the Transformed Labour Force Survey
   • Comments on the Labour Force Survey recovery plan
2. In addition, groups considering the Dynamic Population Model, and Migration statistics have been formed.
3. The sub-groups membership consists of:
   • Dynamic Population Model
     • Dr Fred Piel (Imperial College London)
     • Prof Jon Forster (University of Warwick)
     •  Prof George Ploubidis (University College London)
     •  Prof Arkadiusz Wisniowski (University of Manchester)
   • Migration
     • Dr Marie McAuliffe (International Organisation for Migration)
     • Prof Arkadiusz Wisniowski (University of Manchester)
     • Prof Jakub Bijak (University of Southampton)
     • Georgina Sturge (Migration Observatory)
   • Labour Market
     • Prof Ray Chambers (University of Wollongong, Australia)
     • Prof James Brown (University of Technology Sydney, Australia)
4. From the main MARP panel, Ana Basiri will be contact for the Dynamic Population Model subgroup. Nik Lomax will be contact for the Migration subgroup.
5. It was noted the 2024/25 annual report is due at the end of September 2025. The panel agreed the report should include sections covering activity of MARP subgroups, and requested the Secretariat to begin drafting the report.

Action:

Secretariat to provide chair, and sub-group primary contacts, with draft of annual report.

3. Census 2031 Planning (Owen Abbott)

1. ONS noted the publication of the National Statistician’s recommendation for a census in 2031.
2. A census taskforce will run until January 2026 to plan the critical path toward 2033 (the production of final outputs).
3. The objectives of the taskforce were laid out, noting an aim for an integrated UK census design.
4. The panel commented that stakeholder trust and confidence was important to the process, requiring that stakeholders be informed and consulted. ONS agreed, adding relationships with devolved governments was also a focus.
5. The panel asked ONS to ensure MARP is involved in good time, ahead of critical requirements, to ensure sufficient time for review to be provided where there may be contentious or difficult choices and trade-offs.
6. ONS highlighted innovation events in July discussing ideas such as a paperless census, and use of administrative data. The panel requested a session be arranged in August to discuss ideas amongst the panel members. ONS agreed, and will seek to schedule a meeting.
7. The panel concluded that ONS would need to remain flexible to subsequent innovation, noting rapid technological change occurring including AI. It also added that a census benefits report would be helpful to mitigate against shifts in the external context which may affect funding.

Action:

ONS to hold a MARP session in-person discussing innovation topics for Census in August 2025.

4. EAP219 Statistical Design for Census 2031

1. ONS presented a paper on the underpinning statistical design for Census 2031. It welcomed the panels comments on the paper, and invited further discussion, noting a more refined paper will be brought to the panel in the future.
2. The panel noted the need to rework the approach for measuring communal establishments. It also suggested the use of administrative data to target follow-up though emphasised that a Census Coverage Survey of some form would need to exist as administrative data alone was not sufficient.
3. The use of pre-populated information and generating ideas to reduce paper responses reiterated the benefit of holding a meeting with MARP members to brainstorm innovative approaches in August.
4. Optional questions, and specific questions for geographic areas were discussed. ONS noted that legislation only allowed for socio-economic topics to be mandatory, so some additional topics could only be voluntary.
5. The impact on response rates (including bias) of optional questions, and topics such as income, was discussed. The panel commented that where administrative data was rich, certain topics need not be asked as questions. However, some topics such as disability and health could benefit from census provision to address unmet user need for such information. The panel discussed the benefit of functional questions over self-identification for such topics.
6. The panel noted that optional questions for different regions would lead to areas having different statistics available to them. This might lead to improved policy decisions for some areas that would not be replicable in other areas. It was discussed whether differing quality of statistics across areas could result, which the panel were not in favour of. The panel added that providing ‘prefer not to say’ options did not have the same impact as a question being optional.
7. The panel thanked ONS for the paper, and topics will be further discussed in the August meeting.

5. Future of Population and Migration Statistics update (Mary Gregory)

1. ONS gave an update on the FPMS programme, in context of the recent census recommendation. ONS stated it is still committed to making the most of administrative data.
2. Updates to the panel covered progress on the admin-based census, and additions of Ethnicity and Veteran Status attributes.
3. ONS confirmed household estimates would be published as previously planned.
4. ONS stated it would avoid publishing administrative based statistics too soon, where revisions would be likely be required.
5. The panel thanked ONS for the update.

6. EAP220 2031 Census Topic Consultation evaluation criteria (Cath Hollyhead, Joy Preece, Michelle Waters)

1. The panel commented on this paper prior to the meeting and thanked ONS for the constructive responses. ONS now summarised the paper, responding in more detail to panel comments.
2. The approach to evaluating topics for inclusion in the 2031 Census has made use of reviewing the approach and topics chosen for 2021 Census. Also proposed is a zero-based approach (starting from a blank sheet). The panel agreed with, and recommended, proposals for a zero-base approach. ONS agreed to consider this proposal and confirmed they are assessing the availability and quality of other sources of data.
3. The panel acknowledged the desire for continuity, but disagreed it was difficult to remove questions previously asked. The proposed ONS framework went some way towards this but should question the inclusion of existing topics. It recommended ONS examine other rigorous high-quality surveys, and their alignment of user needs and questions. The panel recommended in addition to a wide user consultation, a comprehensive literature review should be carried out. ONS accepted these comments.
4. The panel commented that it would be useful to review the proposed approach to weighting of the criteria for topic selection. This might include the use of negative weights for undesirable criteria like question size, availability of administrative data and response burden. The panel advised that the risks to using ranking for evaluation versus setting a threshold should be considered. ONS agreed and will hold internal workshops to develop the strategy and process for evaluating topics.
5. Concluding, the panel commented that a framework considering multiple weighting approaches allowed for robustness to be considered. The panel agreed this was a useful topic to discuss further. ONS would report back to the panel in a future meeting.

7. Any other business / Close

1. ONS highlighted upcoming openings for the NS Inclusive Data Advisory Committee. Details will be published on the UKSA website, and shared with the panel.
2. The panel expressed a preference for an in-person meeting in August, though noted availability could be difficult in this time. The Secretariat will follow-up and arrange.
3. On growing number of sub-groups, the panel agreed the structure and organisation of these, allowing for review and sign-off, would be important.

Action:

Secretariat to supply details on NSIDAC positions when advert is published.

Action:

ONS to structure and organise running of subgroups, integrating with the main panel to allow for review and sign-off.