In March 2014, the National Statistician made a recommendation that the census in 2021 should be predominantly online, making increased use of administrative data and surveys to enhance the statistics from the 2021 Census. This recommendation was endorsed by the Government’s formal response, which also set out its ambition that “censuses after 2021 will be conducted using other sources of data and providing more timely statistical information. However, any final decision on moving to the use only of administrative data beyond 2021 will be dependent on the dual running sufficiently validating the perceived feasibility of that approach.”
Following the National Statistician’s recommendation and the Government’s response, ONS established a programme in January 2015 to take forward three high level deliverables:
- A predominantly online census of all households and communal establishments in England and Wales
- Development of alternative administrative data census estimates, to be compared to the 2021 Census
- User focused, improved and expanded population statistics through increased use of administrative data and surveys
This document sets out the scope for an external expert panel to undertake a review of methods and research associated with the programme’s deliverables.
The review will provide assurance to the National Statistician that the online census methodology is robust and the resulting 2021 Census statistics will meet the code of practice for official statistics and therefore can be badged as National Statistics.
In 2023, the National Statistician will make a recommendation to Government on the future of the population statistics system beyond the 2021 Census. The review will therefore also provide assurance to the National Statistician that the methods are valid. This will ensure a robust evidence roadmap to the 2023 recommendation.
This review is planned to take place between 2018 and 2023. Note, previous reviews were undertaken for the 2011 Census and Beyond 2011 programme.
As part of the wider transformation of the statistical system, the design of social surveys is also undergoing transformation towards administrative data-first methods with residual surveys. From 2022 the panel will also provide assurance to the National Statistician and stakeholders that these methods are robust and the statistics resulting from household social surveys will meet the code of practice for official statistics and therefore can be badged as National Statistics.
Occasionally, ONS carries out deep dive quality reviews as a way of focusing on one area and identifying the risks to quality. The National Statistician has asked the panel to extend its review to oversee the assurance of some future deep dives.
Purpose of the Review Panel
The purpose of the review panel is to:
- Provide external, independent assurance and guidance on the statistical methodology underpinning 2021 Census outputs and those based on administrative sources.
- Identify significant gaps and risks in methods and make suggestions for mitigation.
- Review the methods being developed for Population and Migration Statistics Transformation and contribute to their continuous improvement.
- Provide external, independent assurance on the overall statistical design of the Labour Market Survey and the methods being developed for the Social Surveys transformation.
- Provide oversight of methodological deep dives which ONS may undertake across its portfolio.
The Chair is appointed by the National Statistician. The Chair and other panel members should:
- Be available to attend face-to-face or online review meetings as scheduled, and have preparation time in advance to read papers, compile questions and contribute virtually in between meetings.
- Be independent of the Office for National Statistics throughout the review.
The total estimated time required from the review panel members is around the equivalent of between 15 and 20 days per year, which can be made up of smaller pieces of time if required.
The ONS SRO is the Director of Methodology and Quality Directorate. The panel will advise and assure the National Statistician and the Census SRO.
No single person will have the expertise or experience required to conduct the review alone.
The panel will consist of the Chair and up to five other members, selected to give appropriate breadth of experience and expertise. If specific expertise is required for a particular review, additional experts may be co-opted, at the discretion of the Chair and the SRO, to take part in discussion and meetings.
The appointment of Panel members will be made, and if needed terminated, by the ONS SRO, normally acting in consultation with the Panel Chair.
All panel members will be remunerated for their time and reimbursed for their travel and subsistence expenses.
The panel member skills and expertise should complement each other to ensure coverage of all the topics within scope.
The core panel (plus additional experts, if needed) will need the following experience and expertise:
- Census methodology, edit and imputation, estimation and disclosure control
- Survey design in the context of population and socio-demographic estimation
- Use/analysis of administrative data
- Statistical modelling
- Record linkage
- Combining administrative and survey data/small-area multivariate estimation
- Demography and specifically population estimation
- Experience in the use of population and small area socio-demographic statistics
- Experience in Data Science techniques, including Big Data
- Experience in questionnaire design, data collection and registers
Over the full period the review should consider the methodological aspects of the end-to-end design for the 2021 Census from data collection to outputs including additional outputs using administrative data.
The review should also consider the stages of the Population and Migration Statistics transformation project and its evaluation including impacts on the Population Statistics System and 2023 Recommendation.
From 2022, the review should also consider the overall statistical design and methods developed through stages of social survey transformation and their impact on a future Administrative data-first Population Statistics system.
From 2022, the review should also occasionally consider some deep dive quality reviews into specific methodological areas outside of the business-as-usual assurance practices of the panel. Where required, these assurance exercises should be supported by specially selected expert sub-panels raised internally by ONS which will report to the main MARP.
The focus of the reviews undertaken each year will change dependent on the Programme phase. The work programme will be agreed between the Chair and the ONS SRO, with the ONS SRO having the final say.
The Panel will be invited to give feedback on recommended methods, including requests for further work or information, and ONS will accommodate these wherever possible. ONS will then decide on the methods to implement given the feedback received.
Not In Scope
It will be out of scope to review:
- Detailed operational issues related to the Census and the Population and Migration statistics Transformation.
- Work undertaken within the stakeholder engagement project of the programme other than research on public acceptability
- Work undertaken on the Census programmes in Scotland and Northern Ireland but review findings will be shared with colleagues in NRS and NISRA
Organisation / Transparency / Confidentiality
MaRAG – The review panel will be closely aligned to ONS’ Methods and Research Assurance Group which will be the primary source of papers summarising proposed Census, Population and Migration Statistics Transformation and Social Survey Transformation methods. A significant part of this programme will also link into MaRAG and the panel.
Each review will be organised internally by the ONS Methodology and Quality directorate. This will include:
- Organising an ‘induction’ briefing session for panel members to introduce the programme and provide an overview of the breadth of work the review will cover
- Setting the agenda for each review meeting
- Ensuring appropriate papers are provided by areas responsible for each item on the review agenda
- Gaining sign off of papers at MaRAG in advance of each review
- Sending papers to the review panel ahead of the review
- Taking minutes of the panel discussions
- Briefing Census leadership on issues raised
- Monitoring progress on agreed actions from the recommendations made by the review panel
- All communications regarding the panel, the discussions and methods reviewed, by all involved, must be coordinated with ONS
The review panel will compile a report on an annual basis to reflect their findings, which will then be published. The report will include the action taken throughout the year to improve the census methodology.
Following each meeting, the panel minutes will be published alongside the papers considered by the panel.
The panel will be bound by the Census Confidentiality Undertaking, which they will sign as part of their appointment.