Ed Humpherson to Mary Gregory: OSR response on accredited official statistics status – ONS mid-year population estimates for England and Wales

Published:

Sent from: Office for Statistics Regulation

Sent to: Office for National Statistics

Dear Mary,

OSR response on accredited official statistics status: ONS mid-year population estimates for England and Wales

Thank you for your letter detailing the actions that ONS has taken to address the four requirements in our compliance review of mid-year population estimates for England and Wales (MYEs). Our review raised concerns about the quality of subnational estimates, particularly in local authority areas that are subject to high population churn.

We have reviewed ONS’s actions and conclude that the statistics should continue to be labelled as accredited official statistics. The detail supporting our judgement is included in the annex to this letter.

We recognise the significant amount of work that ONS has done to meet the requirements. This has included improving published quality information, for example by highlighting uncertainty associated with local authority estimates. ONS has also made significant efforts to explain any likely continued effects on the MYEs from data collected during the COVID-19 pandemic. These actions demonstrate important progress and the next step for ONS is to test, as planned, how well these improvements meet user needs in practice.

We have, therefore, identified an additional recommendation which ONS should address to further enhance the statistics: to support understanding, ONS should action its plans to seek user feedback on published guidance and identify improvements needed. ONS should also engage with the Greater London Authority as part of addressing recommendation four in our review which asked ONS to better understand differences between MYEs for school-age children and the Department for Education’s English school census.

We expect ONS to report publicly to us on its progress against these recommendations, and the other outstanding recommendations in our review, by the end of 2026.

I would like to thank your teams for their open and positive engagement throughout our review and for their ongoing commitment to improving these statistics. I am copying this letter to Stephanie Howarth, Chief Statistician, Welsh Government.

Yours sincerely,

Ed Humpherson
Director General for OSR


Annex – Review of ONS actions taken in response to OSR’s compliance review of ONS’s mid-year population estimates for England and Wales

Requirement 1

To support user confidence in its decision to continue with the MYEs, ONS must be open and explicit about the strengths and limitations of the current methods and what action ONS is taking, with time scales, to address methodological limitations with independent assurance.

Actions taken by ONS: ONS has introduced a new Quality and Methods Guide (QMG) to describe the strengths and limitations of the statistics in more detail. The MYE statistical bulletin now describes what can and cannot be done with the estimates and highlights cautionary notes. ONS has published a mid-year estimates improvement plan alongside the MYEs to outline how it intends to address limitations and improve the quality and reliability of the MYEs. A published blog draws together the key information and clarifies what this means for users of population estimates. ONS has committed to seeking user feedback on its recent publications through stakeholder engagement in autumn 2026.

OSR’s judgement: Requirement met.

ONS is now clearer about the strengths and limitations of the current MYEs across the suite of published articles, including the statistics bulletin, the Quality and Methods Guide (QMG), the Quality overview of data sources used and the recent blog.

The MYE improvement plan clearly sets out the actions that ONS is taking, including priority areas of focus in line with OSR’s review findings and areas of concern. The plan includes timelines for improvements and commitments to user engagement in autumn 2026. The Methodological Assurance and Research Panel (MARP) has oversight of the plan, providing external and independent assurance. MARP has reviewed high-level plans to improve the MYEs and is assured that the right areas are being focussed on to address challenges with population estimates. The new 6-monthly reporting to the Population Statistics System Committee (PSSC) makes used of established ONS governance arrangements to provide an additional layer of transparency and accountability.

Requirement 2

To ensure that users have confidence in the population baseline underpinning the MYEs, ONS must:

  1. Clearly set out the pandemic’s impacts on the MYEs, prominently linking to its published COVID-19 impacts report, and publish evidence to justify its claim that its estimates since 2021 are unaffected by systemic pandemic effects, particularly in light of acknowledged uncertainties for some areas.
  2. Actively engage with users through its established groups to test the adequacy of current guidance and make necessary improvements, ensuring users are explicitly informed about which data and geographies remain vulnerable to pandemic-related distortions.

Actions taken by ONS: ONS has published a new ‘Effect of the coronavirus pandemic on the quality of mid-year population estimates for England and Wales’ article, which describes the effects of the COVID-19 pandemic on the quality of the MYEs. The signposting on the previously published report on the pandemic has been updated to inform readers of the newly published article.

ONS also published two articles, Local authority quality of population estimates for England and Wales and Improving population estimates for England and Wales. ONS engaged with users of population statistics, namely Ceredigion, Cambridge, Cambridgeshire, the Greater London Authority, the Department for Education and the Royal Statistical Society, before publication. ONS has committed to further engage with a wider range of local authority users in autumn 2026.

A published blog draws together the key information and clarifies what this means for users of population estimates. ONS has committed to seeking user feedback on the published materials which accompany the MYEs to ascertain their understanding of the guidance and identify further improvements in the autumn.

OSR’s judgement: Requirement met.

ONS is now explicit and upfront in its communication that it cannot quantify the exact impact of the pandemic and that uncertainty has increased since 2021, for example in the published blog.

The new ‘Effect of the coronavirus pandemic on the quality of mid-year population estimates for England and Wales’ article clearly explains quality impacts on each of the components of population change and acknowledges greater uncertainty for areas with high student and migration populations. Cross referencing and signposting between the historic and new article are clear, with relevant links also included across the broader suite of the latest MYE publications.

In its MYE improvement plan, ONS has committed to explore ways to help reduce errors and uncertainty and develop new approaches to validate estimates between censuses in the longer-term.

ONS has carried out a significant amount of work to improve how it explains pandemic impacts on MYEs. To ensure that this information is accessible and clearly understood by users, it will be important for ONS to engage with users to gather feedback and identify and respond to any areas for improvement. While we consider that this requirement has been met, we have identified the following additional recommendation which ONS should address to further enhance the statistics.

New Recommendation: To support understanding and guide appropriate use, ONS should provide a public update on the outcomes from its planned user engagement to seek views on published guidance and any resulting improvements by the end of 2026. 

Requirement 3

To help users better understand the strengths and limitations of the MYEs, ONS must:

  1. Improve the prominence and clarity of how it communicates uncertainty in its population estimates, particularly for local authority areas with high levels of migration and/or large student populations. This should include enhancing communication for non-technical users so that they can understand what the estimates should and should not be used for.
  2. Improve the accessibility, structure, and navigation of key methodological and quality and methodology information (QMI) documentation to support a broader range of users.

Actions taken by ONS: ONS published the article Local authority quality of population estimates for England and Wales. ONS has reviewed and updated methods materials which accompany the MYEs and introduced a Quality & Methods Guide (QMG). The MYE statistical bulletin now includes confidence intervals, highlighting the existence and scale of uncertainty around the local authority estimates. ONS has committed to engagement with a broad range of users in autumn on its QMG documentation and other technical guidance to ensure user understanding. A published blog draws together the key information and clarifies what this means for users of population estimates, in addition to the new section in the bulletin about what you can and can’t use the statistics for.

OSR’s judgement: Requirement met.

ONS has improved the prominence and explanation of statistical uncertainty associated with the estimates throughout the statistical bulletin, QMG and a new article focusing on uncertainty associated with local authority characteristics. Published research-based measures of statistical uncertainty are now available and linked prominently in the statistical bulletin. The introduction of a statistical uncertainty measure for mid-2025 estimates, the quality indicators and addition of bias-adjusted confidence intervals (including within visual charts and the interactive tool) helps users to see, upfront, the degree of statistical uncertainty associated with the estimates, particularly at the local authority level.

ONS has improved the navigation, structure and accessibility of its quality and methodology information, taking into account how a broader range of users access and use the statistics and commit to further testing of the documentation and guidance in autumn.

More work is required to explore discrepancies and explain the strengths and limitations of estimates of children in the MYEs and how they should and should not be used. Whilst noted in the article Local authority quality of population estimates in England and Wales as a discrepancy, further analysis and explanation are needed to support relevant use.

Recommendation 4 in OSR’s compliance review highlights the need for ONS to work with the Department for Education and the Welsh Government to reconcile school census and MYE data for school-age children and publish clear explanations where differences remain. We understand that the Greater London Authority (GLA) has used its own adjusted estimates for children in its 2024-based Housing-led population projections published 21st August due to concerns with the accuracy of ONS’s official estimates – further evidence that this issue remains unresolved. As part of addressing Recommendation 4, and to support its own plans to explore additional data sources to improve coverage of children, ONS should engage with the GLA to help make future improvements to these statistics.

Requirement 4

To avoid the risk of users picking up the wrong number for the purpose that they need, ONS must update and promote its guidance for users on which population estimates should be used for which purposes, setting out the strengths and limitations of each.

Actions taken by ONS: ONS has updated the population statistics and sources guide, which explains the uses of the different population estimates and has included links to the guidance within the bulletin and supporting articles

OSR’s judgement: Requirement met.

ONS has improved and expanded its population statistics and sources guide, which now also includes additional population estimates, such as population projections. ONS has taken steps to promote the guidance through links to other relevant publications and communication. ONS should now test the guidance with users as part of its planned engagement in autumn, to ensure that users can now successfully identify the most appropriate population estimate for their needs.


Mary Gregory to Ed Humpherson: Mid-year population estimates for England and Wales (July 2026)

Mid-year population estimates for England and Wales (May 2026)

Ed Humpherson to Mary Gregory: Accreditation of the 2021 Census in England and Wales outputs and Review of mid-2024 population estimates for England and Wales (May 2026)