Casework

Stephanie Howarth to Ed Humpherson: Request to suspend accreditation

Dear Ed,

I wrote to you in October to request a temporary suspension of the accredited official statistics status for a number of Welsh Government’s statistical outputs derived from ONS’s Annual Population Survey (APS).

This was in response to the suspension of the accredited official statistics status for labour market statistics from the Labour Force Survey (LFS) in February 2024. Although the larger achieved sample on the APS for Wales means that estimates are not affected to the same extent by the quality issues experienced by the LFS, the APS sample size has fallen and the statistical uncertainty for estimates has increased. In addition, the APS has not been reweighted with more up to date population estimates and therefore it does not take account of the significant changes to the size and structure of the population as suggested by the 2021 Census.

We appreciate that ONS has a recovery plan in place for both the LFS and APS and welcome the improvements that this should bring in the future. Given the recovery plan is still in progress and the APS has not yet been reweighted, I am writing to ask you to consider a temporary suspension of the accredited official statistics status for additional sets of statistics for Wales produced from the APS. This would affect the following set of statistics:

Levels of highest qualification held by working age adults

This approach is in line with the agreement in place for other outputs based on the APS. A temporary suspension of the accreditation in Wales will therefore ensure a clear and coherent approach for users of all APS estimates until such time as the accreditation can be reinstated.

We will continue to ensure we are transparent with users on how reliable the APS estimates are. We will continue to provide appropriate advice and caveats in our publications for users of APS data and will apply suppression where appropriate.

Yours sincerely,

Stephanie Howarth

Prif Ystadegydd, Llywodraeth Cymru

Chief Statistician, Welsh Government

Ed Humpherson to Stephanie Howarth: request to suspend accreditation

Dear Stephanie

Suspension of the accredited official statistics status for the estimates for Wales produced from the Annual Population Survey

Thank you for your letter setting out your most recent decision following the suspension of the accredited official statistics status for labour market statistics from the Labour Force Survey (LFS), on which the Annual Population Survey (APS) estimates are based.

We agree to your request to temporarily remove the accredited official statistics status from the following APS-based outputs:

Levels of highest qualification held by working age adults

As you explain in your letter, this approach is in line with the agreement in place for other outputs based on the APS and will ensure a clear and coherent approach for users of all APS estimates.

We welcome your plans to ensure you are transparent with users on how reliable the APS estimates are and that you will continue to provide appropriate advice, suppress disclosive data when necessary and highlight caveats in your publications for users of APS data.

Our Labour Market and Welfare domain team will continue to engage with your team on this matter.

Yours sincerely

Ed Humpherson
Director General for OSR

Peter Schofield to Rob Kent-Smith: Increase in the number of people on Universal Credit

Dear Rob,

Thank you for your letter of 31 March highlighting the complaints you have received relating to our press release on 13 March entitled ‘Almost two million people on Universal Credit not supported to look for work’.

I accept that the original statement, and the update we made after you alerted us to the concern, was misleading by not sufficiently recognising the impact of the process of migrating people from legacy benefits on the increase in the number of people claiming disability elements of Universal Credit.

As you requested in your letter, we have acted quickly to address these concerns through removing all references to the 383% figure, and associated lines on the increase. We have published an updated statement on 1st April, with a note to advise that this Press Notice has been revised, clarifying the figures related to increased UC LCWRA caseload.

Going forward, we are ensuring that where we use official statistics in our communications that the lead statisticians and analysts are involved at all stages of the process, and that there is appropriate oversight from our Head of Profession for Statistics. We are reviewing the process to make sure that all colleagues involved are aware of this and follow the agreed approach.

Yours sincerely

Peter Schofield

Jen Woolford to Ed Humpherson: Review of fraud and computer misuse statistics for England and Wales

Dear Ed,

Thank you for your letter dated 3 April regarding your review of fraud and computer misuse statistics for England and Wales. I would like to thank your team for their positive engagement throughout the process.

The ONS Centre for Crime and Justice is committed to improving insights into fraud and computer misuse statistics for England and Wales. The review has outlined concerns about some aspects of the quality of the Action Fraud data and we welcome the accompanying recommendations for improving the quality of these statistics. This includes several recommendations for ONS, including expanding our published information on quality; working with the Home Office and the City of London Police to understand the impact of the new fraud recording system on the quality of the data and statistics; and clarifying the definition of ‘police recorded fraud’.

We will work closely with data providers to develop a joint action plan for addressing these recommendations. We will aim to publish this by the end of 2025.

We look forward to continuing to work with you and your team as we draft this action plan and implement these improvements.

Yours sincerely,

Jen Woolford

Director of Public Policy Analysis, Office for National Statistics

Rob Kent-Smith to Peter Schofield: Increase in the number of people on Universal Credit

Dear Sir Peter

We have received a number of complaints relating to a press release published by the Department for Work and Pensions (DWP) on 13 March entitled ‘Almost two million people on Universal Credit not supported to look for work’.

The original version of the press release stated that: “The number [of people on Universal Credit health with no requirement to look for work] has almost quadrupled since the start of the pandemic when 360,000 people were considered too sick to look for work – a 383% rise in less than five years.”

The statement that the number of people claiming disability elements of Universal Credit has increased by 383% presents an entirely misleading picture to the public. The figure does not recognise that the majority of this increase is due to the process of migrating people from legacy benefits, such as Employment and Support Allowance, to Universal Credit over the last few years. When these people are accounted for, the actual increase in the number of people claiming disability elements of Universal Credit is 50%.

After we raised concerns with DWP, the press release was amended on 27^th March. The updated version of the press release includes some references to people moving from other benefits and acknowledges that the number of people with no requirement to look for work across Universal Credit health and other benefits since the pandemic has increased by 50%.

However, we consider that these additions do not go far enough. The press release still places considerable emphasis on the misleading 383% figure and does not provide the necessary context in a way that members of the general public would understand. We also note that the press release does not state that it is been updated which we would encourage as a matter of transparency.

It is vital that statisticians are included in the drafting process for communications using official statistics, including press releases, to ensure that such an issue is not repeated in the future. As the Head of Profession for Statistics, Steve Ellerd-Elliott (copied), should be supported by the department in upholding his responsibility to ensure statistics are used appropriately.

We request that the press release is updated by Friday 4^th April to remove reference to the 383% figure and that this figure is not used going forward.

Rob Kent-Smith

Deputy Head of the Office for Statistics Regulation

Ed Humpherson to Abbie Self: Review of the quality of criminal court statistics for England and Wales

Dear Abbie

We have now completed our review of the quality of the Ministry of Justice’s (MoJ) criminal court statistics for England and Wales. I am grateful for the positive contribution and engagement from your team in MoJ and your colleagues in HM Courts and Tribunals Service (HMCTS) throughout the review.

We undertook our review because of concerns about inaccuracies in the Crown Court data. Having examined the statistics in depth, we are confident in the quality of the criminal court statistics and are satisfied that you have provided sufficient assurances to users about the quality of the Crown Court data. We judge that the statistics can retain their accredited official statistics status.

We commend MoJ and HMCTS on the collaborative approach to investigating and addressing the Crown Court data quality issues. There is now a better shared understanding of the data and case management systems and stronger working relationships between MoJ and HMCTS. It is also welcome that you have aligned the methodologies used to produce the official statistics and HMCTS’s management information, which has enhanced the coherence of Crown Court data.

We identified two recommendations that you should address to enhance the quality of the statistics. These are expanding and routinely reviewing the quality information published alongside the statistics; and adding caveats and explanations of uncertainty to the statistical bulletin. You have committed to implementing these in the June 2025 statistics release. We will continue to engage with you and your team as you make improvements.

I am copying this letter to Antonia Romeo, Permanent Secretary; Ross Gribbin, Director General for Policy, Analysis and Legal Group; Hugh Stickland, Chief Data Officer and Chief Scientific Adviser; and Damon Wingfield, Deputy Director for Courts and People Statistics and Analysis.

Yours sincerely

Ed Humpherson
Director General for Regulation

Review of the quality of criminal court statistics for England and Wales

Ed Humpherson to Mike Keoghan: Alternative Release Times for Economic Statistics

Dear Mike,

Thank you for sharing with us your assessment of your practice of releasing market-sensitive economic data outside of market trading hours at 0700, rather than the default release time for official statistics of 0930.

Under our Alternative Release Times (ART) policy it falls to the producer, in this case ONS, to periodically consider the need and public good case to continue with alternative release times. We requested that ONS carry out this review as our call for evidence as part of our Code Review attracted nine responses, who made the case for return to an 09:30 publication time. Given release times was not the primary focus on the consultation, the evidence was partial. Therefore, we asked ONS to assess the issue in a way that takes account of the full range of impacts, including public benefits. We also asked that this assessment be based on effective engagement with stakeholders.

Your assessment recommends continuing to release market-sensitive data at 0700. This is on the basis that it increases the visibility of key statistics for the public via the media, and so better serves the public good. Additionally, this mitigates the risk that ONS’s website is not robust enough to manage the increased traffic that reverting to 0930 would likely lead to, which could risk orderly access to data for users. You note that ONS intends to reconsider this issue when it is assured that its new website and API are robust enough to manage the increased traffic and this has been demonstrated for a year. ONS is also committed to ensuring that this future review is not restricted to a binary choice of 0700 or 0930 but should look at all options, including times between these options.

We have considered our response to your assessment from the following perspectives:

the importance of maintaining orderly release given the risks around the performance of the ONS website;
the role of enhanced media coverage in making statistics more available and accessible, in turn serving the public good;
the differing stakeholder views, and
the impact on the functioning of the SONIA market.

We support your plans to continue releasing these market sensitive statistics at the alternative time of 0700 instead of 0930 for the moment. We recognise the value of increased visibility of key statistics and are keen to maintain the orderly release of these vital statistics. We note your approach to maintain these alternative release times until ONS is assured that its new website and API are robust enough to manage the increased traffic that moving from 0700 could lead to. At that point ONS should reconsider its approach, including whether there is an alternative option, which better meets the needs to users.

You may be aware that we periodically review ARTs to make sure they are still justified and warranted. To help us consider this one, we would welcome an update from ONS on its future review of an alternative release time for market sensitive economic statistics by 31 December 2025. If you think this date will cause you any problems, please let me know.

Please ensure the time of release, the reasons for using an alternative release time, and your plans for reviewing this, are clearly and widely communicated with users and other stakeholders with an interest in these statistics.

Yours sincerely,

Ed Humpherson
Director General for OSR

Alternative release times requested under the Code of Practice for Statistics – Office for Statistics Regulation

Ed Humpherson to Alastair McAlpine: Scottish Health and Wellbeing Census

Dear Alastair,

Thank you for your response to our letter regarding the Health and Wellbeing Census in Scotland. We welcome the publication of the Health and Wellbeing Census Scotland: Background Report. We consider that this document will provide valuable information to users and meets the requirement for Scottish Government to review its approach to developing questions for the Health and Wellbeing Census as set out in our July 2022 letter.

We appreciate your assurances relating to your role as Chief Statistician for Scotland and Scottish Ministers. We look forward to seeing how the recommendations on decision making with respect to cross-cutting statistical projects, as outlined in your internal review on Scotland’s Census 2022, are taken forward.

Yours sincerely

Ed Humpherson
Director General for OSR

Ed Humpherson to Alastair McAlpine: Scottish Health and Wellbeing Census – Office for Statistics Regulation

Ed Humpherson to Alastair McAlpine: Health and Wellbeing Census in Scotland – Office for Statistics Regulation

Siobhan Tuohy-Smith to Alex Lambert & Liz McKeown: ONS’s transformation of the Labour Force Survey (TLFS)

Dear Alex, Liz

Since our letter in July 2023, we have continued to review ONS’s transformation of the Labour Force Survey (LFS) against the Code of Practice for Statistics, alongside our review of LFS-derived Labour Market Statistics following the accreditation suspension last year. We have today published a progress report which consolidates our work on both the LFS and TLFS, bringing together our judgements to date and providing updates on the remaining open recommendations and requirements.

On the TLFS, we have found that ONS is engaging well with users, and that the scope of this engagement has been widening. ONS has also been increasingly open with its communication. However, users remain frustrated about the lack of clarity about when the transition to the TLFS will be made. We have found that our previous recommendations about the TLFS remain relevant. We have also made two additional recommendations calling on ONS to publish more detail about the plans to transition to the TLFS, and to set out its plans for regular reporting on the progress of its interim action plan from its ‘lessons learnt’ exercise. For the LFS-derived estimates, we have found that ONS has acted to address most of our requirements. We recognise that given the five-wave design of the LFS any changes take time to feed through and that ONS expects changes to be fully included through all five of the survey waves in the May 2025 publication. We will engage with your team and users to understand the extent to which these changes have increased quality sufficiently to meet user needs. In this context, the classification official statistics in development continues to be appropriate.

We also support ONS’s plans to publish a series of articles by June 2025 exploring the coherence between estimates from the LFS, estimates from business surveys and estimates from administrative data as this will support user need during this period of transition.

I would like to thank your teams for their continued positive engagement with us on these reviews and we will continue to challenge and support the teams, reviewing progress against each of the deadlines outlined in today’s report.

Please do not hesitate to get in touch if you would like to discuss any aspects of this letter and accompanying progress report.

I am copying this letter to Rachel Skentelbery, Deputy Head of Profession for Statistics at ONS.

Yours sincerely

Siobhan Tuohy-Smith

Assessment Programme Lead

Siobhan Tuohy-Smith: OSR review of salmon statistics

To:

Stuart Homann, Environment Agency

Simon Toms, Environment Agency

Steven King, Natural Resources Wales

Ben Wilson, Natural Resources Wales

David Maxwell, Centre for Environment, Fisheries and Aquaculture Science

Alan Walker, Centre for Environment, Fisheries and Aquaculture Science

Dear all,

We have recently been in touch with you and your colleagues involved in producing the Assessment of Salmon Stocks and Fisheries in England and Wales statistics. This publication is jointly produced by the Environment Agency (EA), Natural Resources Wales (NRW) and the Centre for Environment, Fisheries and Aquaculture Science (Cefas). Given the proposed changes to methods and reporting for these statistics, we have agreed to carry out a short review to offer advice, support and recommendations to help you manage these changes in line with the Code of Practice for Statistics.

We expect the review to broadly cover:

Changes to methods: How users and stakeholders have been informed of the proposed changes to methods; assured that the proposed methods are sound and provide the best available estimates; told about any impact on the statistics; and given the opportunity to feed in to decision making.
Changes to reporting: How users and stakeholders have been considered in the proposed changes to reporting; informed about the changes and how their needs will continue to be met; and how outputs will be accessible and continue to provide clarity and insight that serve the public good.
Institutional governance and statistical leadership: The review will consider aspects of decision-making, transparent processes and the orderly release of statistics. We will focus on salmon statistics, but any recommendations are likely to apply at an organisation level.

If we find that advice or recommendations on areas outside this scope would be helpful to ensure the statistics serve the public good, we will include these in the review.

I look forward to working with you and your teams.

I am copying this letter to Ian Lonsdale, Head of Profession for Statistics, Defra.

Yours sincerely

Siobhan Tuohy-Smith

Assessment Programme Lead