Executive Summary
Drawing on our work as the UK’s statistics regulator, this report sets out our annual view of how the UK statistical system is operating. It brings together evidence from across our regulatory work over the past year to identify the main system-level issues, risks and priorities, while also recognising areas of progress and improvement.
This year’s report is written in the context of continued pressure on the statistical system. Over the past year, the system has faced leadership transition, with interim arrangements in place for the National Statistician and Chair of the UK Statistics Authority, and no National Statistician currently in post. It has also faced ongoing challenges to data quality, difficulties with survey response rates and increasing pressure on resources and capacity. These pressures affect multiple parts of the system and are closely connected.
The report highlights areas of progress and improvement. Producers have improved their statistics and adapted to new demands. It also shows examples of producers being open about quality concerns and delivery pressures affecting their statistics.
The Permanent Secretary and senior team at the Office for National Statistics (ONS) have led recovery during this period, with greater openness about the challenges facing the organisation and the steps being taken to address them. We have also seen stronger governance and more proactive engagement with users. Established Government Statistical Service (GSS) structures and the wider statistical community have supported co-ordination and collaboration during a period of leadership transition. These developments provide a stronger basis for addressing the challenges facing ONS and the wider system.
A central theme running through the report is that the system is increasingly having to make difficult choices. Leaders are operating in a context of transition and constraint, and prioritisation decisions are becoming more visible. This is a necessary response to pressure. It underlines that prioritisation is not simply an operational response to resource constraint, but a core leadership responsibility.
We also find that the quality and resilience of statistics remain under pressure, although there has been progress in some areas. ONS has made progress in improving some economic statistics and in stabilising parts of its survey portfolio. However, important challenges remain, including continuing concerns about survey response and representativeness, quality concerns in local authority population estimates, and the effect that quality problems in key sources can have across a wider set of statistics. The increase in suspensions and cancellations of accredited official statistics over the past two years shows that OSR, as the system regulator, and statistics producers themselves, take the Code of Practice for Statistics seriously and use accreditation to give signals to users where statistics do not currently meet the standards of the Code. At the same time, new accreditations and confirmations have continued, showing that accreditation remains an active part of the regulatory framework.
The communication and use of statistics has remained a major focus of our regulatory work. We have seen some improvement in the way producers explain uncertainty, limitations and comparability, and we have continued to intervene publicly where statistics have been misused or presented without sufficient context. At the same time, public involvement and engagement is now a clearer part of the Code’s expectations for official statistics, but practice is still developing across the system. Progress following the inaugural Statistics Assembly has been slower than users might reasonably have expected, and users need a clearer public plan for taking forward the Assembly priorities.
The report also shows that the system continues to adapt. Activity is increasing in areas such as artificial intelligence and implementation of the revised Code of Practice for Statistics. Innovation is also taking place across the system, including through changes to methods, data access and dissemination. However, progress is not consistent across the system. AI use in official statistics remains limited, and innovation is often shaped by capacity and prioritisation. Implementation of the revised Code is not yet fully in place. The revised Code, including the new Standards for the Public Use of Statistics, Data and Wider Analysis, provides a stronger framework for both the production and the public use of statistics. The task now is to ensure that these expectations are applied clearly and consistently in practice.
Our regulatory work shows that many of the challenges facing the statistical system are interconnected. Supporting confidence in official statistics depends on the quality of individual outputs, but also on how the system responds to pressure and change.
Our recommendations focus on the areas where we consider clearer system leadership is needed. Most are directed to the National Statistician once appointed, because the role provides leadership across the GSS and the wider statistical system. There is currently no National Statistician in post and at the time of writing the role has been vacant for six months. Some of the actions to implement these recommendations can only be taken forward once the substantive appointment has been made.
Our recommendations
The National Statistician should publish a baseline system-wide view of resources and capacity across the statistical system, including workforce numbers, key skills gaps and major capacity constraints. This should be updated regularly to support greater transparency about current pressures and how priorities are being managed (target within 6 months of being in post).
The National Statistician should ensure that public reporting on key household and business surveys gives users a clear explanation of survey quality, so they can understand what the estimates can support and where the evidence is not strong enough for particular uses. This should include achieved sample sizes, response rates and representativeness, and explain where survey quality limits the use of estimates for particular geographic areas, population groups or types of analysis (target within 6 months of being in post).
The UK Statistics Authority, working with the National Statistician once appointed and statistics producers across the system, should publish a clear, time-bound plan for taking forward the four Statistics Assembly priority areas. This should show what action will be taken on the four high-level priorities and the main commitments arising from the Assembly report, who is responsible for them and how progress will be reported publicly (target by end March 2027).
The National Statistician’s Office, working with the GSS, should set out a clear approach for how existing guidance and standards apply to the use of AI in official statistics. This should help producers apply the Code of Practice consistently and explain how AI tools and models are being used (target within 6 months of being in post).
The National Statistician should set out how the GSS will support comprehensive application of Code 3.0. This should include clear expectations for ongoing self-evaluation by producers and for building a system-wide view of how the revised Code is being applied in practice (target within 6 months of being in post).