Communicating statistics

This section shares our current views on embedding intelligent transparency, our flagship public campaign, as the norm, and provides examples of good communications, focusing on the communication of core statistical outputs and uncertainty.

Embedding intelligent transparency

Statistics and data should serve the public good. They should allow individuals to reach informed decisions, answer important questions and provide a mechanism for holding government to account. Statistics and data also underpin the successful implementation of government policies, and individuals’ views on the effectiveness of policy decisions.

In last year’s report, we outlined how we wanted the government to bolster public confidence in its use of analytical evidence in the public domain by adopting an open, clear and accessible approach to the release and use of data and statistics. Our overarching ambition is that when data and statistics are used publicly to inform parliaments, the media and the public, they are made available to everyone and published with appropriate explanation of the context and sources.

Transparency and clarity support public confidence in statistics and the organisations that produce them and minimise the risk of the misinterpretation of statistics and data. To support this aim, since 2022, OSR has promoted the principles of intelligent transparency – equality of access, enhancing understanding and independent decision making and leadership – through both its regulatory activities and wide-ranging engagement across government.

Over the last year, our work in this area has continued at pace. Ahead of the 2024 UK General Election, the Chair of the UK Statistics Authority, Sir Robert Chote, clearly set out our expectations for intelligent transparency in his letter to political party leaders. In October 2024, we published a blog on embedding the habit of intelligent transparency, looking at the importance of the concept for governments across the UK and reflecting on our work in this area to date. The Scottish Government has been particularly proactive and has established a group of intelligent transparency champions across different departments and agencies to be a point of contact between the Office of the Chief Statistician and analytical areas. Central government departments have also responded well to this initiative; for example, the Department for Education has targeted internal guidance to reinforce intelligent transparency in relation to the increasing publication of management information.

The issues highlighted through our casework function have been key to refining our approach to intelligent transparency, and in 2024/25 we made several interventions to uphold the principles of intelligent transparency with regard to the cases we investigated.

In February 2025, we applied the principle of independent decision making and leadership to our casework regarding the Scottish Health and Wellbeing Census. Our public letter requested assurance on the processes and governance within the Scottish Government regarding the independence of Scotland’s Chief Statistician. The Chief Statistician subsequently wrote to us on 3 March to reassure OSR that the importance of independent decision making around statistics is fully recognised across the Scottish Government.

Separately, in March 2025, we publicly intervened to challenge the misleading use of a figure by the Department for Work and Pensions (DWP) in a press release on Universal Credit. After applying the principle of enhancing understanding, the department corrected the press release, with the DWP also committing to involve statisticians at all stages when communicating official statistics. Further examples from our casework are highlighted in our Annual Review of Casework: 2024/25, which was published in June 2025.

Although there is now a greater understanding of the need for transparency among producers, analysts and the statistical system, more work is needed to fully embed the principles of intelligent transparency consistently across government. Among the projects being undertaken by OSR to support this aim are:

• an ongoing review of intelligent transparency within ministerial departments and devolved governments. We intend to report on how well the principles are embedded in these areas of government, what risks and challenges are being presented and positive case studies. We hope to publish our high-level findings in autumn 2025.
• research on the feasibility of producing intelligent transparency ‘report cards’ for public bodies, building on a recommendation in PACAC’s 2024 report, Transforming the UK’s Evidence Base. This has involved looking into how other organisations measure compliance using score cards.
• ongoing engagement with external stakeholders, producers, departments and other partners to deliver training sessions and seminars on intelligence transparency.

Most importantly, OSR is embedding the principles of intelligent transparency in the Code of Practice for Statistics as part of our 2025 revision of the Code. This work is ongoing at the time of publication; we are currently engaging with a wide range of stakeholders, including senior statisticians and policy professionals, on the proposed additions. Once finalised, the standards will become the basis for our wider work on intelligent transparency, providing a strong tool for implementing and regulating the concept.

More broadly, we recognise that the wider landscape in which statistics and data are used is continually evolving. We reflected on these changes and the implications for OSR in a recent think piece, Reflecting on the role of statistics and OSR in an evolving public sphere, which was developed with the help of several external philosophers. This work will complement the development of intelligent transparency and our research on how official statistics shape personal decisions to improve the knowledge base on the statistical user landscape.

We want the principles of intelligent transparency to become the norm across all of government. Statistics producers have an important role in helping to achieve this aim and will be bolstered by intelligent transparency being a core element of the revised Code of Practice. Beyond the Code, we will continue to raise the profile of intelligent transparency with senior leaders, policy and operational professionals, and communication teams.

By applying the principles of intelligent transparency across government, we consider that public confidence in data, statistics and the public bodies which use them is supported. When the principles of equality of access, enhancing understanding and independent decision making and leadership are upheld, public conversations can focus on the important issues rather than the validity and transparency of the data. Ultimately, we believe that intelligent transparency supports democratic debate, sound decision making and a functioning public sphere.

Communicating core statistical outputs and the associated uncertainty

Good communication is essential for statistics to be used and understood effectively. When good-quality statistics are communicated well, public good is increased, people make better data-based decisions, and confidence in the statistical system is enhanced. Our engagement with producers and organisations continues to reinforce the importance of giving attention to factors such as balance, data visualisation, overemphasis, the use of dashboards and technical language in the communication of uncertainty in statistical publications.

Many producers are demonstrating good practice in their routine publications. For example, the DWP employed the Quality Assurance of Administrative Data (QAAD) framework to inform its users of survey data in the Family Resources Survey, making its findings on the strengths and limitations of the methodology clear in its Quality Assessment Report. Separately, in October 2024, the Scottish Government published its Agricultural Statistics Hub, transforming how agricultural data in Scotland are shared by bringing together datasets in one user-led, accessible platform. Statisticians in the Department for Culture, Media & Sport have told us they have developed their partnership with their communications team, writing an internal statistics communication strategy and making communication a key theme of its work for the coming year.

Over the last year, many statistics producers have proactively engaged with the issue of effectively communicating uncertainty.

In May 2025, we wrote about how effectively communicating uncertainty in GDP can benefit both the public and government decision-makers in a crucial area of government activity. We highlighted that by effectively communicating uncertainty, statistical producers can enhance trustworthiness and credibility.

Over the past year, we have been looking in detail at concerns raised with us on how well the gender identity census question in England and Wales was understood, and the effects of possible misinterpretations on the quality and usability of the gender identity data from the 2021 Census for England and Wales. Our final report on gender identity data in the Census was published in September 2024. We have endorsed ONS’s decision to publish these statistics as official statistics in development rather than accredited official statistics. We also made recommendations for ONS to improve the guidance on the use of the census identity data, including providing more practical information to help users understand the uncertainty inherent in the data and how this affects their use and interpretation.

In March 2025, ONS published a new research publication providing additional guidance on the appropriate use of the gender identity estimates from Census 2021 in England and Wales and information on the uncertainty associated with them. The publication includes example use cases at different levels of geography and population, and addresses likely anomalies and implausibilities. We consider that this additional information will support users to better understand the uncertainty in the gender identity census data and its implications for use, and we wrote to ONS setting out the details of why we consider that our recommendations related to guidance for use of the census gender identity statistic are now actioned.

An important part of our work over the coming year will be giving consideration to what new training should be offered within the GSS, particularly to communications professionals, while also seeking to embed good practice. To help highlight positive examples, we have been developing a series of case studies which aim to support statistics producers around various aspects of managing and communicating uncertainty. These include Effectively Communicating Uncertainty in GDP and Communicating uncertainty for Scotland’s Census 2022.

We are also working with the Durham University academic network to explore how statistics are communicated to different audiences for different purposes – a core consideration of effective communication. This summer, we intend to publish our guidance on producing dashboards in line with the principles of the Code, building on positive liaison with the GSS presentation champions. Our aim is to bring this work together in a new OSR webpage on communicating uncertainty, which will signpost relevant guidance, blogs and case studies.

The role of management information (MI) in the communication of government information has been an important area of OSR’s regulatory work over the last year. In January 2024, we published guidance for statistics producers on the essential characteristics of official statistics and MI and what separates the two. Departments often release operational data (management information) to provide transparency about their services and delivery. The data collected are representative of the systems involved and present important indicators of interest to the organisation. MI can fill a gap by providing statistics on progress towards a specific policy. Statistics published as MI do not need to follow the Code of Practice for Statistics; however, producers are encouraged to voluntarily adopt the Code. With the data needs of the government missions, we expect that MI may become more commonplace as part of the evidence base published by departments. Like with official statistics, we expect MI publications to note limitations and quality concerns to enable their appropriate use, especially given their potential as gauges for government ambitions.

From our interactions with departments who produce MI, we have been impressed with the commitment to ensuring MI publications meet many aspects of the Code, including standards related to the communication of statistics. While MI has a role in aiding transparency, we encourage producers to ensure that the publication of MI is done appropriately. MI should not replace official statistics as the primary source of public statistics published by government.

We want to see the Cabinet Office work with statistics producers to support understanding of the government’s missions, whilst also ensuring that best practice on the communication of quality and uncertainty continues to be shared across government. Statistics producers should aim to use the most effective and transparent ways of communicating statistics, noting the benefits to both producers and users of statistics.

Recommendations