Abstract

This paper explores how the UK Code of Practice for Statistics (the Code) provides an ethical and practical framework for building confidence in statistics and combating misinformation through the principles of Trustworthiness, Quality and Value (TQV). Drawing on the 2025 refresh of the Code and the Office for Statistics Regulation’s regulatory practice, including interventions on misleading uses of statistics and public engagement work, this paper demonstrates how “thinking TQV” supports better production, communication and use of statistics. It argues that TQV offers a transferable model for strengthening democratic discourse in increasingly complex information environments. supports better production, communication and use of statistics. It argues that TQV offers a transferable model for strengthening democratic discourse in increasingly complex information environments.

Keywords: public good, statistics, trustworthiness, quality, value

Introduction

Across modern societies, statistics play a central role in shaping public understanding. They inform political debate, underpin policy decisions, and influence how individuals interpret the world around them. Yet this influence comes with a growing challenge: statistics now circulate within a crowded and fast-moving information ecosystem where messages are distilled, repackaged and shared at speed. In such an environment, even well-produced statistics can be misunderstood, selectively presented, or deliberately misused. Statistics should not simply be accurate but must also be communicated and used in ways that build public confidence

The Office for Statistics Regulation (OSR) addresses this challenge through its vision (OSR, 2019) that statistics should serve the public good, acting as public assets that inform understanding and shape action (OSR, 2024). This framing is important because it explicitly connects statistical practice with democratic values: trust, accountability and informed debate.

OSR as the regulatory arm of the UK Statistics Authority has the statutory objective of “promoting and safeguarding the production and publication of official statistics that serve the public good” (SRSA, 2007). The UK’s Code of Practice for Statistics (Edition 3.0) (the Code) (OSR, 2025a) is central to this responsibility. The Code is not a rule book or narrow technical manual; rather, it is an ethical and practical framework that recognises the full lifecycle of statistics, from production through to communication and use.

This paper argues that the Code’s framework of Trustworthiness, Quality and Value (TQV) provides a powerful and adaptable model for addressing and anticipating misinformation. By examining both the TQV principles themselves and their application in practice, the paper shows how “thinking TQV” supports a more resilient and trustworthy statistical ecosystem.

TQV as an ethical framework

Overview of TQV

The Code is built on a simple insight: confidence in statistics depends on more than methodological rigour. It arises from the interaction of three interdependent elements:

• the behaviours and culture of those producing statistics
• the robustness of the data and methods
• the way statistics are communicated and used

These elements are captured in the three core principles of Trustworthiness, Quality and Value, outlined below. The TQV framework is holistic in recognising that process, context and communication are integral to the meaning and use of statistics. As the Code emphasises, statistics are only effective when both the numbers and the ways they are handled serve the public good (OSR, 2025b).

Trustworthiness: confidence in people and institutions

Trustworthiness addresses a fundamental question: why should users trust the statistics they see?

The answer lies not only in formal safeguards but in the behaviours and culture of the producer organisations. The Code deliberately emphasises trustworthiness rather than ‘trust’, following the work of Baroness Onora O’Neill who said that “if we want others to trust us, the first step is to be trustworthy” (O’Neill, 2013). Trustworthiness within the Code requires integrity, transparency and independence, ensuring that statistics are produced free from undue influence and presented impartially.

In practice, trustworthiness is often most visible when it is under pressure. Decisions about timing of releases, handling of revisions or responses to criticism all signal whether an organisation is genuinely committed to openness. Importantly, the Code extends this concept beyond statisticians themselves. Trustworthiness applies equally to those who use statistics in the public domain, recognising that confidence can be undermined if figures are presented selectively or without context. The incorporation within the Code of OSR’s concept of ‘intelligent transparency’, which means proactively taking an open, clear and accessible approach to the use of data, statistics and wider analysis in the public domain, has strengthened this expectation on public bodies in the UK (OSR, 2025c).

Quality: beyond accuracy to transparency

Quality is often understood as technical accuracy. The Code broadens this perspective, defining quality in terms of fitness for purpose, transparency and continuous improvement.

This reflects an important reality: all data have limitations. Administrative data may be incomplete and survey data are subject to sampling error, while new types of data sources and methodologies are evolving rapidly. The risk arises not solely from these limitations, but from failing to communicate them clearly.

By emphasising openness about uncertainty and limitations, the Code focuses on the importance of communication. Producers are expected not only to ensure robustness, but also to explain where caution is needed. This approach plays a critical role in combating misinformation. When limitations are clearly stated, it becomes harder for statistics to be misinterpreted or overstated.

Value: enabling understanding for public use

While Trustworthiness and Quality establish credibility, Value ensures relevance and impact by informing and supporting decision making, action and debate. Statistics that are technically sound but poorly communicated or misaligned with user needs will fail to inform decisions.

The Code frames value as an ongoing relationship between producers and users, requiring engagement, clarity and accessibility. Producers are encouraged to anticipate how statistics might be interpreted, to present them clearly, and to support appropriate use.

This proactive approach is particularly important in a misinformation context. Misinterpretation often arises not from bad intent, but from gaps in understanding. By designing outputs with users in mind, producers can reduce these risks.

TQV and the public good in the context of misinformation

The Code acknowledges the challenge of the changing information environment where data are abundant but variable in quality and communication is rapid and often simplified. Misinformation can spread quickly and widely, and corrections or clarifications do not keep pace. Within this context, trust is fragile. Users must constantly assess which sources are credible and which claims are reliable.

TQV provides a structured way for producers to respond to this threat by promoting an environment and culture that supports robust production and use of statistics, strengthened by a central focus on serving the public good. The Code designs the three core principles to be “resilient safeguards” against threats to the public good (OSR, 2025b). This resilience is key: it shifts the focus from reactive interventions to system-wide confidence building.

Extending the Code: Standards for the Public Use of Statistics

A significant development in Code 3.0 is the introduction of the Standards for the Public Use of Statistics, Data and Wider Analysis (OSR, 2025c). These standards reflect a growing recognition that the point of greatest risk for misinformation is often communication, not production. Statistics may be robust, but when summarised in speeches, infographics or social media, their meaning can change.

The standards therefore set expectations for all public bodies, emphasising:

• Equality of access, ensuring figures are available to everyone
• Supporting understanding, through clear and accurate communication
• Decision-making and leadership, grounded in expert advice

At their core is the concept of intelligent transparency: a proactive commitment to openness that anticipates how information will be used and ensures it is accessible and understandable from the outset (OSR, 2025c). This represents a shift from passive transparency (making data available) to active transparency (making data usable and interpretable).

Applying TQV in Practice: OSR Case Studies

Challenging misleading use: the inflation infographic

A clear illustration of actively applying the TQV principles is OSR’s intervention concerning a UK Treasury inflation infographic (Humpherson, 2023).

The social media post included a graph intended to communicate progress on inflation but did so by giving a misleading impression of the scale of the deceleration in inflation. The bar chart was presented with the y-axis beginning at a selected data point rather than zero. OSR identified risks that the presentation could be misleading without sufficient context, particularly in how comparisons were framed in relation to the Government intention to halve the rate of inflation.

In its correspondence, OSR did not dispute the underlying data. Instead, the concern centred on how the statistics were selected, framed and communicated. This distinction is crucial: misinformation does not always arise from incorrect data, but from incomplete or selective presentation.

The intervention highlighted several key principles:

• Value (clarity and context): Users need sufficient explanation to interpret statistics correctly. Without this, even accurate figures can lead to misunderstanding.
• Trustworthiness (integrity): Public bodies must avoid presenting statistics in ways that could give a partial or overly positive impression.
• Supporting understanding (public use standards): Communication should anticipate how audiences might interpret the message, not just what is technically true.

The case reinforced the expectation that statistical communication must be as robust as statistical production. It also demonstrated OSR’s role as an active regulator, engaging publicly and constructively to improve practice. It also shows how TQV operates in real time: identifying risks, clarifying expectations and strengthening trust through transparent challenge. As a result of the intervention, the department committed to ensuring analysts are consulted in the preparation of social media posts using statistics.

Misreporting and reinterpretation: Universal Credit statistics

An example of where OSR has intervened to uphold the use of statistics in public debate involved a government statement that exaggerated the impact of a welfare benefit scheme. A press release had suggested that the number of people receiving a benefit for people out of work or on low income (‘Universal Credit’) had increased in a way that implied a worsening labour market or increased dependency following the COVID-19 pandemic (Kent-Smith, 2025). OSR identified that these claims did not adequately reflect the underlying drivers of the change.

In its intervention, OSR highlighted that increases in Universal Credit caseloads can arise from multiple structural and policy-related reasons and not solely from changes in employment conditions. The figures were being used without sufficient context and so were potentially misleading. OSR also found that interpreting headline increases without explaining the nature of the composition of the claimant population, such as employment status of claimants, risked oversimplification and misunderstanding.

This case demonstrates how misinterpretation can emerge from complexity. Universal Credit is a multifaceted system, and the statistics reflect a range of factors including eligibility changes, policy reforms and economic conditions. From a TQV perspective, Quality was not the issue since the statistics themselves were robust. Trustworthiness though came into question in how the figures were being presented publicly, while the central concern was Value as the statistics were not being communicated in a way that enabled proper understanding.

OSR’s intervention therefore aimed to re-anchor the discussion in the evidence by drawing attention to the broader context. It helped ensure that both policymakers and the public could better understand what the statistics meant in context.

Misinformation does not always involve incorrect numbers but often involves incomplete stories. TQV provides a framework for identifying and correcting these partial narratives. By insisting on context, clarity and integrity, the Code helps shift the focus from simplistic interpretations to more evidence-based understanding.

Supporting understanding: election explainers

While the inflation case illustrates responsive regulation, OSR’s election explainers represent a proactive application of TQV (OSR, 2026).

Ahead of elections, public debate intensifies and statistical claims become more frequent and more contested. In this context, the risk is not only deliberate misuse but also genuine confusion among audiences trying to interpret competing claims.

The election explainers were designed to address this challenge by providing accessible explanations of commonly cited statistics. The explainers clarify how measures are constructed, what they do (and do not) show and they also highlight limitations and areas of uncertainty. This initiative embodied the Value principle in a particularly tangible way. Rather than waiting for misuse to occur, OSR anticipated areas where misunderstanding could occur and provided tools to support interpretation.

Importantly, the explainers are written for a broad audience, not just experts. This reflects the recognition of varying levels of statistical literacy within the public, and that supporting informed debate requires clearly communicating statistics to as broad an audience as possible. The explainers also reinforce Trustworthiness, as they demonstrate independence and impartiality.

By focusing on explanation rather than advocacy, OSR positions itself as a trusted intermediary in public debate. This case reflects a development in OSR’s regulatory practice in supporting users, enabling them to engage critically with statistics.

Regulatory casework as an ongoing narrative

Beyond individual cases, OSR’s casework function represents an ongoing process through which TQV is applied across the statistical system. Through investigation, engagement and public correspondence, OSR creates an accountability mechanism (OSR, Interventions Policy). Through the challenges made through our casework function and establishing the new Standards for Public Use, producers and communicators within government across the UK have become more aware of the need to provide context alongside numbers and to consider how their messages may be interpreted. We always encourage them to engage openly when concerns are raised. In this way, casework reinforces the norms that underpin Trustworthiness, Quality and Value.

Discussion: TQV as a transferable framework

Although developed within the UK statistical system, TQV has broader relevance. The Code explicitly notes that the framework can be applied by any organisation producing or using data. This universality is one of its strengths. In an environment where data are produced across government, academia, media and private organisations, a common framework helps establish shared expectations of good practice. Internationally, TQV is consistent with the UN Fundamental Principles (UN General Assembly, 2014), and offers a practical and accessible articulation of how ethics can influence everyday statistical practice.

TQV provides a strong foundation for guarding against and standing up to misinformation using official statistics by integrating technical rigour (Quality) with ethical behaviour (Trustworthiness) and user-centred communication (Value). It is an essential regulatory tool and provides a critical steer for any analyst seeking to navigate the challenges of misuse and pre-empting misunderstandings.

Conclusion

The challenge of misinformation cannot be solved through technical fixes alone. It requires a broader commitment to how evidence is produced, communicated and used. The Code of Practice for Statistics, through its TQV framework, provides exactly this: a coherent ethical model that connects statistical quality with trust and public value. The addition of standards for public use further extends this responsibility, recognising that everyone involved in communicating statistics plays a role in shaping understanding.

Through its regulatory work, both reactive and proactive, OSR demonstrates how these principles can be applied in practice, from challenging misleading claims to supporting public understanding. Ultimately, “thinking TQV” represents a shift in mindset. It is an approach that places the public good at the centre of statistical practice and recognises that confidence is built not just through numbers, but through transparency, clarity and integrity in how those numbers are used.