Digital, Culture, Media and Sport Committee

Office for National Statistics written evidence to the DCMS Sub-committee on Online Harms and Disinformation’s inquiry on misinformation and trusted voices

Dear Mr Knight,

I write in response to the Digital, Culture, Media and Sport Sub-committee on Online Harms and Disinformation’s call for evidence for its inquiry, ‘misinformation and trusted voices’.

As the Committee may be aware, the Office for National Statistics (ONS) is the UK’s National Statistical Institute and largest producer of official statistics. We place an enormous value on being a trusted source of information. We work tirelessly to ensure that our engagement with the public is not only trusted, but actively combats misinformation by making our communications clear, and providing statistics with context.

The Office for Statistics Regulation complements our work. They assess official statistics against the Code of Practice for Statistics, assigning them National Statistics status if they meet the requirements. This makes the public aware which statistics can be trusted.

This submission goes into further detail on these points. I hope this is useful, and please do let me know if we can provide further evidence or discuss directly with the Committee.

Yours sincerely,

Sam Beckett

Second Permanent Secretary and Deputy Chief Executive of the UK Statistics Authority

**Office for National Statistics written evidence ‘misinformation and trusted voices’, September 2022**

Summary

It is of the utmost importance that the Office for National Statistics (ONS) is regarded as a trusted voice in the UK. The growing use of data in public debate in recent years has emphasised the need for official statistics and analysis that can be relied upon. As the UK’s National Statistical Institute, our role is to provide trusted and accurate data.
According to the Public Confidence in Office Statistics (PCOS) 2021 report, the ONS had high levels of trust from respondents (89%), which rises if they are a frequent user of our statistics (97%). The public trust us with their data and our statistical outputs, at 90% and 87% respectively. Compared to other institutions in public life, including Government and the media, we have the highest levels of trust.
With this trust comes a large responsibility to ensure firstly that we are meeting the data needs of the public, identifying and responding to any data gaps rapidly, and secondly, that our statistics are communicated well, to reduce the risk of misinterpretation.
On the latter point, we do this by proactively engaging with the public directly and through media to improve the clarity and messaging of analysis, building our own trusted social media presence, and using innovative engagement methods to increase the public’s use of our statistics, highlighted during the census and recently with our Personal Inflation Calculator.
Misinformation can still occasionally occur. We actively monitor media channels and respond rapidly to provide clarity where it is needed, for example during the coronavirus (COVID-19) pandemic regarding the number of COVID-19 deaths. We also have strong relationships with media outlets to ensure corrections happen swiftly.

Trusted data sources and institutions

As the UK’s National Statistical Institute, our outputs are regularly assessed by the Office for Statistics Regulation (OSR) against the Code of Practice for Statistics. If fully compliant, they are accredited as National Statistics, with the quality mark on all associated releases on our website to reassure users that they can be trusted.

Public Confidence in Official Statistics Survey

Since 2004, the UK Statistics Authority has commissioned research from the National Centre for Social Research (NatCen) on levels of trust in, and the awareness and use of official statistics in Britain. The latest results of this research were published in April 2022, in the Public Confidence in Office Statistics (PCOS) 2021 report.
Respondents reported high levels of trust (89%) in the ONS. Trust is high regardless of whether people were previously aware of the ONS or not. However, those who have used official statistics are more likely to trust the ONS than those who have not used them, with 84% of non-users saying they trusted the ONS compared with 97% of frequent users of ONS statistics.
In 2021, 87% of people surveyed said they trusted statistics produced by the ONS. 90% also agreed that they trusted the ONS with data the provided them, and that it would be kept confidential.
PCOS also asked respondents about their level of trust in the ONS compared to other institutions in British public life. Of the institutions listed on the survey, the ONS has the highest levels of trust, similar to that of the Bank of England and the courts system. Figure 1 shows a comparison of levels of trust in different institutions, as reported in 2018 and 2021.

Figure 1: Proportion of people that trust different institutions in British public life

Source: Public Confidence in Official Statistics 2021, National Centre for Social Research

Responding to data gaps

In line with the Authority’s five-year strategy ‘Statistics for the Public Good’, (launched in 2020), we are radical and ambitious in providing analysis in a timely way. For example, during the pandemic, we set up and adapted surveys at pace to inform policy decisions and the public: the Business Impacts of Coronavirus Survey (BICS) (now known as the ‘Business Insights and Conditions Survey’), the Opinions and Lifestyle Survey (OPN) and the COVID-19 Infection Survey (CIS). These assessed the impacts on the economy, businesses, society and on the UK’s health. More recently, we have set up new surveys to assess Over-50s in the labour market, and the experiences of Ukrainian nationals arriving in the UK.
Our agility to respond to emerging demands for evidence means we go some way in avoiding speculation, and therefore potential misinformation, on an issue.

Building trust and tackling misinformation

Proactive Communication – Media and Public Engagement

To avoid misinterpretation, each statistical publication from the ONS is expressed in clear, concise language with summarised findings. Contextual background and commentary are provided when they support wider public understanding of the data and its significance. In recent years, we have focused our efforts on ensuring key findings are reported accurately and lead the coverage, with an emphasis on the use of trusted ONS spokespeople.
We build relations with media producers, editors, lead reporters and subject matter experts in targeted media outlets to encourage clear reporting of our statistics and analysis. This also provides a direct channel back for media to confirm details for immediate deadlines.
The ONS identifies areas where insight could be misrepresented or misunderstood and mitigate through our presentation such as creating reusable and shareable social media posts and content.
The direct contact details of the statisticians are provided with each release so they can be contacted directly for guidance by any member of the public. Since the start of the COVID-19 pandemic, the ONS has answered more than 12,000 different queries from the media.
While the majority of audiences will engage with ONS content through media channels, we also raise the ONS profile directly with public audiences. Our main social media presence is the @ONS Twitter account with 343,000 followers, which achieves good comparable engagement and reach, with threads created to support outputs and to respond to specific trends on social media.
We have created a network of statisticians who converse in dialogue on particular issues, themes, and releases on social channels, providing clarity where discussions take place. In recent months, we have also trialled material and commentary on LinkedIn to better understand opportunities to reach out to business audiences.
We seek creative opportunities to increase the public’s engagement in ONS statistics and analysis as the trusted source, including personalised tools and data visualisations. For example, the ONS Personal Inflation Calculator, a collaboration between the ONS and the BBC, enables individuals to see how increases in the cost of living have affected them. This has resulted in ONS data being more accessible, as well as extending our reach to new audiences. For the first results of the 2021 Census of England and Wales, we developed interactive articles and a game to encourage individuals to actively engage with our data and to discover what the results meant for the population of their local area.

Reactive Communication – Challenging Disinformation

In response to the challenge of disinformation and misinformation on social media, we set up an online monitoring and reporting capability. This proved particularly effective during the 2021 Census. We monitored social channels, to identify misinformation and disinformation, and work directly with social media companies to remove content and accounts.
Where inaccuracy and misrepresentation of ONS statistics are spotted in the media we will seek to challenge them immediately whenever possible. News organisations are typically highly cooperative in amending their online articles or publishing corrections of ONS statistics and analysis. There have been no significant instances when a ‘mainstream’, regulated UK news organisation has refused to engage with us when factual inaccuracy has been drawn to their attention.
We are frequently consulted by both the Office for Statistics Regulation and fact-checking organisations outside of government when others use our statistics. Where there are false and misleading impressions of ONS statistics, we will rebut them: for example, in January 2022 we published a media statement and article explaining why some claims regarding the number of COVID-19 deaths were highly misleading. The rebuttal was itself reported in the news media and attracted wide engagement to social channels, illustrating the impact we can have.

Office for Statistics Regulation written evidence to the DCMS Sub-committee on Online Harms and Disinformation inquiry on misinformation and trusted voices

Dear Mr Knight,

I write in response to the inquiry Misinformation and trusted voices, as conducted by the DCMS Sub-committee on Online Harms and Disinformation.

Which organisations are the most trusted sources of information in the UK?

The Office for Statistics Regulation is the independent regulatory arm of the UK Statistics Authority and provides independent regulation of all official statistics produced in the UK. It aims to enhance public confidence in the trustworthiness, quality and value of statistics produced by government through setting, and assessing compliance with, the Code of Practice for Statistics1. In addition, one of our key roles is to use our voice to stand up for statistics and to represent the public, monitoring and reporting publicly where we have concerns about the dissemination and use of statistics and highlighting good practice.

The Code of Practice for Statistics has three pillars: Trustworthiness, Quality and Value. The three pillars work together to provide the conditions to support public confidence in statistics, which relates directly to the question the Committee is asking. In particular, we distinguish trust – a belief on the part of individuals – from trustworthiness – a property of organisations. Trustworthiness is about providing evidence that the systems, processes and governance surrounding statistics are effective. However, we never consider trustworthiness in isolation. We consider all three pillars to determine whether statistics are fully compliant with the Code of Practice and can be designated as National Statistics. This designation demonstrates to users that they can have confidence in the relevant official statistics.

One source that can give some insight into levels of trust in official statistics is the 2021 study of public confidence in official statistics. It found that, amongst people who responded, there was high confidence in the statistical system. While respondents did not necessarily know about the Authority or the OSR, there was strong support for our role, with 96% of respondents agreeing there should be an independent body to speak out against the misuse of statistics and 94% agreeing that such a body should ensure that statistics are produced free from political interference. Regarding the Office for National Statistics (ONS), the largest producer of official statistics in the UK, 87% of respondents reported that they trusted ONS statistics. The public value of statistics has also been shown through 92% of respondents who had used COVID-19 data reporting them being useful. Although this is only one source, and we are careful not to place too much weight on a single survey result, we do consider that this provides some reassurance around public confidence in official statistics.

In addition to official statistics producers, there is a wider ecosystem of statistics and data. Many of these other sources of statistics and data inform policy and public debate and it is important that they are used for the public good. We encourage producers outside of the official statistics producer community to apply the Code of Practice for Statistics on a voluntary basis. Our annual award for Statistical Excellence in Trustworthiness, Quality and Value recognises those who voluntarily apply the core pillars of the Code of Practice for Statistics.

Is the provision of authoritative information responsive enough to meet the challenge of misinformation that is spread on social media?

Our view is that the best way to combat misinformation is to ensure that information that is trustworthy, high quality and high value is made available to the public. In this way, the good information can drive out the bad.

However, we recognise that it is hard to live up to this ideal. The experience of the pandemic is instructive. As we noted in our recent State of the Statistical System report, there are a variety of organisations and individuals commenting on the use of statistics by government. The COVID-19 pandemic in particular was associated with an increase in the role of citizens as ‘armchair epidemiologists’. We wrote a blog4 highlighting how open data enabled great work to be done to communicate data on COVID-19 publicly from outside the official statistics system, including on social media. This demonstrated the changing statistical landscape of increased commentary around official statistics at its best.

Since the pandemic there has continued to be an increased interest in and scrutiny of statistics. This is a positive for the statistics system but also brings risk. Much discussion of statistics takes place on social media with increased risks around misuse, misinterpretation and ‘echo chambers’. Official statistics producers need to be aware of these changes in the use of statistics.

Areas that we highlight in our report that can help official statistics producers meet the challenge of misinformation that is spread on social media include:

improving how uncertainty in statistics is communicated to bring effective insight;

an increase in government statisticians challenging the inappropriate use of statistics and engaging directly with users to support understanding of statistics; and
intelligent transparency around statistics, data and wider analysis.

Intelligent transparency means proactively taking an open, clear and accessible approach to the release and use of data, statistics and wider analysis. As set out in our regulatory guidance on transparency, intelligent transparency is informed by three core principles: equality of access, enhancing understanding and analytical leadership. It is about more than just getting the data out there. Intelligent transparency is about thinking about transparency from the outset of policy development, getting data and statistics out at the right time to support thinking and decisions on an issue, supporting the wider public need for information and presenting the data and statistics in a way that aids understanding and prevents misinterpretation.

In conclusion, a constant refrain of the OSR is that it is important to ensure that the bad data does not drive out the good. However, as long as producers have the right approach, based on trustworthiness, quality and value, good statistics can thrive.

Please let me know if any questions or if I can support the Committee further in its inquiry.

Yours sincerely 

Ed Humpherson 

Director General for Regulation

Office for Statistics Regulation written evidence to the Digital, Culture, Media and Sport Sub-committee on online harms and disinformation’s inquiry of the same name

Dear Chair,

I write in response to the Digital, Culture, Media and Sport Sub-committee on Online Harms and Disinformation call for evidence.

The Office for Statistics Regulation (OSR) provides independent regulation of all statistics produced by the UK Government, Devolved Nations and by all related public bodies. The OSR is the independent regulatory arm of the UK Statistics Authority (the Authority), which was established by the Statistics and Registration Service Act 2007 (the SRSA).

We set the standards producers of official statistics must meet through the statutory Code of
Practice for Statistics. We assess compliance with this Code, and designate statistics as National Statistics. There are three pillars of the Code:

• Trustworthiness: trusted people, systems and processes
• Quality: robust data, method and statistics
• Value: statistics that serve the public good

Our role is to ensure that statistics serve the public good. In a world in which data and information are abundant, people can feel bombarded by information. We focus on the government as a provider of information and statistics, disseminating a reliable, impartial evidence base.

While we also have an important role in challenging instances of statistical misuse (misinformation), most of our regulatory work focuses on what it means to inform society. We help the public to identify the statistics that meet the highest standards of trustworthiness, quality and value and we challenge producers to fill data gaps to better inform society.

This is a very difficult time for everyone as the UK adjusts to rapid changes in society and the economy. Organisations that produce official statistics are showing flexibility and adapting what they collect and publish to respond to this new environment. The pace at which these organisations have set up new data collection and dissemination processes has been unprecedented and enables timely updates on the number of COVID-19 cases and deaths, as well as the economic and societal impacts of the pandemic.

In response to COVID-19 we have developed a package of measures including guidance on factors that producers should consider when making changes to data collection and statistics. We have carried out short regulatory reviews of new COVID-19 questions added to the Office for National Statistics (ONS) Opinions and Lifestyle survey, and of new experimental faster indicators constructed from rapid response surveys, novel data sources and experimental methods.

In accordance with our interventions policy we have responded to concerns about the publication of data on COVID-19 cases and deaths, and have called on the Department for Work and Pensions to ensure management information on Universal Credit used in daily briefings is published and accessible to the public. We have undertaken a review of all the data releases on COVID-19 cases and deaths – at a UK level and for each country within the UK – to help understanding of the available sources and to highlight strengths and areas for improvement.
Following our interventions regarding data on COVID-19 cases and deaths, there have been improvements in the information provided by government. In particular, there is now much greater clarity that the daily deaths data is incomplete and does not include deaths in all settings. It is a leading indicator, however, with the weekly figures from the ONS (and National Records of Scotland and the Northern Ireland Statistics and Research Agency) providing a more complete picture of deaths associated with COVID-19.

However, in order to maintain public confidence in these crucial statistics, we are encouraging producers to continue to clarify the nature and extent of the uncertainty around the UK estimates of deaths associated with COVID-19, and what the figures do and not include. We also continue to state our expectation that any management information used as part of daily public briefings is published and accessible to the public.

In summary, while combatting misinformation is crucial, it is also essential that the public receives information from government that is trustworthy, high quality and valuable – and enabling that outcome is the heart of OSR’s mission.

I hope the Committee finds this evidence to be helpful. Please do not hesitate to contact me if I can be of any further assistance.

Yours sincerely
Ed Humpherson
Director General for Regulation