Our Head of Research discusses how the findings from the 2023 Public Confidence in Official Statistics Survey are relevant to OSR.
At the Office for Statistics Regulation (OSR), we have a vision that statistics serve the public good. This vision guides our regulatory activities, which means we must have a deep understanding of how statistics can serve the public good, including what actions bring us closer to or further away from our goal.
This understanding is supported by evidence such as UK-wide surveys which gather public views about official statistics. For England, Scotland and Wales, this information is collected by the Public Confidence in Official Statistics (PCOS) survey, which is conducted independently on behalf of the UK Statistics Authority. For Northern Ireland, the Northern Ireland Statistics and Research Agency (NISRA) use the Northern Ireland Continuous Household Survey to explore Public Awareness and Trust in Official Statistics (PATOS). The findings from PCOS 2023, which were published this week (14 May 2024), and PATOS 2022 provide valuable insights into attitudes towards official statistics.
In this blog post, we highlight some of the findings from PCOS 2023 and PATOS 2022, and discuss how they align to our activities in OSR. This demonstrates how we see our work serving the public good.
While we are including data from both PCOS and PATOS in this blog post, the two are not directly comparable. For example, the surveys cover different years, and PCOS removes people from their analysis who said they ‘don’t know’ or didn’t provide an answer to questions, while PATOS includes ‘don’t know’ responses. Because of the difference in approach to which responses are included or not, when we say ‘respondents’ in this blog post for PCOS data we mean people who were able to express a response other than ‘don’t know’, whereas when referring to PATOS data ‘respondents’ includes people who indicated that they ‘don’t know’.
Building confidence through trustworthiness
As the Head of Research at a statistics regulator, you may need to forgive me for being specific and precise about technical statements in this blog post. However, I want to emphasise that these surveys report what respondents say, and in the case of PCOS, people who responded that they ‘don’t know’ or who didn’t respond are not included in reported percentages of respondents. The PCOS survey largely asks questions about the UK Statistics Authority’s production branch, the Office for National Statistics (ONS). As a NISRA-produced survey, PATOS focusses on NISRA statistics.
In both PCOS 2023 and PATOS 2022, the majority of survey respondents reported that they trust these statistics producers, with 87% reporting that they tended to trust ONS or they trusted it a great deal, and 85% reporting the same for NISRA.
At OSR, we assert that trust in official statistics is important if they are to serve the public good – people will not use evidence that they feel is unreliable . However, we also recognise that trust cannot be forced; it must be earnt. Our Code of Practice for Statistics (the Code) includes a pillar called ‘Trustworthiness’, which embodies the idea that producers of statistics must consistently demonstrate that they deserve trust. This pillar encourages honesty, independence, reliability and competence. By holding statistics producers to the standard for trustworthiness set out by the Code, OSR supports trust within our statistical system, which we assert helps statistics to serve the public good.
Addressing concerns with intelligent transparency
Equally important to knowing people’s reasons for trusting official statistics is understanding their reasons for not trusting them. In PCOS 2023, respondents who indicated that they did not trust ONS statistics were asked to select a reason for this. Respondents most commonly attributed their lack of trust to statistics being misrepresented by politicians (49%), although many also expressed that they believe the statistics alone do not tell the whole story (45%) and that the government has a vested interest in or manipulates the results (41%).
To address these reasons for not trusting statistics, we assert that there is a need for producer bodies to follow OSR’s principles of Intelligent Transparency:
- Equality of access: Data used by the government in the public domain should be available to all in an accessible and timely way
- Enhancing understanding: Citations for sources and appropriate explanations of context should be provided alongside the information
- Independent decision making and leadership: Decisions about the publication of statistics and data (such as the content and timing) should be free from political influence
These three principles tie directly to the three most commonly reported reasons for not trusting ONS statistics:
- Equality of access allows people to see the statistics themselves without having to rely on how they are represented by public figures
- Enhancing understanding can help users to understand exactly what parts of the story statistics cover (and what parts lie beyond their scope)
- Independent decision making and leadership may help alleviate concerns around government manipulation of official statistics
Advocating for, and supporting the implementation of, Intelligent Transparency is therefore another route OSR takes to building trust in official statistics. Intelligent Transparency sits alongside our casework function, where we investigate potential issues with official statistics such as how they are used in public debate. Here, we use our voice to stand up for statistics, reporting publicly where we have concerns and highlighting good practice. We anticipate that our work protecting statistics in public debate will allow more people to be confident in the use of official statistics, and therefore allow these statistics to better serve the public good.