How to use the Office for Statistics Regulation's work to judge a statistic
The Office for Statistics Regulation (OSR) is the independent regulator of official statistics in the UK. Its published work tells you three useful things about any figure: whether the statistics are accredited, whether OSR has found problems with them, and whether OSR has commented on how they have been used in public. Checking OSR's site takes a few minutes and often settles a dispute about a statistic faster than anything else.
What OSR does
OSR describes itself as "the independent regulator of all official statistics produced in the UK" and "the regulatory arm of the UK Statistics Authority". It judges statistics against the Code of Practice for Statistics , which is built on three standards: Trustworthiness, Quality and Value. It also investigates concerns about how statistics are used by government, politicians and others. See our UK Statistics Authority guide and what the Authority is .
Source: Code of Practice for Statistics
Step 1: check the status of the statistic
Look at the top of the release for its badge or status:
Accredited official statistics : "independently reviewed by OSR and confirmed to comply with the standards of Trustworthiness, Quality and Value". Until September 2023 these were called National Statistics. See accredited official statistics .
Official statistics : produced by official bodies and subject to the Code, but not accredited.
Official statistics in development : new or changing methods; see official statistics in development .
Management information or ad hoc analysis : not official statistics at all.
OSR keeps the list of accredited official statistics and a separate list of accreditations that have been suspended, cancelled or reinstated. If a release says it is accredited, the list should agree.
Source: list of accredited official statistics
Step 2: check whether accreditation was removed or suspended
Accreditation can be suspended or cancelled "if substantial non-compliance with the standards of the Code of Practice for Statistics is identified". OSR then issues an improvement notice; where the producer can fix things within a year, the accreditation is suspended rather than cancelled.
Example: the ONS bulletin Crime in England and Wales: year ending March 2026 states that police recorded crime data "are not designated as accredited official statistics" because "Their accreditation was removed by the Office for Statistics Regulation (OSR) in January 2014 following an assessment that found evidence that the quality and consistency of the underlying data may not be reliable." That is why the ONS prefers the Crime Survey for trends; see how to check a crime trend claim .
Source: Crime in England and Wales: year ending March 2026
Step 3: search OSR's correspondence and reports
OSR's correspondence page can be filtered by topic (for example Health, Crime and Justice, Population and Migration, Economy and Public Finances) and by organisation. Letters fall into a few types:
Reviews and assessments of a set of statistics, often with recommendations;
Casework letters about a specific public claim;
Letters confirming or removing accreditation.
Search the site for the statistic's name. If OSR has written about a figure being used in public, its letter is usually the most authoritative official account of what the figure does and does not show.
Source: correspondence page
Worked examples from OSR's published letters
A claim about a falling waiting list. In June 2024 the Chair of the UK Statistics Authority wrote that a statement about NHS waiting lists falling for five months "was supported by the data, but greater transparency around the discontinuity in the time period being referred to would have better supported understanding." The letter shows how to weigh a claim that is numerically right but leaves out a change in the series. See how to check an NHS waiting times claim .
Pathways and people. In July 2026 OSR explained that waiting list counts are pathways, which "do not correspond directly to the number of individual people waiting for care", while confirming that a March 2026 compliance review found the statistics "continue to meet the standards of the Code".
Historic economic data. On 12 October 2026 OSR wrote to the ONS that parts of the pre-1997 series for gross fixed capital formation (business and government investment) "are affected by significant quality issues that are large enough to affect interpretation", and asked the ONS to publish a plan within six weeks. Anyone quoting long-run UK investment figures from before 1997 should note this. See how to find historical UK statistics .
Sources: wrote, explained, wrote to the ONS
Use OSR's own list of problems as a checklist
OSR's action guidance on misuse and misinterpretation , published on 9 October 2026, is written for producers, but its description of what goes wrong is a practical checklist for anyone checking a claim. It lists "omitting important context, making unsupported claims, comparing figures that are not comparable, or selectively using data". It describes misinterpretation as understanding statistics "in a way that does not reflect what the data can reliably show, often because definitions, context, uncertainty or limitations are missed or unclear". For any figure, ask: is context missing, is the claim supported by the release, are the two numbers comparable, and was the period or subgroup chosen selectively?
Source: action guidance on misuse and misinterpretation
Step 4: check the casework log, or raise a concern
OSR's casework page links to issues logs listing every concern it has considered, which it aims to update every Wednesday. OSR notes that inclusion in the log "does not necessarily mean that we share the concern". If a claim has already been raised, the log tells you, and any outcome letter will be on the correspondence page.
Anyone can raise a concern about the production or use of official statistics through Raise an issue . Include the exact words, where and when they were said, and the official figure you think they relate to. See how to report misuse of statistics .
Sources: casework page, Raise an issue
Checklist
Status of the statistic noted (accredited, official, in development)
Accreditation list checked
Suspended or cancelled list checked
OSR correspondence searched for the statistic
Casework issues log checked for the claim
Concern raised with OSR if needed, with full details
Official sources used in this guide
Sources: Office for Statistics Regulation, OSR: accredited official statistics, OSR: correspondence, OSR: casework, OSR: raise an issue, Code of Practice for Statistics, OSR: letter to ONS on historic gross fixed capital formation (12 October 2026)
Questions
Is OSR part of the ONS?
No. Both sit under the UK Statistics Authority, but OSR is the regulatory arm and also regulates the ONS. See whether the ONS is a government department .
Source: Office for Statistics Regulation
Does OSR say whether a politician's claim is right?
OSR comments on whether statistics have been used in a way that supports understanding, and publishes its letters. It does not run a rating system.
Source: Office for Statistics Regulation
Does unaccredited mean wrong?
No. It means OSR has not confirmed compliance, or has found issues. Read the producer's quality notes before relying on the figure.
Source: Office for Statistics Regulation