Judgements

Consider the Assessment of Salmon Stocks and Fisheries in England and Wales for publication as official statistics

Publishing these estimates as official statistics has the potential to foster trust within the user community and support their appropriate use. It would provide greater protection to publication slots and for independent decision-making, given the Code requirement to pre-announce publication dates and to limit pre-release access to the statistics. Publishing them as official statistics could additionally help to raise awareness of the Code across the producer organisations, support a quality culture, and improve links and collaboration with the wider Government Statistical Service.

We recommend that the producer organisations consider whether to support the production and publication of statistics from the Assessment of Salmon Stocks and Fisheries in England and Wales as official statistics. Though any decision on official statistics status rests with the producer organisations, we would encourage producers to consider our views and to follow The Government Analysis Function’s guidance on labelling official statistics, which recommends “that data published by government bodies should be classified as official statistics unless there are very good reasons not to do so – for example because of serious concerns about quality or lack of transparency that mean compliance with the Code of Practice is problematic”. Users would generally view publications such as these reports as statistical in nature and may not appreciate why they are not published as official statistics unless the reasons are set out clearly.

Regardless of any decision on official statistics status, and in order to support user confidence in the process of updating the methods, we recommend that the planned changes in methods and subsequent reporting should be managed in line with the Code principles of Trustworthiness, Quality and Value. Specifically, we are making the following recommendations for the producer organisations:

Recommendation 1: To protect public trust, support appropriate use, and strengthen independence in line with the Code of Practice, the lead officials should undertake and publish a clear evaluation of whether the Assessment of Salmon Stocks and Fisheries in England and Wales should be produced and published as official statistics. In doing so we suggest that they:

• Draw on the advice of the Department for Environment Food and Rural Affairs (DEFRA) Head of Profession for Statistics.
• Apply Government Analysis Function guidance, on practical considerations when deciding to use the official statistics label, and publish a full evaluation and decision statement with the rationale against Trustworthiness, Quality and Value (TQV) and prominently link the published evaluation from the main statistics page.
• Embed orderly release practices: publish and maintain a 12-month forward publication schedule for these statistics, preannounce release dates, publish pre-release access lists where appropriate, and publish reasons for any date changes.
• Review related releases (e.g. Environment Agency salmonid and freshwater fisheries statistics) for potential official statistics status and publish outcomes to ensure a consistent approach across the family of outputs.

Transparent communication of future developments

To support public trust the producers should be open about how and when they will release any new data and statistics that are based on the new methods. We recommend that information on the planned developments is published in a prominent, accessible and consistent location, with clearly communicated timelines. Where timelines change, the reasons for this change should also be communicated to maintain public trust. This contrasts with the communications to date regarding the consultation on salmon stock estimation methods, where the timeline has changed with little explanation and is located around halfway through the lengthy pdf background document.

To ensure a useful level of engagement where public consultations are held, such as the forthcoming consultation on salmon stock estimation methods, we recommend that the producers promote them transparently and widely, beyond internal and core user forums. As part of the consultation, users should be given opportunities to share their views on what statistics are needed in the longer term, including suggestions for what might be useful if there are significant changes in forecasts, estimates or time series resulting from the proposed methodological changes. Users should also be consulted on the form and content of any future statistical reporting options and given opportunities to provide feedback to maximise the value of these outputs.

To maximise the value of the statistics we recommend that the producers judge what is needed to improve the statistics by listening and responding promptly to user feedback, being transparent about their decisions on how users’ needs can and cannot be met, and by explaining why particular content has been included or excluded from an output. This will be especially important when implementing the results of the public consultation, where it is essential that users are kept informed about changes and their impacts on the statistical series – for example, on potential breaks in the time series or material changes in data quality. The Code places particular focus on public involvement and engagement in statistics, and its supporting guidance may help the producers ensure that users’ needs are appropriately considered during the period of change.

Recommendation 2: To maintain user confidence during changes to methods, reduce surprises, and ensure statistics continue to meet user needs, producer organisations should focus on improving the transparency of their communications around future developments. In doing so we suggest that they:

• Publish an accessible roadmap covering the methods/review programme and reporting changes – including milestones, the 2026 consultation, decision points, and delivery phases.
• Maintain a change log explaining any slippage (for example, the move from a 2025 to a 2026 consultation), with mitigations to sustain user confidence.
• Publish the scope of the 2026 consultation in advance, signalling key technical areas (e.g. rod exploitation rates, effort ratios, egg deposition rates) and flagging potential impacts on time series and data quality.
• Broaden reach beyond core user forums (e.g. smaller angling trusts and the wider public), report who was reached and how, and summarise consultation outcomes to demonstrate transparent decision‑making
• Publish a methods change risk and assurance plan covering key risks (e.g., dependencies on angler‑reported catch/effort data), mitigations, external peer review/sensitivity testing, and post‑implementation review after the first new‑method release.

Improvements to the statistics and supporting quality information

The producers should ensure that data and statistics are easy to use, understandable and tailored to their audience’s needs. We recommend that good practice is drawn from the Environment Agency’s separate but related publication, Salmonid and freshwater fisheries statistics, which is published in html and has a more accessible layout. The Environment Agency told us that it had previously undertaken a successful exercise with the Department for Environment, Food & Rural Affairs (DEFRA) to improve the accessibility of its flood-related outputs. We recommend that the producers explore seeking a similar form of peer review to improve the clarity and accessibility of these statistics and combine this feedback with views received from users on their preferences.

We recommend that the producers consider separating technical and policy information, such as the factors considered in the stock assessment and their assurance, into separate documents. The producers should clearly describe the quality of data and statistics, including any uncertainty and bias in estimates and the implications for interpretation and use.

When updating background documentation, the producers should be open and clear about quality and ensure that information on data sources and their strengths and limitations is clear and accessible. It will be particularly important for the producers to manage and communicate data quality through the method change process. Tools such as OSR’s Administrative Data Quality Assurance Toolkit may provide a useful framework for assessing the quality data sources during the transition.

We welcome the move to make more-timely provisional data available through the Cefas portal to support further reuse. Should a new data reporting tool be developed, we recommend ensuring that it is compatible with commonly used assistive technologies, with supporting documentation and related statistics/datasets clearly signposted from the main publication page. Key quality information, including information on data sources and methods, should be prominent and easily accessible in any future reporting tool to support appropriate interpretation of the statistics.

Recommendation 3: To further enable appropriate interpretation and reuse, ensure inclusivity, and be transparent about methods and uncertainty, producer organisations should prioritise making improvements to the clarity and accessibility of the statistics and supporting quality information. In doing so we suggest that they:

• Move key outputs to accessible HTML with a concise up-front summary; reduce reliance on long PDFs and provide clear navigation to methods and policy context (drawing on the Environment Agency’s HTML fisheries statistics as an example of a more accessible publication).
• Separate technical/methods materials from policy/legislative context into distinct, audience‑targeted documents, each short, task‑oriented and clearly signposted.
• Prominently highlight quality caveats (uncertainty, potential bias, data source strengths/limitations) in each release.
• Use OSR’s Administrative Data Quality Assurance Toolkit to structure quality assessments during the methods transition; document how data limitations are being addressed as methods evolve.
• Ensure any new data tools/portals (e.g. R‑Shiny‑style approaches) are compatible with assistive technologies, signpost related datasets/statistics from the main page, and include inline explanations of methods and quality.
• Continue and formalise provision of more‑timely provisional data via the Cefas portal with clear labelling (provisional vs final), versioning, and supporting documentation to aid appropriate interpretation and reuse.

Broader engagement with users and the statistical community

We recommend that the producers take steps to strengthen their engagement with users, sponsor departments and the wider Government Statistical Service (GSS). The benefits of this would include enhancing experience with, and raising awareness of, the Code across the producer organisations; supporting a strong quality culture; and generally improving collaboration and user engagement.

In particular, the producers should ensure that users, especially non‑technical users, are considered central to decision‑making. To support this, the producers should seek advice and challenge from colleagues with Code experience. This could be supported by appointing Code champions, inviting peer review, making greater use of the support and advice available from sponsor departments, and by engaging with teams in other nations that produce similar official statistics to learn from good practice. Such external challenge would also allow the wider community to learn from good practice within the producers, such as methods being published in peer-reviewed journals.

As part of implementing these recommendations, we encourage the lead official within each organisation to consider whether other releases of statistics within their organisations should be produced and published as official statistics – for example, the Environment Agency’s related statistics on salmonid and freshwater fisheries.

Recommendation 4: To embed the Code, strengthen quality culture, and improve collaboration and challenge, producer organisations should focus on strengthening their engagement with users, sponsor departments and the wider Government Statistical Service. In doing so we suggest that they:

• Name a senior lead official/Head of Profession (HoP) sponsor for the cross‑organisation salmon statistics programme to coordinate Code compliance and oversee delivery during transition (including on orderly release).
• Engage actively with GSS networks and producer teams in other UK nations producing similar statistics to share and adopt good practice (methods, tools, accessibility, quality documentation).
• Appoint Code champions in each organisation to promote TQV, coordinate peer review and external challenge, and act as focal points for improvements.
• Consider how to make users central to decision‑making (e.g. a user reference group); record and publish how user input shapes content and methods, and where needs cannot be met, explain why.