Chapter navigation
This article is for statistics producers and analytical leaders involved in producing and communicating official statistics. It explores the Quality standards in the Code of Practice for Statistics: Prioritise Quality, Be Rigorous and Be Open About Quality. It does not introduce new requirements. It is intended to support producers in considering how these standards apply in practice.
Fit for purpose not perfection
The starting point for quality is that statistics are never perfect measurements. They are estimates shaped by data sources, methods, assumptions, systems and professional judgement. Quality therefore depends not only on how well a statistic is produced, but on how well it represents the real-world phenomenon it is intended to describe. That is why quality cannot be reduced to a final check before publication. It has to run through the whole statistical system: the culture that encourages challenge and learning, the rigour applied to data and methods, and the openness with which strengths, limitations and uncertainty are explained to users. These ideas are central to the Code of Practice for Statistics and to OSR’s vision that statistics should serve the public good.
This means uncertainty is not an occasional problem to be mentioned only when it is large or difficult. It is a normal and unavoidable feature of statistical evidence; recognising it is central to judging quality. The three Quality standards can be understood as a practical response to that uncertainty: creating a culture that is willing to recognise it, applying rigour to understand and reduce it where possible, and being open so users can interpret the statistics responsibly.
Openness is therefore part of how quality is created, not just the final act of publication. By being open about data sources, methods, assumptions, limitations and uncertainty, producers make it possible for users, colleagues and other experts to scrutinise the evidence and understand what the statistics can and cannot support. In practice, openness means showing the workings: not only the final statistics, but the choices, trade-offs and areas for improvement that sit behind them.