​FOI ref: FOI/2022/4218

You asked

Please provide me with copies of your current policies in the following areas:

(1) data quality,

(2) ensuring that statistics are accurate,

(3) avoiding bias in statistical information

(4) ensuring statistics are communicated clearly.

We said

​Thank you for your request.

Information relating to statistical quality is publicly available on the Analysis Function website.

Information on accuracy, bias and communicating statistics can be found here alongside a wealth of other information. The website can be easily searched and browsed for required information.

Some particular items to draw your attention to are:

The overall guidance on Quality in Statistics, which offers practical advice on how to design processes that are efficient, transparent and reduce the risk of mistakes when producing statistics. It covers the dimensions for measuring the quality of statistical outputs.

The GSS Quality Strategy, which sets out best practice to improve the quality of our statistics and manage the processes surrounding their production and is designed to ensure that we comply with the quality requirements of the Code of Practice for Statistics.

The Code of Practice is based on three pillars: Trustworthiness, Quality and Value. Each pillar contains principles that producers of statistics should commit to. The document also contains guidance, resources and links to case studies. The Code covers three areas of practice - collaboration, coherence and transparency and cuts across the three pillars. The Office for Statistics Regulation reviews compliance with the Code.

The ONS Statistical Quality Improvement Strategy, which lays the quality objectives, as well as how to implement those, to improve quality across the ONS in the production of statistics and analysis that serve the public good.

The Communicating quality, uncertainty and change guidance, which sets out principles on how to communicate information about quality, uncertainty and change to users.

As well as the comprehensive guidance available, further support for statistical producers can be identified from the GSS methodology support page.

For more on the quality of data either not being used for statistical purposes, or data that is pre-analysis, the Government Data Quality Framework sets out the importance of understanding whether data is fit for purpose, and the need to tackle the root causes of poor quality.