Where can I find out more?

If you require any further support in addressing ethical issues in your planned project or incorporating ethical principles, please contact the UK Statistics Authority Data Ethics team or visit our webpages for further information on the support services that we offer.

This guidance has focused on ethical issues as they arise in the context of synthetic data for research and statistical purposes, however there is a wealth of other sources available that relate to synthetic data more widely. You can find out more information at the following sources:

• The Behavioural Insights Team (BIT) and Administrative Data Research UK (ADR UK) have produced research on the behavioural barriers to synthetic data for government research. This includes a case study example of synthetic data usage.
• The ONS have produced a pilot study exploring the demands and requirements of synthetic data, which includes the ONS Synthetic Data Spectrum for analysing the quality and risk of synthetic datasets.
• A framework for assessing synthetic data generation tools was developed by The Cyber and Data Science Capability of BAE Systems AI Labs.
• The ONS Data Science Campus has produced the SynthGauge Python package for evaluating synthetic data; this includes some methods to test privacy/disclosure of synthetic data.
• The ONS Data Science Campus produced an interesting blog on “synthetic data for public good and art”, and a report on synthetic data for public good.
• The Department for Work and Pensions Digital have produced a blog outlining how synthetic data can benefit government departments.
• The Counter Trafficking Data Collaborative have developed a new algorithm to derive “synthetic data” from CTDC’s sensitive victim case data, and in turn, a global synthetic dataset.
• The UNECE have created a “starter guide” for using synthetic data for official statistics.

Feedback on this guidance

This guidance has been published as an open draft for comment and feedback. It aims to provide a practical resource for the research and statistical community regarding ethical considerations in the use of synthetic data in research and statistics.

We would welcome views and feedback on this guidance from interested parties, including how it can be further improved or how useful it has been for your project. If you would like to share your thoughts with us, or are interested in collaborating, please contact the Centre for Applied Data Ethics team, or visit our webpages. We look forward to hearing from you.