There are a growing number of tools to help manage ethical considerations in data projects that are suitable for different stages of the design process (see Recommendation 7).
At the conception and design stage of a project, LOTI recommends using the Open Data Institute’s (ODI) Data Ethics Canvas or the UK Government Digital Service’s Data Ethics Framework.
Impact assessments are also useful and include Data Privacy Impact Assessments (DPIA) and Human Rights Impact Assessments (HRIA), Equalities Impact Assessments (EIAs) as well as Algorithmic Impact Assessments (AIA). The Ada Lovelace Institute provides a useful AIA template.
Audits, including technical and regulatory audits, can also be used to evaluate both the data systems themselves and the potential broader consequences of a data project (during or after the fact).
It should be noted that different tools require different skillsets within organisations. For example, a technical audit of an algorithm requires more advanced data skills than a Data Privacy Impact Assessment. Equally, data experts may need greater understanding of the social elements of risk associated with their models if they are to adequately complete the ODI’s Data Ethics Canvas. Recommendation 13 contains more detailed information on training for staff.
Brent Council identified the ODI’s Data Ethics Canvas as a useful tool in the design stage of the projects to help map the ethical dimensions of their data projects.
