Five things we learned at the Administrative Data Research UK 2023 Conference
As part of our leadership of the social care theme of the Administrative Data Research (ADR) Wales programme, a group of us attended the ADR UK 2023 Conference in Birmingham in November 2023.
Wales punches above its weight when it comes to linked data research
The conference covered the whole UK, but Wales was frequently highlighted as a trailblazer in linked data research. The Secure Anonymised Individual Linkage (SAIL) Databank was greatly admired by all.
Researchers from Wales delivered excellent presentations, on topics including the Supporting People Programme, Welsh Government’s fuel poverty scheme, the impact of the environment in and around the home on health and social care outcomes for older people, and exploring the use of health services by unpaid carers in Swansea and Neath Port Talbot.
People were also interested to learn about what we're doing at Social Care Wales to promote linked data research. Our workshop on the subject was a big success with great attendance and feedback.
Using linked data can identify group patterns but it isn't so helpful when it comes to individuals
One of the excellent keynote presentations was from Marion Oswald, Professor of Law at Northumbria University. Drawing on her own experiences working in the criminal justice system, she took us through the origins of the right to privacy. And how this was included in the European Convention on Human Rights after the Second World War in response to the targeting of populations in Nazi Germany.
We can use algorithms to identify risk factors for certain behaviours in data. But this presentation brought to life the fact that using them at the individual level is problematic.
An algorithm can tell you that a certain section of the population might be more inclined to behave in a certain way, but this doesn't help much with identifying individuals. Why? Because we're much more complex than the labels or attributes we end up with.
Researchers need to think about the potential impact of their research from the very beginning
Marion Oswald highlighted the dangers of using data inappropriately to identify individuals. And of using algorithms to make life-changing decisions, for example, about things like potential criminality.
She stressed how important it is that researchers consider from the outset how their research could be interpreted for policy or practice decisions, and to take the right actions to make people aware of this. This includes engaging with practitioners and people who may be affected by policy decisions.
As we lead on the social care theme in the ADR Wales programme of work this type of engagement will help us set research priorities. These can inform both social care policy and practice.
Public engagement is everything
The importance of public engagement was a consistent theme throughout the conference. Our Research Officer Lynsey Cross co-delivered a workshop about engagement. She talked about how SAIL works with a panel of public members to review and give feedback on the quality of linked data research projects.
Ed Humpherson, the head of the Office of Statistics Regulation (OSR), talked about his organisation’s 2022 report Data Sharing and Linkage for the Public Good, and the critical importance of engaging with the public about data questions.
This is something we’re passionate about at Social Care Wales and the OSR report is one we refer to frequently in our work. We need a good understanding of how people can react to different types of data use. This helps us when we make decisions about how widely we share data or how we use it for research. This is an important consideration for our ADR Wales social care leadership work.
Find out more:
Learn more about our work leading on the social care theme for ADR Wales. Or visit our information page on linked data research and blog on linked data research.