To mark the launch of the ADR UK (Administrative Data Research UK) website, Dr Emma Gordon, Director of the ADR UK Strategic Hub, discusses why this is exactly the right time for the partnership to gain traction.
Since December 2018 I have been leading the Strategic Hub team at the Economic and Social Research Council (ESRC) to set up and take forwards the ADR UK programme. As you will see from our new website, this is all about working with government to create linked, research datasets from administrative data sources, so impactful research can be delivered both by government and the wider research community to inform policy decisions. If we are able to harness more of the richness of these datasets, ultimately this will lead to better, more informed policies being delivered across the UK, potentially leading to improvements for society.
I can already hear the shouts of ‘we have tried this all before, and it didn’t work’. ESRC did indeed run an Administrative Data Research Network (ADRN) programme, from 2013 to 2018, which was working hard to deliver many of the same aims as ADR UK, but which was halted before the planned end date.
There were many good, bad and clunky issues along the data use pathway for ADRN, which have been well documented. However, what is also documented is that the devolved administrations of Scotland, Wales and Northern Ireland were able to make a lot of progress under this investment, building productive relationships between civil servants and academics, proving the value of collaborative working in this area.
Building on what did and didn’t work previously, we have designed ADR UK very differently. Most importantly, we have acknowledged the costs to government departments of improving access to administrative datasets. Firstly, there are the financial costs; in many cases, administrative data are drawn directly down from operational systems. These data aren’t of suitable quality for onward linkage and analysis without a lot of extra work. Secondly, there are the reputational costs if something goes wrong, in terms of data security and making sure data shares are legal and ethical.
If I had to sum up why I think now is exactly the right time for ADR UK, it is because of the way we are now able to respond to these challenges.
On the first, we are now working in partnership with the devolved administrations and Whitehall departments to ask: ‘what linked, administrative datasets do you wish you had access to, to support your policy decisions?’; providing financial support to allow these datasets to be created, if this is the blocker.
On the second, we now have the Digital Economy Act (2017), providing a permissive legislative route for departments to share data. Crucially, we are also not asking departments to share data with researchers in the physical sense, but by agreeing to their data to be accessible via a safe setting, such as the Office for National Statistics (ONS) Secure Research Service (SRS), or similar within the devolved administrations.
This is really important, because it de-risks data sharing for government departments, researchers and the public, as data never leave the safe setting; they are just accessed, either in the same building or remotely through a secure portal. As has been summarised very neatly to me, this is the difference between a lending library – where the book leaves the building and the library loses control of it – and a reading library, where control is retained, but the words are shared. It also means the risks of people misusing de-identified data, for instance by linking attribute variables via ‘fuzzy matching’ to identify individuals, as was recently discussed in a journal article in Nature, are mitigated.
ADR UK is investing heavily in ONS as part of this investment. Much of this is so that ONS can have a well-resourced Secure Research Service, both in terms of technology and people. We are also working closely with the UK Statistics Authority to ensure the comparable services run by the devolved administrations are working to the same standards and accreditations.
These changes have already led to some extremely exciting research projects being initiated across UK and devolved government, as a direct result of ADR UK. Some of these are described on our new website, and I look forward to writing many more blogs to update you on progress, on these and other new data-driven projects with the potential to deliver tangible benefit for people in the UK.