PUTTING DATA TO WORK FEATURE
data. Matters there are simplified by the fact that there is a full population register in nations such as Norway. “That means that it is possible to gather administrative data on people throughout their life course, unlike the UK where we take snapshots at specific points.”
She adds that in the UK, individual government departments have a lot to gain from allowing the ONS to be the repository for the data they gather. “In most of government, data sharing is one of the activities that tends to suffer when spending is tight. It can never be a priority compared to work related directly to policy.” By contrast, she characterises the new ADRP approach as “de-risking” data issues for government. It provides it with a trusted route for data to be curated and made available, whether for academic users, the public at large, or indeed government itself. Vignoles and Gordon also share an enthusiasm for making the ADRP visible throughout the UK, not just along the corridors of power in London. Vignoles says: “The ADRN has had huge success in Scotland, Wales and Northern Ireland. In each country there have been exciting projects with impact. ADRP has funding for this to continue and grow, and indeed to show the rest of the UK the right way to do this.” As Gordon sees it: “The devolved administrations have proved the benefits of analysing linked data. We intend to champion the success we have had in the devolved UK, where we have seen new policy insights coming from datasets that had never been linked before.” There is no lack of interest in linked data within central government, but it inevitably has longer chains of command which call for more people to sign up to any new process.
Linked data showed that rich and poor children with the same level of prior school achievement were equally likely to go to university – the real issue was major differences in school
achievement.
The ADRP is only now getting started, but
Professor Vignoles and Dr Gordon have been active in this field for some time. What is their favourite example of linked data in action? Vignoles works in the field of education data,
and points to a case in which she and colleagues were able to illuminate the contentious area of school achievement and university entry. “The HESA [Higher Education Statistics Agency] data had never been linked with data on school achievement. Once we did that it was possible to see deep inequity in university access.” Linked data showed that rich and poor children with the same level of prior school achievement were equally likely to go to university. The real issue was major differences in school achievement. This work has now been extended to look at graduate earnings from a full range of courses and universities. Emma Gordon points to work early in her career,
on the Avon Longitudinal Study of Parents and Children. She did research on linking these data from Bristol to hospital episode statistics, looking for links to physical or mental disability. If a child of 12 is diagnosed as being on the autistic spectrum, what were their hospital notes saying about them at the age of two, compared to the surveys the mothers were filling in at that time as part of the study? “This analysis was a very powerful one and produced a lot of primary research. It was the spur for me to get more academics to use administrative data on everything from birth registrations to cancer diagnoses, and illustrates the advantages of making this data more accessible.” n
i
Administrative Data Research Partnership Web
esrc.ukri.org/adrp
SOCIETY NOW WINTER 2018 27
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
