This research is funded by Economic and Social Research Council as part of the Administrative Data Research Centre for England (ADRC-E). https://adrn.ac.uk/centres/england Administrative data sets have the potential to be a valuable resource for research, with a significant impact on policy making and evaluation. ADRC-E is a UK-wide partnership between academia, government departments and agencies, national statistical authorities, funders and the wider research community that will facilitate research using government administrative data (ie information collected primarily for administrative purposes, by national and local government and other public sector organisations as part of service delivery).
This is a very large project running over five years, within which we are investigating the requirements for information governance which ADRC-E should reflect in its national e-infrastructure policies and practices. Our research studies the relationships between information and records management, open government and transparency, privacy and security, and the re-use of government administrative data, from the viewpoints of information professionals who provide or deny access, researchers seeking access to data, and the citizens, who are often the data subjects and on whose behalf data is created and preserved. Our findings will also contribute to the wider public debate about government information and data management.
A risk management approach to disclosure review
Taking the view that disclosure is a product of (a) sensitive attributes of the data and (b) appetite for risk, this study seeks to capture the multiple perspectives and purposes as a combination of which data might be considered disclosive (citizens' personal privacy; trade, economics and public finance; public order, safety and law enforcement; defence, security and international relations, etc.), and to investigate the applicability of a risk management approach to disclosure review of government administrative datasets used in academic research.
This study will probe, and build out of, the opacity in any distinction between cohort and administrative data. If cohort data are collected with participants' explicit consent and with a specific statistical or research purpose in mind, whereas administrative data are collected (and are often legally required to be collected, without data subjects' explicit consent for the data to be used in research) in order to deliver a government service in respect of the individual data subject, what are the implications of this distinction for the research use of administrative data?
Trusting government administrative data
This research will apply theoretical framings on data provenance, transparency, and trust to practical issues concerning the contexts and circumstances in which administrative data are collected and processed by government, and the implications that this might have for research use. What is the impact of the inherent instability of administrative datasets (day-to-day changes to data, changes to variables collected, changes in government policy, machinery of government changes) on the reliability, validity and trustworthiness of government data and consequently of research based upon such data?