Using administrative data to deliver impactful research

We live in a society rich in data and the opportunities which this brings. However, to make sure we benefit from these opportunities we need to ensure our research and its methodology can keep up to date with new data sources and our ever evolving user needs. The Methodology Division, in the Office for National Statistics, leads a research programme to develop new methodology and research ensuring that we can continue to deliver impactful research in our increasingly data driven society.

Our research programme focuses on priorities across government ensuring many can benefit from our research and it is impactful. A core element of this programme centres on developing research methods so we can use administrative data for statistical purposes, in a secure and safe way that protects people’s personal information. Administrative data can be defined as data which are “initially collected for some administrative purpose – to run an organisation, such as a company, government, charity, school, hospital, and so on”. Read Statistical challenges of administrative and transaction data (this paper is only available to subscribers).

Across the Government Statistical Service (GSS) we are increasingly using administrative data in the production of official statistics but, unlike with survey data, there isn’t a well-established and agreed statistical theory to underpin its use in official statistics. The aim of the Administrative Data Methods Research Programme  is to develop this statistical framework and overcome the many challenges of using administrative data for statistical purposes.

To progress our research programme we are keen to collaborate on the cross cutting methodological challenges across government, academia and internationally. By working together we are more powerful to find solutions for this and it ensures we all benefit. Over the past 18 months we have already established many collaborative research projects ranging from projects focusing on developing a quality framework for non-survey data to developing methods to derive estimates from administrative data sources.

One of our main projects is working with academic experts from Southampton University to understand how to estimate and communicate error of non-survey data sources. Results from this project will be published in the next couple of months, so keep an eye out!

We are really keen to understand similar research taking place across government. Please get in touch to let us know about any relevant research you are doing or even if you would just like to find out more about the programme, email Admin.Data.Methods@ons.gov.uk.

Charlie Wroth-Smith
Alexander Amaral-Rogers
Charlie Wroth-Smith is head of the Methodological Research Hub within the Office for National Statistics.