Advising on the ONS Dynamic Population Model

CPC researchers are advising the Office for National Statistics (ONS) in their quest to transform the way population statistics are produced.

The work on the Dynamic Population Model (DPM), led by Dr Louisa Blackwell (Project Lead - Integrated Statistical Design, ONS) and Dr John Bryant (Bayesian Demography), aims to make subnational population estimates more accurate and timely, and to allow their production to take into account different sources of data in a coherent way. CPC-CG members, including Dr Jason Hilton, Professors Peter Smith, Jakub Bijak and Eren Dodd, as well as Dr Joanne Ellison and PhD student Andrew Hind, are providing advice to the DPM team, particularly in the areas of Bayesian statistical modelling and the estimation of demographic rates.

At the centre of the DPM is the demographic account: a description of population dynamics in which the change in populations over time is determined by the net effect of births, deaths and migration. This fundamental population relationship provides a framework through which different sources of data can be combined to estimate population change. Estimating a system of demographic accounts using a Bayesian statistical model means that the accuracy of different data sources can be taken into account in the production of population estimates. This also implies that the DPM can better incorporate administrative data sources on NHS patients, pupils and students, and taxes and benefits. None of these data sources alone provide complete information on the UK population, but by taking them together and accounting for their biases, robust estimates can be produced, together with estimates of the uncertainty surrounding such estimates.



As well as administrative data, estimates of fertility, mortality and migration rates are required to complete the demographic account. The CPC-CG team has extensive experience in this area, and has provided expert advice to the ONS on incorporating models of rates into the dynamic population model. The DPM is still developing, but the initial published results show the potential of this approach for future population statistics production. This approach will change the way in which official population statistics are derived, from counting of people to using statistical models to reconcile different sources of demographic data.
Population and migration statistics underpin decisions and policies right across our society and economy at national and local levels and for different communities. It is therefore vital they keep up with society’s changing needs.

On 29 June 2023, the Office for National Statistics launched a public consultation on its proposals for a transformed population and migration statistics system. Understanding user needs will provide essential evidence in making its recommendations to Government on the future of population statistics. Join the ONS consultation webinar on 13 July to find out more.

Further reading

Consultation on the future of population and migration statistics in England and Wales (ONS consultation)

Admin-based population estimates: updated estimates for local authorities in England and Wales, 2021 to 2022 (ONS)

Dynamic population model, improvements to data sources and methodology for local authorities, England and Wales: 2011 to 2022 (ONS)

Growing confidence in our ability to move to a new population statistics system (National Statistical, ONS blog)

New Data: Transforming how we count the population (National Statistical, ONS blog)

Posted 03/07/2023 08:06