Our research to transform population and migration statistics has focused on producing our best understanding of the population from administrative data. We anonymously link individual records from administrative data sources and removing records of those unlikely to be a usual resident.
In the past, our admin-based population estimates (ABPEs) were produced directly from these linked data. However, as our methods have evolved these linked data have become an input to our improved method, and we now refer to them as Statistical Population Datasets (SPDs). Our SPDs are also used as a basis for producing admin-based household estimates (ABHEs). The objective of the SPDs and ABHEs is to help us to understand the usually resident population at both national and local level.
Building upon our SPDs, we have developed a dynamic population model (DPM). For more information on our DPM, see our Dynamic population model for England and Wales: July 2022 article. This statistical modelling approach uses a range of data inputs to measure the population and population changes in a fully coherent way. SPDs are one of the data sources that are used in the DPM, alongside components of population change including:
The DPM will allow us to assess historical trends against emerging patterns and behaviours in the population, so that we can produce information about the population more quickly. This will also allow us to better understand how it changes over time. This approach will allow us to produce provisional estimates from the DPM, which can then be updated at a later date as more information becomes available. It is an important milestone in our journey towards providing the best estimates of the population in a timely and responsive way. You can find further information around the different population estimates we plan to publish in our Population statistics and sources guide.
Our work towards improving our migration statistics is ongoing, introducing improved methods using new data sources, as well as reviewing definitions of migration to ensure they meet the needs of users. We are routinely producing admin-based migration estimates (ABMEs), as official, albeit experimental, statistics on international migration.
We are also regularly updating on our progress towards producing timelier and more detailed migration statistics.
We have continued to publish feasibility research and experimental statistics on the use of administrative data for population characteristics such as income, and ethnicity. We have demonstrated the potential for producing some of these statistics down to sub-regional geographies including lower-layer super output areas (LSOA). We are also exploring how multivariate combinations of topics can be created from administrative data.
Our research using Valuation Office Agency (VOA) data to understand housing characteristics has shown how we can produce more frequent and granular statistics on housing characteristics for sub-regional geographies (currently local authorities), including information that has not previously been readily available (such as number of bathrooms).
News and reports
We published our Developing admin-based ethnicity statistics for England and Wales: 2020 article by combining ethnicity data from eight administrative data sources and the 2011 Census. Using this approach, we were able to establish an ethnicity for 84.9% of people in the 2020 SPD V3 for England and 88.5% of people in the 2020 SPD V3 for Wales.
We produced our latest multivariate case study on developing admin-based income by ethnicity statistics. We have also just produced a case study on housing by ethnicity, including exploring methods to produce occupied address level ethnicity measures.
We published research showing how we have continued to refine our admin-based income statistics (ABIS) for England and Wales, bringing together data from the Pay As You Earn, Self-Assessment and benefits systems to derive experimental measures of gross and net individual and occupied address income. This has included publishing income statistics for lower layer super output areas (LSOA). Our latest research identified income information from at least one ABIS data source for 92.5% of individuals aged 16 years and over in our population base, whilst 98% of occupied addresses in our population base had some income information identified.
We produced research showing the feasibility of using administrative data to measure labour market status. This research showed the potential to produce admin-based labour market status (ABLMS) at a local authority level, with 95% of individuals aged 16 years and over (on our population base) being assigned a status.
As we aim to deliver more responsive and high-quality population and social statistics you can find out more on the Get involved page.