Updating the snapshot: Making sure local population statistics remain accurate and reliable
Accurate and truly representative population statistics are central to the Office for National Statistics’ mission of data for the public good. Following the successful census in March 2021, now we are pulling out all the stops to make sure our numbers reflect recent and ongoing population changes. Jen Woolford explains how we are prioritising this important work and collaborating in new ways to deliver it.
Sense checking what we do against local insight is extremely valuable to the ongoing development of our population statistics. For this reason, and following recommendations from the Office for Statistics Regulation review of population estimates and projections in May 2021, we are today launching a new, robust way of receiving user feedback.
The local population insight feedback framework is a tool for users to let us know about local data sources which have the potential to provide additional insight and allow comparison with our population estimates. Where appropriate, we will use this information to improve our population estimates or to provide contextual information about our estimates.
Using this mechanism, we will assess whether data sources can be used to improve our estimates, as an indicator of quality or whether they can help inform future research. You can read more about the tool and find out how to provide feedback on local level population estimates here.
In a similar vein, we’re opening up new ways for users of data and statistics to engage with us across the Government Statistical Service by theme. The aim of the collaboration includes identifying and addressing data gaps and bringing together a diverse community of users. To provide detail on the 12 themes, including population, migration and census led by ONS’ Becky Tinsley, we have published an explainer here.
Using the census to understand population change since 2011
Following the publication of the first results from Census 2021, we are now carrying out reconciliation and rebasing of the mid-year population estimates (MYE) for the past decade. This process happens every 10 years following the census but is particularly important this time given that the coronavirus (COVID-19) pandemic is likely to have had an impact on many people’s movements. The detailed snapshot the census gives us may not reflect these longer term trends.
For this reason, and for the first time, we will produce a ‘Census Day minus one year’ estimate to allow us to best understand how much change happened in the year leading up to March 2021. We have planned a series of provisional publication dates to share more information later this year and in 2023 as the work progresses. If you’re interested to understand more about the reconciliation and rebasing process and why we do it, we have published an article today.
We are also making plans for the methods we will use in producing the official mid-year population estimates for mid-2021, which will benefit from the Census 2021 results and will reflect the number of births, deaths and estimates of migration that took place between Census Day on 21 March 2021 and the mid-year point at 30 June 2021.
Local insight deep dive
Another area of work underway is a pilot study of 14 local authorities to help inform developing work to transform population statistics, including a route to share data and insights directly with us and increase our understanding of changes that have taken place since Census Day. This builds on the extensive collaboration with local authorities right across England and Wales who supported our rigorous quality assurance processes for the first time in a census earlier this year.
This new pilot study will particularly help us test our methods relating to social and geographic features such as high student numbers, coastal, rural, and metropolitan city locations. We will be reviewing administrative data sources relating to what they show about the size and nature of different population groups and publishing our results as case studies. The insights that we gain from the pilot study will be used to help to develop the data sources and the assumptions we will use in the Dynamic Population Model (DPM), launched in July. We will publish new outputs from the DPM for these local authorities in the autumn.
Looking to the future
All of this work feeds into our learning as we forge ahead with our ambition to transform our population, migration, and social statistics. Our overarching aim is to meet the need for more timely and frequent statistics that enable understanding of changing populations. Last week we published new feasibility research on developing subnational multivariate income by ethnicity statistics from administrative data for England and further updates will follow later this year.
Our Population statistics and sources guide is regularly updated to provide information on the different types of population estimates we have published and plan to publish in 2022 and 2023. For example, it includes information about population estimates from Census 2021, Labour Force Survey weights and the official mid-year population estimates for 2021 which we are working towards publishing in late 2022. This also contains information about whether estimates are ‘official’ or the result of wider research, a summary of the main uses and comparability issues for the estimate.
The next stage of census results includes an exploration of data at lower geographies and on specific topics from the autumn. Timings will be updated on our release plans pages soon.
In the meantime, I’m excited about the prospect of meaningful engagement and sharing of learning and thank all users who are already contributing. Feedback is always welcome and if you want help to understand which population statistics you should be using at different times, please email email@example.com.