The Census Analysis Programme
Here at the ONS, we’ve already started drafting that new chapter. Census Day on March 21st, 2021, was only the start for ONS staff. We don’t just crunch the numbers, we plan a programme of analysis and outputs. And we want your input on both. In July, we published our proposals for the Census Analysis Programme, as part of a wider Census 2021 outputs consultation, which is set to tease out the unfolding story of Britain’s population in the 21st century
In this programme of social research, commencing in mid-2022, after the initial census results have been published, we are currently planning 162 articles across 12 topics.
As an organisation, we’ve learnt a lot since the last census in 2011. Our programme then was primarily aimed at our expert audience of professional statisticians and social researchers. Our Census 2021 programme will – of course – do so again, but this time we’re keen to bring together the widest possible audience to engage with our statistics.
The fundamental aim of the Census 2021 Analysis Programme is to provide an inclusive, trusted and engaging narrative on the UK’s economic and social fabric and trends. In doing so we’ll address cross-cutting issues and provide analysis to inform and engage governments, policy makers and the wider public.
It’s important that our analytical programme has longevity well beyond publication day. We know that users don’t just engage with our statistics on a single day, it’s something returned to again and again. Our programme must therefore also be sustainable.
To achieve this, we have set out proposals for all our Census 2021 outputs and analysis and have opened them out to consultation so we can document the full range of user needs.
New questions, new answers
Above all, we’re seeking to deliver an analytical programme which answers the most important questions in society today, producing analyses that are essential for influencing policy and ensuring the right decisions are made to improve people’s lives.
Ethnicity: The 1991 Census was the first to introduce the ethnicity question, allowing for comprehensive analysis of the distribution of different ethnicities around the country. This is an example of how census questionnaires evolve from decade to decade, based on what is important to users and opening new opportunities for granular analysis of the population.
Sexual orientation and gender identity: For those over 16, the Census 2021 is the first where we’ve asked voluntary questions on sexual orientation and gender identity. We will be able to produce statistics on gender identity for the first time and more detailed statistics on sexual orientations.
Armed forces: There is also a new compulsory question on whether respondents had previously served in the UK Armed Forces. We will be able to understand the life outcomes of the armed forces, whether it be concerning health, housing, education, or employment.
Income: Innovations in data linkage open up analytical opportunities even further. In particular, methodological developments in linking census and income data have shown great promise, meaning analysis on topics such as ethnicity by income level are feasible in a way that they haven’t been before.
In the programme we’ve also proposed a piece of work which explores the characteristics of the population living in high-rise towers, another which explores the demographic make-up of those working in the NHS labour force and another which looks at deindustrialisation in Britain and its effect on the population.
Our consultation is open to your proposals for questions and analysis.
How will we publish census analysis?
In the documents we’ve recently published as part of the Census 2021 outputs consultation, we included the ‘shape’ of the analysis and outputs release schedule, as shown in figure 1 below.
From an analysis perspective, the release schedule follows one key principle, that we start with the simplest findings, and then move towards more complex analyses as the programme progresses.
Phase 1: Our earliest findings will be our univariates of the population (e.g. the number of people attributed to a given characteristic, such as health status). We’re keen to hear from our users how these univariate releases should be ordered, so don’t hesitate to let us know in our consultation.
Later publications will describe the population using multivariate data:
Phase 2 of our release schedule will see more analyses added to the catalogue. Here, deeper analyses will crosscut with other topics and provide detailed evidence on key issues.
Phase 3 and beyond: The later phases of the release schedule will see our more sophisticated analyses. Here, projects involving data linkage and longitudinal data are planned, seeking to answer more complex research questions on topics such as generational social mobility and long-term outcomes of the disabled population.
But what do you think?
Feedback from our users shapes what we produce and how we build our products. For us, it’s the most reliable way for us to know whether we’re delivering the right statistics in the right way. Our census outputs and analysis consultation is currently open and will close on the 5th October. Our consultation questionnaire can be found here and asks questions on:
- the shape of the outputs and analysis release schedule
- main changes to variables compared to the 2011 Census
- proposals for feasibility work to derive new variables
- population-base specifications
- taking a census in a period of change
- paradata
You can find out more about our proposed analyses here. We would love to hear what you think of our programme: whether it’s too much or too little, what we’ve missed or what should be left out.
Help us write the most recent chapter of Britain’s demographic history by responding to our consultation and tell us how it should be told:
Webinar: Writing the next chapter of Britain’s demographic history
Thursday, 23 September 2021
14:00 – 15:00 BST
Booking via Eventbrite
Biography
Edward Morgan is the Head of Census Analysis Coordination at the Office for National Statistics. A demographer by training, Edward has broad research interests across the discipline, having written on topics such as mortality, longevity, ageing, historical demography and Sub-Saharan Africa. Edward has degrees from St Andrews University and the LSE. He is also a graduate of the European Doctoral School of Demography.