Assessment of the 2021 Censuses
The Office for Statistics Regulation (OSR) is assessing the 2021 Censuses in the UK. These are conducted by the Office for National Statistics (ONS), National Records of Scotland (NRS) and Northern Ireland Statistics and Research Agency (NISRA), in England and Wales, Scotland and Northern Ireland (NI) respectively.
The purpose of the assessment is to inform the OSR recommendation on whether Census 2021 outputs should be designated as National Statistics, in accordance with the requirements of the Statistics and Registration Service Act 2007, when they are first released.
Census statistics on people and households in the UK are an important resource for a variety of users including government, local authorities, academics, the commercial sector and the general public. Therefore, it is essential that the data meet the standards set out in the Code of Practice for Statistics (hereafter referred to as the Code). Compliance with the Code provides users and the public confidence that England and Wales Census 2021 statistics are of public value, are high quality and are produced by an organisation that is worthy of trust.
The entire England and Wales Census 2021 programme will be assessed. The process involves three phases covering planning and consultation, plans for developing and disseminating outputs and, finally, an assessment of user needs versus outputs, post-publication. Each phase comprises the following elements:
- the ONS compiles and submits the report
- the ONS publishes the report on the website
- the OSR assesses the report, speaks to census teams and census data users, and publishes requirements
- the ONS acts on the requirements
At the end of the Phase 2 assessment period, the OSR will consider whether the England and Wales’ Census 2021 can retain the badge of National Statistics accreditation. If we are successful, our accreditation will be confirmed before the first census outputs are released in 2022.
This report forms part of the evidence base being supplied by the ONS to the OSR. In June 2019, we published our initial report How the ONS is ensuring the 2021 Census will serve the public, which was a first step towards measuring progress against the three pillars underpinning the Code, namely Trustworthiness, Quality and Value. This coincided with similar reports published by NRS (PDF, 1.52MB) and NISRA (PDF, 916KB).
Alongside consideration of the initial report, the OSR engaged with users and other stakeholders about their views on Census 2021 plans, progress and outputs. We wish to thank those individuals and organisations who have willingly engaged with the OSR and provided feedback for the assessment.
The information presented to, and gathered by, the OSR informed its first report published in October 2019, which detailed preliminary findings from the assessment to that point. The report, entitled 2021 Censuses in the UK – Preliminary findings (PDF, 294KB) considered the extent to which planning, engagement and development activities undertaken by the three census offices in preparation for Census 2021 were compliant with the Code.
While the OSR identified areas of strength and were encouraged by progress and future plans, it highlighted specific areas for the census offices to focus on in the future. These are set out in a number of actionable findings, which the ONS, NRS and NISRA are required to address to improve compliance with the Code. This report (see Section 2) describes how the ONS is building upon existing plans and practices to address the OSR recommendations, which will help to deliver a high-quality census for users. It also highlights the strong working relationships between the three census offices, which harness the collective expertise and experience within each organisation to benefit the Census 2021 programme.
This report also provides further evidence (see Section 3) of how we are ensuring Census 2021 will serve the public and demonstrates our compliance with the Code.Back to table of contents
This section is structured around the seven actionable findings presented by the Office for Statistics Regulation (OSR) in its report on preliminary findings from Phase 1 of the assessment. Each section heading is the finding as documented in the report. This is the Office for National Statistics (ONS) response, which has been produced in collaboration with our colleagues in National Records of Scotland (NRS) and Northern Ireland Statistics and Research Agency (NISRA).
Actionable finding 1
Census offices should consider the accessibility of research and other census information on their websites and consider aligning website design and content where possible to provide a common user experience.
The ONS website is the principal channel for disseminating England and Wales census information. The main census page, which is accessible from the ONS home page, provides links for the 2021 and 2011 Censuses, as well as earlier censuses. The Census transformation webpage consists of links to other pages, which provide information on the main elements such as:
We update the Census 2021 webpages to provide users with the latest information on plans, development and progress. In response to the actionable findings from Phase 1 of the OSR assessment, we have added other useful content, namely:
Much of this content was developed via consultation with the other UK census offices, with the aim of using consistent layout and wording where possible. This will provide a common experience for UK-level users who access information on Census 2021 from each of the UK census offices websites. We continue to look at ways of improving both the content and accessibility of information about Census 2021 on the ONS website.
As we move towards and beyond census day, the content on the Census 2021 webpages will continue to expand. With methods and processes being finalised and implemented, there will be scope to provide more definitive and clear information to users. In particular this will include information about proposed output specifications, and ways in which users can help influence those specifications to ensure that they are as useful as possible once the outputs are released. We will aim to co-ordinate activities on this as much as possible across the census offices to minimise the burden on users.
Work has already started and will continue through to the release of outputs, on user research about the tools and methods of access that will be required for users once the outputs are available. Again, information on how to get involved and to share current thinking and plans, will be made available on the Census 2021 webpages.
That development of a modern and fit-for-purpose outputs dissemination platform, accessible via our website, is an important area of activity, with options currently being drawn up for consideration. The needs of users will be central to decisions made. Like NISRA, we aim to integrate the new outputs dissemination platform into our existing online presence. The intention is to use a common set of channels, products and tools for census alongside other ONS data, rather than have it exist as a standalone service. This enables users to be aware of the full range of potential data sources to meet their data needs. Any developments and solutions will be delivered in line with the Government Digital Service.
NRS has enlisted the services of a Scottish digital transformation consultancy, to deliver the new Scotland’s census outputs website. Its design will be strongly influenced by user input to ensure it best meets their information and data requirements. The new outputs website will be delivered in line with the Digital Scotland First Service Standards model.
Although the websites disseminating census data from the ONS, NRS and NISRA will most likely have a different look and feel, and some functionality differences, we will aim to adhere to the same design principles and will look to build similar user journeys to provide some sense of familiarity across the different systems. Furthermore, to ensure we have considered all possible ways to align website design, within the constraints of our own digital landscapes, there will be engagement between the ONS, NRS and NISRA’s website design teams to discuss plans and approaches and to identify specific areas where alignment in design and content may be possible.
Harmonising much of the website content is simpler to achieve and will continue to be a considerable focus of efforts across the three census offices. As an example, we are working closely with colleagues in NRS and NISRA to maximise the harmonisation of content across the three UK census websites. To achieve this, we are using similar page titles, sub-titles, headings, language, and descriptions where possible. This should minimise the learning curve for users moving between other government webpages to census or between the three census offices webpages.
There is strong focus on the harmonisation of UK-relevant content for any new updates to the existing websites. We have published a harmonised UK census harmonisation webpage that has broadly similar content on the respective websites (see Actionable finding 3.2).
It is important that when providing harmonisation information to users that the different UK census websites, products and tools use appropriate cross-site linking and signposting to facilitate a more consistent user journey. Included in the ongoing work on harmonisation of content is the increased signposting to the other UK census office websites to maximise ease of navigation for users and pinpoint equivalent information across the different countries. Work will continue to develop further harmonised content across the respective websites and to maximise the accessibility of information and enhance the user journey.
Actionable finding 2
Census offices should be open and transparent on their decision-making processes and in their decisions on census questions and guidance, particularly in relation to any areas of contention.
In June 2015, the Office for National Statistics (ONS) held a formal, 12-week consultation process asking stakeholders for their views on the topics that were required in the census in England and Wales. The aim of the consultation was to promote discussion and encourage the development of strong cases for topics to be included in Census 2021.
In May 2016, the ONS published its response to the 2021 Census topic consultation. This set out our updated view on the topics to be included in Census 2021, including:
- a summary of proposals for new topics
- next steps
- an overview of our plans
The ONS published a Census topic research update in December 2017 and a further Census topic research update in December 2018. In December 2018, the ONS published its recommendations for the 2021 Census in the White Paper Help Shape Our Future: The 2021 Census of Population and Housing in England and Wales (PDF, 885KB). The recommendations in the White Paper are based on the extensive programme of consultation, stakeholder events, evidence gathering and research.
To support the extensive information already published on question development we are publishing a series of Census 2021 question development reports. These will provide a detailed explanation of the research, testing and evaluation we have carried out to arrive at our question designs for Census 2021.
The first report Question and questionnaire development overview for Census 2021 (published March 2020) explains:
- our approach to question development
- the design process of the online and paper census questionnaires
- how we ordered the census questions
- the criteria we used to evaluate the final question designs
- our next steps as we prepare for Census 2021
This report complements the question development reports on the topics that make up Census 2021.
The question development reports will provide links to previously published research and the findings of additional testing, which led to the final recommended questions for Census 2021 in England and Wales.
The publication of these reports supports our commitment to being open and transparent on the actions and decisions taken when developing the questions and questionnaire. The publication date for these reports has been impacted by the need to reprioritise our work and ensure we are focusing our efforts on publishing statistics in response to the coronavirus (COVID-19). The timetable for publishing these reports is now summer 2020.
Two further reports have also been published to explain changes to the enumeration base between Census 2021 and the 2011 Census and to summarise every test conducted between 2016 and 2020 that informed the question design for Census 2021.
Actionable finding 3
We consider that there should be more focus on the needs of census users at a UK level, in three areas.
3.1 Census offices should consider how best to engage with users and stakeholders of UK census data and statistics users, and coordinate activities as appropriate.
The three census offices have built a strong evidence base on the requirements around UK census data via consultation and engagement with users and stakeholders on this subject. The ONS (Word, 46KB) and NISRA (PDF, 48KB) included sections on UK harmonisation and statistics in their Census 2021 outputs consultations, while NRS did likewise in its Outputs Stakeholder events. This highlighted the considerable value and utility placed on UK census data by a wide range of users.
It is the responsibility of the ONS to coordinate the production of census statistics for the UK. So, to supplement the engagement, we have conducted further research on UK census data requirements.
Firstly, a sample of users who responded to the UK census data questions in the output’s consultation were interviewed, using questions developed by the three offices. The purpose was to gain an insight on how UK data are applied and preferences with regard to accessing these data and supporting information.
Secondly, as part of our census roadshows during 2019, we included an interactive study for attendees to capture their user journey from accessing to downloading UK-level census data. Our research findings have been shared with NRS and NISRA to build a greater collective understanding of the needs of census users at a UK level.
In addition, the ongoing assessment of the 2021 UK censuses by the OSR has provided the three offices with valuable intelligence on the requirements of users interested in UK-level data. As part of the assessment, the OSR received feedback from a range of UK data users, which has been shared with the relevant census office, where authorised to do so.
In response to considering how best to engage with users of UK census data and coordinating activities as appropriate, the three census offices have introduced a few initiatives.
Establishment of the UK Census Data Working Group
This group has the specific aim of collaborating and aligning activities to address Actionable finding 3.1 and 3.3. The group meet on a monthly basis to identify ways in which the findings can be addressed, ensure consistency in approach and discuss progress.
Addition of UK census data webpage
Each of the census offices have added a UK census data section to their respective websites. This includes background information on Census 2021 in the constituent countries and provides an overview of the wide range of harmonisation activities. Plans for UK-level outputs will also be outlined as they are developed and finalised. In addition, the webpage promotes engagement with UK census users by encouraging feedback on their specific requirements for Census 2021 outputs.
Cross-office attendance at user events
The three offices have committed to attending each other’s stakeholder and user events where relevant and practical. This will involve introducing census colleagues from other parts of the UK to attendees and encouraging engagement to discuss their respective countries’ census or the census from a UK perspective. For example, colleagues from the ONS and NISRA attended the NRS Statistical Methodology Stakeholder event in Edinburgh in February 2020 and were introduced to the attendees.
Potential UK-wide stakeholder event
The UK Census Data Working Group has held discussions about organising a UK-level stakeholder event in each of the constituent countries. Group members have reviewed the UK user research undertaken by the ONS to date, as well as findings from discussions with UK data users after the OSR shared user feedback from the Phase 1 assessment and concluded that additional UK-level events at this time would not add value to understanding user needs.
In addition to these initiatives, the ONS has contributed to the newsletter for local authority partners. This updated local authorities about ongoing work in relation to Census 2021 outputs. Important dates to note were highlighted, namely:
- autumn 2021 – publication of a Census Outputs Prospectus describing the full range of Census 2021 products and outputs, and user access
- March 2022 – publication of first results from Census 2021 within one year of census day
- March 2023 – publication of main Census 2021 outputs within two years of census day
We are also in the process of updating the Census 2021 outputs web presence. We are updating existing pages and developing new pages to ensure our stakeholders and users are kept fully informed of our work.
There is a growing body of evidence demonstrating the needs of UK-level data users in England and Wales. We recognise however that at the time of publication, no UK data users in Scotland and Northern Ireland (NI) have made themselves known to NRS and NISRA. The ONS is responsible for disseminating UK data and are considering, with NRS and NISRA, how best to take forward engagement with UK data users.
3.2. Census offices should be clear about the impact of country-specific decision-making for UK census data and statistics and work together to provide greater transparency around their plans and decision-making in meeting the needs of users interested in UK census outputs.
While the ONS, NRS and NISRA are responsible for and focusing on delivering Census 2021 in England and Wales, Scotland and NI respectively, the three organisations are working in unison to ensure that the censuses are successful in:
- providing high-quality population and housing statistics for the UK
- meeting the needs of data users
- satisfying international obligations
This is demonstrated by the statement of agreement between the National Statistician and the Registrars General for Scotland and NI about the conduct of the 2021 Censuses in the UK. It sets out the elements of the census that the three offices are aiming to harmonise, where it is practical and in the interest of users and the public good more generally.
The initial statement of agreement (PDF, 164KB) was published in October 2015, and this has been followed by progress updates for November 2016 (PDF, 320KB) and 2019. The updates are structured around three broad aspects of Census 2021 that are a strong focus of harmonisation, namely outputs, census procedures and governance. Within each section, clear information is provided highlighting to users the considerable progress made to date on harmonisation. Some of the main elements are:
- common topics and questions
- statistical methods
- use of administrative data
- publicity campaigns
The UK Census Committee (UKCC), which the ONS, NRS and NISRA are members of, has an important role with regard to harmonisation of Census 2021 activities across the UK. Among its main purposes is to agree the census delivery approaches and plans in a way that takes full advantage of harmonisation of working practices and sharing of research and resources. The UKCC also agrees the scope of cross-UK working and co-ordination and provides strategic input to this work, considering a number of factors including the need to meet users’ requirements and international obligations through the harmonisation of UK outputs.
Ongoing effective collaboration provides a platform for understanding the impact of country-specific decision-making on UK outputs from Census 2021. Whilst also ensuring that opportunities for harmonisation are identified and implemented.
Harmonisation working groups (HWGs) are a significant part of the daily workings of the three census offices. They support all levels of the census programmes from management through to specific work areas. Covering most census topics, design and delivery, HWGs aim to maximise harmonisation through sharing ideas, best practice and methodologies. Importantly, HWGs help to understand any impacts on UK-level data where harmonisation is not possible.
The UK Census Data Working Group are compiling a descriptive list of all the HWGs and activities contributed to by the UK census offices to capture the magnitude of work that goes into UK harmonisation. Once compiled, it is intended that this information will be shared on our respective websites to aid transparency in the work being done on UK harmonisation. Similarly, the three census offices will work together to review the harmonisation of definitions, classifications and outputs. Upon finalisation of harmonisation approaches in these important elements of Census 2021, we will be better placed to communicate firm plans and impacts to users and stakeholders, especially those interested in UK outputs.
As explained in Actionable finding 3.1, the three census offices worked together to design and develop content for a UK census data section on the ONS, NRS and NISRA websites. New content will be added, and existing content updated as we move towards and beyond March 2021. Regarding UK harmonisation, there is information on:
- why we harmonise
- what we are doing to harmonise
- further relevant information
- how to contact us
In relation to UK census data, the webpage will evolve to provide information on:
- plans for UK outputs from Census 2021
- how UK data are used
- 2011 Census UK data and supporting information
3.3. Census offices should provide users, stakeholders and decision-makers with information on harmonisation of census questions and the impact on outputs at UK level to help inform users and support decision making.
There has been effective collaboration in developing questions for inclusion in Census 2021 between England and Wales, Scotland and NI. While each census office strives to meet the specific user and respondent needs in the relevant part(s) of the UK, there is a collective recognition of the importance of harmonising census questions and topics, where possible and practical, to produce consistent and comparable UK-level statistics.
A prime example is the area of census content and questionnaire design, which has seen the three offices build a strong working relationship via the Question Product Working Group, established early in the Census 2021 programme.
Work is ongoing to produce a clear understanding of the harmonisation of Census 2021 questions across the UK; this includes highlighting where differences exist, the underlying reasons, and the resultant effect on UK-level outputs. This work could not be completed until the respective Census Orders and Census Regulations for each part of the UK had become law, upon which the Census 2021 questions would be finalised.
With the legislation in place, the three census offices will continue to work together to review question harmonisation. The expectation is for this information to be published in 2021 so data users are fully informed at the earliest opportunity and well in advance of the publication of first census outputs in 2022. In the meantime, a summary of the main differences between the questionnaires for England and Wales, Scotland and NI are explained in the question development reports mentioned in Actionable finding 2.
Actionable finding 4
Census offices should build their awareness of the relative strengths and limitations of any administrative, commercial or other data sources used in the production of census outputs, by regular engagement with suppliers. This should be undertaken on an ongoing basis and as part of a normal way of working.
We are using administrative data to support and enhance key elements of Census 2021, as outlined in our White Paper Help Shape Our Future: The 2021 Census of Population and Housing in England and Wales (PDF, 885KB). This approach builds upon the successful employment of administrative data in the 2011 Census and aligns with the approach being adopted by many traditional census-taking countries internationally.
All research into the use of administrative data is being conducted in accordance with our policy on data security, disclosure and ethics. The different areas in which the Census 2021 programme are investigating the use of administration data is explained in How the Office for National Statistics is ensuring the 2021 Census will serve the public (see sub-heading “Use of administrative data”).
Requirements for administrative data and the arrangement for supply of data from other government departments and commercial organisations is coordinated centrally in the ONS. There is a standard robust requirements and stakeholder engagement process. This reduces the room for error, streamlines the process and builds user trust. Requirements gathering and exploration of the available sources are the first steps of the quality assurance (QA) process.
We have built strong and trusting working relationships with administrative data suppliers. Technical discussions regarding data systems and data quality are facilitated through workshops, conference calls and other communication that are managed centrally directly with the data supplier.
We maintain a detailed log of quality issues for each administrative source as the data are ingested to its secure IT system. Data quality reports are produced for each main source, which are updated as additional data extracts are received; these inform discussions with data suppliers with the aim of identifying and resolving observed quality issues.
We use the OSR guidance on Quality Assurance of Administrative Data (QAAD) to support QA of administrative data that we handle. This includes acquiring information for our own QAAD reports from each main data supplier and metadata documents where available to get a better understanding of strengths and limitations of both datasets as a whole and individual variables. Quality assurance is carried out on data when they are received, and checks are regularly reviewed and updated. Knowledge obtained through QA of the data is then summarised and disseminated to the rest of the office.
We develop our understanding of data through Statistical Quality Working Groups, where we discuss the data and the detail of the variables with data suppliers such as NHS Digital.
We have worked closely with the Ministry of Defence (MoD) to understand how we can use their data to improve our statistics on veterans, to support local authorities and public bodies to enact the Armed Forces Covenant.
We have also continued our research into understanding how Valuation Office Agency (VOA) data could be used to enhance 2021 Census outputs and produce administrative data-based housing characteristics. More information on this is in Section 3, see subheading “Integrated outputs”.
A paper “How we can ensure the admin data used in census is fit for purpose?” will be presented to the Census Research Advisory Group (CRAG) and the external assurance panel in summer 2020. The paper proposes a quality framework and strategy for measuring the quality of admin data used in Census 2021, which may be applied across the wide range of census use cases. It aims to demonstrate to CRAG attendees that we hold an overarching QA framework and strategy to ensure the quality of administrative data used in census. The framework and strategy are compatible with existing census QA plans and are in line with the OSR QAAD standard. In addition, this proposal is also in line with international best practice (United Nations Economic Commission for Europe (UNECE) 2018, UNECE forthcoming).
We will continue to build strong relationships with our data suppliers and produce QAAD assessments for relevant data sources, which will be made available as part of our statistical quality assurance methodology.
We continue to work closely and to collaborate with NRS and NISRA on the use of administrative data and the 2021 Census. This includes the UK Population and Censuses Strategic Group as the main forum to discuss:
- harmonisation of the use of administrative data and improvements to the existing population statistics system
- use of administrative data in the 2021 Census
- coordinating plans across the UK for future transformation of the population, migration and social statistics systems and improving the provision of UK statistics.
In addition, each of the four nations are represented at other groups, including the Administrative Data Research Group, which meets twice a year to share information and best practice on how to effectively integrate administrative data in the census process. At monthly Devolved Administration Joint Working Groups, the nations discuss the ONS's acquisition of data through the Digital Economy Act 2017 and how this may support the devolved nations.
We have continued to engage with other national statistical institutes on a bilateral basis, through the UN and other specially convened working groups. Many other countries recognise the opportunities in integrating data, and we continue to collaborate to help us address the challenges we face in using administrative data and transforming our statistical system.
We are working to deliver Census 2021 as planned and are reviewing the role of administrative data in the context of the coronavirus (COVID-19) with regard to our operations. This is clearly an emerging and evolving situation and implications are being reviewed on an ongoing basis.
Actionable finding 5
Census offices should make information on the methodology and quality assurance arrangements available to users at the earliest opportunity.
We published our Approach and processes for assuring the quality of the 2021 Census data in January 2020. It describes our initial proposal for how we will check the Census 2021 processes and results to make sure users can rely on them.
We will update our methodology webpage with further information on how we:
- are building on the 2011 approach and this includes coverage estimation
- are working with the Census Methodological Assurance review panel to develop methods
- plan to publish our statistical design overview in autumn 2020
In addition, the GSS methodology webpages are being updated to include pointers to the census methodology work for 2001, 2011 and 2021.
Actionable finding 6
Census offices should provide users with an indication of future census milestones – including future user engagement opportunities, publication of further research or reporting, and legislative milestones – to provide an added level of transparency and support trustworthiness and public confidence.
We have added Census 2021 milestones to the census website to inform users of milestones alongside anticipated dates. This includes completion of the legislative process, essential components of the data collection operation, such as recruitment of field staff, and publication of an outputs release schedule. While the detailed outputs release schedule will not be finalised until later in the Census 2021 programme, the timeline provided gives a broad indication of the period during which the standard and other outputs, as well as specialist products, are expected to be released.
The timeline will evolve over the period up to and beyond census day in 2021. Currently, many of the dates for the various milestones, events and publications are approximate; definitive dates will be added once finalised. In addition, the timeline will be updated to include relevant new elements. This will ensure that users are fully informed in terms of when to expect main milestones and deliverables along the Census 2021 lifecycle, thereby providing greater transparency and encouraging trust and public confidence.
Actionable finding 7
Census offices should be clear to users what assurance mechanisms are in place and be open about identified areas for improvement in a way that is proportionate and accessible to users.
We have added Assurance mechanisms for the Census 2021 Programme to the census website. This describes the different levels of assurance in place to underpin the successful delivery of Census 2021 in England and Wales.Back to table of contents
This section provides evidence of how we are ensuring Census 2021 will serve the public and demonstrates our compliance with the Code of Practice for Statistics.
Laws governing the collection of statistical data for the census
- The Statistics and Registration Service Act 2007 governs the role of the UK Statistics Authority and its executive office, the Office for National Statistics (ONS), specifying how we should use and disclose personal information, with specific regard to confidentiality.
- The Data Protection Act 2018 and General Data Protection Regulation (GDPR) together provide rules regarding the processing of personal data.
- The Equality Act 2010 provides a public sector equality duty on the ONS to eliminate discrimination, harassment and victimisation; advance equality of opportunity and foster good relations between those who share protected characteristics and those who do not.
- The Census Act 1920 provides for a census to be conducted in England and Wales (subject to an Order and Regulations, as set out in this section), and penalties including for failure or refusal to provide particulars required by the census and for wrongful disclosure of census personal data. The 1920 Act was amended by the Census (Return Particulars and Removal of Penalties) Act 2019 to ensure that census questions on gender identity and sexual orientation are voluntary by removing any penalty for not answering these questions; the Census (Amendment) Act 2000 made census questions on religion voluntary in the same way.
- The Census (England and Wales) Order 2020 came into force in May 2020 and specifies that the census will take place on 21 March 2021; the question topics to be included; the people who need to fill in the census questionnaires; and the people to be included on the completed questionnaires.
- The Census (England) Regulations 2020 and The Census (Wales) Regulations 2020 set out operational details about the census and include a facsimile of the paper questionnaires and a description of the online questionnaires.
Compliance with legislation
We will ensure compliance with the legislation using the following tools and methods:
- Equality Impact Assessment
- Data Protection Impact Assessment
- Privacy information
- Human Rights Impact Assessment
Equality Impact Assessment
The ONS published an updated Equality Impact Assessment (PDF, 661KB) for the 2021 Census in March 2020, at the time the Census Order was laid in Parliament. This assessment identifies if the operation of the census meets the requirements of the Equality Act 2010 and if our proposals for questions also meet the requirements of the Act.
A further update to the assessment will be provided in the coming months.
Data Protection Impact Assessment
We are completing a Data Protection Impact Assessment (DPIA) for the 2021 Census; this is a legal requirement as defined within the GDPR, where the processing of personal data creates a risk to the rights and freedoms of data subjects. We published a summary of the DPIA (PDF, 440KB) in March 2020.
The Census 2021 DPIA is an assessment of the risks to personal data that could arise during the census and ways in which we will manage those risks. The DPIA explains how transparency will be incorporated into data collection of the census statistics, highlighting that all respondents will be provided with privacy information about how their personal data will be processed and kept. Transparency is also achieved via the privacy and data collection page on the Census 2021 website.
Data security will be ensured in line with government guidelines. This is covered extensively within the DPIA. This includes the destruction of all paper questionnaires by the Questionnaire Management supplier, and the secure destruction of all systems and storage media holding census data.
ONS incident management will be informed by any and all suppliers where a security incident occurs during census operations. The census will be conducted in accordance with the GDPR Article 5(1)(f) on security of personal data, as set out in the DPIA.
Our Data Protection Officer (DPO) is responsible for the oversight of data protection across the organisations, including the Census 2021. The DPO is in place to deal with any incidents or queries. We have a dedicated DPO inbox where data subjects can send requests related to their rights. These contact details will be provided on the privacy information that will be disseminated during the census.
The DPIA also highlights the importance given to confidentiality with regards to disclosure of data to researchers. All researchers who wish to access secure microdata files will need to be approved by the Secure Research Service. The Approved Researcher Scheme will be used to grant access to data that cannot be published openly, as permitted by Section 39 of the Statistics and Registration Service Act 2007.
The DPIA along with information on the ONS census website detail how users’ information will be kept secure and safe.
All respondents and field staff will be provided with the necessary privacy information as required under the GDPR. For respondents, the information will be in the initial letters sent to households prior to the census and will be available on the ONS website. It will contain details about:
- our identity
- the contact details of our DPO
- the purpose of the processing
- any recipients of the personal data
- the period of storage
- the existence of data subject rights
- the right to complain to the Information Commissioner's Office
- whether the provision of personal data is a statutory requirement
Privacy information was provided for the census rehearsal. Similar information will be provided for the Census 2021.
Human Rights Impact Assessment
This is being written alongside the DPIA. The assessment identifies any potential impacts to private or family life, and how we ensure this is kept to a minimum and is conducted for legal reasons only.
Review of methods and research
The external Census Methodological Assurance review panel has been set up to undertake a review of methods and research associated with Census 2021. The review will provide assurance to the National Statistician that:
- the statistics resulting from the Census 2021 will meet the Code of Practice for Statistics and therefore can be badged as National Statistics
- the census is methodologically robust
- the evidence to show whether an Administrative Data Census approach to census-taking is valid and enables the government to decide after 2021 about the future of the census
The reviews started in 2018 and are planned to take place up to 2023.
The panel is chaired by Sir Bernard Silverman. Other confirmed panel members are:
- Oliver Duke-Williams, Assistant Professor in Digital Information Studies, University College London
- Dr Nik Lomax, School of Geography, University of Leeds
- Natalie Shlomo, Professor of Social Statistics, University of Manchester
- David Martin, Professor of Geography, University of Southampton
The agendas and minutes from the external panel are published on the UK Statistics Authority website. Past papers presented at the panel will be published in 2020 with ongoing publication of future papers.
In addition to the Response Chasing Algorithm (RCA) and Field Operations Simulation model (FOS) products, mentioned in the Phase 1 report, further systems have been developed to underpin the collection activities that support the ONS’s push to a digital-first census. These are:
- Hard-to-Count index
- Field Prioritisation Algorithm (FPA)
- response profiles
- targeted action groups (TAGs)
The Hard-to-Count index identifies areas at risk of census non-response, and comprises two domains, the willingness to self-respond and the digital domain. This will be used as a tool in the Census 2021 to support pre-planning of field follow-up.
In developing the hard-to-count digital index, data from Driver and Vehicle Licensing Agency were used as a proxy, to see how willing people are to engage with the government online. Ofcom data were used to measure internet availability in areas.
Field Prioritisation Algorithm
The FPA is designed to reduce variability within each local authority by specifically targeting low response in Output Areas. The FPA will allow us to use field staff and their hours most efficiently by prioritising addresses to be visited.
Response profiles are a best estimate of the volume of responses we expect to receive during the census period. They will model when, where and how people are going to respond to the census.
Before the census, the response profiles will feed into the Field Operation Simulation Model (FOS), which will be used to help make decisions about:
- the number of field staff required in each area
- where to send reminder letters
- where to send paper questionnaires as the initial contact
During the live census period, the response profiles will be used by the Response Chasing Algorithm tool (RCA) as a basis for comparison against live return data, so that we can tell where we need to put in extra effort to meet our quality targets of 94% overall response and at least 80% response in every local authority.
Targeted action groups
The TAGs consist of a list of groups of people, defined by some characteristic, who are thought to be harder to count in a census than the general population.
Groups are included in the list because not counting them will have an impact, either national, local or reputational, on the census and specific action is required in order to ensure that they respond. This can range from ensuring that specific information is available to respondents on the online guidance, to providing specific material (such as translation booklets or questionnaires in Braille or large print) to having a dedicated community advisor working in the field to engage with that particular community for up to a year before census day.
Assurance of methods
The Response Chasing Algorithm (RCA), Field Prioritisation Algorithm (FPA), Field Operation Simulation (FOS) and Response profiles have all been reviewed and approved internally and externally to ensure that we are using the most appropriate methods. Papers were presented at the external assurance panel and will be published on the UK Statistics Authority website in 2020.
Sharing methods development
We work closely with Northern Ireland Statistics and Research Agency (NISRA) and National Records of Scotland (NRS) on a regular basis sharing data and our methods. We also work closely with international experts from other censuses (Canada, Australia, New Zealand). We have presented our work to national statistics institutes (NSIs) representatives from Ghana, Argentina, Mexico and Namibia.
We have reached out to attendees from the United Nations Economic Commission for Europe (UNECE) with a survey to gain a better understanding of how other NSIs enumerate homeless people.
We have continued to present our work at external events, such as the People Place and Policy (PPP) conference, International Census Forum and the Local Authority Operational Management Group.
As mentioned in our Phase 1 report, the UNECE holds workshops on statistical data editing every 18 months to two years. Our edit and imputation team regularly attend and present papers. The work on linking and imputing admin data on income was due to be presented at the UNECE workshop in April 2020. However, because of the coronavirus (COVID-19) pandemic the workshop was postponed. The UNECE is discussing alternative ways to hold the workshop.
An update for how the three census offices are harmonising on the statistical procedures applied to the census data, such as methodologies for statistical disclosure control and estimation was published in January 2020.
Quality assurance of the Census 2021 data
We published our Approach and processes for assuring the quality of the 2021 Census data in January 2020. It describes our initial proposal for how we will check the Census 2021 processes and results to make sure users can rely on them. The publication describes:
- the quality assurance (QA) strategy
- the initial proposal for how we will conduct our QA activities
- how we will develop this initial proposal
- our plans for publishing information on quality of the results
- how we expect to manage the work during the census
We expect plans to evolve considering engagement with stakeholders. Ways in which we will seek user views and report on our plans include:
- seeking comment on proposals from Census Advisory Groups covering main users of census results
- seeking endorsement of the proposed approach from the Census Methodological Assurance review panel
- attending groups such as the Local Authority Operational Management Group (LAOMG) that are likely to attract users of census results
- establishing a Local Authority Engagement Group of a small number of local authorities representing a range of different characteristics, to understand their requirements and issues around the quality of census data
- inviting comments from all users on the proposals set out in the publication
- publishing a final plan for the approach to QA in autumn 2020, reporting comments and suggestions we have received from users and how these are reflected in the plan
We will make use of expertise within the ONS and look beyond ONS to develop our approach. The Approach and processes for assuring the quality of the 2021 Census data publication provides more detail on how this will be achieved.
2019 Census Rehearsal
In October 2019 we held a census rehearsal for collection. The rehearsal provided an important opportunity for us to test the processes, systems and services we will need to use for Census 2021. It took place in four local authority areas: Carlisle, Ceredigion, Hackney and Tower Hamlets.
Following this we conducted the Census Coverage Survey and a Census Quality Survey. We are evaluating the information gathered through the rehearsal and will publish an evaluation report in summer 2020. The results of this evaluation will be used to inform further improvements to systems and the planning approach for the England and Wales Census 2021.
Following the Collection Rehearsal, we have been carrying out a Processing and Outputs Rehearsal to ensure we run the full end-to-end census operation in advance of doing it for real. This rehearsal is still in process and is expected to complete in July 2020. The purpose is to test the systems, interfaces and processes involved.
Let’s Count programme
As part of the census rehearsal we piloted our primary schools’ education programme, Let’s Count! in four rehearsal areas. Let’s Count! is a free educational programme that aims to reach a wide tranche of the general public, at a local level, through primary schools and their networks. The rehearsal was a success, and 58% of schools signed up to take part and 86.3% of those that signed up said they would take part in 2021.
The campaign positively influenced parents: 78.4% of parents with children who participated in Let’s Count! filled in their rehearsal questionnaire and 64.9% said they would definitely take part in Census 2021, compared with 48.8% of parents outside of rehearsal areas.
We are currently reviewing activities and materials ready to launch the 2021 initiative. We have plans to develop a secondary schools outreach programme to support Census 2021. We are currently defining the scope and activities that this will include.
Saying thanks to rehearsal areas
The rehearsal was voluntary, with the ONS asking residents to give up their time to test the processes and systems ahead of Census 2021. To give something back to the communities, deputy National Statistician Iain Bell completed an epic charity cycle ride – all 500 miles from Carlisle to Hackney and Tower Hamlets in East London, via Ceredigion in West Wales.
The challenge was a huge success and offered some good opportunities to engage local communities and raise awareness of the census rehearsal. During the challenge Iain visited some of the schools taking part in the Let’s Count! primary schools’ campaign, he:
- ran some Let’s Count! assemblies
- talked to the children and staff about the census
- encouraged them to get involved
As well as local schools, Iain met with local dignitaries to help build support. He met mayors, local councillors, representatives of the local charities and census engagement managers. We had some great opportunities for interviews and press coverage to help promote the census. London Live TV and BBC Radio Cumbria were among the various outlets that covered the story, interviewing Iain about the rehearsal.
The challenge was a roaring success and Iain revisited the rehearsal areas in December 2019 to present the respective mayors and council speakers with cheques for their chosen charities and present the prize to the overall winner of the Let’s Count! competition.
Sharing experiences and learnings from rehearsal
NRS and NISRA also each conducted a census rehearsal in 2019. Each census office took a slightly different approach to the extent and timing of the various aspects of the rehearsal. Work is ongoing to share lessons learned with each other in order to maximise benefits of the rehearsal and effectiveness of census procedures. To date several activities have taken place with more planned.
In January 2020, the ONS, NISRA and NRS held a two-day session as part of the Harmonisation Working Group to have a deep dive into the findings from the rehearsal. This face-to-face session provided an opportunity to work through findings and to relate experiences across the census offices. As part of the event we also then went through a number of scenarios that had either occurred in previous censuses, had the potential to occur or had occurred in other countries. We reflected on how this would be handled in the nations individually and then whether there were UK aspects to consider.
We presented the findings from the two days at the UK Census Committee (UKCC) in early February, which provided a more strategic view of our shared learnings from both the rehearsal and the two-day session. Since the two-day session we have collaborated closely on how we take forward findings. As an example, the ONS and NRS had a session to reflect on the Wave of Contact model (WOC model for contacting households) we were proposing to change as a result of what we learnt.
In March, the ONS hosted a detailed discussion on our evaluation with colleagues from ONS Methodology to check that the technical conclusions we have reached are sound. NISRA and NRS attended this session.
The processing and outputs rehearsal are currently underway and should conclude in July 2020. We will be sharing findings and lessons from our respective rehearsal activities across the census offices.
Stakeholder engagement activities for Census 2021
We ran a series of roadshows across England and Wales in spring 2019. These were well attended by representatives of many user groups: academics, businesses, charities and local authorities. Around half of local authorities attended. Topics discussed were the ONS transformation agenda, content of the Census 2021 questionnaire, outputs design and the operation of the Census 2021. Feedback was gathered and used to feed into the design.
We meet with local authorities and community groups regularly to increase their understanding of the census, and therefore increase engagement. Comments and concerns are addressed or fed back to the relevant teams in the Programme.
As mentioned in our Phase 1 report, we hold twice-yearly Census Advisory Groups (CAGs), which represent users from all sectors (including central and local government, academia, commercial, health and diversity groups) in both England and Wales. Agenda items presented on at the last two meetings (June 2019 and January 2020) were:
- 2019 Census Rehearsal
- addressing update
- question development
- population and social statistics transformation
- geography consultation
- marketing campaign on brand
- statistical design of census
- census outputs and disclosure control
- Administrative Data Annual Assessment update
- promoted the Census 2021 campaign website, which has a lot of information about the 2019 rehearsal
- provided updates about Iain Bell’s cycle ride
- supplied case studies of how census data is used
Development of outputs
We are currently defining the output products that will be made available to users, including metadata. This involves carrying out user research, internally and externally, which will continue through to the release of census outputs. This will help inform the products we produce, and when we make them available.
There will also be internal factors that influence a release schedule as well as the UK perspective. Different factors driving the release schedule and the trade-offs such as timeliness have been discussed at our Output and Dissemination Harmonisation Working Group and at our Public Policy Impact Board. We will take these considerations forward as we develop the release schedule.
We are intending to make available an indicative release schedule as part of our Census Outputs Prospectus in autumn 2021 (at which time we will also have initial indications on how the census collection has gone, which will also inform the release plans). We have started to communicate to users our plans for the Census Outputs Prospectus at Census Advisory Groups and Market Research Society Census and Geodemographics Group. We are also drawing together wider engagement and consultation plans between now and census release. Again, we will look to align and join up as much as possible across the UK.
Flexible table builder
We have previously presented a prototype of our flexible table builder (FTB), based on synthetic data, at events for users to trial and give feedback and will continue to do so. We are currently working on our plans on the delivery of the User Interface changes to the ONS website, which will be required to provide user access to the FTB tool.
The ambition is to make early versions of the FTB available to users (in Beta form) ahead of census release, using a non-census dataset. However, those development plans are at an early stage and so no commitments can currently be made.
We have made available a “100% sample” of 2011 microdata in our Secure Research Service. This includes the full suite of data from the 2011 Census, including all variables for all person and household records.
However, only variables needed for specific projects are provided, and researchers may only apply to use these data if they can justify why the 10% household or individual samples are not sufficient (for example, if researching specific small populations). Each project is assessed for disclosure risk and outputs are subject to disclosure control. This approach will be reviewed and considered for 2021, considering changes to the disclosure control policy.
Also for Census 2021 we are working on plans for a safeguarded household file, which was not available for 2011. The Microdata Working Group involving external stakeholders continues to meet three times a year (last meeting March 2020). Feedback has been received from group members on 2011 Census microdata variables and codebooks to inform the development on 2021 products. The sampling methodology has been agreed.
We have continued research, which relates to the White Paper (PDF, 967KB) commitments to “explore the feasibility of supplementing the census questions with administrative data on total number of rooms, living space and property type.”
Two research outputs have been published in relation to this work.
The first publication, Admin-based statistics for property type, feasibility research: England and Wales, compares 2011 Census accommodation type and Valuation Office Agency (VOA) property type using linked data. The purpose is to help users understand how these data sources compare, and to seek feedback on how alternative sources of information may meet user needs for property type statistics.
The second publication, Admin-based statistics for floor space, feasibility research: England and Wales, demonstrates measures of property size (floor space) from administrative data. The purpose is to help users understand what property size statistics can be generated from administrative data, what they can and cannot be used for, and to seek feedback on their usefulness.
To complete our White Paper (PDF, 967KB) commitments and to provide assurance to users for the integrated census output on VOA number of rooms, which replace the previous number of rooms question, we will publish more of our research in 2020.
A summary and a more detailed methodology working paper on donor-based imputation methods for admin data will explore whether number of rooms from VOA data can undergo edit and imputation. This builds on research presented on Donor-based imputation methods for admin data: How to replace the number of rooms question on the Census at the 4th International Conference on Admin Data Research. This was also presented at the external assurance panel in June 2019 (PDF, 153KB).
We will also publish a Quality Assurance of Administrative Data (QAAD) for the VOA data to assure users that the data are of sufficiently high quality to produce a statistical output on the number of rooms.
We continue to review the feasibility of producing integrated outputs on income, which is highly dependent on the supply of data. Decisions will be made once we better understand, in particular, the timings, in relation to census outputs.Back to table of contents