Housing question development for Census 2021

In December 2018, the government presented to Parliament a White Paper Help Shape our Future: The 2021 Census of Population and Housing in England and Wales. This outlined the Office for National Statistics's (ONS's) proposal to collect information on housing to meet the needs for planning, decision-making and resource allocation for future housing needs and public services. At the local area level, the census is currently the only source of nationally comparable information on housing and is used widely in developing local plans and housing projections.

We will be asking questions on the same topics as the 2011 Census on housing for Census 2021, with the exception of number of rooms, which will not be asked. We will collect housing data in a way that is consistent with previous censuses.

The questions used to collect data on the topic of housing are:

• type and self-containment of accommodation
• number of bedrooms
• type of central heating
• tenure and type of landlord
• number of cars or vans

Census 2021 will be an online-first census, with a target of 75% online returns. We have reviewed and tested each question online and have recommended additional design and functionality amends specific to the electronic questionnaire. We have also recommended additional changes to the question designs following our stakeholder engagement, evaluation and testing programmes. These changes will be detailed in this report.

The main recommended changes to questions designs are:

• to remove the number of rooms question
• to add and update the response options to the type of central heating question to better meet users' needs for these data
• to add a response option to the type of accommodation question to better meet users' needs for these data
• to add a response option to the number of cars or vans question to improve data quality
• to re-design the self-containment question to improve the quality of data captured and reduce burden for respondents

This report provides links to previously published research and the findings of additional testing that led to the final recommended questions for Census 2021 in England and Wales. The questions and response options for Census 2021 have now been finalised through the census secondary legislation: The Census (England and Wales) Order 2020 and Census Regulations for England and for Wales.

In June 2015, the Office for National Statistics (ONS) held a formal, 12-week consultation process asking census users for their views on the topics that were required in the questionnaire 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

A detailed summary of the consultation responses relating to the topic of housing can be found in the housing topic report (PDF, 1.7MB). In this report, the ONS made clear commitments to the public to:

• investigate the impact that asking one question on number of rooms or number of bedrooms would have on acceptability and outputs

• investigate using Valuation Office Agency (VOA) data as an alternative to asking questions on number of rooms or bedrooms

• work with stakeholders to review the responses to the "type of central heating" question

• continue to collect information on all other housing topics

We have provided an update on how we met these commitments in Annex 1.

Following this, we began a comprehensive programme of research and development. A full list of the tests used in the development of the questions on housing is shown in Annex 2. Further details are provided in the Summary of testing for Census 2021.

The tests utilised a range of qualitative and quantitative research methods. A short description of the different research methods and sampling techniques is given in the Question and questionnaire development overview for Census 2021. Testing included participants living in varied accommodation and from a wide range of backgrounds.

In December 2018, the government presented to Parliament a White Paper Help Shape our Future: The 2021 Census of Population and Housing in England and Wales. This outlined our proposal to collect information on the topic of housing and stated our recommendation not to collect data on number of rooms. Further information can be found in Annex 1.

The question recommendations for Census 2021 are now finalised. We have evaluated the questions for their potential impact on data quality, public acceptability, respondent burden, financial concerns and questionnaire mode. We have presented details of this evaluation in Annex 3.

As set out in the White Paper Help Shape our Future: The 2021 Census of Population and Housing in England and Wales, we recommended that the census does not include a question on number of rooms. We would investigate the use of alternative sources for these data.

Number of rooms and number of bedrooms

A question on the number of rooms in a dwelling has been asked on the census since 1891. The data allow users to derive measures of under- and over-occupancy. A question on the number of bedrooms in a dwelling was asked for the first time in the 2011 Census, as this was a new measure of overcrowding introduced by the government. The question on number of rooms referenced the question on number of rooms, so the two questions had to be shown together.

2011 Census question

How many rooms are available for use only by this household?

Do NOT count:

• bathrooms
• toilets
• halls or landings
• rooms that can only be used for storage such as cupboards

Count all other rooms, for example:

• kitchens
• living rooms
• utility rooms
• bedrooms
• studies
• conservatories

If two rooms have been converted into one, count them as one room

(Write in) Number of rooms

How many of these rooms are bedrooms? Include all rooms built or converted for use as bedrooms, even if they are not currently used as bedrooms

(Write in) Number of bedrooms

Redesigning the number of bedrooms question

Following the consultation, we removed the number of rooms question and redesigned the number of bedrooms question as an independent question. We tested (2018:35) the new question both with and without the instruction.

How many bedrooms are available for use only by this household? Include all rooms built or converted for use as bedrooms, even if they are not currently used as bedrooms

(Write in) Number of bedrooms

Feedback from testing

All respondents answered the question correctly with or without the instruction. The new question stem included the phrase "available for use", which respondents found helpful and correctly included how many bedrooms that were officially converted or not.

Respondent feedback led to the recommendation to include the first part of the instruction, as this clarified what to do if room conversions had occurred, and not to include the second part of the instruction, as respondents understood this through reading the question and first part of the instruction.

The starting point for testing the questions on housing was the 2011 Census question designs. You can see these questions on the 2011 Census Household Questionnaire (PDF 1.8 MB).

In all questions, we are using name and address piping functionality to simplify the questionnaire and make it easier for online respondents to understand.

Type of accommodation

The question on type of accommodation was first asked in the 2001 Census. This question provides a count of dwelling, including vacant dwellings. To aid respondent understanding, the response options were grouped into subsets.

To ensure that the question could be displayed on a single screen, we adapted it into a two-stage question.

Stage 1

What type of accommodation is [first line of address]?
[ ] Whole house or bungalow
[ ] Flat, maisonette or apartment
Including bedsits
[ ] Caravan or other mobile or temporary structure

Respondents who selected "Whole house or bungalow" at Stage 1 were presented with the Stage 2A question.

Respondents who selected "Flat, maisonette or apartment" at Stage 1 were presented with the Stage 2B question.

The answer option chosen in Stage 1 is repeated in the Stage 2 question stems; this is so that if a respondent mis-clicks in the first stage, they can see the error in Stage 2 and go back and correct themselves.

Respondents who selected "Caravan or other mobile or temporary structure" at Stage 1 proceeded to the next question.

Stage 2A

Is your house or bungalow:
[ ] Detached
[ ] Semi-detached
[ ] Terraced
Including end-terrace

Stage 2B

Is your flat, maisonette or apartment:
[ ] In a purpose-built block of flats or tenement
[ ] Part of a converted or shared house
Including bedsits

[ ] In a commercial building
For example, in an office building, hotel, or over a shop

Testing the two-stage online question

The online question was cognitively tested (2018:35). We also tested the paper version of the question, which was identical to the 2011 question.

Online, most respondents were able to find and identify their accommodation type quickly. However, some respondents who lived in purpose-built flats in a housing-style construction would select "Whole house or bungalow" rather than the correct answer of "Flat, maisonette or apartment".

Based on the test findings, we added a line of guidance under the response option:

[ ] Flat, maisonette or apartment
Including purpose-built flats and flats within converted or shared houses

Following the test, we recommended replacing the partial question stems with full questions, in line with Government Digital Service's Service Standard:

• Which of the following is your house or bungalow?
• Where is your flat, maisonette or apartment?

Adding a "Part of another converted building" response option

Feedback from the 2019 Rehearsal (2019:15) found that the type of accommodation question did not cover respondents who live in some kind of converted building. Therefore, we have recommended adding a further response option to the question: "Part of another converted building (for example, former school, church or warehouse)" within the subset for "A flat, maisonette or apartment". The response option was adopted from the Government Statistical Service (GSS) Harmonised Principle.

Self-containment of accommodation

A question on self-containment of accommodation was first asked in the 1971 Census. This question provides a count of dwellings shared by two or more households.

Reviewing respondent understanding of "self-contained" accommodation

The 2011 Census question was "Is this household's accommodation self-contained?" The question stem was followed by a definition of "self-contained". The "Yes" response reiterated this definition: "Yes, all the rooms are behind a door that only this household can use".

We tested the 2011 question in cognitive interviews (2018:35) to see if "self-contained" was widely understood. We also tested a version that incorporated the instruction into the question stem. The testing did not include anyone whose accommodation was not self-contained.

Participants answered both question variants correctly, but overall participants preferred the question stem that provided a definition: "Are all the rooms in this accommodation behind a door that only this household can use?". They found the 2011 Census wording too technical.

We also tested a version of the response options where the "No" option was expanded rather than the "Yes" option. Participants preferred the longer "No, one or more rooms are shared with another household" response option helpful compared with repeating the definition on the "Yes" option or removing it completely.

Recommendation

Following the testing, we removed the term "self-contained" from the question and included the full definition in the question stem. We kept the longer "No" rather than the longer "Yes" response from 2011 as respondents expect minority categories, rather than majority categories, to require further explanation.

For online, because of this being the first question using the term "household", we also added a guidance accordion that defines a "household".

Type of central heating

A question on type of central heating was introduced in the 2011 Census. This question supports the monitoring of energy efficiency and fuel poverty policies as well as the allocation of resources to support policy-related interventions.

In our response to the 2021 Census topic consultation (PDF, 1.7MB), we committed to exploring additional response options and to re-design the type of central heating question to meet users' needs and capture better quality data.

2011 Census question

What type of central heating does this accommodation have?
Tick all that apply, whether or not you use it
Central heating is a central system that generates heat for multiple rooms
[ ] No central heating
[ ] Gas
[ ] Electric (including storage heaters)
[ ] Oil
[ ] Solid fuel (for example wood, coal)
[ ] Other central heating

Testing the central heating question

We conducted a cognitive test (2018:35) of variants of the question stem, instructions and response options.

Question stem

We tested two alternative question stems:

What fuel does the central heating in this accommodation use?

How do you mainly heat this accommodation?

The question stem using the term "fuel" led to a number of incorrect responses as it prompted respondents to consider their home's fuel in a broader sense. This led to them including all the energy sources the accommodation used as well as what fuels the central heating, incorrectly.

The question stem "How do you mainly heat this accommodation?" was preferred for ease and simplicity. However, it collected a different concept, namely what is mainly used to heat the accommodation, rather the central heating energy source.

Respondents considered the 2011 Census question stem "What type of central heating does this accommodation have?" to be much easier to understand and less likely to drive error. It had a clearer focus on the type of central heating the property has installed and will be retained for Census 2021.

Instructions

We also tested whether the instructions defining central heating and clarifying that it does not need to be used to be included were required. We found that for those with central heating that they regularly used, the instructions were not required. However, it was helpful for those without central heating and those who did not regularly use it, so we recommend that these instructions are retained.

The instruction to "Tick all that apply, whether or not you use it" cannot be replicated on the electronic questionnaire, as "Select all that apply" forms part of the online template. Therefore, we amended the first online instruction to read "Include all central heating systems that generate heat for multiple rooms whether or not the system is used".

Response options

In our response to the 2021 Census topic consultation (PDF, 1.7MB), we committed to exploring additional response options to the type of central heating question to meet users' needs. Following consultation with data users, we recommended increasing the number of response options by:

• separating "mains gas" from other forms of gas to identify areas that could benefit from expansion of the gas grid

• separating "wood" from other solid fuels to help monitor decarbonisation goals

• adding "renewable energy" heating systems to monitor the increasing number of users of this type of heating

We investigated what the wording of the new response options should be.

Gas

In 2011, there was a single response option for gas. In 2021, this will be split into "Mains gas" and other. We tested naming this option "Tank or bottled gas" and "Other gas (for example Liquid Petroleum Gas and Biogas)". Both versions were understood; however, when considered in context, "Tank or bottled gas" was preferred because it avoided the use of the word "Other", which is visually similar to the "Other central heating" option.

Wood

There are different definitions of "renewable energy" in use, some include "wood" and some do not. Therefore, as well as adding an option for renewable energy, we have added one for wood, taking it out of the "solid fuel" category, and allowing multiple users to collapse the data and use them in different ways, dependent on their needs.

In testing, respondents were not sure how "wood" was defined, so examples were added; the response option now reads "Wood (for example, logs, waste wood or pellets)", and it has been positioned before either solid fuel or renewable energy so that respondents do not select one of those in error.

Renewable energy

Different wordings for the renewable energy response option were tested, with and without examples. We concluded that the examples were needed to define what this option meant but that using "solar panels" was not viable as participants with solar panels selected this option whether or not they were part of their central heating system. The final response option was agreed by the Department for Business, Energy and Industrial Strategy (BEIS) and is "Renewable energy (for example, solar thermal or heat pumps)"

District or communal heat network

In addition to the proposed changes from the topic consultation, we added "District or communal heat network" to monitor the increasing number of users of this form of central heating. "District or communal heat network" was the most popular heating written in the "Other" write-in box (2018:41).

Tenure and type of landlord

A question on tenure of accommodation was first asked in 1961, and a question on the type of landlord was first asked in 2001. The data collected by these questions allow local authorities and other organisations to meet future housing needs and understand changes in the private rental sector.

2011 Census questions

Does your household own or rent this accommodation?
Tick one box only
[ ] Owns outright
[ ] Owns with a mortgage or loan
[ ] Part owns and part rents (shared ownership)
[ ] Rents (with or without housing benefit)
[ ] Lives here rent free

Who is your landlord?
Tick one box only
[ ] Housing association, housing co-operative, charitable trust, registered social landlord
[ ] Council (local authority)
[ ] Private landlord or letting agency
[ ] Employer of a household member
[ ] Relative or friend of a household member
[ ] Other

Using the terms "household" or "householder" in the tenure question

Feedback from the 2011 Census identified a small number of respondents who did not know how to answer the tenure question where there were different levels of ownership within the household. Therefore, we wanted to test if the term "householder" would be better understood than "household".

We conducted a review (2018:9) to understand how participants understood the difference between "household" and "householder" in the question stem. All participants rented from a landlord.

The use of the term "householder" caused some confusion, as participants were unclear if this referred to themselves or their landlord. Participants in shared accommodation preferred the use of the term "household" and were able to answer correctly; therefore, this term will be retained.

Confirming the need for explanatory text on response options

We cognitively tested removing "shared ownership" and "with or without housing benefit" in the response option example texts (2018:35). However, the lack of the "with or without housing benefit" example made the response less clear for those on housing benefit. Both examples were considered helpful, and participants correctly answered the question when they were present, so they will be retained.

Type of landlord question

We tested (2018:35) the 2011 Census type of landlord question to confirm respondent understanding. Respondents found this question clear, straightforward and easy to answer. Respondents were clear which categories they fell under.

We considered collapsing "Council (local authority)" into the top response option "Housing association, housing co-operative, charitable trust, registered social landlord" as there is not a separate user need for the response option. However, when tested it increased burden on respondents as the proposed combined response option was complex.

Therefore, we have recommended using the 2011 Census question with one minor modification in Census 2021. We have removed the brackets in the second response option to read "Council or local authority" to comply with the Government Digital Service's Service Standard.

Number of cars or vans

A question on the number of cars or vans available for private use by members of the household has been included in the census since 1971. The information is used across central and local government and other sectors to support transport modelling, transport planning and associated resource allocation.

2011 Census question

In total, how many cars or vans are owned, or available for use, by members of this household?
Include any company car(s) or van(s) available for private use
[ ] None
[ ] 1
[ ] 2
[ ] 3
[ ] 4 or more, write in number (Write in)

Investigating simplifying the number of cars or vans question

The 2011 Census Quality Survey (PDF, 1.4MB) and Census 2021 topic consultation did not identify any issues with the 2011 question. However, the question includes a long question stem, an additional instruction and a write-in response. We wanted to investigate if a simplified question would reduce respondent burden without affecting outputs.

We conducted cognitive interviews (2018:35) to investigate if the instructional text was required. We found that the instruction prompted participants to think more broadly about company cars, vans and leased vehicles.

To reduce respondent burden further, increase data quality and reduce the coding costs associated with the write-in response, we have recommended increasing the number of response options to five. We have split "4 or more" into "4" and "5 or more", with the write-in remaining attached to the "5 or more" response option. This is because in the 2011 Census, 378,000 households ticked the "4 or more" response. Of these, 44.3% wrote in "4" and 41.5% left the field blank.

We also removed the bracketed "(s)" to comply with the Government Digital Service's Service Standard.

Question design for online

In addition to the testing described earlier, the housing questions have undergone significant user experience (UX) testing (2017:2, 2018:2, 2019:1 and 2020:2). UX testing focuses on understanding user behaviours as people interact with online services. Through observation techniques, task analysis and other feedback methodologies, it aims to develop a deep knowledge of these interactions and what it means for the design of a service.

UX research has taken place on a rolling basis since 2017. Before the Census Rehearsal in October 2019, 458 interviews were conducted at 99 events. The UX testing programme will continue through 2020. All participants are purposively selected to include a wide range of ages and digital abilities.

UX testing included various iterations of the housing questions described elsewhere in this report. Feedback from this research informed decisions made on the design of these questions. For more information on UX testing, see the Question and questionnaire development overview for Census 2021.

Welsh language question development

In 2018, an external agency with Welsh-speaking researchers was commissioned to undertake a series of cognitive interviews (2018:40). The qualitative research tested public acceptability and comprehension of amended and newly designed census questions in Welsh. The questions were tested with people across Wales with varying dialects and Welsh language proficiencies.

Subsequent changes to the text or questions across the census questionnaires were translated by our contracted specialist Welsh language translation service provider. The recommended questions adhere to Cymraeg Clir guidelines and were quality assured by the Welsh Language Census Question Assurance Group (the Assurance Group). This group includes Welsh language and policy experts from the Welsh Language Commissioner and the Welsh Government and was convened to give advice on the accuracy, clarity and acceptability of the language as well as other policy issues pertaining to the Welsh language and bilingual design.

Across the suite of housing topic questions, we have adhered to Cymraeg Clir guidelines and made minor grammatical changes across the questionnaire.

Central heating

In English, we added the response option "renewable energy". The findings from testing (2018:40) recommended to use "Amgen" meaning "alternative" alongside "Ynni adnewyddadwy" meaning "renewable energy" in Welsh because of different ages being familiar with different terminology.

The Assurance Group discussed these terms and it was decided that the "amgen" term ("alternative") is not conceptually the same thing and could be interpreted as "other". The decision was to remove "amgen" and use "Ynni adnewyddadwy" only. We also recommend using "Ynni" (energy) rather than "Egni", as although "egni" is used in schools, "Ynni" was more universally known.

We engaged with Welsh Government stakeholders on the appropriate translations for "solar thermal" and "heat pumps". The following translations were the most appropriate respectively - "solar thermol" and "pympiau gwres". It was also agreed that, for the Welsh language questionnaire only, the examples should be reversed so that heat pumps is before solar thermal to avoid mutations in the Welsh language.

The translation "Rhwydweithiau gwres rhanbarthol neu gymunedol" meaning "District or communal heat networks" was agreed.

The design of the questions recommended for Census 2021 was informed by the research and testing detailed in this report. The questions and response options for Census 2021 have now been finalised through the census secondary legislation.

The paper questionnaires used in Census 2021 are available.

Guidance text and instructions are not part of the legislation, but we consider these to be finalised as well. However, it is possible that guidance text or instructions may change if there is enough evidence to support doing so.

Type of accommodation

Figure 1: Online question on type of accommodation in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 1: Online question on type of accommodation in English and Welsh

.png (133.7 kB)

Figure 2: Online question on type of house or bungalow in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 2: Online question on type of house or bungalow in English and Welsh

.png (85.3 kB)

Figure 3: Online question on type of flat, maisonette or apartment in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 3: Online question on type of flat, maisonette or apartment in English and Welsh

.png (165.8 kB)

The address entered in the first part of the questionnaire is piped through in the type of accommodation question for English and Welsh designs.

Routing

If you select the response “Whole house or bungalow” in the first stage of the question, you proceed to the second stage of the question for houses and bungalows.

If you select the response “Flat, maisonette or apartment” in the first stage of the question, you proceed to the second stage of the question for flats, maisonettes and apartments.

Those who select “Caravan or other mobile or temporary structure” continue to the next housing question.

Self-containment of accommodation

Figure 4: Online question on self-containment of accommodation in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 4: Online question on self-containment of accommodation in English and Welsh

.png (194.5 kB)

Guidance accordion

What we mean by “Household”

This is one person living alone or a group of people, not necessarily related, living at the same address, who share cooking facilities and share a living room or sitting room or dining area.

Number of bedrooms

Figure 5: Online question on number of bedrooms in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 5: Online question on number of bedrooms in English and Welsh

.png (114.1 kB)

In the electronic questionnaire respondents must put a numerical value in the write-in box rather than word values, for example, “0” rather than “zero”. On the paper questionnaire, there is only space to write two digits. Respondents can enter up to a maximum value of 99 bedrooms for online and paper.

Type of central heating

Figure 6: Online question on type of central heating in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 6: Online question on type of central heating in English and Welsh

.png (275.9 kB)

The address entered in the first part of the questionnaire is piped through in the central heating question for English and Welsh designs.

The paper questionnaire includes two instructions: “Tick all that apply, whether or not you use it” and “Central heating is a central system that generates heat for multiple rooms”. In the electronic questionnaire, these two instructions are split differently: “Include all the central heating systems that generate heat for multiple rooms whether or not the system is used” and “Select all that apply”.

Tenure and type of landlord

Figure 7: Online question on tenure in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 7: Online question on tenure in English and Welsh

.png (178.5 kB)

Figure 8: Online question on type of landlord in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 8: Online question on type of landlord in English and Welsh

.png (152.9 kB)

On the electronic questionnaire, the first line of the address is piped into the question stem, whereas on the paper question, the question stem is "Does your household own or rent this accommodation?"

Routing

In the electronic questionnaire, where respondents state that no-one is living in the household in the first section "Who usually lives here?", they are automatically routed out to the end of the questionnaire before this question.

On the paper questionnaire, there is a separate instruction box where respondents are instructed to "go to the declaration page" if "no-one usually lives at the accommodation".

Respondents who select the last three response options: "Part-owns and part-rents"; "Rents"; or "Lives here rent-free" are routed to the landlord question (Figure 9).

Number of cars or vans

Figure 9: Online question on number of cars or vans in English and Welsh

Source: Office for National Statistics – Housing question development for Census 2021

Download this image Figure 9: Online question on number of cars or vans in English and Welsh

.png (167.7 kB)

In the electronic questionnaire, the write-in box "Enter the number of cars or vans" is shown when respondents select the response option "5 or more". Respondents must put a numerical value in the write-in box rather than word values, for example, "0" rather than "zero". This is to speed up the processing on this question. The range of values respondents can enter are between "0" to "20". On the paper questionnaire, there is only space to write two digits.

Question order

Since the 2011 Census, the only change to the order for Census 2021 is the removal of the number of rooms question from the questionnaire.

On paper, a routing question has been inserted before the tenure question. This ensures that households with no usual residents do not answer the questions on tenure, type of landlord and number of vans. This routing was provided in 2011 but as instructions on the relationship matrix question instead.

Question evaluation

The question designs put forward in this report are based on extensive research and assessment using evaluation criteria that were set out in the publication The 2021 Census - Assessment of initial user requirements on content for England and Wales: Response to consultation (PDF, 796KB).

The evaluation considered the potential impact that including a topic on the census would have on data quality, public acceptability, respondent burden, financial concerns and questionnaire mode. The evaluations were used in conjunction with the user requirements criteria to steer the development of the census questions and questionnaire.

A topic that has been assessed as having a "High" potential for impact is closer to the threshold for exclusion from the census than a topic that has been assessed as having a "Low" potential for impact.

After completing the research and development phase, we evaluated the recommended questions against the same criteria using an updated tool that considers the type of evidence we have available and the Census 2021 context. A description of this updated evaluation tool is provided in the Question and questionnaire development overview for Census 2021.

All questions meet our thresholds to ensure reliable information will be collected in Census 2021.

Table 2 provides the updated evaluation scores for the questions on housing. We have presented the evidence used to assess questions as having a "Medium" or "High" potential for impact on the evaluation criteria in Annex 3.

Question harmonisation

As in previous censuses, there will be separate censuses in Scotland and Northern Ireland. The questions for England and Wales have been developed through close collaboration with National Records of Scotland (NRS) and the Northern Ireland Statistics and Research Agency (NISRA), which are responsible for conducting the censuses in Scotland and Northern Ireland, respectively.

We recognise that each country has its own user and respondent needs. However, we aim for harmonisation of census questions and topics where possible to produce UK-wide statistics that are consistent and comparable.

On the topics of landlord, central heating, self-containment of accommodation, type of accommodation and tenure, there are several differences in the questions asked on Census 2021 in Scotland and Northern Ireland because of differing data user needs. These are described in Annex 4.

The housing questions were developed for use in the context of Census 2021 in England and Wales, a mandatory household form. Therefore, it is possible that in different contexts, such as social surveys, a different approach may be more suitable. Government Statistical Service (GSS) Harmonised standards exist for household accommodation and tenure.

In July 2020, we published a summary and methodology paper that provide an overview of how we plan to address missing values when replacing the number of rooms question in Census 2021. We also published a quality assessment of administrative data for the use of VOA data in Census 2021 to provide further assurances to our users. We will conduct further analysis to ensure we are operationally ready for Census 2021, and to provide further assurance for the replacement of the number of rooms question in Census 2021. Our next publication will explore differences in occupancy ratings derived from Census 2011 and linked Valuation Office Agency number of rooms data.

In the Census 2021 topic consultation response, the Office for National Statistics (ONS) committed to continuing to ask questions on the housing topic and develop these for online. We made three further commitments related to housing.

Explore the impact of removing one of the questions about rooms in the household

In our response to the Census 2021 topic consultation (PDF, 1.7MB), we stated that we did not believe it was appropriate to ask two questions designed to meet a single information need if there is no clear requirement to do so. In this case, the data collected by both number of rooms and number of bedrooms were used to identify over- and under-occupancy.

The 2011 Census Quality Survey (PDF, 1.4MB) asked the same questions as the 2011 Census again to measure agreement between the two sources. If all respondents provided the same answers to the Census Quality Survey that they provided in their census return, then agreement would be 100%.

This identified that the quality of data was higher for the "number of bedrooms" question, with 91.4% agreement, compared with just 66.5% agreement for the "number of rooms" question. The findings indicated that the number of rooms question was not a source of reliable data, indicating respondents may include or exclude certain rooms when asked at different times.

We therefore revised and tested a standalone "number of bedrooms" question to ensure the data would continue to meet the user need for counting under- and over-occupancy as discussed in this report.

Explore the potential of using Valuation Office Agency data as an alternative source of information about number of rooms

In our response to the Census 2021 topic consultation (PDF, 1.7MB), we outlined our intention to investigate the use of Valuation Office Agency (VOA) data to provide information on number of rooms and bedrooms.

European Regulation (EC) No 763/2008 requires the ONS to provide census-derived statistics, or equivalent data, on number of rooms to Eurostat, cross-tabulated with a limited number of other sub-topics.

In March 2015, Parliament approved an Information Sharing Order to allow the VOA to share information about addresses with the ONS, including physical attributes, which covers the number of rooms.

Consultation on using Valuation Office Agency data

In June 2017, we published Estimating the number of rooms and bedrooms in the 2021 Census: An alternative approach using VOA data. This research looked at VOA data as a potential alternative source for estimating the number of rooms and bedrooms.

We then held a public consultation from 28 June to 25 September 2017, inviting users to respond to this publication.

The consultation revealed that data users were able to calculate under- and over-occupancy from either the number of rooms or the number of bedrooms question. More users relied on the data from the number of bedrooms question. We found no evidence that both questions were required on the census.

Respondents to the consultation highlighted several concerns with using VOA data rather than census data:

• there was a significant difference between the 2011 Census and VOA data for the number of properties classified as over-occupied

• there were concerns with how often the VOA data are updated

• VOA records data at the address level and the census records data at the household level; respondents were unclear how this difference would impact data use

We concluded that relying on VOA data alone would likely impact data use and comparability with census data. Therefore, we will continue to ask the "number of bedrooms" question to derive data on under- and over-occupancy in Census 2021.

We will continue to work with VOA data to enhance outputs for Census 2021. Further research and an update on achieving this will be published in 2020.

2021 Census White Paper commitments

In the December 2018 White Paper Help Shape our Future: The 2021 Census of Population and Housing in England and Wales, we further committed to continue work with the VOA to explore the feasibility of supplementing the census questions with administrative data on total number of rooms, living space and property type. We have published research on the latter two. We do not propose including questions on living space or property type.

Engage with stakeholders to explore requirements for the type of central heating question and ensure it fully meets user needs

We worked with various stakeholders to update the list of response options to the type of central heating question, including the Department of Energy and Climate Change, the Welsh Government, and local authorities. The final list of response options was designed to meet the data needs of these stakeholders.

Recommended Census 2021 response options that were tested, and the findings of testing, are discussed in more detail throughout this report. The examples of central heating types that are shown alongside the response options were provided by the Department for Business, Energy and Industrial Strategy (BEIS).

References to tests take the form (Year: Test number). "Year" refers to the calendar year the test was undertaken in, and the test number is the position of the test within the year considering all testing that took place in that year. For example, the fifth test conducted in 2017 would be referenced as (2017:5).

A full description of each of these items can be found in Summary of testing for Census 2021.

Evaluation of the type of accommodation question

The type of accommodation question has been assessed as having a "Low" potential for impact on data quality, public acceptability, respondent burden and financial concerns.

Potential for impact on questionnaire mode: "Medium"

On the electronic questionnaire, type of accommodation was split into a two-stage question to reduce the need for scrolling on a mobile device, in line with the Government Digital Service's Service Standard. The electronic questionnaire contains "address piping" functionality to improve the respondent journey through this question, which is not viable on the paper questionnaire.

Evaluation of the self-containment of accommodation question

The self-containment of accommodation question has been assessed as having a "Low" potential for impact on data quality, public acceptability, financial concerns and questionnaire mode.

Potential for impact on Respondent Burden: "Medium"

The self-containment of accommodation question has a long question stem of over 20 words. In the 2019 Rehearsal, respondents spent a longer time on this question (14 seconds) than on the evaluation threshold (10 seconds).

Evaluation of the number of bedrooms question

The number of bedrooms question has been assessed as having a "Low" potential for impact on all five assessment criteria: data quality, public acceptability, respondent burden, financial concerns and questionnaire mode.

Evaluation of the type of central heating question

The type of central heating question has been assessed as having a "Low" potential for impact on public acceptability, financial concerns and questionnaire mode.

Potential for impact on data quality: "Medium"

The central heating question asks for information that cannot be observed in some cases. Testing, including the 2019 Rehearsal, has shown that some respondents have difficulty understanding the concept of central heating.

Potential for impact on respondent burden: "High"

The type of central heating question requires respondents to read the question stem and additional instructions. They must also read a list of 10 response options, with some examples, as this is a select all that apply question.

Evaluation of the tenure question

The tenure question has been assessed as having a "Low" potential for impact on public acceptability, respondent burden and financial concerns.

Potential for impact on data quality: "Medium"

The central heating question asks for information that cannot be observed in some cases. 2019 Rehearsal feedback showed that some respondents had difficulty in understanding this question.

Potential for impact on questionnaire mode: "Medium"

The tenure question on the electronic questionnaire has added functionality to improve the data collected by users and improve user experience of the questionnaire. These functionalities are "address piping" and validation forcing respondents to select a response option before they can continue to the next question.

Evaluation of the type of landlord question

The type of landlord question has been assessed as having a "Low" potential for impact on all five assessment criteria: data quality, public acceptability, respondent burden, financial concerns and questionnaire mode.

Evaluation of the number of cars and vans question

The number of cars and vans question has been assessed as having a "Low" potential for impact on all five assessment criteria: data quality, public acceptability, respondent burden, financial concerns and questionnaire mode.

Throughout the development of the Census 2021 questionnaires, the Office for National Statistics (ONS), National Records of Scotland (NRS) and Northern Ireland Statistics and Research Agency (NISRA) have worked together to ensure harmonisation of the outputs from the housing questions where possible.

Based on the proposed designs, there are some important differences between UK statistical offices, resulting from differing data user needs.

Question design

Landlord

The Northern Ireland question stem "Who do you rent from?" differs from the England, Wales and Scotland question stem "Who is your landlord?". The response options regarding Housing associations, Council or local authorities in Northern Ireland, England and Wales, and Scotland all differ. Scotland have fewer response options on the landlord question compared with Northern Ireland, England and Wales.

Central heating

All three statistical offices ask the same question with the same instructions. However, the list of response options is different in each country.

In Northern Ireland only, there is a question on renewable energy systems.

Type of accommodation

Scotland and Northern Ireland differ in the type of accommodation question response option "in a purpose-built block of flats or tenement". Scotland uses "in a tenement or purpose-built block of flats (including '4-in-a-block')", while in Northern Ireland they use "in a purpose-build block of flats".

Tenure

In Scotland only, in the question on tenure, they have an extra response option that is "Owns with shared equity (for example, LIFT, Help-to-Buy)" because of differing user needs. Northern Ireland's response option order differs with "Owns with a mortgage or loan" being first.

Self-containment and number of bedrooms

In Scotland only, the self-containment question uses the 2011 Census design question stem, "Are all the rooms in this accommodation behind a door that only this household can use?" compared with the England and Wales question stem "Are all the rooms in this accommodation, including the kitchen, bathroom and toilet, behind a door that only this household can use?".

Scotland recommend using the 2011 Census design "No" response option, while England and Wales recommend "No, one or more rooms are shared with another household".

Northern Ireland do not ask questions on number of bedrooms or self-containment of accommodation.

Northern Ireland are using administrative data to gather date on number of bedrooms. There is less user need for a question on self-containment in Northern Ireland than in England and Wales and Scotland.