Second address, migration and citizenship 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) recommendation to collect information on the topics of second address, migration (including short-term, long-term and internal migration) and citizenship.

This would meet the need for population estimates and projections down to the local level that underpin planning and resource allocation. These question also contribute to our understanding of internal and international migration by helping us to arrive at a usual resident base.

The questions used to collect data on the topics of second address, migration and citizenship are:

• second address and second address type

• full-time education and term-time address

• country of birth, date of arrival in the UK and intended length of stay in the UK

• usual address one year ago

• passports held

The main changes to questions designs from those used in the 2011 Census are:

• adding guidance to the second address question to help respondent understanding

• adding a "Partner's address" response option to the second address type question

• combining the response options "Less than 6 months" and "6 months or more but less than 12 months" into a single option "Less than 12 months" on the intended length of stay in the UK question

As Census 2021 will be online-first, the question development also covers the online development research and testing of these questions.

Since the publication of the White Paper, the Office for National Statistics (ONS) has conducted and concluded the final phase of testing on the topics of second address, migration and citizenship.

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.

The evidence base for the finalisation of the questions for Census 2021 is discussed in Section 4: Research that led to the recommended Census 2021 question designs.

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 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

This response included a commitment to continue to collect information on migration (including long-term, short-term and internal migration) and citizenship. We also made a commitment to maintain the same enumeration bases as used in 2011. These included:

• the second residence population

• the out of term-time population

• short-term residents

Production of these enumeration bases requires the continued collection of information on second address, including term-time address.

A detailed summary of the consultation responses relating to the topic of migration can be found in the migration topic report (PDF, 1.7 MB). In this report, the ONS made clear commitments. 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. We have provided a full list of the tests used in the development of the the questions on second address, migration and citizenship 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 from a wide range of backgrounds, including short-term and long-term international migrants, schoolchildren and students, and people who stay at a second address.

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

The questions on second address, migration and citizenship are fundamental to achieving census definitions of a usual resident base. You can find out more about these definitions in Output and enumeration bases: residential address and population definitions for Census 2021.

This section provides details on the testing of the questions on second address, migration and citizenship.

The starting point for the testing on the questions on second address and migration was the 2011 Census question designs, as shown in 2011 Census Household Questionnaire (PDF 1.8 MB).

References to tests are provided in the form (Year: Test number). Year references the calendar year the test was undertaken in. We have presented the full list of tests used in the development of the topics in Annex 2.

Second address and second address type

Questions on second address and the reasons for staying at that address were first included in the 2011 Census. The data are used to improve coverage of the census and reveal changes in society by showing more complex living arrangements.

This question worked well in the 2011 Census; however, we needed to review the response options to ensure they were still appropriate in 2021, and we also needed to ensure that the question worked well on the electronic questionnaire.

2011 Census questions

Do you stay at another address for more than 30 days a year?
[ ] No
[ ] Yes, write in other UK address below
(Address write-in)
OR [ ] Yes, outside the UK, write in country
(Country write-in)

What is that address?
[ ] Armed forces base address
[ ] Another address when working away from home
[ ] Student's home address
[ ] Student's term time address
[ ] Another parent or guardian's address
[ ] Holiday home
[ ] Other

Developing the questions for online collection

These questions cannot be shown together on the same screen page as it makes them inaccessible for screen readers and other assistive technologies. Therefore, we display these on separate screens. In this instance, this means that where a person has a second address, the capturing of the details of that second address will be done on a separate display screen.

The online write-in response for "Country" features search-as-you-type functionality, where suggestions are displayed as the respondent types in their answer. In accordance with the Government Digital Service's Service Standard, if respondents select a response option that indicates they live outside the UK, the country write-in response will appear on a separate screen.

The online questions about the second address type also include the first line or country of the second address. For example, the question might be "What type of address is 1, New Street?" or "What type of address is your address in Spain?".

Evaluating the second address type “Other” response option

Testing conducted prior to the 2011 Census identified a need for seven response options for the second address type question, including "Other". In 2011, 19.2% of second addresses were classified as "Other", and for Census 2021 we wanted to understand why there were so many of these responses.

In the 2017 Census Test (2017:7), we added an open write-in text box alongside the "Other" response. In total, there were 7,965 responses to the second address type question, and 21.3% of those used the write-in response. After discounting around 350 invalid responses (including multiple addresses and existing tick-box responses), the remaining 1,339 write-in responses were re-categorised against a list of second address types.

Of these write-in responses, 18.4% were categorised as a partner's address (3.1% of all responses to the question) and 16.9% were categorised as another relative's address (2.8% of all responses to the question).

These two response categories had a higher response rate than the "Armed forces base address" tick-box (2.1% of all responses). These findings led us to investigate whether there was a need for data for "Partner's address" or "Another relative's address".

We evaluated the proposed response options against the need for internal use of the data for alternative population estimates and the impact the new response would have on existing tick-boxes where the second address serves more than one purpose.

There is evidence that the number of people who stay at a partner's address for more than 30 days a year is increasing as the number of people in England and Wales aged 16 years and over who have never married or never registered a same-sex civil partnership has increased from 29.6% in 2002 to 34.6% in 2016 (Population estimates by marital status and living arrangements, England and Wales: 2002 to 2016), while the 2011 Census identified that 3.7% of the married population were not living together.

This evidence highlighted a clear need for a "Partner's address" response to ensure that couples who are living apart are counted accurately on the census.

The Census 2021 topic consultation report on demography (PDF, 1.6 MB) also identified a high user need for accurate data on household relationships for household projections and service planning. The data on couples living apart would strengthen these data by identifying individuals in relationships that are not necessarily living in the same household.

We did not identify a user need for data on residents staying at another relative's address. In addition, if we were to add this response option, there is a risk of transfer from the existing response options where the second address serves more than one purpose. For example, if the respondent stays with a relative while working away from home, they might only record the second address as "another relative's address". This could have a negative impact on second address outputs. Therefore, we do not recommended collecting data on "Another relative's address" in Census 2021.

We conducted cognitive interviews (2018:38) with participants who could speak, read and write in Welsh to assess the understanding of this question, including the proposed response option of "Partner's address", in Welsh. There was consensus among participants that the new response option "Cyfeiriad partner" (Partner's address) was appropriate.

We commissioned a quantitative survey (2018:44) to test the new question design with participants who either live at a second address or have a household member who lives at a second address. Over 24% of participants ticked the new "Partner's address" response option. The number of "Other" responses fell from 21.3% in the 2017 Census Test to 12.9% in the quantitative survey. These findings suggest the inclusion of "Partner's address" will contribute substantially to a reduction in the number of "Other" responses in Census 2021.

The outcome of our evaluation was to recommend the inclusion of a "Partner's address" response option to the second address type question for Census 2021.

Adding guidance to aid understanding of the question

In a cognitive test (2018:29), some participants hesitated to work out if they had been away from home for 30 days. They would usually resolve this by re-reading the question and interpreting it correctly to mean a single address other than their usual address.

Participants answering the electronic questionnaire took longer to understand the question because of the separation of the indicator question and address write-in. Participants answering the paper questionnaire were able to look ahead to see that the question refers to one other address as well as the list of expected second address types in the next question.

Following the findings of the cognitive test, we wanted to investigate whether additional instructions and guidance would help respondents understand the question.

For the electronic questionnaire, we added an instruction to clarify the intention of the question: "This could be more than 30 days in a row or divided across the year". Given the space constraints on the paper questionnaire, the instruction had to be short enough to fit on one line. Therefore, we shortened the instruction to "These days could be in a row or separate".

On the electronic questionnaire, we also added guidance in an accordion.

We commissioned a small-scale quantitative survey (2018:44) to test the new question designs with participants who either live at a second address or have a household member who lives at a second address. Around 79% of participants found the instruction "This could be more than 30 days in a row or divided across the year" somewhat or very useful. While only 9% of participants clicked to open the accordion "What do we mean by 'another address'?", 95% of those who did found the guidance somewhat or very useful. We have recommended adding both the instruction and accordion to the question.

Full-time education and term-time address

Questions collecting data on term-time address were first included in the 1991 Census. These questions ensure that full data from schoolchildren at boarding school and students are collected only once, from the address where they spend the most time. The data are used to create out-of-term population bases for local authorities with boarding schools and universities.

2011 Census questions

Are you a schoolchild or student in full-time education?
[ ] Yes
[ ] No

During term time, do you live:
[ ] at the address on the front of this questionnaire?
[ ] at the address in question 5?
[ ] at another address?

Note: The response option "address on the front of this questionnaire" referred to the residence where the census questionnaire was delivered. The response option "address in question 5" was the answer to the question "Do you stay at another address for more than 30 days a year?".

Developing the questions for online collection

Accessibility

To make the question accessible for screen readers and other assistive technologies, we have revised the wording of the 2011 Census question slightly from “During term time, do you live:”, to ”During term time, where do you usually live?”. This change has also been made on the paper questionnaires.

Variants of the full-time education question

We conducted testing (2018:44) on the question “During term time, where do you usually live?” to understand how respondents interpreted the phrase “usually live”. This was to investigate whether the question would need additional guidance.

Participants who were in full-time education, or those answering on their behalf, were asked a follow-up question: “In the current school or university term, how many days per week do you usually stay at your term-time address?” The aim was that “usually live” would be interpreted to mean four days or more per week. The results of the testing were that 90% answered that they stay at their term-time address for four days or more per week and 10% answered that they stay at their term-time address for three days or fewer per week.

These results confirmed that respondents generally understood the question as intended.

Adding a term-time address as a second address

We have added new online functionality to ask people with a different term-time address to add this as their second address. The two questions that capture the term-time address are based on the second address questions. This improves the data quality of the second address question.

Country of birth, date of arrival and intended length of stay

The question on country of birth has been on the census since 1841. This is a core question and part of the Census Act 1920. Questions on date of arrival in the UK and intended length of stay in the UK were introduced in 2011 to reach a usual resident base by counting usual and short-term residents.

2011 Census questions and routing

What is your country of birth?
[ ] England
[ ] Wales
[ ] Scotland
[ ] Northern Ireland
[ ] Republic of Ireland
[ ] Elsewhere, write in the current name of country
(Write in country)

Note: If the respondent answered England, Wales, Scotland or Northern Ireland to this question, they were routed around the next two questions.

If you were not born in the United Kingdom, when did you most recently arrive to live here?
Do not count short visits away from the UK
(Write in month and year)

Note: If a respondent arrived before 27 March 2010, they were routed around the next question.

Including the time you have already spent here, how long do you intend to stay in the United Kingdom?
[ ] Less than 6 months
[ ] 6 months or more but less than 12 months
[ ] 12 months or more

Reviewing the response options for intended length of stay in the UK

The Office for National Statistics (ONS) uses the data collected from the intended length of stay in the UK question to count short-term residents, allowing us to arrive at the usual resident base. We use the definition of a short-term resident to mean someone who intends to stay in the UK for less than 12 months. In the 2011 Census, two response options were used to count the short-term migrant population: “Less than 6 months” and “6 months or more but less than 12 months”. We wanted to review the reasons for having two responses to meet a single data need.

The number of respondents who intend to stay for less than 12 months is relatively small. In the 2011 Census, only 9.6% of respondents to this question selected “Less than 6 months” and 17.3% selected “6 months or more but less than 12 months”.

The census is only intended to be completed by those who have been in the UK for at least three months. Therefore, the response “Less than 6 months” only counts people who intend to stay between three and six months.

We proposed to stakeholders that we replace the two response options with a single response option: “Less than 12 months”. There was no user need provided for the two separate response options for counting short-term migrants. Therefore, we have recommended this change for Census 2021.

Developing the questions for online collection

Simplifying the date of arrival question

The date of arrival question includes the phrase “If you were not born in the United Kingdom”. On the paper questionnaire, this provides an additional measure to prevent respondents who were born in the UK from answering this question when they do not need to.

On the electronic questionnaire, this question will only be seen by respondents who answer “Republic of Ireland” or “Elsewhere” on the country of birth question. Therefore, we have shortened the question to read: “When did you most recently arrive to live in the United Kingdom?”

Automatic routing for the intention to stay in the UK question

The intention to stay in the UK question is only asked to respondents who arrived in the UK in the year prior to the census date. On the paper questionnaire, this is handled with a routing instruction. If the respondent arrived before the 21 March 2020, then the routing skips around the intention to stay question; if the respondent arrived on or after 21 March, then they are routed to the intention to stay question.

To reduce respondent burden on the electronic questionnaire, the date provided in the date of arrival question will automatically route respondents who arrived more than a year prior to the census date past the question on intention to stay.

The date of arrival question only collects data on month and year of arrival. However, if a respondent arrived in March 2020, then they are asked an addition question to determine if they arrived more or less than 12 months prior to the census reference date (21 March 2021):

Did you arrive in the UK on or after 21 March 2020?
[ ] Yes
[ ] No

Usual address one year ago

A question on usual address one year ago has been included on the census since 1961. The data contributes to understanding internal and international migration patterns.

2011 Census question

One year ago, what was your usual address?
If you had no usual address one year ago, state the address where you were staying
[ ] The address on the front of this questionnaire
[ ] Student term time/boarding school address in the UK, write in term time address below
[ ] Another address in the UK, write in below
(Write in address)
OR [ ] Outside the UK, write in country
(Write in country)

This question has not changed on the paper questionnaire in Census 2021.

Developing the questions for online collection

A two-stage question for usual address one year ago

For accessibility reasons, we have split the usual address one year ago question on the electronic questionnaire into multiple stages. The address entry form and “country” write-in response appear on separate screens to the initial question.

If a respondent selects “Student term-time or boarding school address in the UK” or “Another address in the UK”, they will be presented with the second-stage question, with address lookup functionality, where they can enter their address details. If they select “Outside the UK”, they will be presented with a second-stage question where they can write-in their country. The “country” write-in also includes search-as-you-type functionality.

Passports held

The question on passports held was introduced in the 2011 Census, as testing showed it was the most successful way to collect information on citizenship. The same question will be asked in Census 2021.

2011 Census question

What passports do you hold?
Tick all that apply
[ ] United Kingdom
[ ] Irish
[ ] Other, write in
(Write-in passport)
[ ] None

Updating the “Irish” response option

The 2011 response option “Irish” has been changed to “Ireland” for Census 2021. This aligns with the response used by the Northern Ireland Statistics and Research Agency (NISRA) in the census for Northern Ireland and uses the issuing country of the passport, matching the other response options.

Updating the instructions

During the 2019 Rehearsal (2019:15), the question included guidance about which documents can be included in the upper accordion.

However, feedback following the 2019 Rehearsal (2019:15) showed that some respondents asked if they could include expired passports. The data collected from the question about passports held are used as a proxy for citizenship. Since information about what to include is important in maintaining data quality, we decided to move this information out of the accordion, where it is hidden unless a respondent clicks to expand it, and onto the page to make it more visible.

Given the space constraints on paper, these instructions only appear on the electronic questionnaire. A screenshot of the final question and include panel is shown in Section 5: Questions recommended for Census 2021.

Question design for online

In addition to the testing described earlier, the 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 had been 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 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.

Search-as-you-type functionality

We have recommended that search-as-you-type functionality is applied to the following second address and migration questions in Census 2021:

• second address (when outside of the UK)

• country of birth

• usual address one year ago (when outside of the UK)

• passports held

Work on this functionality is ongoing. We present more details in the Question and questionnaire development overview for Census 2021.

Welsh language question development

Between 2017 and 2018, we commissioned an external agency with Welsh-speaking researchers to undertake focus groups (2017:17) and a series of cognitive interviews (2017:18). In 2018, the same agency undertook further 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.

To ensure questions adhere to Cymraeg Clir guidelines, some changes to the text or questions across the census questionnaires were translated by our contracted specialist Welsh language translation service provider. These changes 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 Welsh Government. The group 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.

Some of the changes we made to the Welsh translations are presented here.

Second address

Test findings (2018:40) suggested the translated phrasing in the instruction “olynol” meaning “in a row or separate” was not understood well by participants. The Assurance Group decided to use “un ar ôl y llall” (one after the other) instead.

Second address type

For the electronic questionnaire, we changed the question stem on the country variation of the second address type to avoid mutations in Welsh. To mitigate the mutation, we added a colon on the online question "Pa fath o gyfeiriad yw eich cyfeiriad yn: [Name of country]?".

The last response option in English is "Holiday home". In the 2011 Census, the second address type question includes "Cyfeiriad" (Address) and "Cyfeiriad tŷ gwyliau" (Holiday home address). The Assurance Group decided to keep this for Census 2021 to be consistent with the other response options that include "Cyfeiriad".

To be consistent with the labour market suite of questions, we decided to use the translation "i ffwrdd o'r cartref" for "working away from home".

The Welsh translation for the new response option "Partner's address" is "Cyfeiriad partner".

Intention to stay

The Census 2021 intention to stay question tested well (2018:40). We changed “ragor” to “fwy” in the response option “12 mis neu fwy “ (12 months or more) for consistency across the forms.

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.

We have published the paper questionnaires for Census 2021 separately.

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.

Second address

Figure 1: Online question on second address in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 1: Online question on second address in English and Welsh

.png (150.2 kB)

If a respondent selects “Yes, an address within the UK” or “Yes, an address outside the UK”, they are taken to a screen where they can enter the address or country.

Second address type

The second address type question is only presented to respondents who answer “Yes” they do stay at another address for more than 30 days a year.

Figure 2: Online question on second address type in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 2: Online question on second address type in English and Welsh

.png (149.6 kB)

Full-time education

Respondents aged under five years old do not have to answer this question or the subsequent term-time address question.

Figure 3: Online question on full-time education in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 3: Online question on full-time education in English and Welsh

.png (53.3 kB)

If the respondent is 19 years old or over, then the question becomes “Are you a student in full-time education?” or “Ydych chi’n fyfyriwr mewn addysg amser llawn?”

Term-time address

The term-time address question is only presented to respondents who answer “Yes” to the full-time education question.

Figure 4: Online questions on term-time address in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 4: Online questions on term-time address in English and Welsh

.png (83.3 kB)

The first line of the respondent’s main address will be piped into the first response option.

If a respondent has reported that they live at a second address in the UK, then the first line of that address will be piped into the second response option.

If a respondent selects “Another address”, then this address, or country if the address is outside the UK, will be collected.

Country of birth

Figure 5: Online question on country of birth in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 5: Online question on country of birth in English and Welsh

.png (90.6 kB)

On the questionnaire in England, “England” will appear first. On the questionnaire in Wales, “Cymru” (Wales) will appear first. This is the same as in the 2011 Census.

Date of arrival in the UK

The date of arrival question is only presented to respondents who select “Republic of Ireland” or “Other” at the country of birth question. The second stage of the date of arrival question is only presented to respondents who enter a date of arrival of “March 2020”.

Figure 6: Online question on date of arrival in the UK in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 6: Online question on date of arrival in the UK in English and Welsh

.png (72.0 kB)

Figure 7: Online question on arrival in the UK within the last year in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 7: Online question on arrival in the UK within the last year in English and Welsh

.png (54.7 kB)

Intended length of stay in the UK

This question is only presented to respondents who have been in the UK for less than 12 months.

Figure 8: Online question on intended length of stay in the UK in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 8: Online question on intended length of stay in the UK in English and Welsh

.png (100.7 kB)

Usual address one year ago

Figure 9: Online question on usual address one year ago in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 9: Online question on usual address one year ago in English and Welsh

.png (145.6 kB)

The first line of the main address will be piped into the first response option. On the paper questionnaire, the first response option is "The address on the front of this questionnaire".

After the first respondent has answered this question, their name will be piped into an additional response option for all other respondents in the household. This response option will be "Same as [Person 1]'s address one year ago" where "[Person 1]" is the name of the first respondent. This response option will appear first in the list of response options.

If a respondent answers "Student term-time or boarding school address in the UK" or "Another address in the UK", then they will be taken to a second screen where they can enter the full address.

If a respondent answers "Another address outside the UK", then they will be taken to a screen where they can enter the country of the address.

Passports held

Figure 10: Online question on passports held in English and Welsh

Source: Office for National Statistics – Second address, migration and citizenship question development for Census 2021

Download this image Figure 10: Online question on passports held in English and Welsh

.png (162.3 kB)

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 the 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.

Table 1 provides the assessment scores for the topics of migration and citizenship.

The long- and short-term international migration topics include the country of birth, date of arrival in UK and intended length of stay questions. The topic of internal migration includes the address one year ago question. The topic of citizenship includes the passports held question.

The initial consultation did not include the topic of second address.

After completing the research and development phase, we evaluated our recommended questions against the same criteria using an updated tool that considers the types 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 migration.

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.

We recognize 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 second address, migration and citizenship, there are several differences in the questions asked on Census 2021 in Scotland and Northern Ireland:

• in Scotland and Northern Ireland, the questions on second address and second address type will not be asked

• in Northern Ireland only, the question on date of arrival will be asked to all respondents from outside of Northern Ireland, rather than outside of the UK

• in Scotland and Northern Ireland, the question on intended length of stay in the UK will not be asked

The questions on second address, migration and citizenship 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.

The Government Statistical Service have published harmonisation principles on collecting data on migration, citizenship and country of birth. The questions on Census 2021 follow these principles.

Address look-up

We are developing address lookup technology for the electronic questionnaire. This will allow users to enter their postcode or first line of their address, then select their house from a list. This will reduce respondent burden, particularly when entering an address that is not the respondent's current permanent address. The functionality will also help to reduce input errors, which will reduce the processing required.

We have recommended that the technology is used for the following second address and migration questions in Census 2021:

• second address (when in the UK)

• term-time address (when "another address" and no second address has been provided)

• usual address one year ago (when in the UK)

The exact wording of the address look-up will be finalised as part of this work.

Search-as-you-type

We are developing search-as-you-type technology for the electronic questionnaire. We present more details of this functionality in the Question and questionnaire development overview for Census 2021.

The exact wording of the search-as-you-type will be finalised as part of this work.

In our response to the 2021 Census topic consultation, the Office for National Statistics (ONS) committed to collecting data on long-term migration, short-term migration, internal migration and citizenship. We also made the commitment to explore how the online question set can improve the collection of the data while still meeting user needs.

The data collected on the electronic questionnaire are the same as on the paper questionnaire. However, where relevant, we have provided additional guidance, updated question designs and implemented automatic routing to improve accessibility and reduce respondent burden. The easier a respondent finds the questions to answer, the better the data quality will be.

We are developing address look-up and search-as-you-type functionality for the electronic questionnaire. Although this functionality increases the differences between the electronic and paper questionnaires, this will mean that respondents completing the questionnaire online will find it easier and quicker to answer questions.

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.

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

A question 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.

Second address

The second address question has been assessed as having “Low” potential for impact on public acceptability and financial concerns.

Potential for impact on data quality: “Medium”

This question collects information that cannot be directly observed, which can cause data quality issues when answering on behalf of another person. Feedback from the 2019 Rehearsal (2019:15) showed that some respondents found this question hard to answer. Around 10% of online respondents, well above the average seen in other questions, used the “Previous” button to go back and review their answer to this question.

Potential for impact on respondent burden: “Medium”

This question contains a write-in field, which can increase respondent burden. However, in Census 2021 on the electronic questionnaire, respondents will be able to use standard postcode look-up functionality to find addresses without typing them out in full, which can help reduce respondent burden.

Feedback from the 2019 Rehearsal (2019:15) showed that some respondents found this question difficult to answer, and use of the “Previous” button on the questionnaire was high.

Potential for impact on questionnaire mode: “High”

On the paper questionnaire, respondents are able to write in the address or country of the second address within the response options. On the electronic questionnaire, respondents enter the address or country on a separate screen. When entering a UK address on the electronic questionnaire, respondents will be able to use standard postcode look-up functionality to enter addresses using the postcode and street address.

Second address type

The second address type 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”

This question collects information that cannot be directly observed, which can cause data quality issues when answering on behalf of another person.

Feedback from the 2019 Rehearsal (2019:15) showed that some respondents found this question difficult to answer, and this was backed up by quantitative data that showed respondents spent more time answering this question than was expected for a question of this type.

Potential for impact on respondent burden: “Medium”

This question has more response options than in the 2011 Census, which has increased respondent burden. Results from the 2019 Rehearsal (2019:15) showed that respondents found this question difficult to answer and, on average, took a longer time than expected to answer this question.

Evaluation of the full-time education and term-time address questions

The full-time address type and term-time address have been assessed as having “Low” potential for impact on data quality, public acceptability, respondent burden, financial concerns and questionnaire mode.

Evaluation of the country of birth question

The country of birth question has been assessed as having “Low” potential for impact on data quality, public acceptability, respondent burden and financial concerns.

Potential for impact on questionnaire mode: “Medium”

On the paper questionnaire, respondents write in their country of birth within the “Elsewhere” response option. On the electronic questionnaire, if respondents choose the response option “Elsewhere” then they will enter the country on a separate screen. The write-in response option has search-as-you-type functionality; a list of suggested countries will appear once the respondent starts to type a response.

Evaluation of the date of arrival in the UK question

The date of arrival in the UK question has been assessed as having “Low” potential for impact on public acceptability and financial concerns.

Potential for impact on data quality: “High”

This question requires respondents to recall a date in the past, which can cause data quality issues. In addition, this question is not observable and can be considered a sensitive subject and so may cause issues when answering on behalf of another person.

Data from the 2019 Rehearsal (2019:15) showed that respondents took a longer time, on average, to answer this question and this question had a relatively high, above 7%, non-response rate.

Potential for impact on respondent burden: “Medium”

This question has a write-in option and requires respondents to recall a date in the past, which can increase respondent burden. Respondents must answer this question so that they can be correctly routed to the next question. Feedback from the 2019 Rehearsal (2019:15) showed that some respondents reported that the question was difficult to answer, expecially when the date was a long time ago, and quantitative data showed that non-response and time taken to answer this question were above the expected levels.

Where the month of arrival is one year prior to the census date, 21 March 2020, respondents will be required to answer a follow-up question to identify whether they have stayed over one year or less than one year.

Potential for impact on questionnaire mode: “High”

The question stems on the paper and electronic questionnaire differ slightly. Also, on the electronic questionnaire the respondents are asked a separate question (“Did you arrive in the UK on or after 21 March 2020?”) to enable routing to the next question on intended length of stay. This question does not appear on the paper questionnaire except as a routing instruction (for example, “If you arrived before 21 March 2020, go to 13”).

This question also has hard validation on the electronic questionnaire, meaning respondents must give an answer; however, on the paper questionnaire this type of validation is not possible. Results from the 2019 Rehearsal (2019:15) showed that this question had a non-response rate of over 1% on paper.

Evaluation of the intended length of stay in the UK question

The intended length of stay in the UK question has been assessed as having “Low” potential for impact on public acceptability and financial concerns.

Potential for impact on data quality: “Medium”

This question is difficult to answer on behalf of another person. Feedback from the 2019 Rehearsal (2019:15) showed a high non-response rate, above 7%, and that some respondents reported difficulties in answering this question.

Potential for impact on respondent burden: “Medium”

Some respondents in the 2019 Rehearsal (2019:15) reported that they found this question dififcult to answer. The question had a high non-response rate and more than 2% of respondents used the “Previous” button to go back to review their answer, suggesting this question was burdensome to respondents.

Potential for impact on questionnaire mode: “Medium”

The electronic questionnaire uses radio buttons to ensure respondents only give one answer, but this is not possible on the paper questionnaire. Respondents are required to give an answer on the electronic questionnaire; however, this type of validation cannot be enforced on the paper questionnaire.

In addition, this question could be subject to social desirability bias, which is where people respond in a way that they consider socially acceptable, rather than what they believe to be true. Responses to the paper questionnaire are more at risk from this type of behaviour as they are more easily read by other members of the respondent’s household than those on the electronic questionnaire.

Evaluation of the usual address one year ago question

The usual address one year ago question has been assessed as having “Low” potential for impact on data quality and public acceptability.

Potential for impact on respondent burden: “Medium”

The question has long response options and may require the respondent to write in a response. In addition, this question could be difficult to answer on behalf of another person.

Potential for impact on financial concerns: “Medium”

All write-in responses require additional processing, which adds cost to the census. In the 2011 Census, 11.5% of respondents used the write-in response for the question on usual address one year ago. However, the introduction of postcode look-up functionality on the electronic questionnaire should reduce the manual processing required for this question.

Some respondents may have privacy concerns about sharing their address one year ago with other members of their household. This could lead to requests for individual forms, which come with additional postage and processing costs for the census.

Potential for impact on questionnaire mode: “High”

This question has multiple stages on the electronic questionnaire but appears as one question on the paper questionnaire. On the paper questionnaire, respondents have to write in the address but on the electronic questionnaire respondents are able to use postcode look-up functionality.

On the paper questionnaire, it is possible to select two or more mutually exclusive options, while the electronic questionnaire only allows for one option to be selected.

Evaluation of the passports held question

The passports held question has been assessed as having “Low” potential for impact on data quality, public acceptability, respondent burden and financial concerns.

Potential for impact on questionnaire mode: “Medium”

This question has instructions, in an include panel, helping respondents to answer this question on the electronic questionnaire. Because of space constrants, these instructions do not appear on the paper questionnaire.

On the electronic questionnaire, respondents can write in the country of their passport or select from a list of suggested countries that appears after they have typed a few characters (search-as-you-type functionality).

Second address

A second address is any single address where a person has stayed for more than 30 days in the past year that is not their usual residence. This can be 30 days in a row or 30 days separated across the year.

Schoolchildren at boarding school and students

Schoolchildren at boarding school and students who stay at their term-time address for four or more nights per week (in general) should be counted as usually resident at their term-time address and complete the census form at that address. These students and schoolchildren should also be counted as usually resident at their permanent or family address (where this is not their term-time address), and only some information will be collected on the census form.

Schoolchildren at boarding school and students who stay at their term-time address for three nights or fewer per week (in general) should be counted as usually resident at their permanent or family address. They should not complete the census form at their term-time address.

This definition has changed from the 2011 Census to accommodate an increase in part-time boarders and flexi-boarders. We detail this change in definition in the Output and enumeration bases: residential address and population definitions for Census 2021 report.

Migration

The UN definition of a long-term migrant is a person who changes his or her country of usual residence for a period of at least 12 months. The country of destination effectively becomes the country of usual residence. The Office for National Statistics (ONS) uses this time period to distinguish short- and long-term residents.

The ONS has adopted these definitions for counting long- and short-term migrants in Census 2021 in England and Wales.