Sexual orientation 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 (PDF, 967KB). This outlined the Office for National Statistics's (ONS's) proposal to collect information on sexual orientation in Census 2021 to meet the needs for better quality information for equality monitoring.

ONS research and consultation showed a clear need for information on sexual orientation, to support work on policy development and service provision and to allow local authorities to meet and monitor their requirements under the Equality Act 2010.

Although survey-based estimates at national level are regularly published, there are currently no reliable data on sexual orientation at local authority level.

For Census 2021, the recommended question on sexual orientation is:

Which of the following best describes your sexual orientation?
This question is voluntary
[ ] Straight or Heterosexual
[ ] Gay or Lesbian
[ ] Bisexual
[ ] Other sexual orientation
(Write in sexual orientation)

The sexual orientation question, which has not been asked in previous censuses, is voluntary and will only be asked of those aged 16 years and over.

Ongoing work around this topic will focus on understanding and meeting user requirements for outputs.

Since the publication of the White Paper, the Office for National Statistics (ONS) has conducted and concluded the final phase of testing on the sexual orientation topic. This report provides links to previously published research and also provides the findings of additional testing that led to the final recommended question for Census 2021 in England and Wales. These recommendations will be subject to parliamentary approval through the census secondary legislation.

The questions and response options for Census 2021 have now been finalised through the census secondary legislation: The Census (England and Wales) Order 2020 as well as Census Regulations for England and for Wales.

The evidence base for the recommendations made in the White Paper and the finalisation of the question for Census 2021 is discussed in the section, Research that led to the 2018 White Paper recommendations and final recommended question for Census 2021.

The 2011 Census topic consultation, conducted in 2005, identified a strong user need for information on sexual orientation. However, the topic was not included in the 2011 Census. At the time, we had significant concerns about privacy, acceptability, conceptual definitions and the effect that the question could have on overall census response.

We established the Sexual Identity Project in 2006 to meet the user requirements for information on sexual identity, the component of sexual orientation most closely related to experiences of disadvantage and discrimination. The project led to the publication of the Government Statistical Service (GSS) harmonised standard for asking questions and reporting statistics on sexual identity. Sexual orientation estimates published by the Office for National Statistics (ONS) are currently based on social survey data from the Annual Population Survey (APS), which uses the GSS questions.

In June 2015, the 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

This response included commitments to include a question on sexual orientation in the 2017 Census Test to assess the impact of including a question on sexual orientation on overall response and data quality and to assess the public acceptability of asking a question on sexual orientation.

The topic consultation revealed a clear requirement for information on sexual orientation for policy development, service provision and monitoring, and fulfilling duties under the Equality Act 2010.

A detailed summary of the consultation responses relating to the topic of sexual orientation can be found in the sexual orientation consultation report (PDF, 662KB). In this report, the ONS made clear commitments to the public. We have provided an update on how we met these commitments in Annex 1.

Despite a strong user need for the data, our assessment indicated that a sexual orientation question may not be acceptable to the public. To assess these concerns, we developed a research and testing plan. The details of this plan are in the next section, Research that led to the 2018 White Paper recommendations and final recommended question for Census 2021.

The plan was presented to sexual identity data users at a workshop in August 2016, where it received positive feedback.

Following this, we began a comprehensive programme of research and development. We provide a full list of the tests used in the development of the topic of sexual orientation in Annex 2. Further details are provided in the summary of testing for Census 2021.

In December 2017, we published a further census topic research update, which included the findings of the 2017 Census Test. The positive results suggested that we could develop a question that was acceptable enough to provide sufficient quality of data to meet user needs without damaging the overall quality of data on the census.

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 (PDF, 967KB). This outlined our proposals for the sexual orientation topic:

• we should collect information on sexual orientation
• the question should be voluntary
• the question should only be asked of those aged 16 years or older

Alongside the White Paper, we published a further census topic research update in December 2018. This update provided additional details of the research that supported the recommendations announced in the White Paper.

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

This topic was originally referred to as "sexual identity", but the name was changed to "sexual orientation" after the 2017 Census Test. We provide details of the definitions and terms used in this report, including reasons for the change in terminology, in Annex 4. Some early publications use the term "sexual identity".

As set out in the White Paper, Help Shape our Future: The 2021 Census of Population and Housing in England and Wales (PDF, 967KB), with further details provided in the December 2017 and December 2018 census topic research updates, our testing on the sexual orientation topic involved a research and testing plan with three strands of work:

• inclusion of a question in the 2017 Census Test in England and Wales
• a public acceptability survey in England and Wales
• development of statistics from Office for National Statistics (ONS) social surveys (alternative sources)

We provide a full list of the tests used in the development of the sexual orientation topic in Annex 2 and further details 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. This included those who were bisexual, gay, heterosexual, lesbian, and other sexual orientations.

References to tests are provided throughout in 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).

Developing a question for the 2017 Census Test

Following the topic consultation, our main aim was to assess the potential impact that including a sexual identity question might have on response to the census.

The starting point for developing a suitable question was the Government Statistical Service (GSS) harmonised standard principle question (self-completion version). This was developed from the ONS Sexual Identity Project (2006 to 2009) and is:

Which of the following options best describes how you think of yourself?
[ ] Heterosexual or Straight
[ ] Gay or Lesbian
[ ] Bisexual
[ ] Other
[ ] Prefer not to say

The GSS harmonised guidance advises that this question should only be asked of respondents aged 16 years and older and should not be asked about household members who are not being interviewed.

We tested the question by conducting a series of focus groups (2016:2), followed by cognitive interviews (2016:4), with members of the public who had a range of sexual orientations. Based on the recommendations from this testing, we made the following modifications to the question:

• added a write-in field for the "Other" response option; this allows respondents to report their sexual orientation rather than being classified as "Other"
• individuals can answer on behalf of other people aged 16 years and over; this brings the question in line with the census, which allows for proxy responses so individuals can answer the question on behalf of other household members
• the question was made voluntary; this was based on our view of potential legislative requirements at the time, and it was achieved by including a voluntary statement below the question stem

The testing found that participants considered it appropriate to ask this question to all those aged 16 years and older. The write-in field for the "Other" response option was well received by participants, including those who selected this option and wrote in answers such as "Asexual" and "Pansexual". This change was generally supported by stakeholders at the sexual identity workshop.

In 2016, we commissioned independent public acceptability testing (2017:1) of the modified sexual orientation question with households in England and Wales. Participants were shown a copy of the question and asked about their attitude towards it. In this test, 80% of participants stated that they could accurately answer a sexual orientation question for all members of their household aged 16 years and over. Additionally, 8% of participants stated that they would not be able to answer a question on sexual orientation accurately for any other members of their household. This was consistent with the findings of qualitative testing.

Our initial research findings, and the results from the 2017 Census Test (2017:7), suggested that we could develop a question that provides sufficient quality of data to meet the user need without damaging overall data quality in the census.

These findings also identified other avenues of research to improve data quality and refine the question design. These were:

• question wording: referencing "sexual orientation" in the question and shortening the question stem
• response options: investigating the need for additional options and reversing the order of the "Heterosexual or Straight" response option
• age limits: considering asking the question of respondents aged 15 years or under

Developing the question wording

The omission of the term "sexual orientation" from the GSS harmonised standard question was raised in qualitative research in both positive and negative terms. Participants in the focus groups (2016:2) commented that the meaning of question relied on the response options, increasing the burden on respondents. Other participants had reservations towards changing the question, as they were concerned respondents' understanding of the specific terms "sexual orientation" or "sexual identity" would vary.

Feedback from the focus groups showed that referencing the topic in the question, rather than relying on the response options, benefits users by providing a term they can reference if they need support on this question or are reviewing their answers. It also supports respondents who use assistive technologies, such as screen readers.

This approach is more closely aligned with Stonewall's recommended question (PDF, 4.53MB). This has been published since the GSS harmonised standard question was initially developed.

We also explored changing the "Other" response option to "Other sexual orientation" to minimise the number of write-in responses that related to other topics (for example, marital status and gender identity). There were a substantial number of these write-in responses in the 2017 Census Test (2017:7).

In the focus groups (2016:2) and cognitive interviews (2016:4), there were some negative views towards the "how you think" part of the questions stem. Some participants stated that this could be thought of as implying a person's sexual orientation is not factual.

We therefore tested:

• adding "sexual orientation" to the question wording
• removing "how you think of" from the question wording
• adding "sexual orientation" to the "Other" response option wording

Method of testing the question wording

These changes were tested through qualitative (2018:21) and small-scale quantitative research via an omnibus survey (2018:23). The tests looked at acceptability, comprehension, response and data quality. The sexual orientation question wording variants tested were:

• Variant 1: Which of the following best describes how you think of yourself? (GSS harmonised 2017 Census test question stem.)
• Variant 2: Which of the following best describes your sexual orientation? (Similar to the Stonewall recommendation.)
• Variant 3: Which of the following best describes how you think of your sexual orientation?

The "Other" response option wording for the different variants tested were:

• Variant 1: Other, write in.
• Variant 2 and 3: Other sexual orientation, write in.

Following this, Variant 3 of the question wording and Variant 1 of the "Other" response option wording were included in a large-scale quantitative household test (2018:24). The design of the large-scale test meant that the question would be tested in conditions closer to that of the census.

Results for developing the question wording

Participants in the cognitive testing (2018:21) felt that Variant 1 was vague, and some reported having to read the response options before knowing what the question was asking. The phrase "how you think of" was not well understood, and participants had mixed views regarding its acceptability. Several LGB+ participants (that is, those who have a minority sexual orientation) provided feedback that asking someone how they think of their sexual orientation suggests that it is not real.

The results of the omnibus survey (2018:23) showed that neither including "sexual orientation" in the question, nor removing the phrase "how you think of", impacted on response or data quality. All three variants had similar response rates and there were no significant differences in the proportion of people selecting "Gay or Lesbian" or "Bisexual". The average time spent completing the question was the same in each of the three variants.

The omnibus survey (2018:23) showed little variation in the write-in responses to the "Other" response option and very few examples of write-in responses that were not a sexual orientation. For example, all three variants had participants write in "Asexual", and both Variants 1 and 2 had participants identify as "Pansexual". As there was such little variation in responses, we concluded that there was no evidence that including "sexual orientation" as part of the response option would negatively impact the quality of data collected.

In conjunction with this work, we included the sexual orientation question in a large-scale quantitative test (2018:24), which included the term "sexual orientation" in the question stem and "Other" response option.

The results of this test confirmed the findings from the other question wording development tests. The proportion identifying as "Lesbian, Gay or Bisexual" was 2.3%, the same as a previous large-scale test (2017:22). The levels of non-response to the question were also similar across the two surveys. This previous test had used the GSS national harmonised standard question (self-completion), which does not contain the term "sexual orientation".

The evidence from testing led to us recommending the question stem, "Which of the following best describes your sexual orientation?". This is complemented by the final substantive response option beginning "Other sexual orientation".

Developing the response options

At the sexual identity workshop held in August 2016, there was a user need for further breakdown of identities within the non-heterosexual population beyond the existing "Gay or Lesbian", "Bisexual" and "Other" categories. This was reiterated in 2017 qualitative testing, where it was recommended to consider an "Asexual" response option.

However, further engagement with groups representing some of these identities did not result in evidence of a user need for an estimate of the asexual population. The respondent need is met using the write-in field in the "Other" response option.

Evidence from testing indicated that the general public were unfamiliar with the term "Heterosexual". The issue was first identified in the Sexual Identity Project (2006 to 2009) and then again in cognitive interviews (2016:4). Analysis of the 2017 Census Test (2017:7) showed that "Straight" was one of the most common write-in responses to the sexual orientation question. This indicated that some respondents did not understand the term "Heterosexual" or notice that it was followed by the term "Straight".

To resolve this issue, we proposed changing the response option from "Heterosexual or Straight" to "Straight or Heterosexual" for the electronic questionnaire and "Straight/Heterosexual" for the paper questionnaire. The slash was proposed for the paper questionnaire to align with Stonewall's recommendation. For the electronic questionnaire, we are unable to use a slash as it does not meet accessibility requirements for screen readers.

Prior to testing, we discussed this proposal with other government departments and lesbian, gay and bisexual organisations such as the Government Equalities Office, the Equality and Human Rights Commission and the LGBT Foundation. There were no concerns raised with the acceptability of placing "Straight" before "Heterosexual".

Qualitative testing suggested that there were mixed views about the acceptability of placing "Straight" before "Heterosexual". Cognitive interviews (2018:21) found that the term "Straight" has had negative connotations in the past owing to its opposite being "bent", and that it is more acceptable to have less emphasis on the term by placing "Heterosexual" first in the response option. However, iterative testing of the electronic questionnaire found no issues when the "Straight or Heterosexual" response option was introduced.

The introduction of the slash on the paper form was supported by participants. It was seen in the context of the "Gay or Lesbian" response option as highlighting that these are different identities, whereas straight and heterosexual are different words for the same identity.

We tested the impact of placing the term "Straight" before "Heterosexual" as part of a small-scale quantitative test (2019:4) using an online panel survey. The test included a question on sexual orientation with the alternative response option.

We analysed the responses and found no evidence of anyone misunderstanding the question. No participants wrote in either "Heterosexual" or "Straight". The sampling frame and context of the survey meant we were unable to compare the distribution of responses to other surveys.

From this evidence, we recommended changing the response option to "Straight or Heterosexual" on the electronic questionnaire, and "Straight/Heterosexual" on the paper questionnaire.

Confirming that the question will only be asked of those aged 16 years and over

Our testing (2017:1) prior to the 2017 Census Test had suggested that a sexual orientation question was most acceptable if asked of respondents aged 16 years and over. The GSS guidance about asking questions on sexual orientation also recommends that the question should only be asked of those aged 16 years and over.

In May 2018, we held a meeting with stakeholders to test the assumption that the minimum age at which the sexual orientation question should be asked on Census 2021 is 16 years old. Although the stakeholders reiterated a need for data on sexual orientation at lower age groups, there was general support for asking the question on the census to those aged 16 years and over only.

They agreed on two important justifications for this decision relating to data quality.

First, responses for children under the age of 16 years are likely to be completed by the householder on their behalf. The householder may not know the child's sexual orientation, which will reduce the quality, and therefore usefulness, of the data collected.

Secondly, children under the age of 16 years may not yet know their sexual orientation. This becomes more likely the younger the child. At age 16 years, children reach the age of consent and the minimum age for marriage. They can therefore, in most cases, be assumed to know their own sexual orientation by that age.

In the White Paper, Help Shape our Future: The 2021 Census of Population and Housing in England and Wales (PDF, 967KB), we recommended that a question on sexual orientation should only be asked of respondents aged 16 years and older.

There is still likely to be under-reporting of the LGB+ population among respondents aged 16 to 18 years because, as with children aged under 16 years, the householder will generally complete the questionnaire on their behalf.

Respondents will be able to request an individual access code or individual paper form if they wish to respond separately to the rest of their household. Individual answers will override any answers submitted on the household form.

Making the sexual orientation question voluntary

In early quantitative testing (2016:2, 2016:4), participants who identified as LGB+ stated that it was important not to force individuals to disclose that they are LGB+, as it may not be safe to do so or they may want to maintain their privacy. This led to the recommendation that respondents should have the option to not provide an answer.

For the 2017 Census Test (2017:7), the question was made voluntary by including the statement "This question is voluntary". This was the same method used for the voluntary religion question in the 2011 Census. The religion question was the only voluntary question in previous censuses.

We established three approaches to making the question voluntary:

• Parliament passing primary legislation to amend the Census Act 1920 and remove the penalties for not responding for the question collecting sexual orientation and adding the instruction "This question is voluntary"
• adding a "Prefer not to say" response option
• implementing both methods; Parliament passing primary legislation, adding an instruction stating "This question is voluntary", and adding a "Prefer not to say" response option

We conducted a public acceptability survey (2017:1) that included a question containing the statement "This question is voluntary" and asked participants about the acceptability of including a "Prefer not to say" response option. The second and third approaches were established based on findings from this survey.

The results from the survey suggested that the "Prefer not to say" response category would make the sexual orientation question more acceptable. However, we considered that communicating that the question was voluntary by two methods could lead to a large proportion of respondents not answering the question, resulting in reduced data quality.

Therefore, we designed a three-way split-sample test (2017:22) to assess the impact of making the question voluntary through a "Prefer not to say" response option only. The three forms of the question were:

• with a "Prefer not to say" option for the response
• with an instruction that the question was voluntary
• with neither a "Prefer not to say" option nor a voluntary instruction

This test found that the sample with the "Prefer not to say" option had the highest response to the survey (34.8%) and the lowest incidence of failure to answer the question, although neither of these differences were statistically significant at the 95% confidence level. This sample had an LGB+ population estimate of 2.3%. This was nearest to the expected figure, with the other samples yielding lower estimates. Further details of this test were published in the December 2018 topic update. At this time, we therefore recommended making the question voluntary through a "Prefer not to say" response option.

Later, we returned to the legislative option of making the question voluntary. A small-scale online quantitative test was conducted to examine the impact of including both a "Prefer not to say" response option and an instruction statement, "This question is voluntary", compared to just the instruction statement.

The quantitative test (2019:4) found that non-response was 4.1% for the question with both a "Prefer not to say" option and a "This question is voluntary" instruction. This non-response percentage includes "Prefer not to say" responses. Non-response was 2.7% for the question with only the "This question is voluntary" instruction.

For the maximum quality of data to be collected, the non-response should be minimised. We therefore recommended making the question voluntary by Parliament passing primary legislation to amend the Census Act 1920 and remove the penalties for not responding to a sexual orientation question. This legislative change was made in the Census (Return Particulars and Removal of Penalties) Act 2019. On the census, there will be an instruction stating, "This question is voluntary". The word "voluntary" will appear in bold text on the paper and electronic questionnaires.

Question design for online

In addition to the testing described earlier, the question has undergone significant user experience (UX) testing (2017:2, 2018:2, 2019:1, 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 testing for the electronic questionnaire 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 throughout 2020. All participants are purposively selected to include a wide range of ages and digital abilities.

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

Guidance accordion

Feedback from the 2019 Rehearsal (2019:15) showed that some people questioned why the census asks a question on sexual orientation. As a result of this feedback, and the question being voluntary, we added guidance text to the electronic questionnaire on why we ask this question to encourage respondents to answer the question.

The text, which is shown when a respondent clicks "Why your answer is important", under the question and response options, reads:

"Your answer will help your local community by providing organisations, such as charities, with an understanding of the services people might need.

Councils and government will use this information to:

• monitor equality to ensure that everyone is treated fairly
• provide services and share funding"

Welsh language question development

Between 2017 and 2018, an external agency with Welsh-speaking researchers was commissioned to undertake focus groups (2017:17) and a series of cognitive interviews (2017:18). In 2018, further cognitive interviews (2018:40) were undertaken by the same agency. 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. 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.

The findings of this research led to the following recommendations for the Welsh language question design:

• the term "cyfeiriadedd rhywiol" should be used for sexual orientation; this was generally understood by participants and was the preferred term of Stonewall Cymru
• the term "bi" should be included alongside "deurywiol" (bisexual), as this is generally how the term is used colloquially
• the more familiar term "strêt" (straight) should be placed before "heterorywiol" (heterosexual); although the literal meaning of "strêt" in Welsh is "in a straight line", participants understood the word in this context and generally considered it helpful

A further suggestion, to aid understanding, was to include the term "rhywioldeb" (sexuality) in the question stem. This was not included because this term is conceptually different to sexual orientation.

We sought to minimise the use of slashes on the electronic questionnaire owing to accessibility reasons. However, the recommendation was to use a slash for the response options of "Heterorywiol/Strêt" (Heterosexual/Straight) and "Deurywiol/Bi" (Bisexual/Bi), as these were considered a more accessible translation because of mutations within the Welsh language.

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.

Figure 1: Online question on sexual orientation, asked if aged 16 years or over, in English and Welsh

Source: Office for National Statistics – Census 2021

Download this image Figure 1: Online question on sexual orientation, asked if aged 16 years or over, in English and Welsh

.png (127.3 kB)

Design features of this question

Providing information on a person's sexual orientation is voluntary. An instruction on the question reads, "This question is voluntary". Other guidance materials will emphasise this. In Census 2021, the Census (Return Particulars and Removal of Penalties) Act 2019 removes any penalty for not responding to a question or questions on sexual orientation in the census.

The question is asked only of those aged 16 years and older. On the electronic questionnaire, respondents aged 15 years and under are not presented with this question. On the paper questionnaire, they are directed around it by instructional text.

The question makes explicit reference to the topic "sexual orientation" within the question wording. The term is included in the question stem and the "Other" response option.

The question includes a write-in option so that individuals who do not identify with any of the orientations in the response options can write in their answer.

Respondents aged 16 years and older are able to request an individual access code or paper form if they wish to respond separately to the rest of their household. This enables people to answer the census privately, without having to tell the person completing the household form they have done so. Individual answers will override any answers submitted on the household form. This is vital to protect people's privacy and ensure good quality data.

There are small differences to the wording of the online questions for sexual orientation depending on context. For example, if a respondent is replying on behalf of someone else, their name, rather than "your", is included in the question.

Question order

Census 2021 will be the first census to ask questions on both gender identity and sexual orientation, so their position within the form needed to be considered.

There is a risk that respondents will stop answering the form at the sexual orientation question. The results of our testing, including the 2019 Rehearsal (2019:15), suggests this risk is low. We have therefore positioned this question at the end of the socio-cultural questions but before the qualifications section. It is positioned next to the gender identity question.

To assess whether the data quality would be better if the gender identity topic is asked first or if the sexual orientation topic is asked first, a small-scale online quantitative test (2019:4) was undertaken. This showed the question that was asked first had a higher rate of non-response than the second, and that the rate of non-response across the two questions was smaller when sexual orientation was first compared to when gender identity was first.

We therefore recommended that the topics should be ordered with sexual orientation before gender identity.

Question evaluation

The question design put forward in this report is 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 sexual orientation 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 question on sexual orientation. We present 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), who are responsible for conducting the censuses in Scotland and Northern Ireland, respectively. The NRS are including the same question and response options as England and Wales in Scotland’s census. NISRA are also including a question on sexual orientation in their census; the NISRA question will include the same response options as those in England, Wales and Scotland, with an additional "Prefer not to say" option. The question is voluntary on all censuses in England, Wales, Scotland and Northern Ireland, and there will be no penalty for non-completion.

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

The sexual orientation 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.

We considered using the Government Statistical Service (GSS) harmonised question on sexual orientation, but this was developed for interviewer-led data collection. Our testing (2016:2, 2016:4, 2018:21, 2018:23) suggested that for a self-complete data collection instrument such as Census 2021, using a different question stem would be more acceptable to respondents and would not impact on data quality. The data collected through this modified question are comparable with that collected using the GSS harmonised question.

In the Census 2021 topic consultation response, we made clear commitments to the public. As we had identified a clear user need for data on sexual orientation, we included a sexual orientation research and testing plan (PDF, 517KB) in the consultation response. This committed to three strands of research to begin our investigation into whether to include a question on sexual orientation in Census 2021. These are listed here, alongside an update on our progress towards meeting these commitments.

Inclusion of a question in the 2017 Census Test across England and Wales

The 2017 Census Test showed that the inclusion of a question on sexual orientation did not influence overall response to the survey. The proportion of participants who did not answer the sexual orientation question was 8.4%, which was below our threshold of 10%. Further details on the design of the test and summary of the findings are available in the 2017 Census Test report.

A public acceptability survey in England and Wales

The independent public acceptability testing (2017:1) of the Government Statistical Service (GSS) harmonised standard showed that 70% of respondents found it acceptable to include a question on sexual orientation in Census 2021. Subsequent testing of the redesigned question showed that 76% of participants found the question acceptable to include in the census.

Development of statistics from our social surveys

This strand of research examined the feasibility of using alternative data sources to obtain information on sexual orientation that meets the data user need.

In 2017, we published a series of experimental research estimates of sexual identity for the UK. A later publication examined sexual orientation by demographics using the Crime Survey for England. We have published an article detailing our work on Exploring existing data for gender identity and sexual orientation.

Neither of the studies, nor any administrative data sources, could meet the user need of estimating the lesbian, gay and bisexual population at local authority level. This reinforced the need to ask this question in the census.

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 sexual orientation question

The potential for impact created by the inclusion of the topic of sexual orientation is now lower for all criteria than in May 2016, except for questionnaire mode which has been assessed as “Medium”. The potential for impact on financial concerns has been assessed as “Low”.

Potential for impact on data quality: “Medium”

Our testing has shown that this question will deliver good quality data. However, the question has a write-in option and collects data that is both sensitive and not observable. Some respondents to the 2019 Rehearsal (2019:15) found this question difficult to answer.

Potential for impact on public acceptability: “Medium”

This question is new to the census. Testing (2018:24) showed that including a sexual orientation question on Census 2021 was acceptable to 76% of participants. However, this question collects sensitive data and respondents may not want a proxy to answer this question on their behalf. It is also possible, due to the sensitive nature of the question, that answers will be affected by social desirability bias. Meaning that respondents may answer in a way they consider socially acceptable instead of answering accurately.

Some respondents to the 2019 Rehearsal (2019:15) reported that they found this question unacceptable. However, although the question is voluntary the non-response rate was only 9.1%.

Potential for impact on respondent burden: "Medium"

An analysis of feedback from the sexual orientation question during the 2019 Rehearsal (2019:15) showed that some respondents found the question difficult to answer.

This question also contains a write-in response, is a new question on the census, and asks for information that cannot be directly observed.

Potential for impact on questionnaire mode: “Medium”

This question has a write-in response. This question also asks for sensitive information and responses can be subject to social desirability bias, meaning respondents may change their answer to one they consider a more socially acceptable. This bias is more of a risk on the paper questionnaire than the electronic questionnaire since it is easier to see the answers of other members of the household on the paper questionnaire.

Sexual orientation

Sexual orientation is an umbrella term that encompasses sexual identity, attraction and behaviour. It is a subjective view of oneself and may change over time and in different contexts.

"Sexual identity" or "sexual orientation"

In May 2018, the Government Statistical Service (GSS) harmonised standard for reporting statistics on this topic changed its recommended terminology from "sexual identity" to "sexual orientation". This aligned with the protected characteristic of sexual orientation that is defined in the Equality Act 2010.

Other organisations advocate collecting data on sexual orientation. For example, the NHS has developed the Sexual Orientation Monitoring Information Standard and Stonewall provides guidance on collecting information on sexual orientation.

Based on this updated view from stakeholders, we changed the label of this topic and question from "sexual identity" to "sexual orientation" following the 2017 Census Test. Prior publications still reference sexual identity. The recommended question design makes no reference to either term and was not changed.

LGB+ and LGBT

The term "LGB+" is used to describe those who have a minority sexual orientation. This includes those who are lesbian, gay, bisexual or any other sexual orientation other than heterosexual.

"LGBT" is an umbrella term used to describe people who self-identify as lesbian, gay, bisexual or transgender, or as having any other minority sexual orientation or gender identity.