Sexual orientation, UK: 2023

• Annual Population Survey (APS) estimates have been affected by increased volatility, resulting from smaller sample sizes, meaning that estimates of change should be treated with additional caution; APS estimates have also not yet been reweighted to more recent population estimates, so estimates of levels should be treated with additional caution, while estimates of percentages remain more robust.

• Of the UK household population aged 16 years and over, 93.6% identified as heterosexual or straight in 2023; this is down from 94.6% in 2018.

• Of the UK household population aged 16 years and over, 3.8% identified as lesbian, gay or bisexual (LGB) in 2023; this is up from 2.2% in 2018.

• Younger people were more likely to identify as LGB than older people; around 1 in 10 (10.4%) of those aged 16 to 24 years identified as LGB, compared with around 1 in 100 (0.9%) of those aged 65 years and over.

• Most people who identified as LGB in 2023 had never married or formed a civil partnership (77.2%), compared with 36.8% of people who identified as heterosexual or straight.

• Men were more likely to identify as LGB (4.2%) than women (3.4%); specifically, men were more likely to identify as "gay or lesbian" (2.8%) than women (1.2%), while women were more likely to identify as "bisexual" (2.2%) than men (1.5%).

In this bulletin, we explore the sexual orientation of people living in households in the UK based on responses to the sexual identity question in the 2023 Annual Population Survey (APS). People aged 16 years and over were asked this question.

We discuss possible reasons for changes in sexual orientation in 2023, compared with 2018. We make five-year comparisons, rather than year-on-year, because the APS data typically vary between years as the survey sample changes. We are better able to identify trends in the data by comparing across a longer timescale.

An estimated 2.1 million people (3.8% of the UK household population) identified as lesbian, gay or bisexual (LGB) in 2023. This was an increase from 1.2 million (2.2%) over the five years since 2018.

Figure 1. Percentages of people who identified as gay or lesbian or as bisexual increased in 2023

Sexual orientation of the UK household population aged 16 years and over, 2018 and 2023

Notes:

1. Totals may not sum because of rounding.

The bar charts in Figure 1 show that a similar number of people identified as gay or lesbian (2.0%, or 1.1 million) or bisexual (1.8%, or 987,000) in 2023. More people identified as gay or lesbian than as bisexual in previous years. The proportion identifying as bisexual doubled between 2018 and 2023, from 0.9%, or 457,000. The proportion identifying as gay or lesbian also increased but more slowly, from 1.4%, or 723,000, in 2018.

London had the highest regional proportion of people in the UK identifying as LGB, at 5.2% in 2023. It also had the lowest regional proportion identifying as heterosexual or straight, at 91.2%. While these findings did not quite reach statistical significance in the APS data, they mirror similar findings from Census 2021 described in our Sexual orientation: age and sex, England and Wales: Census 2021 article.

Most people in the UK (93.6%) identified as heterosexual or straight in 2023. This is a decrease from 94.6% in 2018. A small proportion answered "other" (0.7%) or "don't know/refused" (1.8%) when asked this question in 2023.

While the proportion of the population identifying as LGB has increased, the proportion identifying as heterosexual or straight has decreased.

The increase in LGB responses may be attributed to more people being more open to identifying their own sexual orientation as being LGB, together with changing societal attitudes meaning a greater acceptance of different sexual orientations.

Recent results from the British Social Attitudes Survey support the view that society is becoming more welcoming towards diversity in sexual identity, as described in the National Centre for Social Research's A liberalisation in attitudes? report (PDF, 655KB). This view is also supported by the recognition of same-sex relationships in law through same-sex marriage and same-sex civil partnership legislation. A question on sexual orientation was included in the census for the first time in 2021, as discussed in our Sexual orientation, England and Wales: Census 2021 bulletin.

Men and women answered the same question in the Annual Population Survey (APS); one of the response options is "gay or lesbian". We therefore use the term "gay or lesbian" when describing both men and women who identified with this response option.

Men were more likely than women to identify as lesbian, gay or bisexual (LGB) in 2023. An estimated 4.2% (or 1.1 million) men identified as LGB, compared with 3.4% (or 938,000) women.

Figure 2: Men were more likely to identify as gay or lesbian, while women were more likely to identify as bisexual

Sexual orientation of the UK household population aged 16 years and over, split by sex, 2023

Notes:

1. Totals may not sum because of rounding.

The bar charts in Figure 2 show that men were more than twice as likely to identify as gay or lesbian (2.8%, or 730,000) than women (1.2%, or 341,000). Women were more likely to identify as bisexual (2.2%, or 596,000) than men (1.5%, or 391,000). This continues a pattern seen in 2018.

Sexual orientation varies by age. This survey question was asked of all people aged 16 years and over.

Younger people were more likely to identify as lesbian, gay or bisexual (LGB) than older people in 2023. Among those aged 16 to 24 years, around 1 in 10 (10.4%, or 715,000) identified as LGB, compared with around 1 in 100 (0.9%, or 113,000) of those aged 65 years and over (Figure 3).

The proportion of young people identifying as LGB has more than doubled since 2018, when it was 4.4% (or 310,000) among those aged 16 to 24 years. In contrast, the proportion of those aged 65 years and over identifying as LGB has stayed similar over time (0.7%, or 82,000, in 2018).

Figure 3: Younger people were more likely to identify as bisexual than gay or lesbian

Percentages of the UK household population aged 16 years and over identifying as gay or lesbian, or as bisexual, split by age, 2023

The stacked bar chart in Figure 3 shows that identifying as bisexual was more common than identifying as gay or lesbian among younger age groups (16 to 24 years and 25 to 34 years) in 2023. This difference only reached statistical significance for the youngest age group of people aged 16 to 24 years; it was not statistically significant for those aged 25 to 34 years. The opposite pattern was seen among the older age groups.

There were large increases in the proportion of younger people identifying as bisexual between 2018 and 2023. This more than doubled from 2.8% to 7.5% among those aged 16 to 24 years and increased from 1.1% to 3.4% among those aged 25 to 34 years. Percentages of those identifying as bisexual were smaller (less than 1.0%) in older age groups in both 2023 and 2018.

Women were more likely than men to identify as bisexual in 2023. Young women had the highest proportion identifying as bisexual (9.2% in the 16 to 24 years age group).

In this section, we look at how legal partnerships vary by sexual orientation. Marriage and civil partnership became available at different times to opposite-sex and same-sex couples. We consider this when exploring changes in 2023, compared with earlier years.

Most people who identified as lesbian, gay or bisexual (LGB) had never married or formed a civil partnership (77.2%, or 1.6 million) in 2023. This percentage was more than twice that of people who identified as heterosexual or straight (36.8%, or 18.7 million).

Figure 4: People who identified as bisexual were most likely to have never married or formed a civil partnership

Percentages of the UK household population aged 16 years and over in each legal partnership type, split by sexual orientation, 2023

Notes:

1. Totals may not sum because of rounding.

The stacked bar chart in Figure 4 shows that opposite-sex marriage was the most common legal partnership type among people identifying as heterosexual or straight in 2023, followed by those who had never married or formed a civil partnership. The proportion of people identifying as heterosexual or straight in opposite-sex marriages has declined from 49.7% (25.0 million) in 2018 to 46.2% (23.4 million) in 2023. The proportion who had never married or formed a civil partnership has increased from 34.2% (17.2 million) in 2018 to 36.8% (18.7 million) in 2023.

People who identified as bisexual were most likely to have never married or formed a civil partnership (84.7%, or 837,000). This figure partly reflects their younger age profile, but is higher than other groups even after accounting for age. More than half (52.0%) of all those who identified as bisexual in 2023 were aged 16 to 24 years. People in this age group were least likely to be in a legal partnership, as described in our Marriage and civil partnership status in England and Wales: Census 2021 article. The legal age of marriage or civil partnership increased to 18 years of age in England and Wales in February 2023, under the Marriage and Civil Partnership (Minimum Age) Act 2022.

Opposite-sex marriage was more common among people who identified as bisexual (10.5%, or 104,000) than same-sex marriage (1.1%, or 11,000) or civil partnership (0.2%, or 2,000).

Figure 5: Same-sex marriage has overtaken civil partnership as the most common type of legal partnership for people identifying as gay or lesbian

Percentages of the UK household population aged 16 years and over identifying as gay or lesbian, by selected legal partnership types, 2014 to 2023

The line graph in Figure 5 shows that same-sex marriage was more common (13.2%, or 142,000) than civil partnership (8.0%, or 85,000) among those who identified as gay or lesbian in 2023. This reflects a steady increase in the percentage of people in a same-sex marriage from 2014, the year marriage became available for same-sex couples. This also reflects a decrease in the percentage of people in a civil partnership, which has been available for same-sex couples since 2005.

More information on legal partnerships in the UK is available in:

• our Population estimates by marital status and living arrangements, England and Wales: 2022 bulletin

• our Civil partnerships in England and Wales: 2022 bulletin

• our Families and households in the UK: 2023 bulletin

• our Marriages in England and Wales: 2021 and 2022 bulletin

• our Marriage and civil partnership status in England and Wales: Census 2021 article

Sexual orientation

Sexual orientation is an umbrella term that includes sexual identity, attraction, and behaviour. The Annual Population Survey (APS) question informing the statistics in this bulletin is designed to capture sexual identity. The question was not designed for specific or detailed studies of sexual attraction or behaviour where a series of more detailed questions and answer categories might be more appropriate.

Sexual identity

Sexual identity has been identified as the component of sexual orientation most closely related to experiences of disadvantage and discrimination, as explained in our Measuring sexual identity: A guide for researchers (PDF, 117KB). Sexual identity does not necessarily reflect sexual attraction and/or sexual behaviour. These are separate concepts not currently measured by the APS. Sexual identity may change over time.

Sexual orientation categorised as "don't know/refuse"

"Don't know/refuse" is not a specific response option on the survey. The interviewer will code this if the person indicates that they do not want to answer the question. We cannot break this category down any further. In 2023, 1.8% of people did not know how they identified themselves or refused to answer the question.

"Don't know/refuse" is distinct from "other", which is a response option that a person may choose if they do not identify with the "heterosexual or straight", "bisexual", or "gay or lesbian" response options.

Sexual orientation categorised as "other"

The "other" category captures people who do not consider themselves to fit into the "heterosexual or straight", "bisexual", or "gay or lesbian" categories. It might also include people who responded "other" for different reasons, such as those who did not understand the terminology or who are against categorisation.

Annual Population Survey

Sexual orientation estimates are based on data from the Annual Population Survey (APS), which collects information on self-reported sexual identity from the household population aged 16 years and over in the UK. The household population excludes people living in communal establishments, except for those in NHS housing and students in halls of residence, which are sampled via the private households of their parents.

The coefficient of variation (CV) included in the published tables indicate the robustness of each estimate. The CV is categorised from "A" to "D". "A" is considered precise and "D" is considered unreliable for practical purposes. Where estimates are considered unreliable for practical purposes, these estimates are based on small sample sizes. These results should be interpreted with caution and considered alongside the confidence intervals published with our estimates.

Official statistics in development

APS estimates are currently not considered accredited official statistics and are designated as official statistics in development. Until September 2023, these were called "experimental statistics". Read more about the change in our Guide to official statistics in development.

More information about the APS being classed as "official statistics in development" can be found in this Office for National Statistics (ONS) letter to the Office for Statistics Regulation about the Annual Population Survey.

Labour Force Survey and Annual Population Survey reweighting

Labour Force Survey (LFS) estimates have been reweighted to more recent population estimates, as described in our Impact of reweighting on LFS key indicators: December 2024 article. APS estimates have not been reweighted in line with this recent LFS reweighting exercise, and therefore remain on population estimates as described in our Impact of reweighting on Labour Force Survey key indicators: 2022 article. APS estimates of levels should therefore be treated with caution, while estimates of percentages remain more robust.

More quality and methodology information on revisions, strengths, limitations, appropriate uses, and how the data were created is available in the Sexual orientation Quality and Methodology Information.

Census 2021

There are differences between survey-based and Census 2021-based estimates of sexual orientation. This is because of differences in the questions asked, including the available response options, and the larger number of responses to the Census 2021 question overall.

The method of delivery can also affect engagement. For example, it is a legal requirement to complete Census 2021, though the sexual orientation question was voluntary. This may yield different results to a voluntary survey.

Additionally, it was possible to provide a proxy response on someone else's behalf on the census. The APS does not permit proxy responses for the question on sexual identity. Users should not make direct comparisons between Census 2021 and APS estimates.

More information can be found in the Cabinet Office's Help Shape Our Future white paper and in our Sexual orientation question development for Census 2021 article. 

Opposite-sex civil partnerships

Civil partnership became available in 2019 for opposite-sex couples. APS estimates of civil partnerships among those identifying as heterosexual or straight are high for recent years, compared with data from Census 2021 and civil partnership registrations for England and Wales. This may be because respondents mistake the term "civil partnership" for other non-legal partnership types, such as long-term cohabitation. For this reason, we have not discussed civil partnerships among those identifying as heterosexual or straight in this bulletin.

While APS estimates of civil partnerships among those identifying as heterosexual or straight were high, compared with other data sources, they only represented a small proportion (0.5%) of people in this group in the UK. APS estimates of civil partnerships among those identifying as gay or lesbian are in line with estimates of same-sex civil partnerships from other data sources.