This article presents experimental research estimates of subnational sexual identity for the combined years of 2013 to 2015. These estimates are based on a 3-year Annual Population Survey (APS) pooled dataset, which has been used for the first time to produce detailed estimates of sexual identity for the UK, constituent countries and English regions and also basic analysis of sexual identity for some English counties and local authorities of Great Britain. The estimates are also displayed within the datasets associated with this release.
Publishing Experimental Statistics, including measures of quality, allows us to engage users with the research that we are currently undertaking and display the evidence that we are using to assess new methods.
This article draws conclusions on the effectiveness of using this method to produce subnational sexual identity estimates.Back to table of contents
The subnational sexual identity estimates published here are experimental research statistics and are therefore not official statistics. The latest experimental official statistics are the Sexual identity estimates for the UK, published in October 2016, which present estimates for the UK for individual years from 2012 to 2015. It is recommended that the publication Sexual identity estimates for the UK should be used for any analysis of sexual identity in the UK.
The estimates presented in this article will not match or be comparable to the official estimates – this is because these estimates were produced using a 3-year Annual Population Survey (APS) pooled dataset and the official UK estimates were produced using a single-year APS dataset.
We are only able to publish estimates for a limited number of local authorities and counties – this is due to small sample sizes resulting in unreliable estimates.
Sexual identity is one part of the umbrella concept of “sexual orientation”. Sexual identity does not necessarily reflect sexual attraction or sexual behaviour – these are separate concepts which we currently do not measure.
Sexual identity estimates are based on social survey data from the APS. The questions asked in the APS collect information on self-perceived sexual identity from the household population aged 16 and over in the UK.
Currently no further breakdown of the category “Other” is collected.
This article presents percentages. Estimates of the population numbers and measures of quality (to show the levels of uncertainty associated with survey estimates) are presented in the datasets. You are advised to consult the quality measures, when interpreting the estimates.Back to table of contents
These estimates have been produced using an Annual Population Survey (APS) 3-year pooled dataset, which covers the period from January 2013 to December 2015.
The single year APS is constructed by combining data collected from waves 1 and 5 of the quarterly Labour Force Survey (LFS) plus annual Local Labour Force Survey (LLFS) “boosts” for England, Wales and Scotland. The single year APS is published quarterly for overlapping annual periods.
The design of the 3-year pooled APS dataset means that no person appears more than once and cases included are spread approximately equally across the 3 years.
The 3-year pooled dataset is constructed using:
wave 5 LFS for year 1
wave 1 and 5 LFS for year 2
wave 1 LFS for year 3
wave 1 and 4 APS “boost” for all years
The method used in constructing the 3-year pooled dataset and therefore the sexual identity estimates is new and experimental; it should not be seen as the defining method in producing sexual identity estimates.
It has not been possible to produce estimates for all of the local authorities within England, Scotland and Wales or the counties within England. This is due to small sample sizes resulting in unreliable estimates.
For counties and local authorities we have applied a rule to the estimates that requires a sample of three or more within each sexual identity category for each county or local authority in order to publish the sexual identity estimates for that particular area.
The rule has been applied to ensure that any estimates that may have been suppressed as unreliable cannot be deduced from any remaining estimates and that unreliable distributions, for example, 100% of the population reporting as heterosexual or straight, are not published because we cannot be sure if these distributions occur because nobody identifying as lesbian, gay or bisexual were recorded in the survey or because none exist in the population.
It is important to note that if the survey and construction of the dataset were carried out again, the result may produce a different selection of counties or local authorities that we can publish after applying the rule noted above.
Estimates are not available on request for the local authorities or counties that are not included in this publication.
For more information on quality, please refer to the quality indicators that accompany each table within the datasets.
the strengths and limitations of the data and how it compares with related data
users and uses of the data
how the output was created
the quality of the output including the accuracy of the data
Figure 1 and Figure 2 show the proportion of males and females within each constituent country of the UK who identify as gay or lesbian and bisexual. The charts also display the confidence intervals for the percentages, which provides an estimated range of values in which the actual percentage is likely to fall 95% of the time. This quality measure shows the robustness of the percentage estimate.
In England, 1.5% of males identified as gay or lesbian, this had a 95% confidence interval of plus or minus 0.1%. Fewer males identified as bisexual. An equal proportion (0.7% (plus or minus 0.1%)) of females identified as gay or lesbian and bisexual in England.
Sexual identity distributions in Wales and Scotland were quite similar. Although Northern Ireland figures for males show relatively similar patterns, those for females display a reverse trend, noting the relatively large confidence intervals associated with these figures.
Figure 3 and Figure 4 show the proportion of males and females within each English region who identify as gay or lesbian and bisexual. The charts also display the confidence intervals around the percentages.
In London, 3.0% (plus or minus 0.4%) of males identified as gay or lesbian, with 0.5% (plus or minus 0.2%) identifying as bisexual. The proportion of females in London identifying as gay or lesbian was smaller than males, whereas the proportion of females in London identifying as bisexual was larger than males.
Back to table of contents
Figure 5 and Figure 6 show the proportion within each age group for each UK country who identify as gay or lesbian and bisexual. The chart also displays the confidence intervals around the percentages.
Across all four UK countries, proportions of those identifying as gay or lesbian and bisexual decreased with age for each consecutive age group. For example, in England, 1.5% (plus or minus 0.2%) of 16 to 34 year olds identified as gay or lesbian, of those aged 60 and over only 0.5% (plus or minus 0.1%) identified as gay or lesbian.
Figure 7 and Figure 8 show the proportion within each age group for selected English regions who identify as gay or lesbian and bisexual. The charts also display the confidence intervals around the percentages.
All regions of England saw a decrease in gay or lesbian proportions for each consecutive age group as demonstrated by the East Midlands in Figure 7; the only exceptions were London where the proportion of 35 to 49 year olds identifying as gay or lesbian (2.4% (plus or minus 0.4%)) was the same as the proportion of 50 to 59 year olds (2.4% (plus or minus 0.5%)) identifying as gay or lesbian and the South East where 35 to 49 year olds and 50 to 59 year olds showed the same proportions identifying as gay or lesbian, with slightly higher proportions of those aged 60 and over identifying as gay or lesbian.
Only London, Yorkshire and The Humber and the North West saw a decrease in bisexual proportions for each consecutive age group. The other English regions showed more diverse patterns by age group for those identifying as bisexual.
Back to table of contents
For English counties, we are only able to publish estimates for 31 out of the 35 recognised counties. The four remaining counties (Buckinghamshire, Cumbria, Lincolnshire and Suffolk) did not have estimates that were based on sample sizes considered robust enough for publication. Please see the Quality and methodology section for more information.
Figure 9 shows the proportion within each county who identify as gay or lesbian. The chart also displays the confidence intervals around the percentages.
Inner London had 3.1% (plus or minus 0.4%) of its population who identified as gay or lesbian.Back to table of contents
Using the 3-year Annual Population Survey (APS) pooled dataset to produce subnational estimates of sexual identity does not seem to be a robust method, particularly for English counties and local authorities of Great Britain. The method means that for a few counties and a substantial number of local authorities we cannot publish a robust estimate of their sexual identity distribution.
Increasing the sample size, either through a larger surveyed sample or through increasing the number of pooled years with a dataset could improve sample sizes enough for us to produce and publish robust estimates for all counties and local authorities. However, rolling together more years will affect the timeliness of the estimates.
We will monitor the sample sizes for sexual identity at the subnational level within the next APS 3-year pooled dataset to assess whether these estimates are robust enough to publish.
We are also conducting research towards making a recommendation regarding the inclusion of a sexual identity question on the 2021 Census. This research update provides further information about the progress of that work.Back to table of contents
Contact details for this Article
Telephone: +44 (0)1329 444661