Small populations, England and Wales: Census 2021 

• Small population datasets provide census data for people in specific small population groups, defined by ethnic group, religion, national identity, main language, or country of birth.

• For the first time, we are publishing data to show the number of people across England and Wales who identified their ethnic group as Eritrean (26,100, 0.04% of all usual residents), their ethnic group as Romanian (366,300, 0.6%), either their ethnic group or religion as Valmiki (1,100, 0.002%), and their religion as Alevi (25,700, 0.04%).

• The small population groups with the highest proportion of residents aged 65 years and over included those who were born in Ireland (46.5%), Jamaica (40.2%) and Cyprus (35.8%).

• The highest proportion of a small population group in any local authority was in Leicester in the East Midlands, for those born in India (16.2%).

What are small population groups?

Small populations are defined by characteristics such as:

• ethnic group

• religion

• national identity

• main language

• country of birth

• or a combination of religion and ethnic group

The majority of small population groups are based on "write-in" responses to the census questions for these characteristics, where the respondent has chosen to provide more detailed information about how they identify. For more information on question structure, see Section 9: Measuring the data.

Because of the relatively small population size in certain groups, confidentiality constraints limit the release of data about these groups in our standard census outputs. Instead, we create bespoke datasets for small populations about which users have specific needs. This may be to better understand that small population or to analyse potential inequalities between that small population and the wider population. Read more in our blog post on the Census 2021 small populations release.

These data are produced only for geographic areas in which the small population being counted exceeds a minimum population threshold. This means that different areas will be suppressed in different datasets.

Which groups have we released data for?           

The Census 2021 small populations release includes data on 43 groups in total. A list of the groups, and the data we have published, can be found on the Nomis small populations page. Nomis is a service provided by the Office for National Statistics (ONS), for statistics related to population, society and the labour market.

The list of small population groups includes:

• 18 populations defined by ethnic group only

• 17 populations defined by country of birth only

• one population defined by national identity only

• one population defined by language only

• one population defined by religion only

• five populations defined by both ethnic group and religion

Different datasets have been produced for different small populations, depending on user need and confidentiality constraints.

First, we provide a breakdown by five-year age group and sex for a total of 42 groups, which are listed in full on the Nomis small populations page. It includes the same groups we produced for 2011 Census, as well as seven new groups based on identified user need:

• Caribbean ethnic group

• Eritrean ethnic group

• Romanian ethnic group

• Hispanic or Latin American ethnic group

• Alevi religion

• British Sign Language

• Valmiki ethnic group or religion

Additionally, we have released data on specific characteristics, such as disability, unpaid care, economic activity, employment, qualifications and socio-economic classification, for the following groups:

• British Sign Language, as a main language

• Caribbean, defined using ethnic group

• Cornish, defined using national identity

• Jain, defined using ethnic group or religion

• Kashmiri, defined using ethnic group

• Nepali and Nepalese (includes Gurkha), defined using ethnic group

• Ravidassia, defined using ethnic group or religion

• Sikh, defined using ethnic group or religion

Finally, for the small population groups that can be defined as either an ethnic group or a religion (including Jain, Jewish, Ravidassia and Sikh), we provide breakdowns for the number of people identifying with the group as their ethnic group only, their religion only, or as both their ethnic group and religion.

This bulletin provides a snapshot of the data that have been released, describing the size of different small populations, their age and sex profiles, and where they tended to live within England and Wales.

Readers may be interested in comparing the data with the 2011 Census data on small populations, as well as with the Census 2021 topic summaries covering ethnic group, language, national identity and religion, and international migration (for country of birth).

New data for Census 2021

For the first time we are publishing data on the number of people across England and Wales who identified as:

• Eritrean, ethnic group, which was 26,100 (0.04% of all usual residents in England and Wales)

• Romanian, ethnic group, which was 366,300 (0.6%)

• Valmiki, ethnic group, religion, or both, which was 1,100 (0.002%)

• Alevi, religion, which was 25,700 (0.04%)

Combined indicators - religion and ethnic group

For some groups, people identified through ethnic group, religion, or both. Our small populations data show the breakdown of these categories for four populations:

• Jain

• Jewish

• Ravidassia

• Sikh

Comparable data from 2011 Census showing the overall number of people with that identity have been published through commissioned tables, for ethnic group by religion (including the Jewish identity), the Sikh identity, the Jain identity and the Ravidassia identity.

Jain

A total of 25,100 people across England and Wales (0.04%) identified as Jain, which is a small increase, but similar proportion, compared with 2011, when 20,400, 0.04%, identified as Jain. In 2021:

• 24,500 (98.0% of all who identified as Jain) identified Jain as their religion only

• 400 (1.8%) identified Jain as both their ethnic group and religion

• 64 (0.3%) identified Jain as their ethnic group only

Jewish

A total of 287,400 people across England and Wales (0.5%) identified as Jewish, which is an increase in number, but similar proportion, compared with 2011, when 271,900, 0.5%, identified as Jewish. In 2021:

• 219,200 (76.3% of all who identified as Jewish) identified Jewish as their religion only

• 52,200 (18.2%) identified Jewish as both their ethnic group and religion

• 16,000 (5.6%) identified Jewish as their ethnic group only

Ravidassia

A total of 9,700 people (0.02%) identified as Ravidassia, which is a slight decrease since 2011, when 11,200, 0.02%, identified as Ravidassia. In 2021:

• 8,200 (84.5% of all who identified as Ravidassia) identified Ravidassia as their religion only

• 1,400 (14.5%) identified Ravidassia as both their ethnic group and religion

• 100 (1.0%) identified Ravidassia as their ethnic group only

Sikh

Finally, 525,900 people across England and Wales (0.9% of all usual residents) identified as Sikh, which is an increase from 2011, when 430,000, 0.8%, identified as Sikh. In 2021:

• 426,200 (81.1% of all who identified as Sikh) identified Sikh as their religion only

• 97,900 (18.6%) identified Sikh as both their ethnic group and religion

• 1,700 (0.3%) identified Sikh as their ethnic group only

National identity

Cornish

In total, 108,900 people (0.2%) recorded their national identity as Cornish, an increase from 83,500 (0.1%) in 2011. In 2021:

• 89,100 (81.8% of all who identified as Cornish) identified as Cornish only

• 10,700 (9.8%) identified as Cornish and British only

• 9,100 (8.4%) had a Cornish identity and at least one of English, Welsh, Scottish or Northern Irish identities (with or without a British identity)

The population pyramid in Figure 1 shows the age and sex breakdown for different small population groups.

Figure 1: Small population groups by age and sex

Notes: 

1. For the British Sign Language group only, the youngest age group is "Aged 3 to 4 years" rather than "Aged 4 years and under". This is because census data on language are only reported for the population aged 3 years and over.

Download the data

.xlsx

Country of birth

The data show that most small populations defined by country of birth had a relatively even split of males to females, aligning with the data on all usual residents across England and Wales (51.0% female, 49.0% male). Of the 17 country of birth groups in this data release, those born in the Philippines had the highest ratio of females (65.5%) to males (34.5%), whereas those born in Turkey had the highest ratio of males (53.4%) to females (46.6%).

The data on age show that the England and Wales population born in Ireland had the highest percentage of respondents aged 65 years and over (46.5%) of any small population group. This is considerably higher than the percentage of those aged 65 years and over across England and Wales as a whole (18.6%). Out of the 17 country of birth groups in this data release, the next highest groups were those born in Jamaica (40.2%) and Cyprus (35.8%). The groups with the lowest percentage of people aged 65 years and over were Romania (1.2%) and Bulgaria (2.8%).

The population born in France had the highest percentage of people aged 0 to 19 years (15.6%), which is lower than the percentage of people aged 0 to 19 years across England and Wales as a whole (23.1%). This was followed by those born in Romania (12.3%) and Bulgaria (12.1%). The country of birth groups with the lowest percentage of people aged 0 to 19 years were Jamaica (1.3%) and Sri Lanka (4.3%).

Ethnic group

Of the 43 groups in this data release, 18 are defined by ethnic group only. These are listed in full on the Nomis small populations page. Again, most small populations defined by ethnic group had a relatively even split of males to females. In accordance with the country of birth data, Filipino was the ethnic group with the highest ratio of females (61.8%) to males (38.2%). Comparatively, Kurdish was the ethnic group with the highest ratio of males (57.5%) to females (42.5%).

For age, Greek Cypriot was the small population ethnic group with the highest percentage of residents aged 65 years and over (22.4%), followed by Turkish Cypriot (17.1%) and Caribbean (9.0%). The small population ethnic groups with the lowest percentages were Romanian (1.0%) and Kurdish (1.9%).

The ethnic group with the highest percentage of people aged 0 to 19 years was Somali (46.4%), followed by Afghan (43.3%) and Kurdish (36.6%). The lowest percentages were for the Greek Cypriot (14.2%) and Greek (16.7%) ethnic groups.

Combined indicators – religion and ethnic group

Of the 43 small population groups, 5 were defined by both ethnic group and religion identities. Age and sex breakdowns are available for 4 of these. There was a higher percentage of females (52.3%) among those who identified as Jain. The percentage of males was higher among those identified as Ravidassia (51.8%), Valmiki (51.6%) and Sikh (50.1%).

The Jain group also had a higher percentage of those aged 65 years and over (22.7%), and a lower percentage of those aged 0 to 19 years (16.6%), than the other groups. Conversely, the Sikh group had the lowest percentage of those aged 65 years and over (12.2%), and the highest percentage of those aged 0 to 19 years (25.1%).

National identity, language and religion

Finally, age and sex data are also provided for the one national identity group, one language group, and one religious group in the small populations data.

For those with a Cornish national identity, 50.8% were male and 49.2% were female. A total of 25.6% were aged 65 years and over, whereas 18.4% were aged 0 to 19 years.

For those whose main language was British Sign Language, 52.9% were male and 47.1% were female; 15.3% were aged 65 years and over, and 19.8% were aged 3 to 19 years.

Finally, for those who identified with the Alevi religion, 50.2% were male and 49.8% were female; 4.2% of this group were aged 65 years and over, whereas 28.4% were aged 0 to 19 years.

The interactive map in Figure 2 shows the geographic distribution of each small population group. Population counts are only provided for areas with 200 or more usual residents (for local authorities) or 100 or more usual residents (for middle layer super output areas) from the chosen small population group.

Figure 2: Small populations, 2021, local authorities and middle layer super output areas in England and Wales

Download the data

.xlsx

Many small populations tended to be clustered within London. For example, some of the local authorities with the highest percentage of any small population group were areas of south London, for the Caribbean ethnic group. The highest proportions were in Lewisham where 13.8% of the usual resident population identified with the Caribbean ethnic group, followed by Croydon (12.3%) and Lambeth (12.1%).

There were also relatively high percentages of residents born in India in areas to the west of central London, such as Hounslow (13.2%), Slough (10.8%) and Brent (10.4%). However, the highest percentage was in Leicester in the East Midlands (16.2%). This was the highest proportion of a small population group in any local authority.

Other examples of London local authorities with particularly high proportions of a small population group included residents of Tower Hamlets who were born in Bangladesh (14.0%) and residents of Harrow who were born in Romania (8.1%).

There were also several local authorities with a relatively high percentage of people born in Pakistan, spread across England. For example, the highest proportion was in Pendle (9.9%) in the North West, followed by Slough (9.1%) in the South East, Bradford (8.3%) in Yorkshire and The Humber, and then Luton (7.6%) in the East of England.

Another noteworthy finding is that the proportion of people who identified their ethnic group as Nepali (including Gurkha) was considerably higher in Rushmoor (10.6%) in the South East, than in any other local authority. The next highest proportions were in Ashford and Reading, also in the South East, which both had 3.1%. This is likely because of Aldershot Garrison's history with Gurkha personnel, and the 2008 High Court Ruling, which gave Gurkhas who retired before 1997, and their dependent families, the right to settle in the UK.

Finally, in Wales, the highest proportion of people who identified with any of the 43 small population groups in this release, was in Wrexham for residents born in Poland (2.6%). Merthyr Tydfil also had a relatively high proportion of the population who were born in Poland (2.0%) and who identified with the Polish ethnic group (1.9%). In addition, though relatively low compared with England, there were also high proportions of people who identified with the Caribbean ethnic group in Cardiff (1.6%) and Newport (1.3%).

We will publish Census 2021 analysis articles on Somali populations, Gypsy and Irish Traveller populations and Roma populations in England and Wales in October 2023. Analysis articles for Cornish, Jewish, and Sikh populations will also follow. Read more about our ethnic group, nationality, language and religion analysis plans more generally.

Ethnic group

The ethnic group that the person completing the census feels they belong to. This could be based on their culture, family background, identity, or physical appearance.

Respondents could choose 1 out of 19 tick-box response categories, including write-in response options.

Main language

A person's first or preferred language.

National identity

Someone's national identity is a self-determined assessment of their own identity; it could be the country or countries where they feel they belong or think of as home. It is not dependent on ethnic group or citizenship.

Respondents could select more than one national identity.

Religion

The religion people connect or identify with (their religious affiliation), whether or not they practise or have belief in it.

This question was voluntary, and the variable includes people who answered the question, including "No religion", alongside those who chose not to answer this question.

This variable classifies responses into the eight tick-box response options. Write-in responses are classified by their "parent" religious affiliation, including "No religion", where applicable.

Usual resident

A usual resident is anyone who on Census Day, 21 March 2021 was in the UK and had stayed or intended to stay in the UK for a period of 12 months or more, or had a permanent UK address and was outside the UK and intended to be outside the UK for less than 12 months.

Small populations

Small populations are groups that are defined by their specific characteristics such as:

• ethnic group

• country of birth

• religion

• national identity

• main language

To protect the identity of individuals in these small population sizes, we do not release detailed information on these groups in the standard outputs.

Instead, we have released bespoke datasets for specific small populations, at all geographies where the number of that population exceeds a set threshold.

Reference date

The census provides estimates of the characteristics of all people and households in England and Wales on Census Day, 21 March 2021. It is carried out every 10 years and gives us the most accurate estimate of all the people and households in England and Wales.

We are responsible for carrying out the census in England and Wales but will also release outputs for the UK in partnership with the Welsh Government, the National Records of Scotland (NRS) and the Northern Ireland Statistics and Research Agency (NISRA). The census in Northern Ireland was also conducted on 21 March 2021, whereas Scotland's census was moved to 20 March 2022. All UK census offices are working closely together to understand how this difference in reference dates will affect UK-wide population and housing statistics, in terms of both timing and scope.

Response rate

The person response rate is the number of usual residents for whom individual details were provided on a returned questionnaire, divided by the estimated usual resident population.

The person response rate for Census 2021 was 97% of the usual resident population of England and Wales, and over 88% in all local authorities. Most returns (89%) were received online. The response rate exceeded our target of 94% overall and 80% in all local authorities.

Read more about question-specific response rates at local authority level in Section 4 of our measures showing the quality of Census 2021 estimates.

Census form questions

To see the questions on ethnic group, religion, country of birth, national identity, language, and religion on the household, individual and Welsh questionnaires, visit our Census 2021 paper questionnaires page.

The majority of small populations are based on write-in responses, where the respondent identified in a way not represented in the main tick-box response options. In Census 2021, this was supported by search-as-you-type functionality for online respondents, in which a list of suggested responses was provided as the respondent typed their answer. This sped up responses and assisted with spelling, improving the consistency of how different identities were reported.

For small populations defined by ethnic group in particular, it is worth noting that people might be assigned to the same small population group even if they selected a different higher-level tick-box option. For example, if one person selected "Any other White background" and then wrote in "Greek", and another person selected "Any other ethnic group" and then wrote in "Greek", then both people would be included within the same small population dataset for a "Greek" ethnic group.

Quality considerations, along with the strengths and limitations of Census 2021 more generally, can be found in the Quality and Methodology Information (QMI) for Census 2021.

Further information on our quality assurance processes is provided in our Maximising the quality of Census 2021 population estimates report.