Official Statistics

Community Life Survey 2023/24: Loneliness and support networks

Published 4 December 2024

Applies to England

1. Loneliness

GSS harmonised questions on different aspects of loneliness were included in the 2021/22 survey.

These are:

1. How often do you feel lonely? (direct measure)

2. How often do you feel that you lack companionship? (indirect measure)

3. How often do you feel left out? (indirect measure)

4. How often do you feel isolated from others? (indirect measure)

For further information, see Measuring loneliness: guidance for use of the national indicators on surveys - Recommended measures for adults.

1.1 Headline findings

During October 2023 to March 2024 (2023/24):

7% of people (approximately 3.1 million people in England) reported that they felt lonely often or always, a slight increase of 1 percentage point from 6% in 2021/22. 

As shown in Figure 1.1: Prior to 2023/24, the proportion of adults reporting that they felt lonely often or always has remained broadly stable at between 5% and 6% since 2013/14 when push-to-web data collection commenced for the Community Life Survey.

20% of people (approximately 8.8 million people in England) reported that they never felt lonely. This was in line with levels in 2021/22 (21%). 

Figure 1.1: The percentage of adults (16+) reporting that they feel lonely often or always, England: 2013/14 to 2023/24 (Base: All adults) 

1.2 Who was more likely to experience loneliness?

Age

There was some variation in the proportion of adults of different ages that experienced loneliness, for example adults aged 16 to 24 (10%), and 25 to 34 (9%), were more likely to report that they felt lonely often or always than adults from all other age groups (4% to 8%). 

Disability Status

Disabled adults (15%) were more likely to report that they felt lonely often or always than non-disabled adults (3%).

Sex

Female adults (8%) were more likely to report that they felt lonely often or always than male adults (6%). 

Gender identity ^{[footnote 1]}

Adults whose gender identity was different from their sex registered at birth (19%) were more likely to report that they felt lonely often or always than adults whose gender identity was the same as their sex registered at birth (7%).

Sexual Orientation

Adults in the ‘other’ sexual orientation classification (17%) and bisexual adults (16%) were more likely to report that they felt lonely often or always than gay or lesbian adults (12%) or heterosexual or straight adults (7%). Heterosexual or straight adults were less likely to feel lonely often or always than adults from all other sexual orientation classifications.  

Ethnicity^{[footnote 2]}

Compared with the England average (7%), adults from the Gypsy or Irish Traveller (19%), mixed White and Black Caribbean (13%), ‘any other’ mixed or multiple ethnic background (12%), Black Caribbean (11%), mixed White and Black African (10%) and Bangladeshi (10%) ethnic groups were more likely to feel lonely often or always. Conversely, adults from the Chinese ethnic group (5%) were less likely to state they felt lonely often or always than the England average. 

Religion

There was some variation in the proportion of adults from different religious groups that experienced loneliness, for example adults from ‘any other’ religious group were more likely than adults from all other religious groups to say that they felt lonely often or always (14% compared with 6% to 9%). Non-religious adults (8%) and Muslim adults (9%) were more likely than Christian (6%) or Hindu (6%) adults to say that they felt lonely often or always. 

Socio-economic classification

Adults from the never worked and long-term unemployed (15%) classification were more likely than adults from all other socio-economic classifications to report feeling lonely often or always (5% to 8%). Adults from the routine and manual operations (8%) classification were more likely to feel lonely often or always than adults from the intermediate occupations (6%) or from the higher managerial, administrative and professional occupations (5%) classifications. Adults from the higher managerial, administrative and professional occupations classification were less likely to feel lonely often or always than adults from all other socio-economic classifications. 

Index of multiple deprivation^{[footnote 3]}

There was some variation in the proportion of adults that reported feeling lonely often or always between different deprivation deciles. Generally, as deprivation increased, feelings of loneliness also increased. Adults from the two lowest deprivation (the most deprived) deciles (11%) were more likely to say that they felt lonely often or always compared with adults from all other deciles (8% to 4%). 

Population density

Adults living in urban areas (7%) were more likely to report that they felt lonely often or always than adults living in rural areas (5%).

1.3 Geographical findings

Regional differences

Compared with the England average (7%), adults living in the North East (8%) ITL1 region were more likely to report that they felt lonely often or always while adults in the South East (6%) ITL1 region were less likely.

Local Authority differences^{[footnote 4]}

The proportion of adults who reported feeling lonely often or always was in line with the England average in 263 (89%) local authorities, higher than the England average in 9 (3%) local authorities, and lower than the England average in 24 (8%) local authorities, although some of these differences were small.

In general, patterns of feeling lonely often or always at the regional level are also observed at local authority level. However, levels of loneliness were lower than the England average for adults living in the South East (6%) ITL1 region, but higher than the England average for adults living in the South East local authority of Portsmouth (11%).

Figure 1.2: Levels of loneliness (feeling lonely often or always),by region and local authority, England: 2023/24 (Base: All adults)

2. Indirect measures of loneliness

To address variation in how people understand the term “loneliness” and potential reluctance to report their feelings of loneliness, the Community Life Survey also asks three questions relating to, but not explicitly asking about, loneliness. Responses to these questions were combined to produce a composite loneliness score running from 3 (indicating less frequent loneliness) to 9 (indicating more frequent loneliness).

The three indirect questions asked respondents:

1. How often do you feel that you lack companionship?

2. How often do you feel left out?

3. How often do you feel isolated from others?

For further information, see Measuring loneliness: guidance for use of the national indicators on surveys - Recommended measures for adults.

2.1 Headline findings

During October 2023 to March 2024 (2023/24):

10% of adults (approximately 3.9 million people in England) reported high levels of indirect loneliness (a score of 8 or 9), a slight increase of 1 percentage point from 21/22 (8%).

In 2023/24, the proportion of adults who reported low levels of indirect loneliness (a score of 3 or 4) was 56%, similar to 2021/22 (58%).

Figure 2.1: The percentage of adults (16+) reporting high levels of indirect loneliness (composite score of 8 or 9), England: 2013/14 to 2023/24 (Base: All adults)

The demographic subgroups more likely to report high levels of indirect loneliness were in line with those more likely to have reported feeling lonely often or always across disability status, sex, gender identity, sexual orientation, socio-economic classification, IMD and population density. Slightly different patterns were found for the following demographic subgroups:

• Age: There was some variation in levels of indirect loneliness between adults of different ages, for example adults aged 16-24 (12%) were more likely to report high levels of indirect loneliness than adults from all other age groups (11% to 7%), with the exception of those aged 25-34 (11%). Adults aged 25-34 and 35-49 were more likely to report higher levels of indirect loneliness than adults aged 50-64 (9%), 65-74 (6%) and 75 and over (7%).  Those aged 65-74 were less likely to report higher levels of indirect loneliness than adults from all other age groups. 

• Religion: There was lots of variation in levels of indirect loneliness between adults from different religious groups, for example adults from ‘any other’ religious group (18%) were more likely than adults from all other religious groups to report high levels of indirect loneliness (7% to 12%). Christian adults (8%) were less likely than adults from all other religious groups (10% to 18%) to report high levels of indirect loneliness, with the exception of Hindu adults (9%) and Jewish adults (7%).   

• Ethnicity^{[footnote 2]}: Compared with the England average (10%), adults from the Gypsy and Irish Traveller (25%), Black Caribbean (16%), mixed White and Black Caribbean (16%), Bangladeshi (13%), ‘any other’ mixed or multiple ethnic background (13%) and Pakistani (12%) ethnic groups were more likely to report high levels of indirect loneliness. . Conversely, adults from Chinese (6%) and White British ^{[footnote 5]} (9%) ethnic groups were less likely to report high levels of indirect loneliness than the England average. 

The geographical differences were also similar between direct and indirect measures of loneliness. The following exceptions applied: 

• Regional differences: There was less regional variation in levels of indirect loneliness as compared to direct loneliness. Compared to the England average (10%), adults in the South East (9%) ITL1 region were less likely to report high levels of indirect loneliness. This is in line with the direct loneliness measure where adults in the South East ITL1 region were also less likely to report feeling lonely often or always. However, adults in all other ITL1 regions reported levels of indirect loneliness in line with the national average.

• Local Authority differences^{[footnote 4]}: Similar to the direct loneliness measure, levels of indirect loneliness were in line with the England average for adults in 262 (89%) of local authorities, higher than the England average in 12 (4%) local authorities, and lower than the England average in 22 (7%) local authorities. Levels of indirect loneliness were lower than the England average for adults in the South East (9%) ITL1 region, but higher than the England average for adults in the South East local authorities of Crawley (14%) and Slough (14%).

Figure 2.2: Indirect loneliness score, by region and local authority, England: 2023/24 (Base: All adults)

3. Support networks

The 2023/24 Community Life Survey included three measures to gauge access to support networks ^{[footnote 6]}:

1. Agreement with the statement ‘if I needed help there are people who would be there for me’

2. Agreement with ‘if I wanted company or to socialise there would be people I could call on’

3. Agreement that ‘there is at least one person I can really count on to listen if I needed to talk (Yes/No)’

3.1 Headline findings 

Overall, during October 2023 to March 2024 (2023/24): 

94% of adults “definitely” or “tended” to agree that if they needed help there are people who would be there for them. This is a slight decrease of one percentage point since 2021/22 (95%). 

92% of adults “definitely” or “tended” to agree that if they wanted company or to socialise there are people they could call. This is in line with levels in 2021/22.

95% of adults had at least one person they could really count on to listen when they needed to talk. This is in line with levels in 2021/22.

3.2 How does access to support networks vary by subgroups?

Age

There was some variation between adults of different ages in levels of agreement that if they needed help people would be there for them, that if they wanted company or to socialise there are people they could call, and that they had at least one person they could really count on to listen when they needed to talk. For example:

• Older adults aged 65 to 74 (96%) and 75 or over (97%) were more likely than adults from all other age groups (93% to 94%) to agree that if they needed help there are people who would be there for them.

• Adults aged 65 to 74 (94%) and 75 or over (94%) were more likely than adults aged 25 to 34 (91%), 35 to 49 (91%) and 50 to 64 (92%) to agree that if they wanted company or to socialise there would be people they could call on.

• Adults aged 75 or over (97%) were more likely to say that there was at least one person they could count on to really listen than adults from all other age groups (94% to 96%), with the exception of those aged 16 to 24 (96%).

Disability Status

Disabled adults were less likely than non-disabled adults to agree that if they needed help there are people who would be there for them (91% vs 96%), that if they wanted company or to socialise there would be people they could call on (87% vs 94%) and that they had at least one person they could count on to really listen when they needed to talk (92% vs 97%).

Sex

Female adults were more likely than male adults to agree that if they needed help there are people who would be there for them (95% vs 93%), if they wanted company or to socialise there would be people they could call on (93% vs 91%) and that they had at least one person they could really count on to listen when they needed to talk (96% vs 94%).

Gender identity^{[footnote 1]}

Adults whose gender identity was different from their sex registered at birth were less likely than those whose gender identity was the same as their sex registered at birth to agree that if they needed help there are people who would be there for them (90% vs 95%) and that if they wanted company or to socialise there would be people they could call on (86% vs 92%). There were no differences by gender identity in terms of having at least one person they could count on to really listen when they needed to talk. 

Sexual Orientation

• Heterosexual or straight adults (95%) were more likely to agree that if they needed help there are people who would be there for them than adults from all other sexual orientation classifications (88% to 92%).  

• Heterosexual or straight adults (92%) were more likely to agree that if they wanted company or to socialise there would be people they could call on than adults from all other sexual orientation classifications (80% to 89%). Adults from the ‘other’ sexual orientation classification (80%) were less likely than adults from all other sexual orientation classifications to agree with this statement.   

• Adults from the ‘other’ sexual orientation classification (87%) were less likely to report that they had at least one person they could count on to really listen when they needed to talk than adults from all other sexual orientation classifications (94% to 96%). Heterosexual or straight adults (96%) more likely than bisexual adults (94%) and adults from the ‘other’ sexual orientation classification (94%) to report they had at least one person they could count on.

Religion

There was some variation in access to support networks across adults from different religious groups, including the following:

• Buddhist adults (81%) were less likely to agree that if they needed help there are people who would be there for them compared with Christian (96%), Jewish (94%), non-religious (94%), Muslim (92%) or Hindu (92%) adults. Conversely Christian adults (96%) were more likely than adults from all other religious groups (81% to 94%) to agree, with the exception of Jewish (94%) and Sikh adults (90%). 

• The same pattern was apparent when looking at agreement with the statement ‘if I wanted company or to socialise there would be people I could call on’. 

• Buddhist adults (87%) and adults from the ‘any other’ religious group (88%) were less likely than Christian (97%), Jewish (96%) and non-religious (95%) adults to say that they had at least one person they could count on to really listen when they needed to talk. In addition, Buddhist adults were also less likely than Hindu adults to report that they had someone they could count on. Christian adults (97%) were more likely to say that they had at least one person they could count on to really listen when they needed to talk than non-religious (95%), Muslim (93%) or Buddhist (87%) adults or adults from the any other’ religious group (88%). 

Ethnicity^{[footnote 2]}

• Compared with the England average (94%), adults from the White British ^{[footnote 5]} (95%) ethnic group were more likely to agree that if they needed help there were people to be there for them. Adults from the Black Caribbean (88%), Bangladeshi (89%), Chinese (90%), mixed White and Asian (90%), and Black African (92%) ethnic groups were less likely to report that if they needed help there were people to be there for them than the England average.

• Compared with the England average (92%), adults from the White British ^{[footnote 5]} (93%) ethnic group were more likely to report that if they wanted company or to socialise there would be people to call on. Adults from the Black Caribbean (85%), Chinese (87%), Bangladeshi (87%), Irish (87%), Black African (89%), Indian (89%) and ‘any other’ White background (90%) ethnic groups were less likely to report that if they wanted company or to socialise there would be people to call on than the England average.

• Compared with the England average (95%), adults from the Black Caribbean (88%), Arab (90%) and mixed White and Asian (92%) ethnic groups were less likely to report feeling that they had at least one person they could count on to really listen when they needed to talk.

Socio-economic classification

• Adults from the never worked and long-term unemployed (90%) classification were less likely to agree that if they needed help there are people who would be there for them than adults from all other socio-economic classifications (94% to 96%). Adults from the routine and manual operations (94%) classification were less likely to agree with this statement than adults from the intermediate occupations (96%) or from the higher managerial, administrative and professional occupations (95%) classifications. 

• Adults from the never worked and long-term unemployed (86%) classification were less likely to agree that if they wanted company or to socialise there would be people they could call on than adults from all other socio-economic classifications (91% to 93%). Adults from the routine and manual operations (91%) classification were less likely to agree with this statement than adults from the intermediate occupations (93%) or from the higher managerial, administrative and professional occupations (93%) classifications.

• Adults from the never worked and long-term unemployed (91%) classification were less likely to report that they had at least one person they could really count on to listen when they needed to talk than adults from all other socio-economic classifications (94% to 96%). Adults from the routine and manual operations (94%) classification were less likely to agree with this statement than adults from the intermediate occupations (96%) or from the higher managerial, administrative and professional occupations (96%) classifications. 

Index of multiple deprivation^{[footnote 3]}

There was some variation in adults’ access to support networks by deprivation decile. Generally, across all three statements, as deprivation increased access to support networks decreased:

• Adults in the five lowest (most deprived) deciles (92% to 94%) were less likely to agree that if they needed help there are people who would be there for them than adults living in the four highest deciles (96% to 97%).

• Adults in the two lowest (most deprived) deciles (88% to 89%) were less likely to agree that  if they wanted company or to socialise there would be people they could call on than adults living in the five highest (least deprived) deciles (92% to 95%).

• Adults in the three lowest (most deprived) deciles (93% to 94%) were less likely to report that they had at least one person they could really count on to listen when they needed to talk than adults living in the four highest (least deprived) deciles (96% to 97%).

Population density

Adults living in rural areas were more likely than those in urban areas to agree that if they needed help there are people who would be there for them (96% vs 94%), that if they wanted company or to socialise there would be people they could call on (93% vs 92%) and that they had at least one person they could really count on to listen when they needed to talk (96% vs 95%).     

3.3 Geographical findings

Regional differences

Compared with the national average (94%), adults living in the South East (95%) ITL1 region were more likely to agree that if they needed help there are people who would be there for them, while adults in the London (93%) ITL1 region were less likely to agree. There were no differences by ITL1 region for the proportion agreeing that if they wanted company or to socialise there would be people they could call on, and that they had at least one person they could really count on to listen when they needed to talk.     

Local Authority differences ^{[footnote 4] [footnote 7]}

Agreement with ‘if I needed help there are people who would be there for me’

Levels of agreement that ‘if I needed help there are people who would be there for me’ were in line with the England average for adults in 274 (93%) local authorities, higher than the England average in 16 (5%) local authorities and lower than the England average in 6 (2%) local authorities, although some of these differences were small. In general, the patterns of neighbourly interactions at the regional level are also observed at the local authority (LA) level. However, there are some exceptions:

Levels of agreement that people would be there for them if they needed help were lower than the England average for adults in the London (93%) ITL1 region, but higher than the England average for adults in the London local authority of Wandsworth (97%).

Figure 3.1: Agreement that if people needed help, there are others who would be there for them, by region and local authority, England: 2023/24 (Base: Adults in subsample of approximately a third of online respondents)

Agreement with ‘if I wanted company or to socialise there would be people I could call on’.

Levels of agreement that ‘if I wanted company or to socialise there would be people I could call on’ were in line with the England average for adults living in 276 (93%) local authorities, higher than the England average in 13 (4%) local authorities, and lower than the England average in 7 (2%) local authorities, although some of these differences were small.

Figure 3.2: Agreement that if people wanted company or to socialise there are others they could call on, by region and local authority, England: 2023/24 (Base: Adults in subsample of approximately a third of online respondents)

Agreement that ‘there is at least one person I can really count on to listen if I needed to talk (Yes/No)’.

Levels of agreement with the statement that ‘there is at least one person I can really count on to listen if I needed to talk’ were in line with the England average for adults living in 277 (94%) local authorities, higher than the England average in 14 (5%) local authorities, and lower than the average in 5 (2%) local authorities, although some of these differences were small.

Figure 3.3: Agreement that people have at least one other person they can really count on to listen if they needed to talk, by region and local authority, England: 2023/24 (Base: Adults in subsample of approximately a third of online respondents)

1. Patterns were identified in Census 2021 data that suggest that some respondents may not have interpreted the gender identity question as intended, notably those with lower levels of English language proficiency. Analysis of Scotland’s census, where the gender identity question was different, has added weight to this observation. More information can be found in the ONS sexual orientation and gender identity quality information report, and in the National Statistical blog about the strengths and limitations of gender identity statistics ↩ ↩²

2.   The Community Life Survey is a sample survey, so estimates are subject to an associated sampling error that decreases as the sample size increases. To avoid reporting on estimates that are not sufficiently robust a threshold has been applied whereby estimates are excluded from the written analysis if they have an unweighted base size of 100 or less.  ↩ ↩² ↩³

3. Analysis of the Index of Multiple Deprivation was conducted using deciles. These range from group 1, the most deprived decile, to group 10, the least deprived decile. In some instances, the analysis refers to a group by number e.g. ‘group 8’, in this case the third least deprived decile. ↩ ↩²

4. The 2023/24 Community Life Survey collects over 170,000 responses, so confidence intervals are generally very narrow. Whilst this reflects a strength of the data, when highlighting differences, some local authorities may be shown to be above/below the England average, but are only a percentage point more/ less than the average. There will be other local authorities who are much more above or below the England average, but will both be shown in the same shading within the map. ↩ ↩² ↩³

5. ‘White British’ incorporates anyone who self-identifies their ethnicity as White English, Scottish, Welsh, Northern Irish, or British. ↩ ↩² ↩³

6. This question was mixed mode in 2021/22 but a web only question in 2023/24.  Analysis of the impact of mode has shown no significant mode effect for this question. ↩

7. The questions on support networks were asked of adults in a subsample of approximately a third of online respondents.  The smaller sample sizes for this question mean that estimates by local authority are less precise, and that we are less likely to see variation in the data above and below the England average.  For one local authority, the Isles of Scilly, the base was below 100, and findings have therefore been excluded from written analysis as per footnote 2. ↩