Census and national data

Most people in the UK are now non-religious and the religious population is becoming increasingly diverse. However, there are lots of different ways of measuring religion or belief and they each show different things – so it’s worth unpacking what each shows and what these results mean for the real world.

The best source of information is the annual British Social Attitudes survey. It asks, ‘Do you regard yourself as belonging to any particular religion?’ In 2024, it showed just 49% of British adults are non-religious, 11% as Church of England/Anglican, 8% Roman Catholic, 21% other types of Christian, and 8% other religions. In Northern Ireland, the 2024 Life and Times Survey found 25% saying they belong to no religion, versus 34% Catholic, 15% Presbyterian, 12% Anglican, 11% other Christian, and 2% other religions.

Apart from religious belonging, other measures include religious practice, with monthly church attendance at under 5%; religious beliefs – with around a third having humanist beliefs and values; and loose cultural affiliation, which shows 47% of the population identifying with Christianity and 38% with no religion.

For policymakers, the important question is what to do with this information. Our advice is that it is best to measure directly what people actually say they want or need because this often differs from people’s stated religions or beliefs.

In depth

How religious the UK population appears to be depends upon the question that is asked. But broadly speaking there are four different ways of measuring religiosity:

1. Based on ‘belonging’ to a religion, or identifying as religious;
2. Based on loose cultural affiliation;
3. Based on believing in the core tenets of a particular religion
4. Based on levels of religious practice (whether self-reported or observed).

All of these measures need to be taken into account by anyone trying to come to a full picture of the religiosity of the UK. Regrettably the most high-profile demographic survey, the Census, only asks about loose cultural affiliation, but its results are often presented as being a measure of a closer religious identity. It also doesn’t measure how many humanists there are.

On top of that, religious individuals hold a range of attitudes towards ethical issues which are frequently out of line with the position of their religion’s hierarchy and closer to Humanists UK’s.

1. Belonging and identifying: Social attitudes surveys

In Britain, the percentage of the population which describes itself as belonging to no religion has risen from 31% to 49% between 1983 and 2024 according to the British Social Attitudes survey. Conversely, the report found that only 40% of people in the UK identify as Christians compared to 65% in 1983. The Church of England has seen the greatest decline in its numbers; belonging has gone from 40% of the population in 1983 to just 11% in 2024.

Among people aged between 18 and 24, 1% say they belong to the Church of England, 4% Roman Catholic, 15% other Christian denominations, 15% other religions, and 64% say they belong to no religion. It is only among over-55s that most respondents are religious.

In Northern Ireland, the 2024 Life and Times Survey found 25% saying they belong to no religion, versus 34% Catholic, 15% Presbytarian, 12% Anglican, 11% other Christian, and 2% other religions.

A 2014 YouGov poll found that only 3% of British adults consider themselves to be ‘very religious’ and only 20% ‘fairly religious’, while 37% consider themselves to be ‘not very religious’ and 40% ‘not religious at all’.

2. Loose cultural affiliation: Census data

The English and Welsh Census uses the highly leading question ‘What is your religion?’ Similarly, the Scottish and Northern Ireland Censuses ask ‘What religion, religious denomination or body do you belong to?’ By assuming that all participants hold a religious belief, the question captures some kind of loose cultural affiliation, and as a result the censuses recorded a far higher percentage of the population as ‘Christian’ than other ways of measuring it. The placement of the question alongside questions of ethnicity and national identity only further compounded this issue.

According to the 2021-2 UK censuses, those of no religion are the second largest belief group, making up 38% of the population. 47% described their religion as Christian, 6% Muslim, 3% other religions, and 6% no response. This represents a massive change from earlier censuses, where the figures were 72% Christian and 23% No religion in 2001; and 59% Christian and 26% no religion in 2011.

However, at the time of the last Census, a poll conducted by YouGov on behalf of Humanists UK found just 34% of English and Welsh adults who ticked ‘Christian’ said they did so because they ‘believe in the teachings of Christianity’, 27% did so because they ‘believe that Jesus Christ was a real person who died and came back to life, and was the son of God’, and just 14% did so because they ‘go to religious services at a church other than weddings, christenings/ baptisms or funerals’. 55% did not select any of these three options.

Instead, people in England and Wales were more likely to tick ‘Christian’ because they were christened (59%) or brought up to think of themselves as a Christian (49%), because one of their parents is/was a Christian (44%), or they went to a Christian or Sunday school as a child (42%). 26% also said they had ticked ‘Christian’ because ‘this is a Christian country’, while 12% said ‘it reflects my ethnicity’. None of these reasons reflect these people’s religious beliefs or practices today.

Most British adults who ticked ‘Christian’ said they either never attend a place of worship (27%) or do so less than once a year (24%). Similarly, 29% of those who ticked another religious answer said they never attend a place of worship, while 14% said they do so less than once a year.

When asked how religious they are, 26% of those ticking ‘Christian’ said that they are very or somewhat non-religious, as did 20% of those ticking another religious option. 30% of those ticking ‘Christian’ and 14% of those ticking another religious answer said they are neither religious nor non-religious.

The Office for National Statistics uses the religion question as a proxy question for ethnicity. This is in order to capture the Jewish and Sikh populations, both of which are captured under equalities legislation as ethnic groups but are not included in the ethnicity category in the Census, as they should be, rather than the religion category. The result is that a very loose cultural affiliation is ‘measured’ by the Census in terms of religion or belief, with particular over-inflation of the Christian figure, and an undercounting of the non-religious population. As a result, the census data on religion is most definitely not suitable for use by employers, service providers or politicians as a proxy for religious belief, belonging or practice.

Another problem with the Census is its misleading results around humanism specifically. Under ‘Other religion’ people could write in their own answer. 10,000 people wrote in ‘Humanism’. But this is meaningless. Almost all humanists ticked the ‘No religion’ box instead, an option more suitable for them. Humanism is not a religion. The Census has a write-in option where people can write in any words they wish but this doesn’t mean very much. The fact that 10,000 wrote in ‘Humanism’ means as much as that only 2,414 people (0.00% of the population) wrote in ‘Believe in God’.

3. Beliefs and values

There have also been opinion polls seeking to measure how many people hold humanist beliefs and values. A 2019 YouGov poll of British adults commissioned by Humanists UK found that 51% of British adults say they do not belong to any religion, and 7% identify ‘humanist’ as the word that best describes them (the other options presented to the remaining 44% being atheist, agnostic, spiritual, naturalist, none of these, and don’t know). Some of those who chose other options would likely also subscribe to the label ‘humanist’, but even setting that aside, the results suggest that around 7% of British adults primarily identify as a humanist.

With that said, humanism is different from the major religions in being a descriptive label for a set of beliefs that have existed throughout history and across the world. Often when people come to self-identify as a humanist they say they have ‘discovered’ a term that has long applied to them. This doesn’t happen with religions but is more akin to sexual orientation, for example. To try to get closer to understanding this phenomenon, the poll then asked a series of questions about their beliefs about religion, ethics, morality, and reason. The results found that 29% of the population has a non-religious outlook on life that matches the humanist one.

Surveys show a lower level of belief in the core tenets of Christianity than do measures of cultural affiliation or belonging. For instance, a 2013 YouGov survey found that just 27% of the population believe that Jesus Christ was the son of God, just 26% believe in the Biblical account of the crucifixion, just 22% believe in the devil and just 33% believe in life after death.

4. Attendance at places of worship

The 2024 British Social Attitudes survey also found that 66% of the population never attend religious services while only 9% of people report going to a religious service once a week or more, down from 12% six years earlier; and 14% attending monthly or more.

However, social desirability bias means the self-reported Church attendance is invariably higher than actual recorded attendance. Estimates from recorded attendance put the monthly figure at about a third of the reported figure.

The Church of England’s own attendance figures also attest to decline; in 2024 average monthly attendance figures were around 1 million. Average Sunday attendance figures were 581,000, a third of the number that attended in 1968 and slightly over half the 2002 figure of 1,005,000.

And on top of all this, recorded Church attendance is lower than the proportion of parents who admit to attending church regularly simply in order to get their children into church-controlled schools. The Church of England’s own research has found that the presence of nearby oversubscribed Church of England schools has a bigger impact on church growth than anything else.

In 2025 the Bible Society published a report claiming a revival in church attendance. This was based on two large YouGov polls. But other, even larger YouGov polls showed decline over the same period. YouGov ascribed these differences to question wording and weighting. In other words, it said it didn’t know which polls were right. Other sources – such as recorded church attendance and the British Social Attitudes survey – instead show decline over the same period.

Policy attitudes

Attitudes towards belief

A 2014 Survation survey found that 68% of the British public think atheists are at least as moral as the religious, with one in eight thinking they are more moral and just 6% think they are less moral. Meanwhile, according to YouGov in 2025, 58% of British adults think religion has a negative influence on the world, versus 19% positive. Another YouGov poll found that just 25% of British adults have a favourable view of the Church of England, with 49% having an unfavourable view.

The 2023 British Social Attitudes Survey found that just 19% of British people think it is ‘very’ or ‘fairly important to be a Christian’ to be ‘truly British’, down from 32% in 1995. This means that Christianity is the least important of the nine factors that were asked about.

Religion and government

A 2024 Whitestone Impact poll for Institute for the Impact of Faith in Life found that 71% of British adults agree that ‘religion and law should be kept strictly separate’, with just 12% disagreeing.

A 2012 YouGov poll found that 67% of people do not think that religion should play any role in public life.

In the British Social Attitudes Survey 2010:

• 75% of those questioned believed their religious leaders should not try to influence their voting behaviour.

• 67% believe religious leaders should stay out of government decision-making.

• 45% of Britons believe that the involvement of religious leaders would have a deleterious effect on policy.

• Only 25% of people believe religious involvement would produce better policy.

• 73% of respondents believe that ‘people with very strong religious beliefs are often too intolerant of others’. This view was held by 82% of people who class themselves as non-religious, and 63% of those who consider themselves religious.

Religion and education

In June 2014, an Opinium poll found that:

• 58% of the British public was opposed to the existence of state-funded faith schools, with just 30% accepting state funding.

• 70% of those opposed to state funding said this because they think the taxpayer should not be funding religion, 60% because they think ‘faith’ schools promote division and segregation, and 41% because they think they are contrary to the promotion of a multicultural society.

• 56% of respondents said that ‘faith’ schools should teach the national curriculum, with only a small minority arguing that they should have significant flexibility over what they teach.

In a ComRes/Accord poll from November 2012, 73% of British adults agreed (and 50% strongly agreed) that ‘state funded schools, including state funded faith schools, should not be allowed to select or discriminate against prospective pupils on religious grounds in their admissions policy’. Just 18% disagreed.

In an Ipsos Mori poll commissioned by the teachers’ union NASUWT and Unison in April 2010, when asked which group is the most appropriate to run state-funded schools, only 4% of British adults answered ‘religious organisations’. When asked which groups should not run state-funded schools, 35% said religious organisations (the highest figure obtained by any of the answers listed).

In a YouGov/Accord poll from June 2009:

• 57% believed that state-funded schools that select students according to their religion harm community cohesion.

• 72% agreed or strongly agreed that all schools should implement recruitment and employment policies that do not discriminate on grounds of religion or belief.

• 74% held the view that all state schools should teach an objective and balanced syllabus for education about a wide range of religious and non-religious beliefs.

Ofsted’s Parents Annual Survey 2021 asked parents to pick their top three factors from a list of twelve for choosing which school to send their children to; only 6% of those with school-age children picked religion. The reasons given as important were location, ethos or values, and Ofsted results.

In 2019, British parents responding to a YouGov poll ranked religious worship last in a list of 13 possible topics that could be covered in assemblies. Just 29% thought worship was appropriate. This compares with 76% who thought ‘the environment and nature’ would be appropriate. 74% who thought ‘equality and non-discrimination’ were appropriate. And 73% who thought ‘celebration of achievements’ and ‘physical and mental health’ should feature. Polling conducted in 2021 also showed that 60% of British parents with school-age children oppose the collective worship law being enforced. Just 24% think it should be.

In Northern Ireland, a 2024 LucidTalk poll of parents found that 70% want schools to be mixed, with 67% wanting integrated education to be the main model. A 2025 poll found that most NI adults of parent age don’t think collective worship should be compulsory, with just a third thinking it should be. Most support inclusive RE, with a third supporting the current Christian instruction approach. 55% of respondents said they do not agree with the legal requirement for church representatives to have seats on school boards, with only 36% backing the status quo.

Assisted dying

Polls taken on the issue of assisted dying consistently demonstrate that most of the public wish the law to be reformed, and to create a humane and ethical law on assisted dying. An October 2024 MRP poll conducted by Electoral Calculus found that 74% of the public supported legalisation for terminally ill and incurably suffering people, compared to just 6% opposed. Furthermore, in every constituency, more people support than oppose assisted dying, and in 631 of the 632, most people support it.

Polls also consistently show that most religious people of all major denominations and most disabled people support assisted dying.

Abortion and contraception

A 2023 YouGov poll for MSI found that 87% of British adults think abortion should be legal, with most supporting decriminalisation, 77% supporting safe access zones outside abortion clinics, and 63% supporting allowing a nurse or another medical practitioner to approve abortions rather than the two doctors needed now.

Those wanting an outright ban on abortion may now be as low as 4% of the population.

Gene editing

According to a 2025 YouGov poll, over 80% of British adults, including most religious respondents, think gene editing should be allowed to prevent passing on hereditary genetic disorders (e.g. cystic fibrosis or Huntington’s disease).

Establishment and bishops in the House of Lords

A 2013 YouGov poll found that 51% of British people think that the Church of England should be separated from the state, while 27% thought that the connection should continue.

A 2017 YouGov poll for The Times found that 62% of British adults think that no religious clerics should have ‘an automatic right to seats’. Only 8% support the status quo, with a further 12% supporting adding leaders of other faiths alongside the bishops.

Charity and volunteering

12% of British adults say ‘it’s part of my religious belief/philosophy of life to help people’, and many non-religious people are also motivated by their own beliefs to do the same. The report Faith and Voluntary Action from the National Council of Voluntary Organisations (2007) finds that ‘Religious affiliation makes little difference in terms of volunteering’. The report also states that motivation for volunteering is complex – and ‘faith’ as a motivator is actually very difficult to prove/assess/measure.

Humanists UK has produced a fuller briefing on this matter.

The international picture

Pew has estimated that between 2010 and 2020, the share of the population that is non-religious went from 23% to 24%. 29% of the population is Christian, but the share of countries that are majority-Christian is estimated at 60%. The UK ranked as the 14th least religious country.

That said, there is reason to think that these figures significantly underestimate the number of non-religious people. This is because severe prejudice against the non-religious in many countries means that people aren’t comfortable with stating their true beliefs. The academic Will Gervais and others have studied this in the United States, using a clever technique to enable people to covertly reveal their actual views. They found that the share of the population who are non-religious rises by 20%. They are now repeating this study in other countries, with results forthcoming.

More generally, in 2020 Pew found that across 34 countries, 51% say belief in God is not necessary to be moral. In 2023, Ipsos found across 26 countries that 47% think religion does more harm in the world than good, and only 37% think religious people are better citizens.

In 2015, Pew produced research projecting that the share of the population that is non-religious would fall from 16.4% in 2010 to 15.6% in 2020, and then further to 13% in 2050. But this research doesn’t take into account those issues with it being impossible to be openly non-religious, or the non-religious facing prejudice. That is something that may change over time. It also doesn’t take into account how changing technology, social norms, and wars, may affect things. Indeed, as mentioned above, Pew’s 2025 Pew work revised the non-religious share for 2010 up to 23%, and furthermore found it grew by 2020 to 24%.