Growing up in the North East of England in the 1980s and 1990s, it barely mattered what type of Christian you were. My dad’s family was Protestant while my mam’s was Catholic; I went to a Catholic secondary school, but a Protestant primary school – mostly because the latter was closer to the Catholic primary school all my cousins went to. Well, not all of them, but my two cousins on dad’s side practically constitute a rounding error. As I said, mam’s family are Catholic.
I did attend Catholic mass – in my village it was on Saturday evening because the local priest was busy on Sunday mornings at the bigger, more important church in a nearby town. But then the BBC started showing Gladiators on a Saturday night and I was old enough to say I wasn’t interested in the men in silly outfits doing silly things, and that I’d rather stay home and watch Gladiators instead.
That was the end of my active participation in religion and, for a lot of people, it’s a similar story – attending a faith school, growing up in a relative desert of non-faith schools, but falling out of religion once they’re old enough to be beyond the instruction of parents and teachers. That has essentially been religion’s role in the UK for most of my life, with each successive generation getting less religious – and less strident about the centrality of religion in their lives – than the last.
You’d be forgiven for thinking that trend had arrested or was reversing, had you been reading the many headlines from the past year proclaiming a growth in religion among the young. These headlines began last April, when the Bible Society published a report headlined “The Quiet Revival”, based on a YouGov poll showing a growth in youth religiosity.
In the study, YouGov compared the results of a 2024 sample of over 13,000 people with the results to similar questions asked of 19,000 people in 2018 and found that the percentage of all adults who attend church at least once a month had risen from 8% to 12%. Among 18-24 year olds, religious attendance had risen starkly from 4% to 16%, with most of that growth being among young men. In the 2024 data, 20% of churchgoers aged 18-24 were Anglican (down from 30% in 2018), while 41% were Roman Catholic, and 18% were Pentecostal. It looked like my mam’s side was winning back a lot of lost market share.
The report caused a flurry of headlines across the press, with the Independent writing about how Gen Z are ‘flocking to church’ because “Christianity is cool again”; the Telegraph declaring “Belief in God doubles among young people”, the Financial Times wrote that Christianity had “regained the underground appeal of its earliest days”, and the Daily Mail pinned part of the responsibility on how “’hot priest’ influencers are drawing young people to the church in their droves”. The Mail also pointed out that sales of the Bible had hit a record high, and Gen Z were behind a huge surge in Christmas churchgoing (the latter being based on an opportunistic PR survey from Christian aid charity Tearfund).
Then there were the follow-up pieces to explain and analyse this new trend. So many news desks sent journalists down to their local church, to interview Christians of the long-established and freshly-minted variety, that they might well have caused a second not-so-quiet spike in attendance. While many of those journalists managed to find someone relatively young to talk to – or, failing that, to settle for a church leader to ask about the new surge in youth attendance – none of the reporting seemed to question whether it was genuinely the case that one in every six young people was regularly attending church.
Christian Concern, partners of the abortion-restricting Alliance Defending Freedom, called it “the return of a prodigal generation” that “has sent ripples ruffling the calm surface of entrenched secularism in the UK”. Meanwhile, Church Times warned that we should “Take care that ‘quiet revival’ is not ‘stolen’ by a form of Christian nationalism”, highlighting that “Tommy Robinson supporters are turning to Christianity, leaving the Church in a dilemma”. While Premier Christianity argued that “The quiet revival has a gender problem”, because of how heavily skewed it is in favour of young men – including an interview with a young man called Tom who had been to a rally in London and heard that our country needs to return to its Christian values, so he’d bought a bible and found a church.
How quiet is quiet?
The repeated reference to the “quiet” nature of this religious revival is somewhat key – because, for all the figures showing a huge surge in belief among the young, there wasn’t a commensurate amount of confirmatory evidence. Admittedly, the Bible was flying off the shelves in record numbers, but if four times as many people under the age of 24 were attending church, why hadn’t churches noticed it until this report was commissioned? Why did the stories about a mass uptick in youth engagement only appear once YouGov discovered that it existed – hadn’t anyone spotted that the size of their congregations had swelled, or that their demographics had radically shifted younger to accommodate the one in six people under the age of 24 who were now regular attendees? Were all of these Gen Z men just too quiet, shy and retiring for anyone to spot?
The answer, of course, is that there is no quiet revival – there never was. As subsequent investigations have shown, the report was wrong because it was based on faulty and fraudulent data.
This ought to have been obvious at the time – after all, YouGov surveys commissioned by the Bible Society are not the sole measure of the religiosity and respective religious demographics of the UK. Data from the Church of England showed an uptick in weekly church attendance of 4.5% between 2022 and 2023, and a further 1.5% from 2023 to 2024… however, those minor revivals didn’t even cover the 19% fall in attendance since 2019. And while the Bible Society’s data suggested most of the growth was more prominent among Catholic and Pentecostal faiths, data from the Catholic Church shows that mass attendance fell from 700,000 in 2019, to 575,500 in 2024. My mam and her family should put the champagne away.
The gold standard measure of religiosity in the UK is the annual British Social Attitudes (BSA) survey, which tracks social trends over decades. It shows that while 67% of the UK claimed to be Christian in 1983, it was as low as 37% in 2018; meanwhile, the percentage of “no religion” rose from 31% to 53% over the same time, with non-Christian faiths making up the remainder. In terms of denomination, Brits identifying as Anglican fell from 14% in 2017 to 11% in 2024, with Catholicism remaining steady at 8%, and ‘other Chrisitan’ growing from 18% to 21%. In essence, over the seven year period, all the data could show was 3% of Brits moving from one category of Christian to another.
We can even check the Quiet Revival’s central claim – not about identity, but about church attendance – in the BSA dataset. In 2017, 18% of the country attended church at least once per month. By 2024, that was… 14%. Quiet, certainly, but no revival. And what about the youth? While BSA doesn’t have data for 18-24 specifically, 8% of 18-34 year olds attended a religious service once a week in 2017, and in 2024… it remained 8%.
Record Bible sales?
If the rise in religious identity and attendance was merely a mirage, what of the headlines about record sales of ‘the Good Book’? It does appear to be true that more copies of the Bible were bought in 2025 than in recent years, but this may essentially be a case of the tail wagging the dog: as a result of the interest generated by the Quiet Revival story, the Bible Society released their own annotated version of the bible, with more than 1,000 notes, prompts, stories, and features to help its readers to address “the nitty-gritty detail of life” – “from work meetings to football games to family dinners”.
Not to miss out, other religious groups like the Society for the Promotion of Christian Knowledge (SPCK) got in on the game, releasing their own tailored versions of the Bible. One Christian online retailer now lists more than 10,000 different editions of the Bible on sale. According to the SPCK, some of the best-selling editions are the ones that are priced reasonably enough to make them giveaways in evangelism. It isn’t that there are more Christians, it’s that the existing Christians are buying more bibles, some with the intention of giving them away to prospective converts in order to turn up the volume on that quiet revival.
Susceptibility of online surveys
So, what happened here? As a result of the pushback on these headlines at the time, YouGov re-examined their data, and in March this year they published their conclusions: a large amount of the responses to their survey were actually fraudulent. As a result, they’ve withdrawn the survey completely.
YouGov is among the more respectable opinion polling outfits. They claim to “track every panellist across time, platform, and at every step of their lifecycle – building a unique profile that evolves with every interaction”. Essentially, if you say you’re an 18-year-old with two kids in one survey, and that you’re 25 and childfree in another, they can compare results to see if their system is being gamed and weed out suspect user accounts accordingly.
However, their checking systems have flaws. One of those flaws is to do with ‘hard-to-reach’ populations. If you’re looking to survey men of any age, they’re easy to find, so any inauthentic accounts will be diluted by the large number of legitimate responses. But if your poll is specifically targeting men aged 18-24, there are obviously fewer of those than men overall. And if you’re trying to poll men aged 18-24 who attend church regularly, that’s an even smaller niche. The smaller the subset, the more impact outlier results can have, and the harder it is to ensure those results aren’t tainted by bogus respondents. That appears to be what happened here.
For the Bible Society’s part in this, it wasn’t actually their fault that this data came back cooked – even YouGov have confirmed as much, taking all responsibility for the errors. The Bible Society’s statement expressed their disappointment, but added:
“YouGov’s error does not mean that all of the findings were wrong – it means that we cannot reliably support those findings on the basis of this survey”.
While this might be an effort to save face, it isn’t one that holds much water: the fact that there is now no data either way from YouGov doesn’t mean that the conclusions drawn from the fraudulent data are definitely wrong… but there was plenty of other data from other sources available at the time of publication that disagreed with the Bible Society’s figures, and more since their report was published.
The Bible Society was given a dataset that appeared to tell them exactly what they wanted to hear. Did they check it against other available data? If they had, they’d have noticed that their conclusions were huge outliers.
Now that we know the data was fraudulent, the Bible Society don’t get to retreat to an “it is impossible to know either way” position. If I told you my cat can speak fluent French, and then I admitted that was a lie, it doesn’t mean that you now can’t know either way whether my cat can speak French. There is enough data on other cats to make an educated assumption.
YouGov’s statement also points out that,
“There is a tendency for fraudulent respondents to have a bigger impact amongst harder to reach groups. Problematic accounts are more likely to be found among these demographics, as there are fewer available genuine respondents and bad actors target those groups where they think they will receive more surveys and more incentives.”
This should be a familiar story to long time followers of my “Bad PR” work on the many, many issues with market research and PR polling. As I’ve written about extensively, because survey respondents are paid a very small amount for every poll they participate in, and cannot cash out their earnings until they’ve earned over a certain amount, users are incentivised to take as many surveys as they possibly can – even when the survey isn’t directly relevant to them. For it to be worth doing at all, it has to be done at scale, answering as many surveys as you can, as quickly as you can. There is therefore an obvious benefit to falsely appearing to sit within a hard-to-reach demographic niche – you’ll get access to more surveys to answer, and a better chance of meeting your cash-out target, because there are fewer of ‘you’ around.
These are criticisms I’ve been making of the PR polling industry for more than 15 years, but here’s a modern twist: each survey you take only pays you a very small amount, which makes it hard to make economically viable for many people – unless you live in places like the global south, where those micro-rewards might stretch much further. Thus, the rise of ‘survey farmers’, who can target the less-diligent end of the polling industry and use VPNs to evade detection by their verification systems.
Survey farmers can even operate multiple accounts from multiple devices, to ensure there’s no hard-to-reach niche that remains un-farmed. And with the use of AI chatbots, it’s easier than ever to be answering surveys as a 45-year-old divorcee, a 36-year-old mother of two, and a grandparent in his 60s – all at the same time. After all, in this connected world, if your local political influencer can actually be working a lengthy shift from India, then why can’t your local churchgoing Gen-Z Christian?
Obviously, every tech exploit is an opportunity for white hats to come in and save the day with detection tools and protocols, so YouGov is doubtlessly working hard to shore up their polling practices to better identify and weed out inauthentic users. YouGov’s business model relies on accuracy; fraudulent and unreliable results are a clear threat to their profitability.
However, the business models of other PR polling firms don’t rely on accuracy – they rely on headlines generated, attention grabbed, and clients promoted. So I’m less optimistic that companies like OnePoll will rush to roll out fixes that might lessen their ability to harvest headline-worthy results from their polling. In a world of survey farmers, such companies are effectively the supermarket conglomerates who rely on a constant supply of cheap produce at scale.
So, the next time you see a surprising headline about how the average British person thinks or acts – or worships their god – bear in mind the findings might not be based on results from actual Brits… or even from actual people.
