The religion question on a census form is, as a rule, designed for people who have a straightforward answer. There is a tick box for Christianity, one for Islam, one for Hinduism, and a small written-in field labelled “any other religion” for everyone else. That field is where Paganism lives, officially, and what it produces is a number that is accurate in the same way a photograph through frosted glass is accurate: clearly something is there, but the detail is lost.
I am Pagan. I have been, officially, since November 2025, though the convictions arrived considerably earlier than the label. I am not going to use this post to explain what that means for me — I have written elsewhere about the symbols I carry and the traditions I follow — but it does mean I have an interest in what the numbers actually say, and whether they tell any kind of coherent story.
It turns out they tell a partial one, with a large footnote attached.
What the Census Says
The 2021 census for England and Wales — the most comprehensive source we have for this, and almost certainly the last national census of its kind — recorded 73,733 people who wrote in ‘Pagan’ as their religion. Add those who wrote in ‘Wicca’ (12,813), ‘Druid’ (2,490), ‘Reconstructionist’ (742), and ‘Thelemite’ (227), and you arrive at roughly 90,005 broadly Pagan-identifying respondents in England and Wales alone.
Scotland ran its census a year later, in 2022, and added 19,113. Northern Ireland contributed 1,334. The UK-wide total, combining all the Pagan-adjacent categories across all four nations, lands somewhere around 126,980 — or about 0.2% of the population. Scotland sits slightly higher at 0.4%, which might surprise people, or might not, depending on how familiar they are with Scotland.
That figure represents genuine growth. In 2001, the comparable total for England and Wales was around 40,000. By 2011 it had risen to 73,010. By 2021 it reached 90,005. Each decade has seen meaningful increases — not explosive growth, but a consistent upward line.
In Cornwall specifically, 1,769 people identified as Pagan in 2021, making it the third most commonly recorded religion in the county. This is not surprising to anyone who has spent time in Cornwall. What is more surprising, perhaps, is that West Northamptonshire also shows up in the data as notably Pagan-heavy, in the same cluster as Wiltshire and Derbyshire. I live close enough to Northampton to have a small collection of Pagan friends there, which is anecdotal rather than statistical, but the census broadly agrees with my social circle on this one. The geographic distribution of Paganism in England and Wales does not always map neatly onto the places you might expect.
The Gap in the Data
Here is where the frosted glass becomes relevant.
Estimates from the 1990s — before any census data existed — converged on a figure of 107,000 to 140,000 Pagans in Britain. When the 2001 census came back with 40,000, that gap needed explaining. The most convincing explanation involves several overlapping factors: the religion question is voluntary; followers of minority religions have specific reasons not to record themselves in state documents; and in 2001, the Satanic Panic was still recent enough to be vivid. Some Pagans simply do not regard their practice as a religion in the formal sense, and would not describe it as one on a form.
Dr Robin Douglas, a historian of religion who published a detailed analysis of the census data in October 2025, put the likely true figure for England and Wales alone in the hundreds of thousands, noting it was plausible that the Pagan community had already grown to a size comparable with the Jewish community — which the census recorded at 287,000. That is a substantial gap between the official count and the probable reality.
An academic study of how Pagans actually responded to the 2011 census found a similar picture, estimating that if the true numbers had grown proportionally to the recorded ones, the actual Pagan population in Britain at that point could have been anywhere between 180,000 and 240,000. The census figure for 2011 was 73,010.
The Pagan Federation, which has been lobbying for accurate census representation since before the first religion question was asked in 2001, made the point explicitly in the run-up to the 2021 survey. The BBC rejected a Pagan ‘Thought for the Day’ on the grounds that Pagans did not constitute even a 365th of the population. If the true number is in the hundreds of thousands, that argument collapses. But the argument was made on the basis of the census data, and the census data undercounts.
Who Self-Identifies, and How
The Pagan Federation’s own analysis of the 2021 results acknowledged a particular complication: many Pagans feel more comfortable selecting ‘No Religion’ than writing in their tradition. The ‘No Religion’ category grew by 12 percentage points between 2011 and 2021, reaching 37.2% of the population — 22.2 million people. Some portion of that group practises something that would, by any reasonable definition, qualify as Pagan. They have just decided not to say so on a form.
Some of this is definitional. The umbrella is genuinely wide. Wicca, Druidry, Heathenry, Asatru, Reconstructionist polytheism, animism, shamanism, eclectic witchcraft — these traditions are different enough from each other that the question of what qualifies as Pagan is not always simple. Shamanism, as a category, increased more than tenfold between 2011 and 2021, from 650 to 8,000. That is the largest proportional increase of any write-in religion in the census. Whether all 8,000 of those people would be counted as Pagan depends on who you ask.
The demographic breakdown, though, is consistent and worth noting. Of those who wrote ‘Pagan’ in the 2021 census, 62% were female and 38% were male. For those who wrote ‘Wiccan’, the split was 78% to 22%. Neither figure is especially surprising given the traditions involved, but the gap has widened compared to 2011, when the split was 57% to 43% across broadly Pagan categories.
Age is where things get more interesting, and perhaps more complicated. The Pagan community, as visible through the census, is heavily adult. There are relatively few minor members — Paganism is, as Dr Douglas put it, largely a religion of adult converts. The peak age range in 2021 was the 50s. In 2011 it was the 40s. The age peak moved forward by a decade in exactly the time a decade had passed. Which means, to put it plainly, that the people who were 40-something Pagans in 2011 became 50-something Pagans in 2021, without a comparable wave of younger people joining to compensate. Whether that indicates an ageing community headed for decline, or simply a community whose members are living their lives without rushing to declare themselves to the state, is genuinely unclear from the data alone.
What the Numbers Do and Do Not Tell You
The honest answer to ‘how many Pagans are in the UK’ is: probably several hundred thousand, recorded at around 127,000, with the gap attributable to a combination of stigma, definitional ambiguity, distrust of state records, and people who simply do not think of their practice as something that belongs on a government form.
The recorded figure has tripled since 2001. That is real. The reasons for that growth — greater social visibility, reduced stigma relative to the Satanic Panic era, the internet enabling dispersed communities to find each other and develop shared identities — are plausible enough that the trend is probably genuine and not just a data artefact. Paganism has been described, sometimes anecdotally and sometimes by people who should know better, as the fastest-growing religion in the world. The census does not confirm that, but it does not contradict it either. It just undercounts it.
For a faith tradition that has no central institution, no authoritative membership roll, and no meaningful infrastructure for knowing how many people belong to it, the census is the closest thing to a headcount that exists. 127,000 official. Several hundred thousand probable. A community that has been quietly growing for half a century and still does not register as more than a rounding error in the national data.
That last part is frustrating. Though I assume it will remain true regardless of what the numbers eventually show.