Writing, life, politics

The Census Religion Question: A Football Analogy

The UK is currently conducting its ten-yearly census. It includes an optional question about religion, but the question’s format has angered some humanists/atheists:

“Instead of asking, ‘Do you have a religion and if so, what is it?’, the question asks ‘What is your religion?’, a closed question that funnels people into giving a religious response, even if they don’t go to a church or a mosque, even if they don’t believe in God.”

The British Humanist Association have followed up these complaints by conducting surveys that have revealed that of those who claim to be Christian when asked the question on the census, 27% of those (when questioned further) said they did not believe that Jesus Christ was a real person who died, came back to life and was the son of God. A further 25% were unsure, meaning that only 47% of those who claim to be Christians actually hold Christian beliefs.

But why does this upset the humanists? What exactly is their beef? Why does it matter? Well this morning it occurred to me that you could create a good analogy using questions about support of football teams.

Imagine Newcastle city council decided to do their own mini-census, and included the following question:

Which football team do you support?

[ ] Newcastle United

[ ] Other, specify _____________

[ ] I don’t support a football team

Now I obviously don’t have figures to hand, but I suspect the answers would be something like this:

Newcastle United: 70%

Other: 10%

None: 20%

Why do I have such a high figure for Newcastle? Two reasons really:

1) There’s only one professional team in that city, and the people of the city strongly identify with it.

2) Most people will, when asked, volunteer a team that they support, even if they’re not all all into football. People tend to pick a team in childhood, or have one picked for them. Often, they will follow the team their parents followed. Even if they’re not at all into football, their default answer will be the local team from where they grew up.

Asking someone which football team they follow tells you very little about the actual popularity of football.

But imagine you asked the following questions?

Are you a football fan?

[ ] Yes

[ ] No

If yes, which football team do you support?

[ ] Newcastle United

[ ] Other, specify _____________

I’m guessing now that the answers would be more like:

Yes: 40%

No: 60%

Newcastle United: 35%%

Other: 5%

Why does this matter?

Well imagine that the owners of Newcastle United decided that they needed a new stadium, and told the council to, say, loan them £30 million of council-tax payers money at preferential rates and help them out with planning permission.

Asked one way, they can say that the club has the support of a majority (70%) of the population.

Asked the other, they can only claim a minority of 35%.

The question you ask dictates the answers you’ll get, and the problem is that having gone to a lot of trouble to get those answers, people will then use them to try and prove various points. (If they weren’t going to do so, then why bother asking the questions?)

We have a census question that is designed not to measure religious belief, but religious/cultural affiliation. That’s fine, as long as people treat the answers as such, and don’t use them to try to prove points about the extent or otherwise of religious beliefs.

But they probably will.


  1. iamreddave

    There is a great Yes minister bit on this survey design mechanism. Survey Design

    • Jonny Nexus

      I remember that from watching it, but I’ve just watched it again. Classic! 🙂

  2. bletch

    Census folk here (Australia) got all upset after the last census when a bunch of people filled in ‘Jedi’ for the religion question.

    … Was about to talk more about implications of this and decided to check facts. Well, check wikipedia anyway. And found this:

    Note the UK 2001 results. 2.6% of Brighton residents are Jedi! Now we know the real reason Jonny moved to Brighton…

    And as for those New Zealanders…

    Well anyway, it doesn’t say so there, so stories of what the religion question may get used for may be apocryphal. But they range from counting religions with a certain percentage as an “official religion” to provision of government funding (for what? Religious-based schooling? And I do recall threats of prosecution for filling in the 2006 census falsely (i.e. Jedi).

    • Jonny Nexus

      A bunch of people have got so upset by it, they’ve created the following website:



      And I seem to recall that Brighton was officially the least religious place in the UK!

© 2021 Jonny Nexus

Theme by Anders NorenUp ↑