Survey: what does census religious identification mean?

Couple of thoughts.

1. The survey refutes conclusively the suggestion that those who tick one of the denominational boxes on the census form are mostly orthodox and observant members of the denomination that they tick. The survey does this for the UK but we would all expect a similar result in Ireland.

But I’m not sure that that’s such a big thing. I mean, we pretty well knew this already, didn’t we? Is anybody, on any side of church-state debates, going to be surprised at this finding? It’s not as though there were powerful forces lined up arguing that everyone who ticks, e.g., “Catholic” is a faithful and observant Catholic. In fact, I don’t think I’ve ever seen that argument advanced. So, to some extent, the survey may be attacking a straw man.

2. Phutyle suggests, in post #7, that census religious identification is a cultural identification rather than an ideological one. I think he’s right, and this survey shows that he is right.

3. Zombrex thinks, in post #18, that the census is asking what people believe (currently, rather than in childhood). I think he’s wrong. The question is simply “what is your religion?”; it’s up to the individual to decide whether he bases his religious identification on belief or on some other factor or on a range of factors. The survey suggests that most people in fact don’t base their identification largely or entirely on belief. That doesn’t make their answers “wrong”. On the principle that what you believe is not the state’s business, I’m inclined to think - or, at least, hope - that they are more likely to be correct in their approach to answering this question, and Zombrex is wrong. ("Correct" in the sense that their approach is meaningful and relevant, having regard to the state's purposes in asking the question.)

4. loconnor1001, in post 15, suggests that “the purpose of the census question was to identify the possible need for non-catholic schools”.

The UK Office for National Statistics has commented on the reasons for asking the question. When they consulted about the contents of the 2011 Census questionnaire, all their users (central and national government; experts, community and special interest groups; local and regional government; local service providers) wanted the form to include information on relation. A number of reasons were offered:

- Policy delivery, including meeting statutory requirements under the Race Relations Act
- Resource allocation and service provision
- Providing a clearer view of society and gaining a better understanding of certain ethnic groups.

The strongest demand was from local and regional government (91% of responders said that they required census information on religion). Given that in the UK it is local governments which provide housing, education and social services, this is unsurprising. It ties in with loconnor1001’s suggestion that the information is used to plan the provision of schools, etc.

5. This again suggests that Zombrex’s view that the question is about belief is probably wrong or, at least, oversimplified. Whether parents want to send their children to, e.g., a Catholic schools is probably not closely related to whether they profess, e.g., a faith in the dogma of the Assumption of the Virgin Mary, or whether they are orthodox in their beliefs generally, or whether they go regularly to mass. There’ll be a better correlation (though not a perfect one, by any means) with whether they identify culturally as Catholics.

6. The survey isn’t all good news for Richard Dawkins. Only 6% of the survey population (that’s all Britons, not just Britons who ticked a religious denomination) were prepared to say “I don’t believe in God”. Even if we throw in all the don’t-knows, none-of-the-aboves and prefer-not-to-answers, it still only comes to 13%. 87% of the population professed a belief in some form of God, supernatural intelligence or creator. And this ties in with 12% who, in answer to a different question, were prepared to say “I wouldn’t really call myself religious at all”. (This figure rises to 15% when you add in the don’t-knows, none-of-the-aboves and prefer-not-to-answers.)

This is striking, because 33% of respondents said that they either did tick the “no religion” box in the census form, or would have ticked it if they had filled out the form themselves (37% if we include the don’t-knows and prefer-not-to-says). Yet, when this survey scrutinize their beliefs, it turns out that probably only a third of these, at most, are unbelievers.

In other words, just as the proportion of respondents ticking the demoninational box overstates commitment to a denominational position, so the proportion of respondents ticking the no religion/no answer boxes overstates commitment to atheist/agnostic positions - if anything, this survey suggests, by an even greater degree.

The real distortion in the census may not be that the religious are overcounted and the irreligious are undercounted. The distortion may in fact come from treating the religion question as an enquiry about belief, and then identifying people's beliefs as theist or atheist/agnostic depending on how the answer it. What this survyey suggests is that, if you do this, you will overcount those of theistic belief signficantly - but you will overcount those of atheistic/agnostic belief even more signficantly. There's a large middle ground whose beliefs are neither conventionally (i.e. denominationally) religious nor conventionally irreligious, and they're being slotted into one category or the other.

muppeteer wrote: »

I Think you may have picked this up wrong, as all the survey results are based on those who identified themselves a Christian on the census, not all Britons.
Page one.

You’re quite right. My apologies.
I was misled by the statement on page 2 that the base for question 19 (and others) was “all respondents (1,136)”. You have to go back to page 1 to find that the 1.136 figure isn’t actually “all respondents”. A total of 2.107 people responded; 1,136 is the subset of those respondents who say they were recorded as Christian in the census, or would have been, if they had answered the question.

muppeteer wrote: »

As for why people ticked the Christian box Q5 seems to answer that most clearly
with 46% answering and only 18%

I think you need to take 3 and 5 together, actually. Both questions offer the same range of options, but in 3 you can choose all that you think are relevant to you, while in 5 you can just choose one.

muppeteer wrote: »

The only other large groupings are because their parents where Christian 13% or that they went to Sunday school 8%.
All of which suggests that most think that they answered Christian because of their own history not their current beliefs.

It’s clear that heritage plays a large part in religious identification. My point is, we knew that already. This is not a surprising result.

muppeteer wrote: »

Unfortunately the survey doesn't have a question on whether or not they would want to send their kids to a Christian school(which seems to be the main reason you suggest that state services want the religion question asked).

No, loconnor1001 suggested that as the main reason. The Office of National Statistics suggests that users want religious data for broader reasons, which include but are certainly not limited to school provision.