I remember an academic friend of mine once telling me that correlations do not make for explanations. He was right and I have been cautious about statistical correlations (in particular) ever since. The phrase came to mind when I read just now an interesting article by Martin Beckford on the Telegraph website about new academic research due to be published in January by the National Centre for Social Research (NatCen).
Based on the results of 4,486 interviews conducted in the respected 2008 British Social Attitudes survey, it notes:
- 50 per cent of respondents now call themselves Christian, down from 66 per cent in 1983.
- the proportion of Britons who say they have “no religion” has increased from 31 per cent to 43 per cent. Non-Christians, including Muslims and Jews, now represent 7 per cent of the population, up from 2 per cent, 25 years ago.
- The steepest fall was among those who say they worship in the established religion, the Church of England, down from 40 per cent of those who call themselves Christians to 23 per cent. (“Official Church attendance figures show that average Sunday attendance was 978,000 in 2007, compared with 1.2m in 1983.”)
It then draws several conclusions:
- More and more people are ceasing to identify with a religion at all. “Indeed, the key distinction in Britain now is between religious involvement and indifference. We are thus concerned about differences in religiosity – the degree of religious commitment – at least as much as diversity of religious identity.”
- “The declining Christian share is largely attributable to a drift away from the Church of England.”
- The decline in faith is largely attributable to children no longer being brought up in a particular religion. (“The results suggest that institutional religion in Britain now has a half-life of one generation, to borrow the terminology of radioactive decay… Two non-religious parents successfully transmit their lack of religion. Two religious parents have roughly a 50/50 chance of passing on the faith. One religious parent does only half as well as two together.”
The statistical problem is simply that different surveys cover different periods of time, ask different questions and use different criteria. So it is difficult to draw conclusions that might show any degree of consistency from the various studies done. The Christian Research data of a couple of years ago was a case study in seriously questionable conclusions being drawn from selective data and based on assumptions that were questionable (for example the use of flat-line projections that assume nothing will change in the next thirty years).
But why should anyone be surprised that people who no longer belong to a church also no longer feel they should use a church’s label to describe their (lack of) allegiance? It is no surprise that the biggest loss should be recorded for the Church of England as it is the only church that does not simply count as its ‘members’ those who consciously commit to attending the church on a committed basis. Clarity in terms of specific commitment is bound to reduce the numbers, but we need to ask what story the particular statistical dynamic is telling – which might not be the obvious one.
However, as Lynda Barley says (at the end of the article):
Statistical comparisons over a long period have the drawback of ignoring recent trends.The Church of England has been carefully monitoring Christian affiliation and churchgoing following the 2001 government census result that 7 in 10 people regard themselves as Christian. Independent surveys continue to show that 7 in 10 people are Christian and approaching half are Anglican in contrast to the British Social Attitudes Survey findings which focus on religious membership.
Local church counts of worshippers throughout October for the last nine years record 1.7 million individual Church of England worshippers each month in each year. At the same time, it has been ordaining some 500 new clergy each year.
The Church of England doesn’t really ‘do’ membership. Signing up to the Electoral Roll can say various things about the commitment or ‘belonging’ of someone. Even paying regularly by Gift Aid doesn’t really tell us a great deal about belief or commitment. It is notoriously difficult to say who is and who isn’t a ‘member’ of the Church of England’; all we can say is that the Church is there for everyone who wants it – a unique vocation of service to the whole community.
The surprise is simply that Terry Sanderson, President of the National Secular Society, is still being consulted for a view on such research. He said:
Last week at a gathering of faith leaders at Downing Street, the Prime Minister said that Christian values were ‘at the heart of national life’. This research shows that this is simply not true. This report shows more clearly than ever that Britain is a post-religious society and policy should reflect that.
Two responses: (a) Mr Sanderson would say this regardless of the ‘evidence’ put to him. If you said the sky was blue he would claim this as evidence of the death of Christianity in Britain. (b) ‘Christian values’ are not the same category as ‘membership’ or ‘commitment’ – which makes his statement a good example of a non sequitur. Even if the conclusion were to be right, you couldn’t draw it from this evidence or the Prime Minister’s statement about ‘Christian values’. Is his ideological prejudice so powerful that it blinds him to anything good about Christian (or other religious) contributions to society?
And, in the light of other discussions going on on this blog, just to confirm that this appears to me to be a good example of good reporting – summarising and bringing to the attention of a wider audience some research that is worth discussing and doing so in a clear and comprehensible way.
(17 December addition: See excellent comment from George Pitcher, too, at www.telegraph.co.uk/comment/columnists/georgepitcher/6830892/The-lost-Christians-have-found-new-homes.html)