Are religious people really more left-wing?

Demos's analysis suffers from some severe methodological problems.

In his Easter sermon, Rowan Williams suggested that “the high watermark of aggressive polemic against religious faith has passed”. Right on cue comes a report from think-tank Demos suggesting that (in the words of author Jonathan Birdwell) the “natural allies” of progressives “may look more like the Archbishop of Canterbury than Richard Dawkins.”

Faithful Citizens (available here) boasts a foreword from the former Labour minister Stephen Timms, who writes that religious faith can provide “a key source for the values the Labour party exists to promote and the aspirations it seeks to realise” and “a large reservoir of potential support for Labour’s aims.”

That might come as a surprise to Oliver James, who wrote at the end of last month that he assumed that Guardian readers were “a largely agnostic or atheist bunch.” But beyond the obvious point that not all practising Christians fit into a US-derived stereotype of the Religious Right, what does the report actually tell us about faith and politics in 21st century Britain?

Based on an analysis of the UK Citizenship Survey and the European Values Survey, the report highlights apparent correlations between religious commitment and social or political activism. It emerges, for example, that 55% of people with faith described themselves as left of centre, that they were more likely than the non-religious to value equality over freedom and less likely to have negative feelings about immigrants.

There was also a suggestion that the actively religious were more active politically, being more engaged in local community work, more likely to sign petitions or go on demonstrations, more likely to belong to a political party than people who weren’t religious. This would not, in itself, prove that believers are more left-wing. Some might, after all, be volunteering for the Conservative party. But taken together with the findings about general social attitudes, at least as reported at the weekend, it would seem to provide encouragement for Labour politicians to “do God” more enthusiastically.

Unfortunately, the analysis suffers from some severe methodological problems.

The report looks at both the UK and some other European countries, but the selection of those countries looks somewhat arbitrary. Switzerland, we are told, was excluded because it was not a member state of the EU, Italy because the patterns it revealed were “anomalous” (significantly, we are not told what was so anomalous about Italy) and the Scandanavian countries because “research consistently shows higher levels of civic engagement… which would have skewed our results”.

Since Scandanavian countries also consistently score low in markers of religious observance and belief in God, one can see how this might have “skewed” the result that Demos wants to present, that there is a natural link between religiosity and social activism.

The report divides the public into three broad groups: “exclusivists” who believe that their religion is the One True Faith; secularists, who regard religions as all equally false; and “pluralists”, less exclusive believers. Unfortunately, the latter category is wholly artificial, as it combines two entirely different stances: followers of one religion who are prepared to concede that other religions might teach some common truths or ethical principles, and vaguely “spiritual” types who believe that no one religion has a monopoly of the truth.

A “pluralist” might be a very devout and committed follower of their faith, or might not be a practising member of any religion at all. The category is meaningless. It is especially meaningless if the aim, as that of the Demos reports appears to be, is to harness the social activism of practising religionists to progressive causes.

What about the key finding that religious believers were “more likely to be left-of-centre”? The way that this finding was reported strongly implied that believers were more likely to be left-wing than non-believers. But this was not, in fact, the case. The figures for the UK showed that while 55% of believers considered themselves left-of-centre, 62% of non-believers did so. Thus believers were actually less likely than secularists to consider themselves left-wing!

With the other headline findings the picture was equally mixed. We read with some surprise that “the extent to which someone feels that religion is important to their sense of identity does not appear to have a positive impact on their civic engagement.” On freedom versus equality, in the UK a clear majority (approaching 60%) of both the religiously affiliated and of secularists prioritised freedom. Support for freedom was almost identical between the groups, though there were slightly less enthusiasm for equality among the non-religious (36% as opposed to 41%). In the continental European sample there was indeed a clear divide, with religious believers more likely to prioritise equality; but here I suspect that not including the Scandanavian countries may have skewed the result.

On most other issues, there’s very little difference between the views of believers and those of non-believers, and while practising religionists were indeed slightly more likely to be active in other ways, the divergence was not great. For example, in the UK 12% of those who belonged to a religious organisation professed themselves “very interested” in politics, as opposed to 11% of those who were unaffiliated. Well within the margin of error.

85% of British respondents, both religious and secular, said that competition was good rather than harmful. There were similarly high levels of support for personal responsibility as opposed to reliance on the state. Such answers raise obvious questions about the self-definition of the majority of people as left-of-centre politically but say nothing at all about the influence of religion.

One thing that the report does concede is that religious observance in the UK continues to decline — 50% of Britains do not regard themselves as belonging to any religion, as opposed to 31% thirty years ago. The trend is especially strong among younger citizens. So politicians both left and right should perhaps think before chasing religious votes with too much enthusiasm.