Surveying the Survey: Anglicans and Same-Sex Marriage [full version]

[toggle title_open="Close Me" title_closed="Open For Executive Summary" hide="yes" border="yes" style="default" excerpt_length="0" read_more_text="Read More" read_less_text="Read Less" include_excerpt_html="no"]The recent survey reporting the views of Church of England people on same-sex marriage needs to be read as descriptive not prescriptive and its description needs a wider context which acknowledges that the method used results in both the larger group (on which the published data almost exclusively focuses) and smaller group it identifies as Anglican (about 30% and 18% of the adult population respectively) being much larger than even Christmas worshippers in the CofE (about 4%). Its method also excludes some who are regular worshippers in the CofE but who don't identify primarily as Anglicans.

Similar surveys show that this sort of group is little different from those identifying as non-religious in how they make decisions about life. This leads to them having views on sexual ethics such as

- two-thirds of these Anglicans say they would not feel at all guilty about extra-marital sex
- a quarter of these Anglicans would not feel at all guilty (rising to over 1/3 when including those “not very guilty”) about committing adultery.

Having made these caveats, the article identifies eight areas where this and similar surveys help us understand changing attitudes to same-sex marriage. These include:

- the significant difference in support among non-religiously identified and religious people,
- the significant and growing number of supporters for same-sex marriage among those who identify with any Christian denomination. In many of these, as with self-identified Anglicans, there may be a simple majority believing it is right, but not yet usually an absolute majority ie more say "wrong" or "don't know" than say "right".
- the continued strong belief same-sex marriage is wrong among those who identify as "Orthodox Christian", "Pentecostal" or "Evangelical".

It concludes by arguing that Christian leaders need to be more confident - in church and society - about explaining the Christian understanding of marriage with which many people still identify.[/toggle]

What follows is available as a PDF and there is also a shorter summary version online here.

Two years ago in a conversation with Canon David Porter, the Archbishop’s Director for Reconciliation, I pointed out one of the challenges as we entered a period of conversation and then decision about sexuality: we have no clear idea what people in the CofE – whether bishops, clergy or laity - really think about any of this. For all its weaknesses, one of the strengths of the Church of Scotland’s process was that it carefully measured the views of those in its leadership, surveying nearly 25,000 elders as it considered whether to allow ministers to be in civil partnerships. Nothing on this scale has been done within the Church of England but Jayne Ozanne has recently commissioned and published a survey. It has gained much media attention as it claims, in the headline of the press release, that in relation to same-sex marriage “Church of England supporters now outnumber opponents, with 3 out of 4 younger Anglicans in favour”. The survey has not been without its critics. Peter Ould and Ian Paul offered a rapid rebuttal and some of the Facebook exchanges have been rather robust and not a model of “good disagreement”.

The theological significance of what Christians believe

Before looking at the details of its method and results it is important to begin with what such a poll can and cannot do it. In summary, it is descriptive - if done well it can give us an insight into the present mind of at least part of the church – but should not be normative. The mind of the church is not theologically insignificant. Theologians talk about the sensus fidelium (the sense of the faithful). However, even research as extensive and careful as that done by the Church of Scotland, has serious limits theologically. It captures only the mind of a sample of a particular church denomination in a particular time and place. The category of sensus fidelium has a much more catholic ecclesiology with a bigger vision of the church across time and space. It is also clear that – especially when taking a snapshot of a small part of God’s people – the majority viewpoint is frequently in error. Scripture is full of prophets who are a faithful minority. Pilate’s crude polling in Holy Week produced a decisive but flawed assessment of Jesus of Nazareth among the Jews in Jerusalem. It is fairly clear that in the pretty small early Christian community in Corinth a majority preference for particular views and practices, including in relation to sexual behaviour (1 Cor 5.1-2), was not viewed as the best guide by the apostle Paul. Christians, particularly those from the evangelical tradition with which Jayne identifies, do not believe the voice of the people – even the people of God - is the voice of God.

The survey’s goal

The survey’s primary aim appears to have been to discover the views of Anglicans in England on same-sex marriage. To do this You Gov (who have done similar surveys in the past, notably for Lancaster University and the Westminster Faith Debates, overseen by Linda Woodhead) polled a large sample of 6,276 adults in Great Britain between January 19th and 21st. The challenge is to find the English Anglicans among them. While not the equivalent of looking for a needle in a haystack this is not an easy task. The first element – the geographical one of a GB sample – is relatively easy to address but the second one – that of what counts as “CofE” is much more complex (for an amusing take triggered by debate over the survey see Are You Really Church of England?).

Recognising this challenge, the survey came up with two answers – one broader and one from within that which was narrower. This distinction explains why Jayne Ozanne has objected to those she thinks are saying or implying that she didn’t address the complexities – she did. The next section looks at the first group in some detail both because most of the figures cited are from this group and it was the pool out of which the smaller group was drawn.

Finding the English Anglicans – I

This section explores who is included and who is excluded in the larger definition of Anglican used in the survey and then explores this grouping in terms of evidence in relation to how they form their moral beliefs, how they evaluate their moral behaviour and to whom they see themselves as belonging in terms of affiliation to a religious community.

Who is included?

The published data subdivides the 6276 GB adults (5213 in England) into subgroups defined as giving their religious affiliation. The exact question which was asked is not stated but given the categories listed and You Gov’s past surveys it is likely to have been [this is now confirmed by Jayne Ozanne]

Do you regard yourself as belonging to any particular religion, and if so, to which of these do you belong?

This showed that a surprisingly large number (3078, or 49%) compared to past You Gov surveys made the strong response “No, I do not regard myself as belonging to any particular religion”. Those not giving such an answer (or preferring not to say) then had to identify with a list of possible religious groupings. A total of 1652 (26.3%) in Britain answered “Yes – Church of England/Anglican/Episcopal”. Of these 1523 lived in England (from an English sample of 5213).

In other words this survey’s definition of “CofE” is such that it includes 29.2% of English adults.  This is a remarkably high proportion compared to attendance figures. It is worth thinking how big a fringe would need to be included in your local parish church for it to encompass nearly 30% of the parish population. The attendance figures for 2014 published a few weeks before the survey in Statistics for Mission found that

As a proportion of the population living within the dioceses of the Church of England (not including the Diocese in Europe), 1.8% attended a Church of England church each week in October 2014, and 4.3% attended at Christmas.

Mapping “Church of England participation” since 1960 (p. 28 of Statistics for Mission) the highest % of population participating on four measures (Electoral Roll, Usual Sunday attendants, Easter communicants and Christmas communicants) is for electoral roll in 1960 and it was just under 7% of the population.

Who is excluded?

Another intriguing feature of this survey is the list of possible options which respondents were offered. You Gov published data appears to have a longstanding standard short list offering answers to “Do you regard yourself as belonging to any particular religion, and if so, to which of these do you belong?”. People can answer “No” (which many would see as a rejection of religion that they may not feel reflects their view) or “I’d prefer not to say”. If they say “Yes” then they have to be more specific. The minimal question divides the “Yes” into 5 Christian denominations, 4 other religions and “Other”. This can be found, for example, in the Jan 2013 survey for Lancaster University on same-sex marriage. Sometimes there are further categories as in full results of a June 2013 survey for Lancaster University which Jayne Ozanne includes in her press release. These add 3 more Christian denominations and 1 other religion.

To these Jayne Ozanne’s survey adds 3 other answers labelled as “Orthodox Christian”, “Pentecostal” and “Evangelical”. These do not appear to have been used much (perhaps at all) in the past – certainly not in the 2013 survey cited as a comparator - although they do appear in a recent December 2015 survey for Lancaster University where the wording is given as:

  • Yes, Orthodox Christian
  • Yes, Pentecostal (e.g. Assemblies of God, Elim Pentecostal Church, New Testament Church of God, Redeemed Church of God)
  • Yes, Evangelical – independent/non-denominational (e.g. FIEC, Pioneer, Vineyard, Newfrontiers)

It is not clear whether this fuller wording lies behind the shorter designations of simply “Orthodox Christian”, “Pentecostal” and “Evangelical” in Jayne’s figures or how people were given these options. A total of 132 opted for these 3 categories – 2.2% of the whole sample and 4.9% of all those who identified with a Christian identifier. They are likely to confuse rather than clarify matters as most would find someone calling themselves a “Baptist Methodist” slightly odd whereas “Evangelical Anglican” is a common designation. Faced with the usual list I could only be “Anglican” but if I could choose “Orthodox Christian” or “Evangelical” might well opt for one of those. If I did then I – an ordained Anglican – would not be in her sample. It is likely that (especially if there was limited explanation) then these 3 groupings include at least some (perhaps very committed) CofE people who are thereby cut out of the figures for Anglicans and, although not a large group, their views on same-sex marriage are very significantly different from those identifying as Anglican. This change from the 2013 survey also makes it more difficult to compare the figures precisely as some of those against same-sex marriage in denominations in 2013 have probably not changed their views but simply been reassigned into these new, strongly conservative, categories.

It is important that an explanation is offered of this new form of question and why it is included, that the terms and their use are given further scrutiny, and the impact of their inclusion on the data for the other Christian categories examined and acknowledged.

How do these Anglicans form beliefs about how to behave?

The published data from the recent survey only tells us about beliefs on same-sex marriage but similar You Gov surveys enable us to piece together how the beliefs of this group of those who identify as Anglicans in this way are shaped. Jayne’s Press Release helpfully directs us to pp13-14 of a 2013 You Gov/Lancaster University survey which asked people

Which ONE, if any, of the following do you MOST rely on MOST for guidance as you live your life and make decisions?

Their sample of 4018 had 1120 identified as CofE/Anglican/Episcopal – a proportion at 27.9% fairly close to that in the recent survey. While caution is needed we can conclude that there is a good chance that the Anglicans found by this method are likely to be not too dissimilar from those in this most recent survey. The Anglicans gave the following responses:

What I MOST rely on MOST for guidance in life % Ang/CofE/Episc % No religion
Own reason and judgement 34 41
Own intuition or feelings 23 21
Family 21 18
Trusted friends 4 4
God or 'higher power' 4 0
The tradition and teachings of my religion 3 0
Science 2 5
A scripture or holy book, e.g. Bible, Qur'an 1 0
The religious or spiritual group to which I belong 1 0
Great literature and art, past and present 0 1
Deceased loved ones 1 1
Religious leaders, local or national 0 0
None of these 3 6
Don't know 3 2

Given this pattern of how people believe it is not surprising that 37% of the “Anglicans” in the sample were honest enough to respond that they were not influenced by any religion at all.   Nor should it be shocking if there is a divergence between these “CofE supporters” and Anglican leaders. In fact it is perhaps more surprising that when 70% of those who say they are non-religious believe same-sex marriage is right, only 45% of “CofE supporters” reach that conclusion and 37% (compared to just 16% of non-believers) actually believe it is wrong.

What do these Anglicans feel about how they behave?

Another January 2013 You Gov/Lancaster University Survey focussed on attitudes to sexual behaviour more widely. Its sample of 4437 had 1261 identify as Anglican in response to the question which at 28.4% of the group again suggests a similar pattern of Anglican identity as in Jayne Ozanne’s recent study. While caution is again needed we can therefore conclude that there is a good chance that the Anglicans found by this method are likely to be not too dissimilar from those in the new survey. This 2013 survey asked people

How guilty, if at all, would you feel if you did one of the following things?

The figures are again illuminating for this group of Anglicans

Very Guilty Fairly Guilty TOTAL GUILTY Not Very Guilty Not At All Guilty TOTAL NOT GUILTY DON’T KNOW
Used pornography for sexual stimulation 14 16 30 26 32 58 11
Used contraception 2 1 3 10 83 93 3
Had sex before marriage/civil partnership 5 7 12 19 66 85 3
Had sex outside marriage/civil partnership 42 18 60 10 25 35 5


Given this pattern of how people would feel about these patterns of sexual behaviour it is once more not surprising if there is a divergence between “CofE supporters” and Anglican leaders. What is even more remarkable is that if around 35% of Anglicans (on this self-definition) would not feel guilty about adultery only 10% more than this (45%) in the recent survey believe same-sex marriage is right.

Where do these Anglicans belong?

Doubtless because of the problems caused by using this very broad form of Anglican self-identification, Jayne Ozanne’s survey asked a second question. The exact wording is only partially reproduced on the data made available but has been confirmed as the standard You Gov question:

Which, if any, of the following best describes the group or community that you are involved with or whose gatherings/services you attend?

Here there is a different set of options made available to respondents than in the other question although with some overlap of category descriptors. It is not clear whether this question was asked before or after the other question but it is likely that if asked after it will exaggerate involvement as many, having identified specifically as belonging to a particular religion, would perhaps feel strange then saying they had “no specific religious affiliation”.

Taking the 1523 people who identified as Anglican in the first question and who lived in England they divided as follows between these:

Anglican/CofE 909 59.7%
RC 12 0.8%
Historic Free Churches 7 0.5%
Muslim/Islamic 1 0.1%
Other Christian 33 2.2%
Church of Scotland 7 0.5%
Baptist 5 0.3%
Jewish 0 0%
Charismatic/Pentecostal 5 0.3%
Hindu 1 0.1%
Sikh 2 0.1%
Buddhist 2 0.1%
Alternative form of spirituality 6 0.4%
Other 28 1.8%
No specific religious affiliation 505 33.2%

In other words, a third of this original group of Anglicans are not involved with any religious group or gathering (they presumably didn’t want to reject religion so opted for CofE even though they have no involvement with it), about 5% are affiliated with other Christian groups, despite initially designating themselves as Anglicans, and about 2.5% actually belong to non-Christian faiths. Thus the group of Anglicans has been reduced by almost 40% when questions are asked about actual involvement or attendance at servicers or gatherings. There is though no objective measure of what counts as being involved with or attending gatherings or services. It remains quite a high figure – about 18% - compared to the official attendance figures cited earlier.

Finding the English Anglicans – II

The answer outlined above gives the new survey a second, smaller possible category of English Anglicans. What is noteworthy is that the published data gives us no information about this group other than its size (909 out of 5123 English respondents so 17.7%) and its support for same-sex marriage (42% think it is right, 39% think it is wrong and 19% don’t know). In contrast, in Jayne Ozanne’s figures specifically for all Anglicans living in England, she focuses on the larger group of 1523 offering a breakdown of their views on same-sex marriage by gender, age, region and social class. They, of course, had slightly stronger support for same-sex marriage (45%-37%-19%). It appears, therefore, that her preference is for the first, larger grouping once limited to England.

What can we say about this second grouping? Again it is worth noting who is included and who is excluded.

Who Is Included?

This group is simply a subset of the larger group. It is those in it who both live in England and who say that “Anglican/CofE” is what “best describes the group or community that you are involved with or whose gatherings/services you attend”. Although only about 60% of the original grouping it is still large (over 17% of the population). Given that in 2014 only 4.3% of the population attended a CofE church at Christmas (up from just 1.8% during October) it means that no more than a quarter of this group attend an Anglican church even at Christmas and a maximum of 10% attend weekly. In fact the figures are likely even less than this because of another factor.

Who Is Excluded?

Those who say that “Anglican/CofE” is what “best describes the group or community that you are involved with or whose gatherings/services you attend” includes more than those who identify as Anglicans in the first question. All of us will know from experience that some of the most committed Anglican worshippers may say “I’m really a Baptist/Methodist/Pentecostal/Evangelical/Roman Catholic”. The data in the survey shows that 5% of those who identified originally as Anglican give a different Christian denomination in answer to where they attend. It is likely that there are those who originally identified as something else who then say that “Anglican/CofE” best describes where they are involved. But at no point do they appear in the calculations in this survey.

In short, even this smaller group includes an overwhelming majority (roughly 3:1) whose commitment does not extend to attending a CofE service every Christmas while excluding a number (perhaps a significant number) of people who are actively involved and regularly attend services in their CofE parish but do not identify as Anglican.

What is their pattern of belief and behaviour?

This is almost impossible to answer as I have not been able to find any other You Gov survey which works with this particular subset. However, if the surveys used above are a reasonable approximation of the larger group then it is clear that, for example, recourse to Scripture or tradition or God can be most relied on by only a small number of these Anglicans. Only 8% of the larger group of Anglicans cited these as their primary guide in the 2013 study described above. This means that, in Jayne Ozanne’s larger sample of 1523 there would, on a generous estimate, likely be no more than 150. Even if all of these also said they attended CofE services then the % in that smaller group (of 909) who described Scripture, tradition or God as their primary guide would only reach 16.5%.

The only thing we really know about this group is that in this survey it too showed a majority for same-sex marriage. On this basis it can be said that both ways of finding English Anglicans – whatever their limits or flaws - do show more “supporters of CofE” thinking same-sex marriage is right than thinking it is wrong. However, with only a 3% majority in a sample of 909, this particular finding is within the margin of error.

So what can we learn from the survey?

The analysis above shows the limits of what can be claimed on the basis of this new research but, particularly given other surveys, it appears that we can be reasonably confident about the following features of our church and society in relation to same-sex marriage.

First, there is probably now an absolute majority in British society (ie over 50%) who approve of same-sex marriage and this is particularly strong among those who reject any identification with religion (where this survey gives 70% thinking it right, 16% wrong and 14% don’t know).

Second, and in contrast, there is not an absolute majority in favour of same-sex marriage among those who identify as Christians (and in almost every denomination - the only exception in this survey is the URC and bizarrely the Free Presbyterians where sampling only found 1 in this small, very conservative Reformed denomination but they were in favour of same-sex marriage!) and among most other major faiths (not Buddhism and Hinduism here). This is the case even though (1) the groups are simply defined in terms of self-identification rather than any objective criteria and so include those who are “lapsed” and/or very loosely involved if at all in the faith community’s common life and (2) past surveys have shown that more regular attendance is linked with a more negative view to same-sex marriage and so these groupings probably over-estimate support among the more committed. This therefore does seem to be an issue where attitudes remain significantly different between those who maintain some sort of identification with a faith tradition – including Anglican - and those who have none.

Third, making this point specific to the focus of this survey, even in this survey and even given the very large group being identified as Anglican (on either method), more who self-label as “CofE” either believe same-sex marriage is wrong or don’t know than believe it is right. The reported lead of “pro-same-sex marriage” over “same-sex marriage is wrong” although new compared to similar past studies and perhaps significant is still small.

Fourth, every Christian denomination has a significant grouping of those who identify with it in some sense and who believe same-sex marriage is right. Although a strict minority (ie under 50%) it is often larger than those who believe it is wrong (even among Roman Catholics) or around the same size.

Fifth, it would appear that over the last 2-3 years the general movement in religious groups, as among non-believers, is towards more of those identifying with them believing that same-sex marriage is right. Here the comparison with the 2013 question, prior to same-sex marriage being legal, using a similar question is interesting with all denominations seeing a shift, some such as Baptists and URC (though with caution on small samples) being quite significant. Across all Christian identifiers about 9% less think it is wrong with 7% more thinking it is right and 2% more not knowing. However, the inclusion of 3 new designations in the question, comprising 5% of the Christians and all being strongly against same-sex marriage, makes precise comparison difficult as these new categories have likely contributed something to the apparent shift in other Christian denominations.

Sixth, in wider society and probably in most faith groups, support for same-sex marriage is higher among women and among younger people.

Seventh, the claim that bishops take a different view could therefore be due, as Jayne Ozanne claims, to the fact they tend to be men over 55. However, focussing on this is to factor out a range of other significant factors, notably that the group of Anglicans identified in this survey as favouring same-sex marriage have been found in other surveys to give much more weight in moral judgments to their own reason, intuition and feelings than to God, Scripture or tradition. It is not unreasonable that bishops – whatever their age or gender – should be different from them in this methodology and perhaps therefore reach different conclusions on justifiable grounds.

Eighth, while there remains a lack of clarity about the 3 new categories included in this survey the results there are particularly stark in relation to same-sex marriage:

Right Wrong Don’t Know
Orthodox Christian 39 46 15
Pentecostal 24 59 17
Evangelical 20 63 16

These Christian groups (likely to be identified with by those who have a stronger affiliation than with some other groups, especially Anglicans) have a much higher proportion of people believing same-sex marriage is wrong than any other Christian group (apart from the small number of Brethren). In the case of Pentecostals and Evangelicals the figure is higher than among Muslims. This raises interesting questions for groups such as Accepting Evangelicals who argue for same-sex marriage to be accepted as an orthodox, evangelical belief.

What should Anglican leaders do?

The poll – despite the various weaknesses noted – was lapped up by the media which generally bought the narrative attached to it by its sponsors without serious question. Once again, as following the Primates’ Meeting and the decision in Jeremy Pemberton’s employment tribunal, there was silence from the bishops of the church in the media which tended to be dominated by those Christians, such as Jayne Ozanne, supportive of same-sex marriage. The overall impression again is that although the church has official teaching, nobody is prepared to defend it let alone explain why it is what the church teaches. It appears that its leaders are either complacent or resigned to support for it draining away. In such situations, with the change in law and wider culture, it is not a great surprise that more people who identify in some sense as Anglican are coming to believe that same-sex marriage is right.

And yet still many Christians – and people of other faiths – even if their allegiance is relatively weak, believe same-sex marriage is wrong and a significant number are, unsurprisingly, left unsure. There is, in other words, a real opportunity to explain why Christians and most human cultures down through the ages have been right to identify and privilege as unique the way of life which unites a man and a woman in an exclusive sexual union intended to be lifelong. That needs to be done within the active life of the church where it is likely that regular worshippers (even though they rarely get Christian teaching in this area) are – like evangelicals and Pentecostals – actually much less persuaded of the rightness of same-sex marriage than those Anglicans whose views are reported in this survey. And it needs to be done too through the media who are clearly interested in Christian views on this subject. That will help those whose instincts remain cautious about this new definition of marriage to understand and remain committed to a Christian vision. It may also persuade those who have changed their views that there remain good reasons – which at least need to be respected even if not accepted – to believe same-sex marriage is wrong and that such views are not to be ridiculed and treated as discriminatory or even extremist by society as a whole.

Leave a comment