What can social surveys tell us about church attendance amongst Catholics in Britain?

Summary

This post examines the evidence from recurrent social surveys bearing upon changing patterns of church attendance amongst Roman Catholics in Britain. It analyses survey data from multiple sources going back several decades. When using data based on self-reports of religious attendance we should be mindful of the limitations of this method – with respondents prone to over-reporting of their actual attendance at services – and be aware of other techniques used for collecting data on the nature and extent of participation in religious activities (such as daily time-use diaries). Also, it needs to be borne in mind that self-identifying Catholics have generally constituted around 10% of the sample in recent social surveys in Britain, based on religious affiliation questions. Therefore, since they comprise relatively small groups of respondents, this may account for some of the fluctuation between individual waves in any particular survey series. This post focuses on summarising the broader picture, rather than commenting on specific instances of change over shorter time periods.

Introduction

Church attendance statistics and the interpretation thereof for Christian denominations and traditions have featured in BRIN on numerous occasions, engendering interesting feedback and debate. Recent figures on church attendance across denominational groups have also been given prominent coverage in the broadsheets. Change and continuity in religious attendance in Britain has been examined in scholarly research, particularly in recent analyses of British Social Attitudes data (for example: Lee 2012; Voas and Ling 2010). It was also the subject a major research report issued by Tearfund in 2007 (entitled ‘Churchgoing in the UK’). Rather than take a broader perspective on patterns and trends in religious attendance in Britain, looking across religious groups, this post primarily examines attendance within a particular denomination, that of Roman Catholicism.

What can survey data from the British context tell us about the religious behaviour of Catholics? First, we briefly compare Catholics with other religious traditions, using the recently-released BSA 2012 survey. There are clear differences between Anglicans and Catholics in their levels of attendance at religious services: 41.5% of Catholics attend church frequently (once a month or more) compared to 17.4% of Anglicans. Anglicans are much more likely to report that they never attend religious services (with the exception of special occasions relating to births, deaths and marriages), at 50.3% compared to 33.6% of Catholics. Differences are less stark for those attending services infrequently (less often than once a month): 32.2% for Anglicans and 24.9% for Catholics. It is also worth noting that other Christians also show a much higher level of frequent attendance than Anglicans but not as high as Catholics. The highest level of frequent attendance is shown by members of non-Christian faiths (at 53.5%), who also exhibit the lowest level of non-attendance (25.4%)

Table 1: Attendance at religious services by religious affiliation

Source: BSA 2012 survey. Weighted data.

What about Catholics’ longer-term attendance at religious services. Can we see any clear patterns – change or continuity – from longitudinal social survey data? Within the contemporary Catholic adult population, moreover, which social groups are more likely to report they attend services? This post presents and discusses the available evidence bearing upon these two questions. It first reports the evidence from a range of national and cross-national survey series and then looks in more detail at levels of church attendance amongst social groups.

Trend data on attendance at religious services

This post uses evidence from multiple surveys in order to try and get a more robust picture of any trends in attendance amongst Catholics in recent decades. We use data from the BSA surveys, British Election Study (BES), Eurobarometer (EB) surveys and the European Values Study (EVS) (the latter two being cross-national in scope). Where applicable, we present weighted percentages from the surveys and, given the caveats outlined in the ‘Summary’, also report the unweighted base (number of Catholics) for each time-point. Generally, when surveys ask about religious attendance, respondents can choose from a range of options to report how (ir)regularly they attend. However, to try to provide greater clarity of presentation and comparability across the surveys used here we generally classify attendance as follows:

• Frequently-attending: at least once a month or more.
• Infrequently-attending: less than once a month (or, where applicable, varies too much to say)
• Does not attend: never attends.

First, we present data from the BSA surveys, covering the period from 1983 to 2012 (with the exceptions of 1988 and 1992, when they were not undertaken). The question wording used in the BSA for measuring religious attendance is as follows: ‘Apart from such special occasions as weddings, funerals and baptisms, how often nowadays do you attend services or meetings connected with your religion?’[1] The data for attendance are shown in Figure 1.

Figure 1: Frequency of attendance at religious services amongst Catholics, 1983-2012 (BSA surveys). 

We can see that there has been a decline in the proportion of Catholics reporting that they attend religious services on a frequent basis: from 55.2% in 1983 down to 41.5% in 2012. There has been a corresponding increase in the proportion who says they attend infrequently (i.e. less than once a month), rising from 19.9% to 24.9% over nearly three decades. However, the larger increase has been amongst those who do not attend services, increasing from about a fifth in 1983 (19.9%) to a third in 2012 (at 33.6%). On the evidence of the 2012 data, then, around two-fifths attend services regularly, around a quarter say their attendance is infrequent and a third do not attend. Is this trend also evident in data compiled from other longitudinal survey series?

We next present over time data from the BES studies undertaken at every general election since 1964, which have asked about attendance in most, if not all, election surveys. Note the variations in question wording used across the BES surveys, particularly the qualification ‘apart from special occasions’ in the question used from 1983-1997, akin to that in the BSA question. We report data for the period of 1964-1997. We can see that the level of frequent attendance has fallen from a high-point of 71.0% in 1964 to just over two-fifths in 1997 (at 41.7%). There have been corresponding increases in the proportions attending either infrequently (from 13.7% to 32.8%) or not at all (from 15.2% to 25.5%). Unfortunately, data on religious attendance were not collected in the 2001 and 2005 BES surveys (two differently worded questions were included in the 2010 in-person survey which asked separately about religious activities undertaken (i) with other people and (ii) by oneself).

Table 2: Frequency of attendance at religious services amongst Catholics, 1964-1997

Source: BES surveys. Weighted data.

Question wordings:

1964: ‘How often do you attend church?’

1979: ‘How often do you attend church, chapel, or other place of worship?’

1983-1997: ‘Apart from special occasions, such as weddings, funerals, baptisms and so on, how often nowadays do you attend services or meetings connected with your religion?’

Next we present data from two long-running cross-country surveys. Firstly, the Eurobarometer surveys (Figure 2), for which we have data covering 1973-1998 (we omit data from more recent surveys due to changes in the response options for the measure of attendance). Because of the attendance categories used in the earlier EB surveys, we have to use a slightly different classification scheme: attends once a week or more; attends less often; or does not attend. The EB question asked: ‘Do you attend religious services …?’.

Figure 2: Frequency of attendance at religious services amongst Catholics, 1973-1998 (EB surveys)

We can see that the proportion of Catholics who report attending once a week or more declined from 57.1% in 1973 to 40.8% in the late-1990s. Those attending less often (which here is more complicated as it includes those attending once a month) increased from 27.1% to 35.0% and those reporting they did not attend services increased from 15.8% to 24.3%.[2] Secondly, the next set of data comes from the European Values Study (EVS), which undertakes periodic cross-national surveys (spaced every nine years). The data here show a slightly different picture from that already discussed, in that, perhaps surprisingly, there is little change over time in the proportion who report never attending services (at 27.6% in 1981 and 27.0% in 2008). The change over time has been in the form of the proportion attending frequently declining (from over half in 1981 to about two-fifths in 2008), with a corresponding increase in the proportion attending less often (rising from 18.7% in 1981 to 33.1% in 2008). The EVS data also shows an evident decrease in the proportion who never attended in 1990, but this ‘blip’ disappears as the levels increase again in 1999 and 2008 readings.

Taken together, it would seem that the surveys tend to show a picture of fairly consistent decline, with – the EVS data partly excepted – clear decreases in the proportion attending frequently and corresponding increases in the proportions attending less often or not at all. Given this historical picture, which social groups within the contemporary Catholic population are more – or less – likely to be frequent-attenders?

Table 3: Frequency of attendance at religious services amongst Catholics, 1981-2008

Source: European Values Surveys. Weighted data.

Question: ‘Apart from weddings, funerals and christenings, about how often do you attend religious services these days?’

Who is more likely to attend frequently?

In this section, we use the best available survey data in order to build up a social profile of those groups within the Catholic population most likely to report attending church frequently. We do this by analysing a survey of adult Catholics in Britain (with a sample size of 1,636), conducted online by YouGov from August 31-September 2 2010 in the run-up to the papal visit. In Table 4, we present self-reported attendance rates for the following characteristics:

• Sex, age group, ethnic group, social class, educational attainment, whether any children in the household, region and political party supported.

First, looking at the overall distribution, we can see that just over two-fifths report attending frequently (43.0%), nearly a third attend less often (31.9%) and those who do not attend services stand at a quarter (25.1%).  This figures approximate those found in the most recent BSA survey (for 2012) shown in Figure 1. So which social groups are most likely to report that they frequently attend services? We can see that there is little difference between men and women (with slightly over two-fifths reporting they attend frequently). There are substantial differences in level of attendance by age group: those aged 65 and over stand out, with 65.8% attending frequently compared to between 33.1%-40.8% for the other age groups. There are also differences by social class, with those in the highest occupational grades more likely to report attending services frequently: 50.4% for those in the AB category (professional and managerial occupations) compared to a range of 36.6%-41.9% for those in the C1/C2/DE groups. This difference by socio-economic status is reflected in the break-down by educational attainment: regular attendance is higher for those with degree-level qualifications than for those without. Attendance is also noticeably higher for those who have one or more children in their household (at 51.1% compared to 39.6% for those with none).

The breakdown by region shows that frequent attendance is most commonly found amongst Catholics who reside in the East of England, Scotland, the South East and the Midlands (and lowest in the Yorkshire and Humberside area). We also given figures for political party attachments amongst Catholics: frequent attendance is highest amongst those who support the Conservative Party (at 49.9%) and lowest amongst those who opt for a minor party or do not support any party. Data from a 2008 survey undertaken by the Pew Research Religion & Public Life Project showed that, amongst Catholics in the US, those more likely to attend church weekly included those aged 65 and over, women and those living in the Midwest and Southern regions.

Table 4: Religious attendance by social group, Catholic adults in Britain

Source: YouGov survey of Catholics adults in Britain, August-September 2010 (n=1,636). Weighted data.

For the purposes of historical comparison, further analysis of the 1978 Roman Catholic Opinion Survey (which sampled 1023 Catholic adults living in England and Wales only), shows a similar social profile for frequency of attendance (based on the following question: ‘’How often do you go to Mass?’). That is, there are some similar differences based on sex, age, social class and party support. For example, 72.5% of those aged 65 and older attended frequently compared to 36.7% of those aged 15-29. In terms of social grade, 62.5% of those in the AB category attended frequently, compared to 46.9% for those in the DE category. There were some interesting differences based on age finished full-time education: the highest proportions of frequent-attenders were found amongst those who, on the one hand, had finished school aged 14 or under (60.6%) and, on the other, those who completed their education aged 17-19 years (63.5%) or aged 20 and over (70.0%). The lowest levels were found amongst those who completed their education aged either 15 or 16 years (respectively, 41.5% and 45.0%). Conservative supporters were more frequent attenders (at 59.8%), compared to Labour supporters (47.5%) or those who expressed support for another party (55.3%). Levels of attendance were not too dissimilar for men (51.8%) and women (54.5%).

Finally, BRIN readers who are interested in looking further at scholarly research on this subject, focusing on Britain or elsewhere, may like to consult the following sources;

• Brenner, P. S. (2011), ‘Exceptional Behavior or Exceptional Identity?: Overreporting of Church Attendance in the U.S.’, Public Opinion Quarterly, 75 (1): 19-41.
• Chaves, M. (2011), American Religion: Contemporary Trends. Princeton, N.J.: Princeton University Press, chapter 4.
• Conway, B. (2013), ‘Social Correlates of Church Attendance in Three European Catholic Countries’, Review of Religious Research, 55(1): 61-80.
• Lee, L. (2011), ‘Religion. Losing Faith?’, in A. Park et al. (eds), British Social Attitudes 28. London: Sage, pp. 173-184.
• Lünchau, P. (2007), ‘By Faith Alone? Church Attendance and Christian Faith in three European Countries’, Journal of Contemporary Religion, 22(1): 35-48.
• Voas, D., and R. Ling (2010), ‘Religion in Britain and the United States’, in A. Park et al. (eds), British Social Attitudes. The 26^thReport. London: Sage, pp. 65-86.

Dr Ben Clements, Department of Politics and International Relations, University of Leicester