by David Voas.
Some of the best statistics we have about churchgoing come from attendance counts. Every year the Church of England, for example, publishes figures on Easter, Christmas and average weekly attendance based on data gathering a year or so earlier.
One challenge in interpreting the numbers is that they tell us about attendances, not attenders. Users of the 1851 Census of Religious Worship have always faced the difficulty that many Victorians attended both morning and evening services, so attendance counts must somehow be scaled down in estimating the number of churchgoers. Today the problem is the reverse: many ‘regular’ churchgoers do not attend every week.
We can use self-reported attendance frequency from social surveys to estimate how many people attend how often. Unfortunately we have a good deal of evidence that the answers people give to these questions are often inaccurate. People who intend to go weekly may claim to do so, whatever their degree of success.
Any estimate of the number of churchgoers is sensitive to the assumptions we make about attendance frequency. If everyone goes weekly or not at all, then one million attendances in a given week translates directly into one million attenders. If people attend just once a month, then a weekly attendance count of one million would suggest that there are four million churchgoers.
One solution is to maintain a register for several weeks in a sample of churches so that we can identify who has or hasn’t come previously. One of the best known studies of this sort was carried out in the diocese of Wakefield in 1997; see
The results of the survey are surprising. Of all individuals attending at least one service during an eight week period, more than half came only once. (It is hard not to suspect data error; one wonders how many people were listed twice on the registers.) If we say that someone should be regarded as a churchgoer if he or she attends at least monthly, then the Wakefield survey (if taken at face value) suggests that there are about 37 percent more Anglican churchgoers than there are C of E attendances in any given week.
Church leaders sometimes argue that even regular attenders now appear more sporadically than in the past. To put it another way, the decline in average weekly attendance exaggerates the decline in the number of churchgoers, because fewer people are coming every week. The conjecture is plausible and deserves investigation (using more than anecdotal evidence).
One possible test is to compare attendance at Easter with that in an ordinary week. The theory would be that all churchgoers still make a serious effort to attend on the holiest day in the Christian calendar, even if the importance attached to regular weekly attendance has diminished. If the ratio of Easter to average weekly attendance has increased over time, it implies that the decline in the number of churchgoers has not been as rapid as the decline in weekly attendance counts.
We have Church of England statistics extending back several decades only for Easter communicants (i.e. participants in Holy Communion); all Easter attenders have only been counted since 2001. In addition, the historical data on ordinary attendance is a count referred to as ‘usual Sunday attendance’; the Church now prefers ‘average weekly attendance’ on the grounds that some people come to midweek and not Sunday services.
The evidence from these statistics is mixed. From 1990 to 2008 there has been no change in the ratio of Easter communicants to usual Sunday attendance: the former is about 20 percent higher than the later, with relatively little variation from year to year. During the 1980s the Easter communicant numbers were relatively higher (at around 28 percent more than ordinary attendance), which would undermine the theory that churchgoers are now attending less often than previously. In 1970, however, the ratio was considerably lower (1.06). If that value is representative of earlier years, then by implication the number of active Anglicans has held up better than the weekly attendance counts.
As for Easter attendance itself, over the past several years it has been about 25 percent higher than average weekly attendance. Some of those in church at Easter will be visitors or infrequent attenders, but one might assume that they are balanced by churchgoers who are away on holiday. Easter attendance is arguably a reasonable proxy for the total number of churchgoers (defined as people who go at least once a month). In 2008 – the latest year for which figures are available –1,415,800 adults and children attended Church of England services on Easter.
David Voas is a sociologist of religion at the University of Manchester.