Churchgoing and the Weather

The most adverse winter for thirty years, currently being experienced in the United Kingdom, will doubtless take its toll on levels of church attendance, not least given the relatively elderly profile of most mainstream Christian denominations. We shall probably never know for sure, since those Churches which make annual returns of their worshippers mostly do so on the basis of counts during the autumn.

One small clue to the impact of bad weather on churchgoing comes from a census taken in the Dunfermline Presbytery of the Church of Scotland in 2009 (and to be repeated this year). This is the subject of a short feature by Peter Brierley in the February 2010 issue of FutureFirst, the bimonthly magazine of Brierley Consultancy. A more detailed report on the 2009 Dunfermline census, including breaks by gender and age, may be obtained from Rev Allan Vint at

The Dunfermline census was conducted over two Sundays. The first, 8 March 2009, ‘proved to be a day of adverse weather conditions which resulted in particularly poor attendance for almost every church’. The census was therefore repeated on 15 March, when the number of worshippers was 7 per cent higher than the week before.

There are also some scattered historical data about the effects of bad weather on churchgoing, especially from 30 March 1851 when there was a Government census of religious worship throughout Great Britain, an exercise which has never been repeated. The day was mostly wet and stormy, as confirmed by extant meteorological readings. Comparison between the statistics for census day and average attendances reveals the former to be significantly lower, particularly in rural areas. In Shropshire, for example, general congregations were reduced by 22 per cent below the norm for places of worship which expressly commented on the state of the weather on 30 March.

Other local counts of church attendance from the late Victorian and Edwardian eras point in the same direction. In Cheltenham congregations were again 22 per cent lower on 29 January 1882 (a very wet day) than on 5 February (a reasonably fine one). In West Cumberland the census of churchgoing was taken on 14 December 1902, an exceedingly stormy day, which, in this predominantly rural area with indifferent transport and roads, apparently reduced congregations by up to two-thirds.

However, the evidence is by no means consistent. At Bradford a census taken on 11 December 1881, immediately after heavy snow, did not produce an appreciably smaller turnout at church and chapel than a replication on 18 December, when the weather was somewhat better, especially in the morning. Similarly, in Carnarvon attendances on a bitterly cold day (26 January 1908) were just 7 per cent below those on a warm sunny one (5 July), the same difference between two Sundays as for Dunfermline in 2009.

Hitherto, there has been little discussion of the topic in the academic literature. One exception is Robin Gill, who has made a short study of the relationship between weather and church attendance during the late nineteenth century (The ‘Empty’ Church Revisited, Aldershot: Ashgate, 2003, pp. 20-3). He concludes that the weather did make a difference to churchgoing, but less than one might expect. Also, according to him, bad weather was more likely to reduce Free Church than Anglican attendances.

British Religion in Numbers: All the material published on this website is subject to copyright. We explain further here.

This entry was posted in Measuring religion and tagged , , , , . Bookmark the permalink.