This appendix discusses the method of correction used in the first part of the book, to adjust 1851 published registration-district data for missing values. For a variety of reasons, a small percentage of omissions occurred on the enumerators' forms. Such omissions were footnoted by Horace Mann in the published district tables of the census, although he did not make any changes to those tables. Mann did, however, provide overall tables in which he included estimates for ‘defective returns’. The problem is one that has commonly been ignored by historians. They have been aware of the difficulties in supplying such missing data, and of the relatively small numbers of census figures affected. The view has usually been taken that these missing figures have little effect on calculations. For England and Wales, omissions of sittings affect 7.3 per cent of places of worship, and of attendances 4.0 per cent. There is no alternative source generally available to assess the ‘real’ historical values of missing figures. Even chronologically very proximate sources of local information – in the rare cases where they exist – may still be misleading in that they do not give precise information as to what happened on Census Sunday.

Nevertheless it is possible to adjust the data to give rather more accurate values, taking some account of the missing figures which Mann alerted the reader to. In the registration-district analysis, on the scale covered here, it was felt that this matter needed to be addressed, to make the figures more accurate.