[note this page forms a slightly updated section of my paper delivered to Manx Methodist Historical Society AGM, Nov 1998 - see index for other sections and actual returns]

General Background to 1851 Religious Census


The unveiling of the newly restored Albert Memorial in October 1998, serves as a useful reminder of the mid 19th century enthusiasm for the sciences and technology coming after the political turmoil of the 1830's and the hungry 40's. The 1851 Great Exhibition is possibly the best remembered, with Albert's memorial standing at the head of the South Kensington complex built from the profits of the exhibition which he inspired. One aspect of this enthusiasm was the collection of data1 on which to base future legislation. A decennial census2 had been taken in the UK from 1801 (1821 in Man) though those until 1841 had not gathered much information other than the total population and some data on the number of dwellings. The 1851 population census, taken on the night of Mothering Sunday, the 30th/31st March, was the first census to ask detailed questions re structure of the household, ages etc. Along with this census two other censuses were taken:- an educational census of all schools (including Sunday Schools) asking about the number of students, teachers and running costs, and a census of Accommodation and Attendance at Worship generally referred to as the Religious Census taken on the 30th March.

Neither this census nor the population census asked people directly for their religious affiliation - a religious census had been carried out in 1834 in Ireland which in part led to the reorganisation of parishes and dioceses. It was proposed to include similar questions in the 1851 population census but vigorous opposition by the Bishops in the House of Lords (led by Bishops of Oxford and Salisbury) successfully prevented this, though many churchmen felt that this would have shown the Established Church in a better light than the scheme adopted, as it would have included many nominal members who did not attend a church service. Questions on personal religious affiliation were difficult at a time when many Anglicans, including the brother of the Bishop of Oxford, had converted to Catholicism and others were considering their position (the future Cardinal Manning moved to Rome the following Sunday). It should also be remembered that 1850 saw considerable anti-Catholic agitation over the decision by the Pope to restore the hierarchy and give bishops territorial designations - the evangelical wing of the Established Church were vocal in their protests, a bill was passed by Westminster to forbid this 'papal aggression' (the last anti-Catholic legislation) and disturbances were fomented in several parts of the British Isles - including, according to High Bailiff Laughton3 Douglas where the windows of St Francis Xaviers Chapel and of Roneys (a Catholic grocer) in Duke Street were broken.

To avoid this storm of protest about asking about personal religious affiliation, the stated purpose was to discover 'how far the means of Religious Instruction provided in Great Britain during the last fifty years have kept pace with the population during the same period'. As Horace Mann put it in his introduction:

At the recent census, it was thought advisable to take the latter course; partly because it had a less inquisitorial aspect, - but especially because it was considered that the outward conduct of persons furnishes a better guide to their religious state than can be gained by merely vague professions. In proportion, it was thought, as people truly are connected with particular sects or churches, will be their activity in raising buildings in which to worship and their diligence in afterwards frequenting them; but where there is an absence of such practical regard for a religious creed, but little weight can be attached to any purely formal acquiescence. This inquiry, therefore, was confined to obvious facts relating to two subjects. - 1. The amount of ACCOMMODATION which the people have provided for religious worship; and 2. The number of persons, as ATTENDANTS, by whom this provision is made use of'.

The provision of a national census inquiring into the provision of religious and educational facilities was welcomed by many - a Times leader4 of 25 March 1851:

'the relation of existing means to existing wants in both these departments; the extent to which existing means are valued and improved; the connexion between the average instruction and the average religiousness of districts; between the quality of that instruction and the modes of that worship; the connexion of both with the general aspects of society in various localities-these are topics in which the light of universal statistics has not hitherto been shed, and which yet cannot be fairly treated without such illumination.

To use an old theatrical term the means adopted was to measure 'bums on seats' as well as asking details about each church, chapel or meeting place. Turning these attendance statistics back into denominational membership figures has exercised all students ever since !

Although it was intended to repeat the Religious Census in 1861 the various denominations could not agree to the form it should take and thus the plan was dropped - hence the 1851 census is the first and only such carried out in the UK (and Man) and as many have described it, somewhat of a failed experiment.


The delivery of the forms, and the discovery of the various meeting places, was the responsibility of the local census enumerators who had to deliver the census forms in the week prior to the census and then to collect them along with the population census forms on the following Monday. These forms were then passed onto the registrars who would elicit further information to fill in missing sections.

Three distinct forms were produced:
(A) black print on a blue paper for Anglican Churches
(B) red print on a blue for Non-Anglican places of Worship
(C) black on white for The Religious Society of Friends or 'Quakers'.

Each form asked slightly different questions -All forms asked about attendance at services on that specific Sunday, forms A and B (but not C) asked about average attendance. Form A, for Anglican churches, asked about endowments, the income of the church and its incumbent - this question in particular had been vigorously opposed by Lord Stanley as too prying; the government capitulated and removed the legal penalties on non-completion of the Education and Religious census forms thus in the eyes of many completely removing the point of the census as returns would be only partial. Others criticised some questions as ambiguous and not capable of easy answer. The Bishop of Oxford, Samuel Wilberforce, led a last minute attempt to prevent the census but failed and thus effectively a 'voluntary' but officially encouraged Religious census went ahead. In practice it seems that most Anglican incumbents were happy to fill in the 'prying' sections of the form - many local studies comment on the sometimes caustic remarks on the same subject given by some poorly paid clergymen.

Those who had commented on the ambiguity of some of the questions were however proved correct and some information, such as the number of children present, was capable of much misinterpretation.

Questions asked:

(based on the section in D. Robinson 1997)

Form A:
I: Name and Description of Church or Chapel
II: Where Situated
with boxes for Parish, Ecclesiastical Division etc; Superintendent Registrar's District and County and Diocese
III: When consecrated or licensed
IV: In case of a Church or Chapel Consecrated or licensed since the 1st January, 1800; state hereafter by Whom Erected, Cost, how defrayed
V: How Endowed
VI: Space available for public worship
required details re free sittings, other sittings and total sittings - however no definition was given and some returns were misleading
VII: Estimated Number of Persons attending Divine Service on Sunday, March 20, 1851
This question was supposed to distinguish between general congregation and Sunday scholars - a distinction that was often confused. It also asked for separate attendance figures for Morning, Afternoon and Evening services (multiple morning services etc. were to be aggregated). The question also asked about average attendance for the same services
VIII: Remarks
IX: Signature
Form B/C:
I Name or Title of Place of Worship
II: Where Situate (as for form A but no question re Diocese)
III: Religious Denomination
IV: When erected
those before 1800 to be indicated as pre-1800 (this question seems to have caused some confusion in the case of rebuilt chapels)
V: Whether a separate and entire building
this and question VI were to distinguish between cottage meetings and those in chapels
VI: Whether used exclusively as a Place of Worship
VII: Space available for Public Worship
similar to that on Form A but clearer instructions (form C asked for total floor space - possibly based on assumption that the Friends stood for their servivces)
VIII: Number of attendants
same question as question VII on Form A but tabular layout of the question differed encouraging mistakes or misinterpretations.
Roman Catholic churches which may have several morning Masses were instructed to aggregate attendances
IX: Remarks
X: Signature


The task of tabulating the returns was given to a young (28 year old) solicitor, Horace Mann, who published his report in 1854 as one volume of the overall census reports5. It turned out to be an unexpected best seller! Over 21,000 copies were sold shortly after publication - one Congregational minister described it as 'at once the best and cheapest book on the religious denominations in England that has ever appeared'6. Part of this was due to Mann's own thorough review of the various denominations and his thoughtful comments on the actual tabulated results which include some quite forceful comments on social inequality. Perhaps the key result for many was summed up in Mann's comment 'it must be apparent that a sadly formidable portion of the English people are habitual neglecters of the public ordinances of religion' and his comments that most of these neglecters are to be found in the labouring classes.

Results and Mann's Analysis

Mann's report covered only England and Wales, results were tabulated for counties and major towns - a similar report was done for Scotland. His first difficulty was to turn attendance figures into membership figures - simple aggregation of attendance figures would overestimate numbers as it was common practice to attend more than one service. His formula to estimate the number of individual worshippers, Ñd, for a given denomination was:
Ñd = Md + Ad/2 + Ed/3 where Md = morning attendance, Ad afternoon and Ed evening.

The number of worshippers was then normalised to the possible church going population of the area, P, to give Ñd /P or as he termed it - the Index of Attendance (IA).

He also added together his estimated number of worshippers across all denominations to give Ñtotal and normalised Ñd to his estimated number of worshippers to give Ñd/ Ñtotal expressed as a percentage - his PS or proportional share.

Mann determined P by estimating that some 30% of the total population had a 'legitimate' excuse for non-attendance (small children, elderly , infirm...) - the total number of attendants was 7.26M out of a total population of England and Wales of 18M

The aggregated figures8 published for England and Wales were






Roman Catholic








27.6 / 47.4%

14.9 / 25.7%

6.6 / 11.4%

5.1 / 8.7%

2.0 / 3.5%


29.5 / 48.6%

15.2 / 25.0%

6.8 / 11.1%

5.2 / 8.5%

2.1 / 3.5%

The first row gives the 'uncorrected figures' as published. The final row gives the corrected figures after further enquiries were made about non-returns (most of which would appear to have been Anglican)

These national statistics had great variation across regions - e.g. in Cornwall the Anglicans had a PS of just 27.2% compared with the Methodist 64.5% whilst Essex had Anglican 57.5%, Methodist 8.5% and Independents 23%. Large towns showed even more variation e.g. in Liverpool the Roman Catholic PS was 32.5% reflecting the large Irish population. Larger towns and industrial areas also showed a much lower IA than rural areas.

Mann's formula of determining Ñd was much criticised as giving undue weight to Anglican practice which would have its main service in the morning whilst many non-conformists, Methodist included, would have their main service in the afternoon or evening. J. Vickers9 gives the following breakdown for attendance figures by time of day:



Roman Catholic













(note in all above tables, row totals will not sum to 100% as minor sects (e.g. Mormons) have been ignored).

To late 20th century eyes the fact that over 50% of 'eligible attendees' had actually gone to church that day would be worthy of note - however in the mid 19th century these figures were seen as an indictment of the ungodly nature of the country.

Subsequent Use

Once the original returns were released in 1951, under the 100-year rule, historians and other researchers could look at detail at the figures. It took a surprisingly long time for the academic community to excavate in this mass of documents, many seemed happy to rely on Mann's published tables. Local Historians have however over the last decade or so found them a fruitful source of data10.

One option was to apply different weightings to the reported attendance figures to arrive at the number of individual worshippers - a refinement was to keep Mann's estimate of multiple attendance but to calculate Ñd as Largest + half of second largest + one third of smallest attendance figures. Another scheme which allowed relative determination of denominational strength was Pickering's Largest Congregation approach (i.e. a Minimum bound on numbers)

Using the largest congregation figures a minimum of some 17% of the total population attended an Anglican church and a very slightly larger number, 17.2%, attended a dissenting service - thus some 34.2% of the population attended church that Sunday which in many parts of the country was a wet and windy day.

Although the IA figures may be defective the Proportional Share (PS) figures are generally accepted as giving a good indication of the very different regional strengths of the various denominations. Gay in 'The Geography of Religion in England' included many maps of the various denominational strength of the country or region, which clearly indicated the strength of Anglicans in the south of England and the strength of Methodists in areas where Anglican church provision had been weak .

Provision of Places

Mann's published figures also included the overall number of places for England and Wales, some 34,467 buildings providing some 10.2M sittings of which the Anglicans provided some 51.9% These sittings were enough to seat some 58% of the population but Mann noted the disparity of provision especially in the industrial towns of the North of England.


Much argument has ensued as to the completeness and accuracy of the gathered data.

It would appear that small Non-conformist cottage meetings were missed (some estimate some 4% of total membership), some Anglican incumbents refused to fill in the returns though in many cases subsequent enquiries filled in the blanks. Many forms gave what appeared to be estimated returns (a suspicious number of figures ending in '0' or '5') and in some cases duplicate returns were made for the same chapel. The registrars seemed to discount most duplicated returns (generally from Methodist Chapels) but some got through - e.g. the Manx returns would appear to include two for Bride WM. The registrars had some difficulty in assigning some returns to specific denominations. These were also errors on the actual returns which were not caught and sometimes it appears that the clerks were not immune from showing their own religious affiliations (especially anti-Catholic).

One problem with interpretation was the aggregation into counties etc which was not a useful one from a sociological viewpoint as parts of a county could be very different in nature from its neighbouring areas (e.g. heavily industrialised vs. rural) Gay's continued use of the county boundary hides some significant differences when the figures are viewed on a finer scale.

Educational Census

This has generally been neglected by historians - much of the information is readily available elsewhere as Education and its provision was a well studied problem at this time. It also appears that unlike the Religious Census the actual returns have not survived.

The census questions asked for considerable details re status of scholars - in vast majority of cases just the number on the roll and the number of teachers were given.

Like the Religious census the first questions were concerned with the name and location of the school, the starting date of the Sunday School- specific questions re scholars and teachers were:

(5): Number of scholars on Roll (male & female to be shown separately)
(6): Number of scholars on Roll receiving tuition on 30 Mar 1851
(7): Number of scholars on Roll receiving books who pay for them
(8): Number of scholars on Roll who provide own books
(9): Number of scholars on Roll who also attend some other day school
(10): Number of scholars on Roll who also formerly attended some other day school
(11):Number of teachers paid and unpaid - again split m & f

Other questions related to fees paid by scholars estimated annual cost of running the S/S and how income, other than fees, obtained


  1. Although censuses had been taken much earlier (e.g. the well known one c. 1AD!) and Political Arithmetic (e.g. for calculation of Life tables) studied from 17th century it was the 19th century that saw the real development see T. M Porter 'The Rise of Statistical Thinking, 1820-1900' Princeton (ISBN 0-691-02409-x) 1986
  2. An excellent survey of various topics connected with the Victorian censuses is given in: Local Communities in the Victorian Census Enumerators' Books ed D. Mills and K. Schurer 1996 (ISBN 0-904920-33-X) Local Population Studies Supplement (available via Dept of History University of Essex, Colchester UK)
  3. A.N. Laughton High Bailiff Laughton's Reminiscences Douglas 1916 see chapter VII
  4. Quoted by David Robinson in 'The 1851 Religious Census: Surrey'
  5. Census of Great Britain, 1851. Religious Worship in England and Wales. Report and Tables (PP1852-3), LXXXIX [1690]
  6. Thomas Davies minister of York Road Independent Chapel, Lambeth quoted by D. Robinson p liii.
  7. See extracts published in Religion in Victorian Britain III Sources pp313-321
  8. Table and other figures extracted from Coleman The Church of England in the Mid-Nineteenth Century
  9. J.A. Vickers The 1851 Religious Census
  10. see list of County Religious Census books - most appear to have been published within the last decade.



index next

Any comments, errors or omissions gratefully received The Editor
HTML Transcription © F.Coakley , 2001