The Work and Premises of the Census Office 1840-1919

Census Clerks

The impermanent nature of the Census Office - over a period of 80 years - meant census clerks were recruited on temporary contracts. For the vast majority, their period of employment encompassed the preparation of the census and the subsequent processing and analysis of results. 

A certain number of experienced staff would be seconded from the General Register Office (GRO). The whole operation generally took three or four years. For example, the Census Office for 1861, which employed 106 temporary clerks, was not wound up until 1863.

Clerks poured over the enumerators' books. By referring to original schedules they sometimes amended entries and would correct additions in columns and tables. Later, specific headings of  information would be abstracted for publication in the Census Reports. To monitor their progress they would often tick of information as they went - these, plus notations can be seen on scans today. 

Using occupational dictionaries they assigned occupations to a particular category; the GRO dictionary for 1881 contained between 11-12,000 occupations. These were divided into 'Classes', 'Orders' and 'sub-Orders'. Sometimes standard headings or their allied code numbers were written in or alongside the occupation column. For more information read our blog article Trades and Professions in the Census 1841-1901.   

Until the introduction of tabulation machines in 1911 (see below) all tables in the Census Reports relied on manual data processing and analysis.  

In 1841 clerks were paid between five and eight shillings, according to their standard of work. 
They were primarily assigned 'task-work', namely piecework. This system of payment was already established at the GRO and George Graham, the long serving Registrar General, wrote in 1870 as to why it worked over a daily wage: 
'If temporary clerks and writers and boys are on day pay, they may be placed at desks; but no amount of supervision can obtain from all of them a good day's work. They know that the more work they execute in a day, the sooner their temporary employment will cease and they will be again turned adrift; therefore it is their interest to do as little work as possible.'   

From 1870 the Census Office were provided with Ordnance Survey (OS) maps to aid in administration. Boundary changes since 1891 meant a new set were issued for planning of the 1901 census. 

Women were first employed as census clerks in 1901. A lady superintendent was seconded from the Post Office and placed in charge of them. One of their number, 28-year-old Charlotte Benstead, was listed in that year's returns as worked as 'Forewoman at Census Office'. 

It wasn't all work and no play as the Office fielded a football team. In 1891 Sporting Life reported they were set to play against the 'Pirates' at Clapton, London on 10 October.

An Office on the Move

It was Thomas Lister, the first Registrar General, who insisted that the abstraction of statistics should be done centrally - namely in London. For the census of 1841 temporary accommodation was secured in Adelphi Terrace; there was no space at nearby Somerset House, the then base of the General Register Office (GRO).

The Census Office, Craig's Court, London, 1861 (Illustrated Times, 13 April 1861)
 

The Adelphi premises had proved unsatisfactory, so for 1851 the Census Office was established at Craig's Court, Charing Cross; they were to remain there for a further three censuses. The new headquarters consisted of three adjoining houses and were not a resounding improvement. The rooms within were cramped and poorly ventilated. The working day could extend to ten hours and the fumes from gas lights triggered headaches amongst many staff.

In 1890 the Census Office relocated from Craig's Court to temporary iron buildings erected in Charles Street, Westminster, opposite the Local Government Board (LGB). A telephone link was installed between them and the GRO in Somerset House. 

The pop-up headquarters were enthusiastically lauded in the The Lakes and Chronicles Reporter (20 February 1891):
'The corrugated iron structure forms really an ideal public office, all on the ground level, spacious and light and airy, and affording facilities for supervision probably unequalled in any Government establishment in the kingdom. The three great rooms, now all ready, and waiting 
for the great staff of clerks to be called in when the enumerators have gathered in the papers, and have entered their particulars in the enumeration books to be passed on by the superintendents to the pigeon-holes now all ready for them here in Charles-street, have a particularly bright and pleasant aspect...Electric bells and first-rate hot-water heating apparatus and thorough ventilating arrangements have been provided throughout.'

In 1899 these buildings were dismantled and re-erected behind the Tate Gallery in Millbank; the site served as census base for the 1901 and 1911 censuses. 

The Census Office at Millbank, London (Illustrated London News, 1 April 1911).
The iron-built premises were equipped with sprinklers in the event of a fire.

Young girls and women working at the Census Office, Millbank, London.

1911: The Adoption of Technology

In readiness for 1911 the Census Office re-opened at Millbank in 1909. In a revolutionary move, the recently appointed Registrar General, Sir Bernard Mallett, instigated the use of machine tabulation to process data.

Four kinds of machines were employed:

• Punching machines for perforating cards (68 Key Punches and 8 Gang Punches)
• 15 automatic sorting machines (for division into 11 groups)
• 8 automatic counting machines (of three types of varying capacities)
• 2 automatic tabulating machines (for tabulating the population and fertility results)

(The Daily Mirror, 28 March 1911)

Cards for 1911, the 'Fertility Census', were of three kinds: Population, Personal and Fertility. An army of coders, clerks and punchers were employed to process and tabulate the schedule data as it arrived to Millbank. That year up to 19 separate prices of information could be recorded for each household individual.

170 staff were employed, working in two shifts of six hours each. The highest grade of clerks were assigned to coding the occupation and industry replies. Punch operators were recruited from young girls just leaving elementary school; they transferred the coded information to the cards. The number of Personal cards punched per hour steadily increased, from 127 (5 August 1911) to 417 (21 September 1912). Experienced clerks were put in charge of counting or tabulating machines. Reflecting the population of England and Wales a total of 36,070,492 Personal cards were punched, whilst 6,136,605 Fertility cards were punched. 

Superintendents of Statistics (1839-1919)