Content associated with: Population tables, Scotland, Vol. II, 1871    Page 78

Questions about housing

Edward Higgs

Questions on housing were always part of the British censuses from their inception in 1801. In that year John Rickman, the clerk of the Houses of Parliament who organised the decennial enumerations from 1801 to 1831, asked the overseers of the poor in England and Wales, and the school masters in Scotland, to indicate the number of uninhabited and inhabited houses in their parishes, along with the number of families occupying the latter. In the years 1811 to 1831, he asked for the numbers of inhabited houses and the number of families occupying them, along with the numbers of houses being built, and of those uninhabited for other reasons (Higgs, 114–17).

The British decennial censuses from 1841 to 1861 were organised by the General Register Office (GRO) after which the Scottish GRO took over the enumeration north of the border. After 1841 the census was also taken on a different basis: household schedules were now given to each household head who had to fill in the required details relating to the members of the household. In the years up to and including 1901, census enumerators collected these schedules and copied them, in the order in which the households occurred in the streets, into special census enumerator's books (CEBs) which were sent to the Register Offices in London and Edinburgh for analysis.

But data on housing were still collected in the period after 1831, even though the censuses were based on households, several of which could inhabit one house. In the CEBs of 1841 there was a special dual column, headed 'Uninhabited or Building' and 'Inhabited' in which the enumerator was to put a '1', '1U', '1B'. In 1851 there was no special column in the CEBs but enumerators were supposed to indicate if houses were being built or were uninhabited. In 1861 the dual column of 1841 returned and continued until 1901, when the column was split into four — 'Inhabited', 'In occupation', 'Not in occupation', 'Building'. From 1891 enumerators were also to give the number of rooms occupied by a household if under five, and from 1911 all householders had to give this information. The CEBs also had columns for house addresses, and special marks were used on the pages of the CEBs to mark the beginning and end of houses, and of the households within them (Higgs, 120–6). The CEBs of the period 1841 to 1901 also contained preliminary tables in which enumerators were to give a summary of the number of houses on each page of their books (Higgs, 106–8).

From 1911 onwards the household schedules were themselves the main source used for statistical analysis and the production of the old CEBs ceased. However, the enumerators had always been required to keep a list of the addresses and descriptions of every building and dwelling in their enumeration districts in memoranda books, and these details were now sent to the central census authorities in summary form (Census of England and Wales, 1911, Vol. VI, 3). These data now formed the basis on which the stock of housing was analyzed in the published Census reports.

Information relating to the number of houses in defined administrative units was one of the first things published by the census authorities in the aftermath of each census. Within about a year of the actual taking of the census in England and Wales a summary volume was usually published giving the number of people and houses in particular areas. These statistics were probably derived from the housing tables in the CEBs. This would be followed up with a more detailed analysis in later publications on the particular census. In England and Wales from 1911 onwards there were special volumes dedicated to an analysis of the housing data collected by the enumerators, which it was hoped were now more accurate than those collected in the nineteenth century (Census of England and Wales, 1911, Vol. VI, 3). In 1911 there was also a separate volume containing an analysis of 'tenements', defined simply as the space occupied by a household (Census of England and Wales, 1911. Vol. VIII, iii). Such a volume did not appear in subsequent years. In Scotland the discussion of housing formed a part of the smaller number of general Census reports issued.

The main problem with all this information on housing was that the definitions of a 'house', 'dwelling', 'tenement' and 'room' were vague. In a rural setting, or the middle-class suburbs of large towns, the identification of a house or cottage might be fairly simple. But in the 'rookeries' of the great cities, where older buildings had been subdivided and rooms partitioned to squeeze in the ever-increasing urban population, there were insuperable problems.

The 1841 instructions to the enumerators were not particularly helpful:

By 'House' is meant Dwelling-House; and every building in which any person habitually sleeps must be considered as a dwelling-house.

But what of buildings which provided temporary shelter for people on census night, and how did one define a 'building'? From 1851 onwards the house was defined as the space between the external and party walls of a building. This was as much a legal as a physical description since a party wall was one between two buildings or pieces of land intended for distinct occupation, in the use of which each of the occupiers had a partial right. The definition depended on finding an 'occupier'. In the maze of dark back-alleys and courts in London and the industrial cities, it may have been difficult for enumerators to do this. The erection of model or industrial dwellings for the working classes also caused problems. Each block could be treated as a single but subdivided house, although each flat was a self-contained unit. In Scotland, where such buildings had been common from a much earlier date, the definition of a house had long since included that of a dwelling with a door opening directly on to a common stair. Some English enumerators, following their common sense but not their instructions, chose to regard every such flat as a separate house. What actually constituted a 'room' also caused endless problems (Higgs, 49–56; Census of Scotland, 1871, Eighth decennial census..., xxxi).

The census authorities themselves became disillusioned with the quality of the data collected in the old CEBs. As the GRO in London put it in 1931, the definition of a house in the Victorian censuses, 'could have done little to advance the practical utility of the records when houses of all kind, large or small, whether utilised as private family residences or as public institutions, were treated as of equal value and included indifferently in a composite aggregate'. It hoped that more accurate data had been collected from 1911 onwards, and especially with the adoption of the definition of a houses as a 'structurally separate dwelling' in 1921 (Census of England and Wales, 1931, Housing report, v-vi).


Census of England and Wales, 1911, Vol. VI. Buildings of various kinds, BPP 1912–13 CXIII. [View this document: Buildings of various kinds, England and Wales, Vol. VI, 1911]

Census of England and Wales, 1911, Vol. VIII. Tenements in administrative counties, county boroughs , &c., and ages and occupations of foreigners, BPP 1913 LXXVII. [View this document: Tenements, England and Wales, Vol. VIII, 1911]

Census of Scotland, 1871 Eighth decennial census of the population of Scotland taken 3d April 1871, with report. Vol. I., BPP 1872 LXVIII. [View this document: Population tables, Scotland, Vol. II, 1871]

Census of England and Wales, 1931, Housing report and tables (London: HMSO, 1935). [View this document: Housing report and tables, England and Wales, 1931]

Edward Higgs, Making sense of the census. The manuscript returns for England and Wales, 1801–1901 (London, 1989).