Selected Subjects: Ages

Next Selection Previous Selection

4.3 Ages

4.3.1 1821-1931

Great Britain 1821 - 1851

A tentative question as to age was included in the census schedule of 1821. Enumerators were asked:

'If you are of the Opinion that in making the preceding Enquiries (or at any time before returning this Schedule) the Ages of the several individ¿uals can be obtained in a manner satisfactory to yourself, and not inconvenient to the Parties, be pleased to state (or cause to be stated) the Number of those who are under 5 Years of Age, of those between 5 and 10 Years of Age, between 10 and 15, between 15 and 20, between 20 and 30, between 30 and 40, between 40 and 50, between 50 and 60, between 60 and 70, between 70 and 80, between 80 and 90, between 90 and 100, and upwards of 100, distinguishing Males from Females'.

Although the question was optional, the returns showed that eight-ninths of the people enumerated had given their ages and the information thus obtained was published in a summary at the end of each county table in the Enumeration Abstract which showed the distribution by thirteen age-groups for every hundred, district, ward or similar division of the county as well as for large towns, cities and boroughs. The question was not repeated at the census of 1831, when only the number of males over 20 years of age was obtained; the numbers returned were shown for every place for which figures were given in the county tables of the Enumeration Abstract . It will be seen from page vi of the first volume of this Abstract that enumerators were instructed to examine age returns which did not conform to the pattern, suggested by 1821 Census experience, that broadly half of the male popula¿tion was under 20 years of age and to correct them where necessary. How many did so is not known. This was probably the only occasion on which enumerators were requested to 'correct' their data.

In 1841 a column was provided on the householder's schedule for a statement of age. The preface to the Enumeration Abstract contained three tables of which the first was a comparative statement for 1821 and 1841 of the proportion of males and females in each of the thirteen age-groups per 10,000 of the population for every county in Great Britain and for each of the Islands in the British Seas; the second provided the same comparison for a number of principal towns, while the third showed the proportion per cent of females to males at each quinquennial period of life in the counties of Great Britain as shown by the returns made in 1841. In the county tables of the Enumeration Abstract information as to age was restricted to the numbers of males and females under and over 20 years of age enumerated in each parish or place.

At the census of 1851 a precise statement of age at last birthday was required and the information published in Population Tables II showed the age distribution by quinquennial groups in all registration districts and sub-districts and principal towns in England and Wales and in counties and principal cities and burghs in Scotland. Comments on the difficulty of obtaining correct statements of age and an assessment of the value of the data, together with a comparison of the results with those obtained in other countries, were given on pp xxiii-xxviii of the first volume of Population Tables II .

England and Wales 1861 - 1931

The age distribution published in the 1861 Population Tables Vol II, in Vol III of the 1871 Population Abstracts and in the third volume of the Reports for 1881 and 1891 followed the same quinquennial pattern as in 1851 with the addition of single ages in the under 5 group. In the Reports on the census of 1901 numbers at each age from 13 to 21 years were also tabulated, as well as the results of an experimental tabulation by single years of ages of about half a million of the population. Volume VII of the Reports on the 1911 Census was the first to contain tables showing the ages of the entire population by single years throughout the whole period of life. This volume contained extensive analyses of the people of England and Wales and comparisons with the age distributions in other countries. Analyses of local populations given in Table 8 of this volume showed the ages of persons, males and females at each age under 21 years and in quin¿quennial age-groups in each administrative county and in the aggregates of urban districts, of county boroughs and of rural districts; the table also gave the same age detail for males and females in each county borough, municipal borough, other urban districts and rural districts.

Statistics of ages of the local populations published after the census of 1921 were more extensive than those of 1911. Tabulations by individual years of age were given in the County Parts for each county borough and for county aggregates of urban and rural districts. Tabulations for every urban and rural district followed the conventional quinquennial grouping. In Table 32 of the General Tables full detail was given for England and Wales as a whole, for London and for aggregates of all urban and rural districts.

Economic conditions prevailing in 1931 were reflected in the census reports and the published statistics of age were slightly curtailed in comparison with those of the previous census. Populations classified by individual years of age were published in Table 17 of the General Tables for England and Wales as a whole, for density aggregates of areas and for each of the twelve regions. For all lesser areas, each urban and rural district, individual age details were restricted to ages below 21 in Table 16 of the series County Parts , with a summary in the conventional age groups in Tables 14 and 15 of the same series.

Scotland 1861 - 1931

Except as mentioned below, the age tabulations were given in the second volume of the series of reports on each successive census in Scotland. From 1861 to 1901 inclusive the conventional quinquennial age grouping was used, with analyses by single years in the under 5 group. The Report for 1871 contained an interesting discussion on the differences in age distribution in the old settled countries of Europe compared with that in the newly settled countries such as the USA. In 1921 analysis by single years of age was given for burghs with over 10,000 population and in 1931 for large burghs.

Accuracy of age statements

Most of the textual parts of the census reports contained comment on the extent to which statements of age given on the schedules were regarded as accurate. The report by George King, FIA, FFA, to the Registrar General on the graduation of ages, contained in Volume VII of the 1911 Reports , should be noted as well as the discussion contained in the 1921 and 1931 General Reports for England and Wales and the second volumes of the Scottish Reports for 1871 and 1911 .[i]

4.3.2 1951 - 66

In the published reports relating to the three censuses which have taken place since the end of World War II, age has been a constant factor in almost all subjects. Only a few tabulations do not use age as an axis; for 1951 only the Workplace Section of the Usual Residence and Workplace volume fell within this category (England and Wales); in 1961 the reports on Workplace , Housing and Socio-Economic Groups, and in 1966 the volume of Workplace/Transport Tables . The position for Scotland was broadly similar.

1951

The particulars which were required to be entered on the schedules for 1951 were age in years and months with infants under one month being so returned. This requirement was unchanged from 1931 and 1921 but prior to that, the statement had been limited to complete years except for children whose age was less than one year. There was no intention at any time, though, to tabulate by months and the extra information was obtained merely to try to get extra precision in the answers to this question.

Accuracy of statements

As has been mentioned in the paragraphs dealing with 1821-1931 there has been comment at each census on the extent of accuracy and similar investigations were carried out in 1951.

In addition, to provide a more absolute test of types of error, it was decided to match a sample of census age statements against the dates of birth of the selected persons as shown in the birth registers. Also, for persons who die sufficiently soon after census date to render a change of home address between enumeration and death unlikely, yet sufficiently long after the census to make it probable that they were enumerated at their usual home address and not in hospital, it is possible to match the census schedule and the information obtained at death registration in order to check the consistency of comparable items of information and to gain some appreciation of the validity of such information. Such an investigation was carried out for deaths registered in the period 1st to 7th May 1951 inclusive. For full details see Chapters II and IV of the General Report .

Publications - England and Wales

Tabulations for 1951 on this topic show some changes from the previous census of twenty years earlier. The analysis of ages by individual years, which still appears as Table 17 of the General Tables volume is restricted to England and Wales only, while the regions of England, Wales and the density aggregates are covered by Table 25 in single years of age up to 24 and quinary groups thereafter. Although the majority of tabulations by age in the three post-war censuses also carry analyses by sex and marital condition, this is not so with Table 25 which contains an age/sex breakdown only. However, Table 26 which again covers the region, etc by quinary age groups (0-4 to 90-94 plus 95 and over) has a full listing of sex/marital condition groups. The County Report series contains two tables (numbers 21 and 22 ) which are identical in the age structure. An age analysis by quinary groups is the common factor, and Table 21 deals with data for the county, county borough and urban areas with populations of 50,000 or more, while Table 22 is for urban areas with populations of less than 50,000 and for rural areas. Table 23 gives an analysis by single years of age under 21 for the county and each of the other local authority areas therein; and Table 4 also contains a restricted age distribution for selected civil parishes. In the report on Greater London and five other Conurbations , the principal items on age are Table 21 (quinary groups as above) and Table 23 (single years under 21) for each area. Mention must also be made of the One Per Cent Sample Tables which deal with age at Section 1 - in Tables 1-4. These tabulations are similar in scope if not identical in content to the General Tables which later superseded them.

Publications - Scotland

There are differences in detail in the Scottish volumes for 1951; the main tables are to be found in the General Volume and the County Reports which are similar in scope to the series for England and Wales. The General Tables contain at Table 21 an age analysis for Scotland as a whole by single years (0-99 and 100+), while Table 25 gives mainly quinary groups (0, 1-4, 5-9 to 90-94 and 95+) for Scotland, cities, counties, regions and regional hospital board areas. Table 30 gives a similar analysis for the cities, counties and large burghs with the addition of a full marital condition breakdown. The County Reports series provided in Table 11 details of the population by single years of age (0-99 and 100+) for each area as a whole and Table lla classified the large burghs by single years under 21. Table 15 uses a version of the mainly quinary group structure (0, 1-4, 5-9 to 80-84 and 85+) applied to the counties, large burghs, small burghs, landward, mainland and insular areas, districts of county or city and wards as appropriate.

1961

In 1961 the schedule question was again unchanged but combined with a statement of sex and the instruction was to enter Sex - M or F and Age, in years last birthday and completed months since then. This was supplemented by appropriate notes also on the schedule.

Accuracy of statements

In addition to the standard evaluations of accuracy, the birth matching exercise of 1951 was repeated and in this instance comparison was also made between the answers given in the post-enumeration survey and in the birth registers. The comparison with death registration records was also repeated using deaths reported during May and June, 1961, of people under 75 years of age at the time of death. For full details see Chapter 3 of the General Report .

Sampling and Bias

Age is an axis used in many tables, including those derived from the ten per cent sample. In the comparison between the full count and the sample the total number of persons showed an excess of 0.5 per cent in the total sample population. But comparison for age showed a systematic under-representation of old people, a small deficiency for children under 5 and too many young people between the ages of 5 and 20. For further details see the General Report and 7.2 and 7.3 of this volume.

Publications - England and Wales

The publications following the 1961 Census included a volume Age, Marital Condition and General Tables , which as far as the section on ages is concerned, covers much the same ground as the preceding General Tables for 1951. Table 8 for 1961 is identical with Table 17 for 1951, (Age by single years - England and Wales only), while Table 15 differs only in slight detail from the 1951 Table 25 (the analysis by single years finishes at age 20 instead of 24 and the areas covered in 1961 include conurbations). Table 16 is identical with the 1951 equivalent (Table 26 ) apart from the additional areas covered (see above). In the County Report series, the 1961 Table 6 is equivalent to the 1951 Tables 21 and 22 combined as it covers all local authority areas plus conurbation centres and New Towns while Table 7 is almost identical with the 1951 Table 23 except that, again, conurbation centres and New Towns are added to the list of areas tabulated.

The Great Britain Summary Tables were introduced in 1961 and Table 4 is very similar to Table 8 of the Age, Marital Condition and General Tables (for England and Wales) in classifying the population by single years up to 109.

Publications Scotland

The Scottish volumes for 1961 included the Age, Marital Condition and General Tables wherein Tables 8 , 15 and 16 are identical, except for areas, with those similarly numbered in the version of this publication for England and Wales. Table 8 refers to Scotland only while Tables 15 and 16 refer to Scotland, regional divisions and sub-divisions and burghal/landward aggregates. In the County Report series, Tables 6 and 7 are identical with their equivalents for England and Wales except that the areas covered are the counties, large burghs, small burghs, burgh wards, districts of county, New Towns or the cities and city wards.

1966

There was a change in the question on the 1966 schedule on this subject; the person completing the form was required to 'state the sex (M or F) and date of birth' of everyone appearing thereon. This instruction was backed up by the general issue of a specimen completed form (in reduced size) which, of course, clearly showed how all questions should be answered.

This was the first time in which date of birth was specifically asked for in the census. For tabulation purposes 'date of birth' was converted to 'age in completed years at census date'.

Publications - England and Wales

For 1966 there were a number of amendments to existing tables on age as well as one new publication dealing with this subject. There are three tables in each County report which should be noted. Table 2A gives the age breakdown by quinary groups 0-4 to 70-74 and 75 and over and the area analysis is for the county, county boroughs, urban areas with populations of 50,000 and over, county aggregates and New Towns (Scale A). The age classification for Table 2B is rather a hybrid mixture and comprises 0-4 years, 5-14, 15-19, 20-24, 25-44, 45-59, 60-64 and 65 and over. The area analysis is for urban areas with populations of 15,000 but less than 50,000 and rural districts with populations of 15,000 and over (Scale B). Table 3 gives single years of age under 21 for the areas defined above as Scale A.

In the Great Britain Summary Tables , Table 2 deals with age in single years 0-94 with a 95 and over group and the areas distinguished are Great Britain, England and Wales, and Scotland. Table 3 gives age by quinary groups (0-4 to 90-94 plus 95 and over), while Table 4 deals with single years of age under 21. The two last-mentioned tables give the appropriate data for Great Britain, England and Wales, Scotland, regions of England and Wales and conurbations. The new publication is the United Kingdom General and Parliamentary Constituency Tables and Part I, Table 2 contains an age break¿down by quinary groups (0-4 to 90-94 years plus 95 and over) for the United Kingdom, Great Britain, England and Wales, Scotland and Northern Ireland.

Publications - Scotland

The 1966 County Reports for Scotland have three tables (Nos 2A, 2B and 3) giving age and area analyses, which in age groupings and format are identical with those bearing the same numbers in the County reports series for England and Wales. The Scottish series use as a basis for area tabulation in tables 2A and 3, each county of city, each county, each county exclusive of large burghs, and large burghs of more than 50,000 population (Scale A). Table 2B is tabulated by each large burgh and each small burgh with a population of more than 15,000 but less than 50,000 and each district of county with an expected population of more than 15,000 (Scale B). The Report on the Special Study Areas has two tables for each area which are relevant; Table 2 gives an age analysis by quinary groups (0-4 to 70-74 plus 75 and over) while Table 3 is concerned with single years under 21. The area base for both tables comprises the Study Area, county (part, if applicable), burghs and districts of county.

Conclusion

In conclusion, it must be emphasised that the foregoing does not purport to be a complete list of tabulations involving the use of age data. It is no more than an arbitrary selection of tables where age is an important component and where substantially the whole age range is involved; some tables using age groups of single years under 21 or under 24 have also been included.


[i] See Observations on Errors of Age in the Population Statistics of England and Wales by V P A Derrick (Journal of the Institute of Actuaries, Vol.LVIII, pp 117-159. July, 1927) and two papers by J C Dunlop, viz (a) Note as to Error of Statement of Ages of Young Children in a Census (Journal of the Royal Statistical Society, Vol LXXIX, pp 309-317. May, 1916) and (b) Mis-statement of Age in the Returns of the Census of Scotland (Ibid Vol LXXXVI, pp 547-569. July, 1923) .

Office of Population Censuses and Surveys/General Register Office, Guide to Census Reports: Great Britain 1801-1966 (London: HMSO, 1977) Crown Copyright. The Office of National Statistics has granted the Great Britain Historical GIS Project permission to computerise this publication and include it in this web site. All other rights reserved.

Next Selection Previous Selection