Selected Subjects: Migration

Next Selection Previous Selection

4.5 Migration

4.5.1 1851 - 1931

1851 - 1901

Although the movement of population was not included as a census question and therefore migration as a subject does not appear in census tabulations there have been a few tables included in various reports. The information generally relates to immigration or emigration and was obtained from other Official sources. Prior to 1911 there are only two tables, Table 30 of Appendix A in the 1881 General Report and Table 27 in the 1891 General Report, which show internal movement from and into ancient counties.

1911 - 1931

In 1911 and 1931 migration is shown for various types of area but only as a net gain or loss after excess or deficiency of births compared with deaths has been taken into account.

4.5.2 1951 - 66


The question was still not asked but some thought was given to the regional distribution of population and the emerging pattern of change. See General Report .


Following an ever growing demand from universities, local authorities and government departments for an inquiry to throw further light on movement within the country, a question on migration was introduced at the 1961 Census. It was designed to obtain information about the amount, direction and characteristics of population movement within the country and the number of years a person had lived at his usual residence. Information was collected only on a sample basis and this enquiry therefore appeared on the E10/W10 schedule (England and Wales) as question N, which reads as follows:

N (for all persons)

Was this person's usual address a year ago (ie on 23 April 1960) the same as that indicated in column C? Write 'yes' or 'no' at (i).

If 'yes' state at (ii) how many years ago he/she moved to that address. (If he/she has lived there since birth write 'birth').

If 'no state at (iii) the full usual address on 23 April 1960.

Notes and examples of completed entries were issued in a separate leaflet and note 11 therein on change of address (column N) instructs:

  1. `Years' means completed years eg for 4 years 11 months write '4 years'.
  2. Give the whole period since the person moved to the address indicated in column C even if he or she has since been temporarily away, eg owing to National Service, other war service or evacuation.

The most substantial product resulting from the answers to the three questions was the Migration Tables with advance tables in, leaflet form. In addition to fulfilling the defined objects of the enquiry, these tables also provided data (as variable factors in cross tabulations) on the age of migrants, their marital condition, socio-economic group, occupation and industry, together with similar information for the remainder of the population classified by the length of time they had lived at their present usual residence.

Tables relating to people refer to the resident population assigned to their area of usual residence. Tables relating to families and households are restricted to persons enumerated at their usual residence. The area of usual residence is naturally dependent on statements made in answer to questions C (usual residence) and N on the schedule and in completing question C, people were instructed to give their home address, not the term-time address for school children, students, etc and members of HM Forces not enumerated in camp or on board ship were told to give the address of their married quarters or other home address as their usual residence.

All tables exclude people who stated that they were usually resident outside England and Wales, and with a single exception (Table 1 ), they also exclude members of HM Forces enumerated in camp or on board ship.


In considering figures of migrants it should be remembered that a 'migrant' is anyone whose usual address at Census date (23 April 1961) was different from the usual address on 23 April 1960 and thus does not, by any means, give a true reflection of the total number of movements during the year. Anyone who moved more than once during the year will have been counted only once and then as moving from the usual address on 23 April 1960 to that on 23 April 1961 irrespective of any intermediate moves he may have made.

An 'immigrant' to an area was resident there on 23 April 1961 but resident outside the area on 23 April 1960.

An 'emigrant' from an area was resident there on 23 April 1960 but resident elsewhere in England and Wales by 23 April 1961; the tabulations exclude any moves elsewhere.

'Wholly moving private households' are those households enumerated at their usual residence of which all members present on census night (if aged 1 or over) had moved in the previous year; but they need not have all moved from the same former area. Whereas in a 'wholly moving family' the members must all have moved from the same area of former usual residence.

It must also be remembered that in tables relating to households and families these groups will not correspond to those in the Household Composition tables which are on a 'de jure' basis.

Also tables classifying the economically active population by industry include the out of work classified by former industry. This differs from the practice in other volumes where usually only persons in employment are classified by industry.

For full definitions see Migration Tables .


A not unimportant point of procedure on the coding of the migration data at the Census Office should be noted. In the course of the processing it was necessary to assign the replies for people who had given apparently inconsistent answers to question N or who had not answered the question at all.

People coded 'not stated' only appear in Table 1 and are excluded from the remainder of the tabulations, and 'migrants (area not stated)' are included in some tables as having duration of residence under one year but are generally excluded.


Persons who stated they were not migrants but failed to give a duration of residence (some four per cent of the sample population) are again generally excluded from the tabulations.

All ten per cent sample figures are subject to sampling error and in the course of comparison of empirical sampling error with conventional sampling error using a small sub-sample of 1,923 households, the effect of clustering was apparent. Clustering is a situation where people of a particular type or with particular characteristics are grouped in households instead of being distributed randomly over the whole of the population. The clearest examples of this were the few figures available relating to migration, which all showed high ratios. This is because migration, or lack of it, tends to be a household characteristic. A lot of migrants were members of migrant families or households. Thus 4.9 million people changed their address in the year before census; 1.1 million complete families moved, which means that at least 2.2 million people moved in families, ie in groups.

The same point is illustrated by Table 14 in the General Report . This table shows, for the households in the sub-sample used for the investigation of empirical sampling errors, the number of people in the households and the numbers of migrants or other people with a specified duration of residence. Thus, of the 81 households containing migrants within the same area 60 were households where the whole household had moved. The same is true for 74 of the 93 households containing migrants between different areas. The feature is less well marked for households where one or more persons had lived there for 5 years or more, but there is still a fairly close association for the smaller households. Such clustering of migrants naturally increases the variance of estimates of migrants which are based on a sample of households, a household usually being equivalent to a family. A similar feature may well affect the figures for socio-economic groups; members of a household are likely to have similar socio-economic groups to each other.


No specific information is available for migrants but users of the tables are reminded that all the data is from the ten per cent sample and subject to bias particularly when another axis eg occupation is involved.

Post-enumeration survey

The post-enumeration survey confirmed the number of migrants and there was no evidence that the number of non-migrants was seriously in error, but the distribution by duration of residence was slightly suspect.


In addition to the main volume and leaflets for England and Wales, tables appear in the Great Britain Summary Tables. And for Scotland, in volume VIII Internal Migration . Most of the Scottish tables are identical with those for England and Wales although the level of thresholding was smaller.


There were several changes in the procedures for the Sample Census of 1966. Possibly the most important of these was the introduction of a new question on five year migration. Details of the usual address five years previously (23 April 1961) were sought on the same lines as the corresponding infor┐mation for one year migration had been obtained in 1961, and was in fact, repeated in 1966. The form of words used in 1966 was not identical with the 1961 questions but the general intention was quite obviously the same. The siting of the one year migration question (with the five years enquiry immediately following) was considered improved in that it now came directly after the basic instruction on usual residence ('if the person usually lives here, write 'Here', etc). The enquiry on duration of residence was not repeated.

Changes in definition

A change of some importance is to the definition of population included in the published tables. In 1961 the rule was quite inflexible in that every┐one who stated a former usual residence outside England and Wales was excluded, which seems perhaps somewhat restrictive. However, in 1966 this was broadened to the extent that migrants moving to and from Scotland are also included and similarly migrants to and from England and Wales are shown in the Scottish tables. Also two tables relate to migrants whose usual address was previously outside Great Britain.

There is also a major change in the persons included in households and families. Because the 1966 schedule contained questions about absent persons usually resident in the household it was possible to retain the household structure and base the tables on enumerated persons present and resident in the household, plus absent persons; whereas the tables relating to people are based on enumerated persons present and usually resident plus persons enumerated elsewhere but usually resident in the specified area.

In the tables the concept of a 'wholly moving family' remains unchanged from 1961 in that all members except children under 1 (or under 5) are migrants and have the same area of previous residence. But the concept of a 'wholly moving household' was dropped and the tables contain information on all members of the household providing one at least is a migrant. Immigrants from abroad are also included in the tables relating to households and families.

Finally, there is a slight difference between the concept of 'economically active' in the Migration Tables and that used in the Economic Activity Tables . Those migrants with a job in the week before census but with a place of work outside England and Wales are counted as 'economically active residents' in the Migration Tables but not in the Economic Activity Tables . They appear classified by their stated occupation but in the industry category 'Place of Work outside Great Britain'.

Publications - England and Wales

In addition to the main Migration Tables , volumes were published at regional level which contain the relevant parts of some of the national tables and a table showing for each local authority area persons usually resident in the area by area of former residence. The only other reference to migration is in the United Kingdom General and Parliamentary Constituency Tables .

Unpublished data

For some tables which were restricted in the published reports data is available in unpublished form.

Publications - Scotland

The tables are broadly the same although in some cases there is less information and in others much more and there was no thresholding on certain tables which for England and Wales were limited either by size of population or number of migrants.

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