In the 1981 Census residents in wholly absent
households ie no person present in the household on census night,
were excluded from the enumeration at their usual address. In an
attempt to improve coverage of the 1991 Census, all households absent
on Census night were left a form with a letter asking them to
complete the form on their return home. For those absent households
for whom a completed form was not received, certain data are imputed
using some basic information (eg number of persons living at the
address and details of the accommodation) collected by the enumerator
and also drawing on information from similar absent households which
did complete and return census forms. See also 1991 Census
Any person who is usually a member of the household
but on Census night was absent on holiday, at school or college, or
for any other reason, even if they are being included on another
census form elsewhere.
The main source of information on accommodation is
taken from Panel A of the Census form and Question H2. Use is also
made of the information provided by the building bracket linking
household spaces in a converted or shared house, bungalow or flat
(Panel A boxes 7 or 8).
An adult is any person who is not a dependent child.
Each household was enumerated on a 12 page H form
(W form in Wales). The names of persons spending Census night in
communal establishments and on HM Ships and other vessels were listed
on an L form with each person enumerated separately on an I form (Iw
form in Wales). Enumerators had the responsibility for recording
certain housing information (Panel A on the H form) and making an
assessment of the nature of the accommodation if it was not occupied
on Census night. As in 1981, the form filler was instructed to
include on the Census form 'every person who spends Census night in
this household, including anyone staying temporarily'; 'any other
people who are usually members of the household but on Census night
are absent on holiday, at school or college, or for any other reason,
even if they are being included on another census form elsewhere';
'anyone who arrives here on Monday April 22nd who was in Great
Britain on the Sunday and who has not been included as present on
another census form'; and 'any newly born baby born before April
22nd, even if still in hospital'.
April 21-22nd 1991.
The term covers all establishments in which some
form of communal catering is provided. Such establishments were
enumerated using the L form on which were listed the names of all
persons present on Census night together with individual (I) forms
containing the relevant census questions.
In 1981 the population in communal
establishments(then known as non-private households) included campers
with communal catering and persons sleeping rough.
In 1991 all those present in the establishment on
Census night and all who arrived on April 22nd who were not included
as present on a census form elsewhere, were listed. The names of any
non-resident staff, on duty on the premises on Census night were not
required to be listed.
See also 1991 Census Definitions, Chapter 4, p. 12-15.
1. Persons quoting a LS date of birth enumerated in
a Communal Establishment are included in the LS sample.
2. No double enumeration in Communal Establishments
is accepted by census processing. In such situations, eg resident in
a Communal Establishment but in hospital on Census night, the form
processed would be that completed where the person was on Census night.
All persons enumerated in a household on Census night.
All persons, present or absent on Census night,
usually resident in a household.
Persons aged 0-15 in a household; or persons aged
16-18, never married, in full time education and economically
inactive. The additional qualification 'and economically inactive'
has been added to the 1981 Census definition, which also included
persons aged 19-24 who were also never married and classified as a
student from the question on economic activity.
Defined as 'structurally separate accommodation.'
See also 1991 Census Definitions, Chapter 5, p.16-26.
The general topic of economic activity covers a
wide range of census characteristics including both those obtained
directly from questions on the Census form, such as economic
position, employment status and occupation, and others which are
derived by combining answers to these questions, for example, social
class and socio-economic group, which are derived from occupation and
All persons aged 16 or over at the time of the
Census were asked Question 13 which provides information on economic
position and employment status. Subsequent questions were addressed
only to those persons who were either in paid employment in the week
before Census day, or who had had a paid job within the previous ten
years. In cases of multiple ticking at Question 13 up to three codes
are entered on the computer file. The three lowest numbered boxes
ticked are given priority except for boxes 5 (On a Government
Employment or Training Scheme) and 8 (At school or in other full time
education) which take precedence over all other boxes.
When interpreting the full 12 fold 'Economic
position/Employment status' classification, Economically inactive
students (Category 9) are those allocated a primary code of 8, and no
other boxes 1-4 were ticked. Persons who ticked any box 1-4 in
addition to box 8 are separately identified as 'Students,
economically active' (Category 8), as well as being included in the
appropriate economically active category. Those describing themselves
as being on a government scheme are coded to box 5 which takes
priority over all other boxes. 'Other inactive comprises persons
looking after the home or family, or for whom the last box was ticked
and are not reallocated to another category, including persons of
Question 13 on the 1991 Census form attempts to
obtain the same information that was asked for in two separate
questions in 1981. Question 10 on the 1981 form had just two 'persons
in employment' boxes, which distinguished only between persons with a
full time and a part time job. The information on employment status
(whether a person was working for an employer or was self employed
with/without employees) was obtained from a separate question,
The introduction of economically active students in
categories 1-4 is also a change from 1981, introduced to follow
International Labour Organisation definitions adopted in the early 1980's.
The 1991 question allows for persons on government
employment and training schemes to be identified. The 1981 question
on employment status included a box for 'apprentice or articled
trainee'. See also 'Hours'.
Qualifications obtained over the age of 18 (see
also Qualified Manpower) are grouped into three Educational Levels.
Level a - higher degrees of UK standard
Level b - first degrees and all other
qualifications of UK first degree standard; and
Level c - qualifications that are (i) generally
obtained at 18 and over, (ii) above GCE 'A' level standard; and (iii)
below UK first degree standard.
A Qualified Person is one who holds at least one
qualification at level a, b or c. Persons holding more than one
qualification are generally analyzed by the highest qualification; if
two or more qualifications of the same (highest) level are held, the
one most recently obtained is used. See Also 1991 Census Appendix 10.
Question 13 used to provide detail on employment
status since there was no separate question. Categories for the self
employed were included in the Question 13. Information on managers/supervisors
is obtained from the occupation question. Persons on a government
employment or training scheme are classified as economically active -
Answers to this question were assigned to one of 35
codes. The full classification incorporates each of the 7 pre-coded
categories from the question, plus another 28 derived from any multi-ticking
of boxes and the written descriptions given in the 'Black-Other' and
'Other ethnic group' boxes. Written descriptions which were the same,
or had the same meaning, as one of the pre-coded categories were
assigned to the relevant pre-coded groups.
Answers to the questions on age, sex, marital
status, long term illness and relationship in household (Question 5)
were used to classify households and groups of individuals into
families. The 'relationship to the first person entered on the Census
form' question differs from that followed in 1981 only in that
cohabitant replaces the former de facto spouse (derived from write in
answers), and that additional codes are introduced for child of
cohabitant and cohabitant of son/daughter in an attempt to identify
'hidden' families within households. The Other unrelated, category,
now includes domestic servants who were separately coded in 1981. See
also 1991 Census definitions p. 37.
Family Unit Type
A computer algorithm was used to allocate
individuals within households to a detailed family type
classification. The algorithm also defines the number of family units
within a household, the relationship of each unit to the head of the
household,and the generation within the family unit to which the
individual belongs. in any family unit within a household where there
are two generations, the younger generation must be single (never
married) and have no obvious partner or offspring. When a person in a
younger generation, has or can be shown to have had, a relationship
to a person other than their parent/s, that person is not placed in
the same family unit as his or her parent/s.
NOTE: The following terms are used in relation to
Head of Household: See under 'H'.
Dependent Child: Is a person in the second
generation of a family. There is no age limit to the term 'child'.
Non-dependent child: is any person in the second
generation of a family who is not a dependent child.
Head of family: is generally taken to be the head
of household if the family contains the head of household, otherwise,
- in a couple family, the head of family is the
first member of the couple on the form;
- in a lone parent family, the head of family is
the lone parent; or
- a no family person ie non-dependent person can be
treated as head of family.
Family size: is the number of residents in a family
as defined above. No person can belong to more than one family.
Head of Household
It should be noted that in 1991 the H form was
addressed 'to the Head or Joint Head or members of the Household aged
16 or over' (thus allowing for households with no head). Furthermore,
the question on relationship in household asked for the relationship
of the second and subsequent persons to the person entered in the
first column on the form. For statistical purposes the 100% processed
Census tables, the head of household is usually taken to be the
person entered in the first column. See also 'Household Composition'.
Although a question on hours worked was included in
the 1971 Census, this item was dropped from the 1981 Census form and
replaced with additional full-time/part time job boxes. A question on
hours usually worked in a persons main job was re-introduced in the
1991 Census, as well as retaining the full-time/part-time categories
in the economic activity question. the maximum number of hours worked
to be recorded is 99, hours worked in excess of 99 are coded as 99.
Fractions of an hour are rounded to the nearest even number.
Full time work: 31 hours or more a week.
Part time work: 30 hours or less a week.
Uses the answers to the questions on age, sex,
marital status and long term illness for each person in the
household. A household is either one person living alone or a group
of people, who may or may not be related, living or staying
temporarily t at the same address, with common housekeeping. As in
1981, enumerators were instructed to treat a group of people as a
household if there was any regular arrangement to share at least one
meal, including breakfast, a day or if the occupants shared a common
living or sitting room. The occupants of one room accommodation or a
caravan are treated as a single household.
The head of household is regarded as the person
entered in the first column of the form, provided that person was
aged 16 years or over and usually resident at the address of
enumeration. If one of these conditions was not met, the first person
aged 16 or over to be entered on the form and recorded as usually
resident at the address of enumeration was coded as the head. In the
last resort the oldest resident aged under 16 was taken as the head.
No head was identified in households consisting entirely of visitors
(same as 1981).
Household Dependent Type
An additional household dependent type
classification has been introduced for the 1991 Census, and is
defined in terms of dependents and non-dependents in the household.
in this classification a dependent is either a dependent child or a
person who both has a limiting long term illness and whose economic
position is either 'permanently sick' or 'retired'.
A non-dependent is any person who is not a dependent.
The accommodation available for a household. Based
on information obtained from Panel A, Question H2 and H4.
Unshared converted accommodation is a household
space with its room(s) including a bath or shower, WC and kitchen
facilities behind its own private door. Unshared not self contained
accommodation is identified by Panel A box 7, Question H2 box 2
(bedsit) or box 4. This latter group is further sub-divided into two
household space types - those with exclusive use of bath/shower and
inside WC (flat) and those without exclusive use of bath/shower and
inside WC (rooms).
Shared dwellings comprise two or more household
spaces sharing an entrance into the building that are not self contained.
Unattached spaces are household spaces also sharing
an entrance into the building but where they are the only one not
Where not self contained accommodation is
unoccupied on Census night and no information on amenities is
available to the enumerator/ the accommodation is classified as not
Because of a basic change in the definition of
self-contained accommodation in 1991, it is not possible to compare
accurately 1991 and 1981 statistics on self contained and not
self-contained accommodation. In 1981 self-contained accommodation
required a household sharing an entrance from outside the building
with one or more other household(s) to have its own rooms, not
including a bathroom or WC, enclosed behind its own front door inside
the building. For 1991 the definition was changed to require the
rooms and kitchen facilities, bath or shower, and inside WC, to be
contained behind the 'private' door.
In 1981, purpose built flats (categories 4 and 5 of
the 1991 household space type classification) were counted as a
single group 'in a purpose built block of flats or maisonettes'. The
separate 1991 categories of detached, semi detached and terraced were
only identified in Scotland in 1981. In England and Wales in 1981
these were included with accommodation in converted or shared houses,
bungalows or flats as a single group 'household spaces in permanent
buildings with separate entrance from outside the building.
The industry in which a person is engaged is
determined by the business or activity in which his or her occupation
is followed. The industrial classification has regard only to the
nature of the service or product to which the labour contributes.
Industry codes are assigned, as far as possible, by
reference to lists from the Dept. of Employment which give names and
addresses of employers by industry code. Where the employer's name is
not listed a code is allocated based on information given in the
description of the employer's (or self-employed person's) business.
There has been no subsequent revision to the 1980
Standard Industrial Classification which comprises the full range of
industries grouped into 10 divisions each denoted by a single digit.
The Divisions are divided into 60 Classes, each denoted by the
addition of a second digit, and divided further into 222 Groups and
334 Activities by the addition of third and fourth digits. Census
industry coding is based on the Activity heading level with some
exceptions where it is not possible to distinguish separate
activities/subdivisions of activities. See also 1991 Census
definitions p.30-40, and 1991 Census Appendix 7.
Lone Parent Family
A father or mother together with his or her never
Single persons are those who have never married -
bachelors and spinsters.
Married, first marriage, persons are those whose
first marriage has not ended by divorce or death of a spouse.
Re-married persons are those who have married again
after their first or subsequent marriage(s) ended in divorce or death
of a spouse and who were still married at the time of the Census.
Widowed and divorced persons are those whose most
recent marriage ended respectively, through the death of a spouse or
divorce. Persons who were separated but not divorced from their
spouse are classified as either married, first marriage, or re-married.
The question in England and Wales is unchanged from
that asked in 1981.
The identification of a migrant is based on answers
to the questions on usual address, and address one year before the
Census. The question is unchanged from the 1981 Census.
A migrant within one year preceding the Census is a
person with a different usual address one year ago to that at the
time of the Census.
A migrant household is a household whose head is a migrant.
A wholly moving household is a household all of
whose resident members aged one year and over were migrants with the
same postcode of usual residence one year before the Census. Children
aged under one are included as members of wholly moving households.
Reflects those situations where a LS member is
entered on more than one census form. May be either a simple
present/absent pair or complex multiple enumeration.
The LS member's Record Type 41 contains single
enumerations, 'present' records from simple present/absent pair
multiple enumerations and 'chosen' records from complex multiple
enumerations. The LS member's Record Type 43 (Absent Usual Residents
file) contains 1991 Census multiple enumeration records not included
in Record Type 41. These are 'absent' records from simple
present/absent pair multiple enumerations and 'not chosen' records
from complex multiple enumerations. There may be more than one 'not
chosen' record in a complex multiple enumeration.
The variable HISCEN91 on the LS member's Core
record contains two codes, 2 and 3, which indicate whether the 1991
Census record is a multiple enumeration or not, and its degree of
complexity. Most Record Type 43s are associated with HISCEN91=2;
approximately 3 % of records will have HISCEN91=3. HISCEN91=1 is
never found with Record Type 43.
The occupation of a person defines the kind of work
performed; this generally defines the assignment to an occupation
group. The nature of the factory, business or service in which the
person is employed has no bearing on the classification of the
occupation, except to the extent that such information may clarify
the nature of the duties. Occupation codes are allocated from the
write-in answers at Question 15, which was asked of all persons who
had had a paid job either in the week before the Census or within the
previous ten years. A new Standard Occupational Classification (SOC)
replaced both the 1980 Classification of Occupations and the
Classification of Occupations and Directory of Occupational Titles
(CODOT). The main concept of the new SOC was to classify jobs as
opposed to classifying persons. The classification is based entirely
on information about the type of work done, as indicated by the job
title and description, and unlike its 1980 predecessor takes no
account of ancillary information on employment status, such as
whether someone is self employed, which is not always available from
non census sources.
The format of the classification is hierarchical,
offering different levels of aggregation suitable for various
analytic purposes. Thus SOC comprises:
- 9 Major Groups, subdivided into
- 22 Sub-major groups, subdivided into
- 77 Minor Groups, subdivided into
- 371 Unit Groups created from
- 3,800 CODOT occupational titles.
Most of the work ln developing SOC went into the
definition of the Unit Groups, with which users are mainly concerned.
The aim was to adapt the structure of the 350 Occupational Coding
Groups used in the C080 to meet the SOC criteria.
Over half (56.3%) of the Occupational Coding Groups
match with the SOC Unit Groups on a one to one basis, and a further
4.9% can be exactly reconstructed by aggregating two or more SOC Unit
Groups. Of the others which cannot be exactly matched, a significant
number differ only slightly from the SOC Unit Groups thus providing a
reasonably good overall fit. Much of the remaining discontinuity
affects the residual 'nec' groups in C080. See also 1991 Census
definitions p.37-39 and 1991 Census Appendix 8 and 9.
The minimum age at which a person may receive a
national insurance retirement pension ie. 60 years for a woman, 65
years for a man.
Combinations of the answers given at Question 6
(wherabouts on Census night) and Question 7 (usual address) form the
building bricks for the various population bases as follows:
Visitors from within GB
Elsewhere in GB
Visitors from outside GB
Absent residents absent within GB
Elsewhere in GB
Absent residents absent outside GB
Categories (d) and (e) are not enumerated in
communal establishments. Persons in category (d) should also be in
category (b) on the form completed where they are present on Census night.
Or persons present, is a count of all the persons
recorded as spending Census night in an area regardless of whether
this was where they usually lived. See also De Facto.
Question 19 on the Census form requested, for all
persons aged 18 and over, details of degrees and professional and
vocational qualifications obtained. The question was in the same form
as the one included in the 1981 Census. The qualification code
indicates the type and level of the qualification eg degree, diploma
etc, and the subject code indicates the subject, or combination of
subjects, in which the qualification was obtained. The awarding
institution and the year of the award are not coded, but the
information is used to improve the accuracy of the coding.
Relationship in Household
Used in the analysis of the composition of
households. The aim of the approach is to classify households taking
account of the inter-relationships between household members. Each
members relationship to the head of household is classified. The head
of household or a joint head, or other member of the household is
entered as person one on the Census household form and the
relationship to that person is answered for all other persons on the form.
The question is very similar to the one asked in
1981 but, additionally, attempts to identify by means of a stated
category on the form rather than by the coding of write-in answers.
Step and adopted relationships, when specified as
such on the form, are not distinguished from blood relationships.
The list of relationship codes differs from that
followed in 1981 only in that cohabitant (code 2) replaces the former
de facto spouse (derived from write-in answers) and that additional
codes are introduced for child of cohabitant and cohabitant of son/daughter.
Since the 1911 Census it has been customary to
arrange the large number of groups in the classification of
occupations into a smaller number of summary categories called Social
Classes. In the 1991 Census persons with a paid job are assigned to a
Social Class by reference to their occupation in the week preceding
the Census, or where there was no paid job, on the basis of the most
recent paid job held within the previous ten years.
Members of the armed forces and those with
inadequately described occupations are not allocated a social class;
persons on a government employment or training scheme are similarly
not allocated a social class.
After a consultation exercise in 1989 OPCS decided
to retain the name 'social class' but to expand it to Social Class
based on occupation in order to make its basis more explicit.
The proposal to change the name of Social Class II
from Intermediate Occupations to Managerial and Technical Occupations
was welcomed and has been implemented.
Notwithstanding these minor changes to the
nomenclature of the classification, the occupation groups included in
each of the social class categories, are, as in 1981, selected in
such a way as to bring together as far as possible people with
similar levels of occupational skill. In general, for the 1991
Census, each SOC Unit Group is assigned as a whole to one or other
social class and no account is taken of differences between
individuals in the same group, such as differences of education or
levels of remuneration. However, for persons having the employment
status of 'foreman' or 'manager' the following additional rules apply:
- each occupation is given a basic social class
- persons of 'foreman' status whose basic social
class is IV or V are allocated to Social Class III; and
- persons of 'manager' status are allocated to
Social Class II with certain exceptions.
Classification by socio-economic group (SEG) was
introduced in the 1951 Census and extensively amended in 1961. This
non hierarchical classification aims to bring together people with
jobs of similar social and economic status. The allocation of
occupied persons to an SEG is determined by considering their
employment status and occupation.
It is not practical to obtain from a census the
degree of responsibility exercised by employers and managers. An
indirect distinction between greater and lesser responsibility is
provided by classifying employers and managers by the size of the
establishment in which they work. SEGs 1.1 and 1.2 comprise
respectively, employers and managers in enterprises employing 25 or
more persons, while SEGs 2.1 and 2.2 comprise those in enterprises
employing fewer than 25 persons. Civil servants, local authority
officials and ships' officers are conventionally regarded as working
in large establishments.
Student's Term-Time Address
A student's usual address is taken for annual
estimates purposes as the term-time address, but in the 1981 Census
this was considered to be the student's home address. Statistics
derived from the 1981 Census were therefore not directly comparable
with the Registrar General's annual estimates. The date of the 1991
Census fell in term time for some educational establishments but in
vacation time in others. The Census included a new question on term
time address of students and school children.
The major subject(s) of each qualification that a
person holds is coded using a standard subject classification,
consisting of 10 subject groups and 108 primary subjects.
The tenure of a dwelling is chosen with reference
to the tenure(s) of the constituent household space(s). For a
dwelling of one household space, the tenure of the dwelling will be
the same as that of the household space. For a multi-household space
dwelling the tenure is chosen according to a priority order. The
highest ranked tenure among the household spaces becomes the tenure
for the dwelling.
Tenure of Accommodation Household Spaces
The tenure of the household's accommodation was
obtained from the answer to Question H3.
The division of owner-occupiers into 'buying
through mortgage or loan' and 'owning outright' represents a slight
difference from the categories adopted in England and Wales in 1981
when owner-occupation was classified as either freehold or leasehold,
irrespective of whether the property was being purchased by mortgage
or owned outright.
In England and Wales the renting categories are the
same as 1981 except that the category 'renting from a New Town
Development Corporation or from a Housing Action Trust' is separately
identified on the 1991 Census form to reflect the increase in this
sector of the housing market.
The journey to work question asked about the
longest part, by distance, of the person's normal daily journey to
work. The categories listed are the same as in 1981 except that the
former category 'car pool /sharing driving' is no longer separated
from 'driving a car or van' or 'passenger in a car or van'. Where
more than one box was ticked the lowest number was coded.
Type of Move
A migrant within a defined area is one whose former
usual address and usual address at Census are within the same defined area.
A migrant to a defined area is a migrant whose
usual address at Census was inside the defined area but whose former
usual address was outside that area.
A migrant from overseas to a defined area is a
migrant whose address at Census was inside the defined area but whose
former usual address was outside Great Britain. (This definition
carries no implication with regard to birthplace or ethnic group of
A migrant from a defined area is a migrant whose
former usual address was inside the defined area but whose usual
address at Census was outside the area but within Great Britain.
Migrants resident in a defined area is the sum of
migrants within the area and migrants into the area, including those
migrants resident in the area with origin not stated. Migrants
from/to contiguous areas are defined as the sum of all migrants
from/to areas which have a shared boundary with the defined area.
Indicates where a person usually lives. For most
people the answer to the question on usual address is
straightforward. For some, however, it is not so particularly when a
person lives at more than one address throughout the year.
Question 7 (Usual address) on the H and I forms
included the instruction that, for students or (school)children away
from home during term-time, their home address should be taken as
their usual address.
The governors of prisons were advised that if a
prisoner had served six months or more of a sentence (though not
necessarily at the same establishment) immediately prior to Census
night, the usual address should be taken to be the address of the
prison. If less than six months had been served, the usual address
before admission should be given. A similar procedure was followed
for children in community homes.
Usually Resident Population
(Topped up present/absent: 1991 base) is a count of
all persons recorded as resident in households in an area, even if
they were present elsewhere on Census night, plus residents in
communal establishments who were present in the establishment on
Census night. This population is 'topped up' with persons from
enumerated wholly absent households and imputed wholly absent
households. See also De Jure and 1991 Census definitions p.7-8.
Those persons present at this address on Census
night whose usual address was not this address.
Question 6 on the Census form. Used in conjunction
with Question 7, Usual address, to define population bases. If
'Elsewhere' was ticked and an address elsewhere in Great Britain was
given, then the input is the postcode of that address; if an address
outside GB was given, then the input is the country of usual
residence (using the same codes as the country of birth
classification). Where there was no answer to the question on usual
address, or where the answer was 'none' or 'no fixed address' the
imputation procedure provides a code of either 'this address' or
'elsewhere, not stated'. See also Population Bases.