2011 Census: Updated view on
content for England & Wales
Office for National Statistics
March 2007
• Share current thinking on likely content of
the 2011 Census questionnaire
• Provide update on development of ethnicity,
identity, language & religion questions
• Provide the opportunity to comment
• NB: Presentation covers Census in England
& Wales
10.00 Introduction
10.10 Background – work done so far
10.35 Population issues
11.00 Likely content – household questions
11.20 Likely content – individual questions
11.50 Panel discussion
12.20 Way forward
• 12.30 Lunch
• 13.30 Ethnicity, identity, language and religion
Census provides statistics on:
• Population units:
– people and housing &
– key demographics (age, sex, marital status, ethnicity)
• Population structures:
– households, families
• More detailed characteristics :
– eg religion, labour market status, industry, qualifications,
health/disability; etc
• Key requirement to ‘Get the Count Right’
– 2011 Census aim : “maximise overall response rates
and minimise non-response in specific areas and
among particular population subgroups”
Key lessons from 2001 Census
• Need to gather information on more than just
usual residents
– Include visitors
• Need to know response rates quickly
– Better management control of the field operation
• Focus more effort in hardest to count areas
• Need to understand coverage
– Comparisons census vs admin data
• Addresses & occupancy status
Elizabeth McLaren
• Work done so far
- 2005 consultation
- evaluation of submissions
- topic groups
- question testing
- other consultation
• Current working assumptions
• Questionnaire design constraints
• Timetable
Work done so far – 2005 consultation
• Over 2000 responses from 500 users
• Prioritised topics based on:
– Strength of user need
– Do alternative sources exist?
– Could we develop a workable question?
– Would public accept question?
– Continuity from previous census
• Allocation of topics to one of three categories
Evaluation of topics
Category 1 Topics likely to
be included
Question testing to
refine questions
Category 2 Topics under
Investigation of
alternative sources
Question design &
Category 3 Topics not to be Topics will not be
considered further
Summary of user requirements
• Most 2001 topics
• New topics including:
- income
- language
- second residences
- national identity
• Difficult trade-offs to be made
Work done so far – topic groups
• Eight topic groups
- Population definitions & bases
- Ethnicity, identity, language & religion
- Labour Market, NS-SEC & qualifications
- Migration
- Income
- Health & care
- Housing
- Demographics
• Membership
- Census
- ONS business areas (DfES for qualifications)
- Methodology
- Harmonisation
- Wales, Northern Ireland & Scotland
Work done so far
• Topic-specific consultation completed/underway
- targeted
- working groups
• Question development and testing
- new questions developed for Test
• Seeking comment from Census advisory groups
• Consulting other groups eg statistical Heads of
Profession in government departments
Current working assumptions
• 24 pages (20 pages in 2001)
• 6 people per questionnaire (5 people in 2001)
Front page
Household members
Relationship matrix
Accommodation questions
Individual questions
Current working assumptions
• 3 pages of individual questions per person
(as in 2001)
• 4 pages of individual questions only possible
if funding secured and impact on response
rates expected to be minimal
• 2-column layout
• Assumptions on length of questions based on
testing completed so far
Questionnaire design constraints
• Public acceptability of questions
• Respondent burden
• Quality of information that can be collected
• Costs eg printing
• Time eg processing
• Scanning limitations
Postal test
March 2007
Census Test
May 2007
Decision on length of Census questionnaire
Summer 2007
Evaluation of Census Test
Winter 2007
Topics and questions for Census agreed
Spring 2008
Questionnaires for 2009 Rehearsal agreed
Summer 2008
Census Rehearsal
Spring 2009
Questionnaires for 2011 Census agreed
September 2009
Census questionnaires approved by Parliament
Spring 2010
Census day
Spring 2011
• Any questions or comments?
Population Issues
Beth Moon
• What are some of the main issues with getting the
population count right?
• Who are some of the difficult to count groups?
– Resident visitors
– People with multiple residences
– Short-term migrants (including migrant workers)
• What information needs to be collected to include
and identify these groups?
Population issues
• Key aim of the Census is to provide a robust
estimate of the population
• Society is becoming increasingly complex
• More information may be needed to identify all
sub-groups of the population
• In 2001, there was a difference between the
Census count and rolled forward population
2011 Population Base
• Agreed population base for 2011 is usual residents
plus visitors
• Usual Resident – someone who spends the
majority of their time residing at that address
• Visitor - a person staying overnight at an address
on Census night at which they are not usually
• Three types of visitor
a) Usual address elsewhere
in the UK
b) Usual address abroad
c) No usual address
• Planning to collect basic
demographic information
– To help with coverage
Resident visitors
• Usual resident in the UK
• No usual address in the UK
– eg ‘sofa surfers’
– eg sold a house and not bought another
• Requirement to collect full
Census return
– must not be recorded as a visitor
• May need to be separately
Second residences
• An address where someone stays for part of the week or
year that is not their usual residence
• More people have more than one residence
– eg residence used while working away from home
– eg holiday or weekend home
• Potential to be:
– Counted at both (or more than two)
– Missed at both
• Ideally want to know
– Second address - so that we can match
– Why they spend time there or length of time at each
Short-term migrants
• Many short-term migrants in the UK who stay or
intend to stay for short periods ie 3 – 12 months
- Using services and may be working
• Increasing requirement to collect information from
these people
– Census uses 6 month UK residence definition
– Population estimates use 12 month UK residence
Short-term migrants
• To identify short-term migrants and understand
differences, ideally want to know
– Year/month of arrival in UK
– Anticipated total duration of stay
• Recognised difficulties asking about intentions on a
self-completion questionnaire
• Questions are being considered for the Census but
further research is required
Short-term migrant workers
• Requirement to collect additional information about
people from overseas who are living and working in
– group is large and growing importance in contribution to
UK economy
• Assist in reducing the undercount of jobs in Census
and Labour Force Survey (LFS) compared to ONS
Business surveys.
• Ideally, this would include all people working
regardless of length of stay
• Difficult to count groups need to be included and
identified in the 2011 Census
• Collecting extra information in 2011 may help to
minimise and explain differences between Census
count and rolled forward population estimates
• Identifying these groups may add complexity to the
Census questionnaire
– May confuse the majority of the population
– Will reduce the number of questions we can ask on
population characteristics
• Work is underway to explore these issues and
understand what is possible in the Census
• Any questions or comments?
Proposed questionnaire content
- Household questions
Jenni Merritt
• General outline of household questions/formation
of ONS Housing Topic Group
• Household and family Relationships
– Relationship Matrix
• Household questions
– Category 1 – likely to be included in 2011 Census
– Category 2 – under consideration
– Category 3 – will not be included
Household Questions
• Working assumption of one page
• Less pressure for space than individual questions
• ONS Housing Topic Group
• User requirements for housing questions
Household & family relationships –
Relationship matrix
Household & family relationships –
Likely to be included
• Relationship Matrix
• UNECE’s recommended
method of data collection
• Additional Categories for 2011
– Step-brother or step-sister
– Same-sex civil partner
• Further research to be
Housing questions - Likely to be included
• Accommodation type
– Inclusion of ‘linked by garage’
• Household tenure
• Type of landlord
• Self-contained accommodation
– Assesses counts of dwelling and
• Number of rooms/bedrooms
Housing questions – Under consideration
• Central heating
– Measure of housing quality and
– 8.5% ‘no’ responses in 2001
• State of repairs
– Additional measure of housing
• Number of vehicles
– Informs transport planning
Household question – Under consideration
• Telephone number
– Enable follow up of
partial responses
– To be tested in 2007
– Inclusion dependent
upon outcome of test
How are we taking this forward?
• Further consultation with users on their
• Outcome and analysis of 2007 Test
– State of repairs question
• Review of alternative sources available
– English House Condition Survey
– General Household Survey
– Integrated Household Survey (2008)
Housing questions – Not included
Topics from 2001
• Access to bath/shower and toilet
– Measure of deprivation?
– 2001 findings – 0.5% no access
• Lowest floor level
– Asked first time in 2001, no plans to include for 2011
• Accommodation on more than one floor
– Asked by NISRA in 2001, no plans to include for 2011
Housing questions – Not included
New Topics
Internet access
– Rapidly changing
– Decrease in utility of outputs
Garden access
Pet ownership
Smoke alarms
Renewable energy sources
– User need not high enough to
justify inclusion
• Any questions or comments?
Proposed questionnaire content
- Individual questions
Peter Stokes
• Factors considered
• Topics likely to be included
– Requirement for information
• Topics under consideration
– Competing requirements
– Additional work planned
• Topics unlikely to be included
– Reasons for exclusion
Factors considered
• Support main objectives of Census
– Maximise overall response rates and minimise
differences across areas / population groups
– Understand and explain differences between Census
and Mid-Year Population estimates
• Meet strongest user requirements
– Government resource allocation / Policy development
• Continuity with existing Census information
Individual questions - Likely to be included
• Basic demographics
– Name
– Sex
– Date of Birth
– Marital/Civil Partnership status
• Other basic information
– Student status
• Term-time address
Individual questions - Likely to be included
• Migration
– Country of birth
• Asked since 1841
• Stock of migrants
– Address one year ago
• Asked since 1961
• Measures internal & international migration
Individual questions - Likely to be included
• Ethnicity, Identity, Language & Religion
– Ethnic group
• Asked since 1991
• Most requested topic
– National identity
• New topic
• Complement ethnic group question
– Welsh language proficiency
• Asked since 1891 – in Wales only
• Policy development and resource allocation
– Religion
• Asked since 2001 – voluntary question
• Resource allocation
Individual questions - Likely to be included
• Health and care
– Health status
• Asked since 2001
• Strong predictor of use of health services
– Long-term illness / disability
• Asked since 1991
• Policy development and resource allocation
Individual questions - Likely to be included
– Occupation
• Asked since 1831
• Required for NS-SEC and widely used separately
– Self-employed or employee
• Asked since 1851
• Required for NS-SEC and used separately
– Supervisor status
• Asked since 1851
• Required for NS-SEC only
– Ever worked
• Asked since 2001
• Routes ‘never worked’ away from NS-SEC
Individual questions - Likely to be included
• Other labour market
– Economic activity – 5 questions
• Asked since 1851
• Policy development and resource allocation
– Address of workplace
• Asked since 1901
• Transport planning
– Travel to work
• Asked since 1971
• Transport planning
Individual questions - Under consideration
• Topics from 2001
– Qualifications
• Asked since 1961
• Policy development and resource allocation
– Industry
• Asked since 1911
• Improves quality of NS-SEC and used separately
– Name of employer
• Asked since 1911
• Improves accuracy of Industry coding
– Carers
• Asked in 2001
• Policy development to support unpaid carers
Individual questions - Under consideration
• New topics
– Second residences
• Help ‘get the count right’
• Improve understanding of difference between
Census and mid year estimates
– Month and Year of arrival
• Improve migration statistics
– Citizenship
• Improve migration statistics
– Income
• Identify areas of deprivation
– Language
• Policy development and service provision
Individual questions - Under consideration
• Possible options
– Prioritise understanding population
• Second residences
• Month and year of arrival
• Citizenship
– Prioritise continuity
• Industry and Carers; or
• Qualifications; and
• One migration question
– Prioritise identifying areas of deprivation
• Income
• Language
Individual questions - Under consideration
• Consultation
– Understand user priorities between topics
• Evaluation of 2007 Test
– Effect of income question on response
– Success of other new questions
• Question testing
– Cognitive and postal tests
• Establish length of questions required
• Acceptability of questions
• Quality of information
Individual questions - Not included
• Topics from 2001
– Hours worked
• Asked in 2001
• Insufficient user requirement
– Number of employees at workplace
• Asked in 2001
• Used to derive NS-SEC using full method
• New topics
– Large number of new topics requested
• Sexual orientation
– Concerns about acceptability & quality of data
• Country of previous usual residence
– Other topics more useful for migration information
• Any questions or comments?
Way forward
• Topic-specific consultation to be completed
• Evaluation of 2007 Test questionnaire
• Cognitive question testing and focus groups
• Postal tests
• Investigation into alternative sources to be completed
• Funding for Census agreed in summer 2007
• Update to Census advisory groups in autumn 2007
• Decision on topics for Census by early 2008 to feed into
White Paper and Rehearsal
• Parliament approves Census questionnaires in early 2010
Contact details
Census Questionnaire Design and Content
Office for National Statistics
Segensworth Rd
PO15 5RR
Development of ethnicity, national
identity, religion and language
Frances Forsyth and Karin Bosveld,
March 2007
• Question development process for ethnicity,
national identity, religion and language
• Main issues for development
– Ethnicity
– National Identity
– Religion
– Language
Question development process
• Consultation with users
– What information is needed and why
• Question testing research
– What works in practice
Consultation with users
• Establish what information is needed and why
– 2005 consultation on Census topics
– 2006/07 consultation on ethnicity, identity, language and
– Aiming to reach a wide range of users
Question testing research
• Establish what questions work in practice
– Surveys, interviews and focus groups with members of
the public
– Main phase of testing: 2007
Objectives of Census questions
• Census questions should:
– Be clear and acceptable to respondents
– Provide information that meets the key needs of users
– Do so within the available space and at a reasonable
• What is measured and why
• 2007 Census Test question
• Issues and way forward
Why ethnicity information is collected
• Race Relations Act makes it unlawful to
discriminate on grounds of:
– Race
– Ethnic or national origins
• Public authorities have duties to:
– Monitor equality
– Give due regard to eliminating unlawful racial
discrimination and promoting equality of opportunity and
good relations between different racial groups
Definition of an ethnic group
An ethnic group has a distinct identity, based on
recognising a long shared history and having
distinct cultural traditions, which may be related to
one or more of the following characteristics:
Geographical origin
Country of birth
Cultural traditions
(House of Lords, 1983)
Race or ethnicity?
• In the UK people sometimes use the terms
‘ethnicity’ and ‘race’ interchangeably but they are
– Race tends to relate to physical characteristics, for
example including skin colour
– Ethnicity relates to social or cultural identification
• Both are covered by equalities legislation
2007 ethnic group question (Wales)
2007 ethnic group question (England)
Issues for 2011: ethnic group
• Maintain consistency with previous information
where possible
– Many users need to track changes over time for the
groups named in the 2001 question
• Update the classification where:
– Strong justification for change
– Lack of other suitable source of information
User requirements for 2011: ethnic group
• Main types of requirement
– Information on additional groups
– Changes to the terminology of the question
User requirements for 2011: ethnic group
• Information on additional groups
– Arab
– Gypsy, Romany, Irish Traveller
– UK identities e.g. Welsh, English, Cornish
– Sikh
– Jewish
User requirements for 2011: ethnic group
• Information on additional groups
– Kashmiri
– African: different nationalities or regions
– Black British
– Eastern European identities
– Mixed: Chinese and White
Way forward on additional categories
• Establish the amount of space available
• Prioritise new categories based on:
– Strong need for information
– No alternative sources of information
– If ‘Other, write in’ is not sufficient
• Test effect of new categories on overall question
Issues for 2011: ethnic group
• Acceptability of terminology
Colour terms Black and White
Mix of colour and geographical labels
Gypsy or Romany
Arab or Middle Eastern
Mixed or Multiple Heritage
Way forward on acceptability
• Question testing
– Understand acceptability among members of the public
– Preferences for terminology and question wording
– Effect of any change on response
• Possible further consultation with users on the
• Working with partners in devolved government
Issues for 2011: multiple response
• Advantages
– Richer data
– Ability to more accurately describe ethnicity
– Around 2% gave multiple responses in 2001
• Disadvantages
– Loss of comparability with 2001
– People may answer inconsistently
– Can be difficult to process and interpret
Way forward: multiple response ethnicity
• Consultation with users
– Multiple response ethnicity appears unpopular with
central and local government users
• Question testing
– Will see if single response ethnicity is adequate for
Q&A: ethnic group
• Comparability v change
• Additional information needed
• Acceptability of terms
• Multiple response
National identity
• What is measured and why
• 2007 Census Test question
• Issues and way forward
Why national identity information is
• Provides information on identification with different
countries of the UK
• Records ethnic group separately from national
• Race Relations Act makes it unlawful to
discriminate on grounds of nationality or national
National identity
• Sense of belonging to a particular country or
• Like ethnicity, based on subjective identification
– may be different from country of birth, legal nationality or
ethnic group
2007 national identity question (Wales)
2007 national identity question (England)
Issues for 2011: National Identity
• Strength of need for question
– Information in its own right
– As improving response to ethnic group question
• How does it work in practice
– Do respondents recognise difference between national
identity and similar concepts?
– Are these identities better measured as ethnic groups or
national identities?
Way forward: national identity
• Consultation with users
– Strength of need for national identity information
• Question testing
– Effect of including/excluding national identity on
Q&A: national identity
• Strength of need for information
• Aspects of national identity required
• What is measured and why
• 2007 Census Test question
• Issues and way forward
Why is religion information collected
• Provides richer information on diversity
• Race Relations Act protects Jews and Sikhs as
ethnic groups
• Equality Act protects from discrimination against
religion or belief
• Records religious identity or affiliation
– Does not measure the number of practising members
of each religion
– Does not aim to collect information on non-religious
belief systems
• Voluntary question, due to Census law
2007 Religion question (Wales)
User requirements for 2011: Religion
• Additional categories
– Pagan
– Humanist
• Measure different denominations
– Christian: Church of England/Protestant, Catholic
– Muslim: Sunni, Shia
User requirements for 2011: Religion
• Measure religious practice
– With our without religious identity/affiliation
• Improve response by making the question
Religion: way forward
• Review evidence from the consultation and
question testing
– Assess strength of need for change
– If need for change is strong, consider whether
amendments possible
– If not, continue with 2001 question
Q&A: religion
• Religious affiliation / identity
• Religious practice / current faith
• Why an expanded language question is being
considered for inclusion
• 2007 Census Test question
• Issues and way forward
Why should language be measured?
• Growing need for information on:
– Level of English language proficiency
– Languages used other than English
• Some languages covered by:
– Welsh Language Act
– European Charter for Regional or Minority Languages
2007 language question (Wales)
Issues for 2011: Language
What additional aspect to measure:
• Languages spoken
• English language proficiency
• Preferred language of communication
– where English is not the first language
• First language or mother tongue
• Main language spoken at home
• Different aspects of fluency: speak, read, write,
Issues for 2011: Language
• Users need many different aspects of language
– What aspects would be most useful
– What Census could provide – quality of information
Language: way forward
• Establish which aspect of language ability is most
• Test questions to see whether Census can provide
language information of sufficient quality
Q&A: language
• Aspect of language needed
• Strength of need for information
• 2006/07 consultation on ethnicity, identity,
language and religion:
– deadline 30 March 2007
– Summary report: Summer 2007
• Question testing and research: main phase 2007
Contact details
Ethnicity, Identity and Inequalities Branch
Office for National Statistics
Room D201, 1 Drummond Gate

Looking for a way - Office for National Statistics