 |
Interaction data |
What are interaction data?
Interaction data involve flows of individuals in the UK between origins and destinations.
These flows are either the residential migrations of individuals from one place of usual
residence to another or of commuters making journeys from home to workplace.
Census interaction data are derived from the questions on the census form relating to place
of usual residence one year ago, and the place of work for the respondent’s main job. These
interaction data are currently available at a range of different spatial scales for the 1981,
1991 and 2001 censuses.
The nature of interaction data mean that datasets are often much larger and more complex than
other census data. For example, a cross-tabulation that showed the counts of migration flows
from all of the 8850 Census Area Statistics (CAS) wards in England and Wales to all of the CAS
wards in England and Wales, would create a table with 78,322,500 cells. A similar table for Output
Areas would have 30,777,088,356 cells!
What can interaction data tell us?
Interaction data are of unique significance as they tell us not only where individuals are
moving and commuting from and to, but also the characteristics of these people.
As an example, data on migration can tell us how the numbers and characteristics of certain
populations are changing in different locations – information not just of academic interest,
but with practical and administrative implications. To take one area, census interaction data
from 2001 reveals that in London, whilst overall the city is losing people to other regions
of the UK, it is gaining far more young adults than it is losing.
As well as area specific patterns, migration data from the census is also able to tell us,
for example, the migration propensities of different ethnic, age, gender, socio-economic
and family status groups in the UK. Similarly, interaction data on commuters can reveal not
just the characteristics of different commuters, but also the locations that these commuters
travel to and from, the distances involved, as well as their preferred method of transport.
Furthermore, interaction data can tell us how these migration and commuting patterns have
changed over time.
Migration to Cardiff (1991 SMS Set 2, Table 1)
1991 Census Special Migration Statistics, Crown Copyright
It is possible to select data for a range of different origin and destination geographies
(which do not need to match) for any areas in the UK.
For example, it is possible to obtain data that provides counts of all males aged
18-24 who moved into the University ward in Leeds, from other wards in the Leeds
district, from other districts in West Yorkshire and from other regions in the UK, in 2001.
Obtaining the data
The
Centre for Interaction Data Estimation and Research (CIDER) facilitates
access to the interaction data for members of UK Higher Education institutions.
Access is provided through
WICID – the Web Interface to Census Interaction Data.
Access is free to all registered users, with WICID enabling users
to download customised data in a number of different formats. For users
who are unfamiliar with the WICID interface, the CIDER web site provides
links to tutorials and training materials. Access to a library of common
queries is also available, which enables users to download data quickly
without the need to go through the query building process.
Links to previous publications and presentations produced by CIDER are
also available on the web site.
What data are available?
Currently CIDER only provides access to census data from 1981, 1991 and 2001.
There are also plans to provide access to other non-census interaction datasets
which will complement the present data holdings.
The sets currently held include:
Census migration datasets:
2001 Census: Special Migration Statistics (Levels 1, 2 & 3 + Level 2 Scotland Postal Sectors).
Migration data at District, Ward, Scottish Postal Sector and Output Area level
1991 Census: Special Migration Statistics (Sets 1 [including re-estimation for 2001 boundaries]
& 2, MIGPOP, SMSGAPS and SAS LBS Table 100 A & B). Migration data at Ward and District level;
Ward level re-estimations for 2001 geographies
1981 Census: Special Migration Statistics (Set C and 1981 county/regional Geography).
Ward level migration data re-estimated for 1991 and 2001 boundaries. County/ region
level data for 1981
Census commuting datasets:
2001 Census: Special Workplace Statistics (Levels 1, 2 & 3). Commuting data at District,
Ward and Output Area scales
2001 Census: Special Travel Statistics (Level 1, 2 & 3 + Level 2 Scotland Postal Sectors).
Commuting data for Scotland at Council Area, Ward, Postal Sector and Output Area Scales
1991 Census: Special Workplace Statistics (Sets A, B and C [including modified set Cs]).
Commuting data by residence (A) Workplace (B) and within & between wards (C) for 1991 and 2001 boundaries
1981 Census: Special Workplace Statistics (Set C and 1981 county/ regional Geography).
Ward-level migration data re-estimated for 1991 and 2001 boundaries. County/region level data for 1981
Variables that can be selected from these datasets may include:
Age
Sex
Family status of migrant
Ethnic group
Whether suffering limiting long-term illness
Whether in household
Economic activity
Moving groups
Moving groups by tenure
Moving groups by NS-SEC of group reference person
Migrants in Scotland/Wales/Northern Ireland with some knowledge of Gaelic/Welsh/Irish
Living arrangements
Method of travel to work
Further Reading
Champion, T. (2005) ‘Population movement within the UK’ in R. Chappell (ed.)
Focus On People and Migration, Basingstoke: Palgrave Macmillan, pp. 92-114.
Champion, T. and Coombes, M. (2007) ‘Using the 2001 census to study human
capital movements affecting Britain's larger cities: insights and issues’
in Journal of the Royal Statistical Society series A (statistics in society) 170(2): 1-20.
Champion, T., Coombes, M., Raybould, S. and Wymer, C. (2006) Migration and socio-economic
change: a 2001 census analysis of Britain's larger cities, draft report to the Joseph
Rowntree Foundation.
Duke-Williams, O. (1998) ‘Interfaces to interaction data’ in P.H. Rees (ed.)
The 2001 census: what do we really, really want?, Working Paper 98/7, School
of Geography, University of Leeds, Leeds.
Duke-Williams, O. (2000) Designing zoning systems for flow data, Innovations in
GIS 7, pp. 115-134, London: Taylor and Francis.
Rees, P. and Duke-Williams, O. (1995) The story of the British Special Migration
Statistics, Scottish Geographical Magazine, 111 (1), pp13-26.
Rees, P. and Duke-Williams, O. (1997) Methods for estimating missing data on
migrants in the 1991 British Census, International Journal of Population Geography 3, pp323-368.
Rees, P., Thomas, F., and Duke-Williams, O. (2002) ‘Migration data from the
Census’ in P. Rees, D. Martin, and P. Williamson (eds.) The Census Data System, London:
Stillwell, J., Duke-Williams, O. and Rees, P. (1995) Time-series migration in
Britain - the context for 1991 Census analysis, Regional Science 74 (4), pp. 341-359.
Stillwell, J., Rees, P., and Duke-Williams, O. (1996) ‘Migration between NUTS
Level 2 Regions I the United Kingdom’ in P. Rees, et al (eds), Population
Migration in the European Union, London: Wiley.
Stillwell, J. and Duke-Williams, O. (2003) A new web-based interface to
British census of population origin-destination statistics, Environment and
Planning A, 35 (1), pp113-132.
Stillwell, J. and Duke-Williams, O. (2007) Understanding the 2001 UK census
migration and commuting data: the effect of small cell adjustment and problems
of comparison with 1991, Journal of the Royal Statistical Society Series
A (statistics in society) 170(2).