Contact us

Faculty of Science, Engineering and Computing.
Penrhyn Road
Kingston upon Thames
Surrey KT1 2EE

Tel: +44 (0)20 8417 9000

Centre for Earth and Environmental Science Research

Creating a consistent census-based geography, 1971-2001

Executive Summary
ResearchersProf. N. S. Walford (Principal Investigator): CEESR
Dr Kelly Hayles: Department of Sustainability and Environment, Australia.
Funding Body/SourceESRC
DurationOct 2004-present
Project SummaryThe central aim of this project was to create a consistent geographical and statistical platform linked to recent British Population Censuses that is sufficiently flexible to form the basis for analyses of past and, with the incorporation of new statistics, future population change.

The project applied areal interpolation and dasymetric mapping techniques to produce digital boundary data files and estimates of census-derived counts for consistent geographical units spanning the British censuses from 1971 to 2001. Further research arising from analysing the datasets produced by the project has included collaboration with Dr Slawomir Kurek at the Pedagogical University of Krakow to compare recent demographic trends in Western and Eastern European countries.


Mapping of demographic and socio-economic information from Population Census sources almost invariably depends upon the presence of Digital Boundary Files (DBF) for the regular or more commonly irregular polygons to which the statistical counts relate. Such national boundary files for the smallest spatial units have been created in association with the release of census counts in the case of the 1991 and 2001 enumerations, although the equivalent DBFs for 1981 enumeration districts and 1971-81 common census tracts were only produced after more than 20 years had elapsed (Walford, 2005). The boundary files for this series of census enumerations currently exist as largely discrete, separate datasets, although work has been undertaken to connect the different geographies by means of look-up tables for all or part of Britain (Atkins et al., 1993; Dorling et al., 2003). Exeter et al (2005) used a series of Look Up Tables (LUTs) to link the 1981 through to the 2001 Census geographies for Scotland. The success of this technique for Scotland is largely due to the nested hierarchy used in Scotland which aggregates unit postcodes, however, this is not the case for England and Wales geography changes.


The techniques of dasymetric mapping and areal interpolation are recognised spatial analytic procedures (Fisher and Langford, 1996), although have rarely been undertaken on a national scale, over three census dates and for a large number of spatial units, over 200,000 in the case of the 2001 British Population Census. The Linking Censuses through Time (LCT) project attempts to deal with census incompatibility through aggregation to a higher ward level geography (Martin et al 2002).

Inter-censal change in Britain has been undertaken at a small scale however this research allows for larger scaled analyses to be undertaken through the creation of comparable geographies at the Output Area level and spanning the 1971, 1981, 1991 and 2001 Censuses.


The central aim of this ESRC funded project (RES-000-22-0668) is to create a consistent geographical and statistical platform linked to the recent British Population Censuses that is sufficiently flexible to form the basis for analyses of past and, with the incorporation of new statistics, future population change. There are four specific objectives for the project:

To produce digital boundary files for sets of consistent, census-related geographical units; one moving forwards from the variably sized 1971-81 Census tracts; one moving backwards from the smaller postcode-based Output Areas of the 2001 Census;

To identify and estimate the values of a consistent set of demographic and socio-economic census type counts taking due account of changes in population bases;

To undertake an analysis of aggregate population change between 1971 and 2001 thus providing an evaluation and comparison of the datasets referred to in the previous two objectives;

To investigate the development of innovative techniques that incorporate a time dimension for visualising demographic and socio-economic change.


The project methodology was based on applying spatial analytic techniques to existing census-related datasets. There were as follows:

Census Geographies

  • Digital boundary files for 1981-2001 (Edina UKBORDERS)
  • 1971-81 Census Tract Digital Boundary Files
  • 1981 Enumeration District Digital Boundary Files

Census Counts

External country boundaries and land use data

The creation of comparable census geographies firstly required census counts to be made comparable across the four census years. Then two techniques were used to create comparable geographies, Areal Interpolation and Dasymetric Mapping.


The project has generated digital boundary files and associated estimates of comparable census counts spanning the British Population Censuses between 1971 and 2001. The projected demonstrates the feasibility of applying areal interpolation and dasymetric mapping techniques at at the small area level on a nationwide basis where the spatial units number in excess of 100,000 in each census. Three sets of boundaries and counts have been created that allow the analysis of demographic and socio-economic change for consistent geographies over a period of 30 years.


2005 Walford, N. S. and Hayles, K. Thirty years of geographical and statistical (in)consistency in the British Population Census: preliminary steps in harmonising small area data resources Royal Geographical Society with the Institute of British Geographers International Conference, London.

2006 (April) Walford, N. S., Hayles, K. Creating consistency in British Census space, GISRUK06, 5-7 April, University of Nottingham.

Associated Research Groups


A new doctor-patient relationship in landform and landscape studies?

Professor Nick Clifford from King’s College London will be giving the 2014-15 CEESR Annual Lecture on …

Awards for Geology graduates

Recent KU Geology graduate,┬áDelano Henry, has been awarded the British Sedimentological Research Group Prize for Best …

Human activity has contributed to increased forestry disease

Connections between climate change and biological trends are often difficult to establish from short-term studies. In …