|Orford, Scott and Radcliffe, Jonathan Stephen 2007. Modelling UK residential dwelling types using OS Mastermap data: a comparison to the 2001 census. Computers, Environment and Urban Systems 31 (2) , pp. 206-227. 10.1016/j.compenvurbsys.2006.08.003|
In the UK dwelling type is a commonly used term to refer to the 'building style' of a residential property. The concept is used frequently in urban analysis, social science research and Government policy, and has been used as a measure of homogeneity in the construction of Output Areas (OAs) in the UK 2001 census. However, there are no formal definitions of dwelling types in the UK beyond a basic statutory definition of dwelling-house and flat. The British Government's Office of National Statistics (ONS) also categorises dwellings into different types but provides no clear guidance to how these are defined. There is no source of dwelling type data for individual properties at a national scale beyond sample surveys. The main source of information is the 2001 census, which provides counts of ONS defined dwelling types for OAs. The objective of this research is to use OS Mastermap, a high resolution topographic database of Great Britain, to model and provide dwelling type information for individual residential addresses. The success of the modelling exercise is measured by comparing the modelled dwelling type information to data on dwelling type collected in the 2001 census, using Chi-square as a goodness of fit measure. Small area differences in the two measures are analysed in order to ascertain where OS Mastermap does not provide a very good estimate of dwelling type. The analysis highlights problems with the modelling process and discusses the potential of OS Mastermap as a source of dwelling related data and also possible small area errors in the 2001 census.
|Schools:||Geography and Planning|
|Uncontrolled Keywords:||OS Mastermap ; Census ; Dwelling type ; Disclosure ; Small area characterisation|
|Last Modified:||19 Mar 2016 22:05|
Cited 8 times in Google Scholar. View in Google Scholar
Cited 6 times in Scopus. View in Scopus. Powered By Scopus® Data
Cited 2 times in Web of Science. View in Web of Science.
Actions (repository staff only)