Research interests
David's core research interests concern the modelling of population in geographical information systems, particularly with regard to census and health applications. He has long standing interests in gridded population mapping, automated zone design, health care accessibility and, more recently, time-space population modelling. He is an expert in georeferencing and modelling small area ppulation data and devised the current system of Output Areas used for the publication of 2001 and 2011 census output data in England and Wales, subsequently developed into 2001-based Super Output Areas and 2011 Workplace Zones. Most recently, he has been involved in work on future population data collection systems, particularly the use of administrative data both as a means of estimating population characteristics and as a research resource. Several of his projects have yielded new software tools and web resources which are linked from the various specific project sites.
Research Projects
Deputy Director of the UK Data Service
http://www.ukdataservice.ac.uk
Co-Director ESRC National Centre for Research Methods
http://www.ncrm.ac.uk
Deputy Director of the Administrative Data Research Centre for England
http://www.adrn.ac.uk/centres/England
PI of ESRC ReStore repository project
http://www.restore.ac.uk
Population 24/7 modelling project
http://www.researchcatalogue.esrc.ac.uk/grants/RES-062-23-1811/read
AZTool automated zone design software
http://www.geodata.soton.ac.uk/softward/AZTool/
Research group
Population, Health and Wellbeing (PHeW)
Affiliate research group
Administrative Data Research Centre for England
Research project(s)
The output areas used to publish data from the UK Census have traditionally been designed based on the locations of residential addresses. The geographical distributions of residential and workplace locations are, however, very different: in city centres, for example, there are often few residential properties but many workplaces, whereas in suburban areas there may be many residential properties but few workplaces. The publication of workplace data by residential geographies therefore presents significant challenges to researchers and practitioners for two key reasons: (i) it can cause some workplace-related data to not be released at all due to confidentiality concerns (as happened for the 2001 Census) and (ii) the residence-based zones are often not an appropriate statistical or geographical base for mapping or analysing workplace-related data.
Professor David MartinUniversity of Southampton University Road Southampton SO17 1BJ
Room Number:44/2099