Ian Gregory (University of Portsmouth, UK)

Career:

My background is in GIS and human geography. I graduated in geography in 1992 from Lancaster University. I then took an MSc in GIS at Edinburgh University. Following on from this I started working for Humphrey Southall in 1994 building a GIS that would link a wide variety of pre-First World War statistical on economic distress with the boundaries they were originally published using. This resulted in the design of the Great Britain Historical GIS, a system capable of accurately mapping almost all of the major datasets published by the government since the early nineteenth century, particularly census and vital registration data. Essentially this research involved devising ways of incorporating changing boundaries into a GIS database. More recently I have been exploring methods of removing the impact of changing boundaries to allow long-term comparison of data using standardised local areas. This will allow long-term demographic change to be explored at a more detailed spatial scale than has previously been possible. It involves using complex methodologies that make maximum use of the available data to interpolate a variable or variables collected and published at one date onto the administrative units used at another date. This methodology will form the backbone of the atlas that is to be written by Southall et al to mark the 200^th anniversary of the British census, first collected in 1801.

Selected Publications:

Ian Gregory: Current Research

I am the principal researcher on the Great Britain Historical GIS Project, having been involved in the project from the outset and having originally designed the system.

• The construction of a Geographical Information System (GIS) containing an integrated assembly of census, vital registration and electoral data for Great Britain over the last 200 years, plus the boundaries of the changing reporting units. For England and Wales, our GIS records the changing boundaries of c. 630 Registration Districts between c. 1840 and 1911; c. 1,800 Local Government Districts between 1911 and 1974; and c. 15,000 Civil Parishes between 1876 and 1974. A full description of this is available from:

• The dissemination of the individual components of this system via Internet-accessible on-line services operated by the History Data Service at Essex and the UKBORDERS census unit, together with network access to the core resource by selected users. See:

• Developing methodologies to explore long-run change. Until recently the most common way of comparing data published at a variety of different dates for different administrative units was to resort to massive aggregation. In Britain this typically involved taking district level data and aggregating to approximately 50 counties. This is highly unsatisfactory as most of the detail contained in the original data are lost and the highly irregular shapes and sizes of counties can affect the results of any analysis. A far better approach is to analyse data at the level they were originally published at, namely district level. The problem with this is that boundaries changed so frequently that this was traditionally impossible for all but localised study areas and a very limited number of dates. The building of the historical GIS, however, allows us to devise new methodologies to interpolate data from all dates onto a single, standardised district-level geography to allow them the be compared over the long-term. This then opens the door to a wide variety of analytic work on long-term change in British society over the last 200 years.
• The communication of these findings to a wide audience. We will create a paper atlas to mark the bi-centenary of the first census of 1801, and may, in the longer term, create an electronic atlas, accessible by users in British schools and public libraries, which will also give access to the underlying data.
• Raising interest in GIS for historical applications. I have taught the basics of GIS to a significant number of historical geographers and some of this is now beginning to feed through to substantive research. In addition, we have been very successful in raising interest in GIS among historians and historical geographers through conference presentations, workshops, web sites, and other means.

Go Back