As most of you will know, a new version of the Vision of Britain site is on its way.
Earlier e-mails have said something about the new content, and there is not much point saying more until it is available. However, we are also working to improve the core information on administrative units and their basic population statistics, especially at parish level. This e-mail is to say that we are making larger changes than we originally envisaged, and that although we think that this will be overall a big improvement there are some drawbacks. What follows is about England and Wales, not Scotland, although we hope to do similar work there.
The system is built around some very formal information about what administrative units existed, what they were called and how they were related. This was taken mainly from Youngs' two volumes on the administrative units of England and Wales, and Melville Richards' broadly similar book on Wales. We then linked to this two sets of information which had in fact been created earlier and independently: (a) boundary polygons for the units, for parishes at census years; (b) transcriptions of the parish-level census tables.
This linkage was done near the end of our national lottery funding in 2004 and was somewhat rushed, although in general our methods meant that data would not be linked at all rather than being linked incorrectly. The early census parish-level tables frequently, and especially in the north of England, provide figures for whole parishes and then for the component townships. In linking these in, we usually computed totals for the parishes as a whole, and did not include the township data. This often led to poor matching with the boundary information, which defined the mother parishes narrowly and had separate polygons for the townships, and also to sudden drops in the mother parish populations in the late 19th century when the townships became separate civil parishes.
This is now all being stripped out. We are now re-linking both the polygons and the population statistics with the aim of providing data for the lowest-level units, and in the process sorting out a mess of minor problems with the information from Youngs and Richards which stops the data linking correctly. Improved polygon linkage will mean far more complete statistical mapping, and as the software driving the statistical mapping is being completely replaced it should all be far more usable.
However, we will be substantially revising some of those parish population time-series. Sorry.
How can we best explain this change in the new site?
The above is very much about the 19th century. We are also working to improve the linkage of census boundary change information into the system, which will lead to much more reliable information on when units were created and abolished during the 20th century.
Reader/Director GB Historical GIS,
Department of Geography,
University of Portsmouth,
PORTSMOUTH PO1 3HE
Historical GIS team: 023 9284 2500