Q: Why are there sometimes holes in the statistical maps?

Previous Item

A: The system does not store individual statistical maps. It stores various kinds of information, and then tries to create the best possible maps from them. In particular, it holds a large list of administrative units, with the dates they were created and abolished; a very large collection of individual statistics, each with a date and the administrative unit it is about; and a big set of computerised boundary lines, each with dates of existence and the administrative unit it was for. There will be a hole in a map unless we can find the statistical data, the unit information and the boundaries for that unit at that date. Reasons include:

  • The unit did not even exist. For example, our maps for "modern counties" are limited to England, as counties no longer exist in Scotland or Wales, and even in England there are gaps for Unitary Authorities, which we treat as a kind of district. In general, we try to define our unit "types" precisely so that mapping all units of a given type for a particular date gives you a complete map of the country.
  • The unit existed but no statistics were reported for it. Many census tables provide district-level information only for the largest towns, other areas being covered only by the county totals. In 1961 and 1971, the census listed information for areas within some towns and cities by ward, not by parish, so these show up as holes in the parish maps.
  • We hold statistics but no boundaries. There are whole categories of administrative units we have yet to map, and with some pre-19th Century unit types systematic mapping will probably never be possible. Statistics for these are presented in graphs but not in maps. There are also individual units whose boundaries remain unknown. This is particularly a problem with parishes as you go back into the 19th century.
  • We could not link the statistics to a unit. We start by computerising a complete table of statistics, but loading the data into the system involves matching up the unit names in the table with our master list of administrative units, and sometimes the statistical reports used different versions of the names. We deal with this by carefully checking our sources and then adding the variant name to our information about the unit. This work is continuing, so there will be fewer holes as the system develops. However, uncertainties both about what units existed and about what their boundaries were mean there will always be gaps in our mapping of the earliest statistics. The census officials themselves reported in 1851 that the 1801 census has missed whole parishes.

Next Item