Bringing historical British Population Census records into the 21st century: A method for geocoding households and individuals at their early‐20th‐century addresses

Author:
Open Access
Publisher website
Google Scholar
Abstract
Increased use of geographical information systems (GISs) for social scientific research has highlighted the analytical possibilities offered by unlocking historical data sources that are impractical with contemporary data. Genealogists and social scientific researchers have used information relating to individuals, households, and their addresses contained in the British historical census records for a number of years. This article takes geographical exploration of these a stage further by developing a method of attaching grid references to the census addresses, thereby opening up new research possibilities including spatial analysis. The resultant geocoded census data enable address‐, household‐, and individual‐level historical research and aggregation to contemporary and historical spatial units for exploration of demographic and socio‐economic change. The paper focuses on the development of a method of geocoding 1901 and 1911 Census data in respect of six study areas in the historical counties of London and Middlesex with over 260,000 individuals within some 60,000 households or communal establishments in both 1901 and 1911. Successful semi‐automated matching of historical census addresses with a contemporary address database is related to population density, change from property naming to numbering, residential development, thoroughfare name changes, and transcription error. Complete geocoding was completed by manual digitising with the aid of ancillary information sources and geographical information embedded in census records. The method outlined has the potential to be replicated in other areas and be adapted to use with other contemporary address databases that contain grid references. The paper outlines the factors that would influence transferability of the geocoding method.
Funding Information
  • British Academy (SF110183)

This publication has 22 references indexed in Scilit: