Combining historic interpretations of the Great Britain population with contemporary spatial analysis: The case of surnames

Abstract

This study applies contemporary geospatial analysis to the 1881 Census of Great Britain to establish, through a process of inductive generalisation, the surname regions in 19th century Britain. All surnames in the census were geo-referenced to the 662 Registration Districts listed. The districts' surname compositions were compared using the Lasker Distance calculation. The resulting similarity values, assigned to each district, were grouped according to Ward's hierarchical clustering. By mapping the cluster allocations a clear regionality, characterised by a high spatial autocorrelation value, emerges that closely matches the historical attempts to establish the surname regions in 19th Century Britain. As one of the first studies of this kind to utilise a complete population register for 1881, with results aligned to historical interpretations, one can think of the maps presented here as a baseline for the regionality of surnames in Great Britain, against which the effects of more recent migrations can be compared.