An investigation into the hierarchical nature of internal migration destination choice in Great Britain

Abstract

This research investigates the psychological basis of the process of migration destination choice. Specifically, it attempts to demonstrate that this process is hierarchical in nature, meaning that migrants group potential destinations into clusters or regions and examine the characteristics of groups of destinations before comparing specific locations within preferred regions. An examination of the variation in this hierarchical behaviour between migrants from different origins also provides valuable insights into the nature and role of cognitive space: that is, each individual's internal representation of the world around them. Four different techniques are used to illustrate hierarchical aspects of migrant's decisionmaking: the competing destinations model; two variants of the nested logit model; and a novel hybrid model. Hierarchical behaviour is demonstrated through comparison of these models with the traditional 'flat-processing' migration model. Migration data used in this research covers moves within Great Britain and is derived from the 1991 UK Census of Population with several other sources providing information describing the various potential migration destinations. This research will contribute both theoretically and methodologically to the field of migration research. It is novel in its hierarchical approach to modelling the UK internal migration destination choice process, and also in its algorithms for generating the discrete and probabilistic regionalisations of space that are used to represent migrants' hierarchical destination choice sets. It is hoped this work will be of practical benefit to the population and migration forecasting community by providing the basis for an inherently more accurate hierarchical approach to migration destination choice modelling.