Longitudinal Approaches to Analysing Migration Behaviour in the Context of Personal Histories

Abstract

Longitudinal approaches to analyzing migration behaviour are reviewed in this paper and used to investigate interdependencies between the migration histories of individuals and their histories of participation in employment. Precise information about the timing and sequencing of events can be especially useful in the analysis of relations between employment and migration and the empirical analysis is based on some longitudinal data that are especially detailed in this respect: The Survey of Income and Program Participation (SIPP) for the United States. Relations between these migration histories and the histories of other aspects of individual lives, including employment histories, can be investigated within a general framework in which discrete-state continuous-time stochastic processes serve as general models for the lifetime migration behaviour of individuals and observed migration histories are treated as particular realizations of these processes. The processes are formally defined, in a general and abstract way, by a set of discrete states and a set of functions that describe the chances that an individual will make a transition between any pair of states at any time. In the case of migration histories the states of the model correspond to particular localities where an individual might reside and the transition functions summarize the chances that an individual will, at any time, move from one locality to another. The observable phenomena that correspond to the operation of these processes are, for any individual, a set of migration events that occur on particular dates during the individual’s lifetime and a corresponding series of episodes of residence in particular localities, with each episode beginning and ending on a particular date.