Migration and marriage in the life course: a method for studying synchronized events

One of the main methodological problems in explaining migration from a life course perspective is the adequate modelling of synchronized events. On the basis of a West German sample of residence histories that is representative of three birth cohorts, of which the one born in 1939-41 is analysed, log-linear techniques were applied to separate the effects of being married, and of getting married, on migration rates. Results show that the dependence of short and long distance moves on age substantially diminish if marriage is considered as a synchronization variable. Moreover, the common finding that married persons move less than the unmarried is reversed at short distances if marriage is taken into account as an event which influences the probability of another (event dependence).