Models of migration between regions are often based on the assumption
that individual moves can be modelled by a Poisson distribution whose
parameter is a function of origin and destination characteristics, and
generalized cost; this is true of Poisson regression models and spati
al interaction models. The Poisson assumption is that each individual
acts independently from others making the same move. In fact, migratio
n is usually engaged in by household groups, not independent individua
ls, making the Poisson assumption invalid. It is possible instead to c
onstruct a model in which the probability of a household moving is giv
en by a Poisson model and the number of individuals in a moving househ
old is given by an observed house-hold size distribution. This general
ized Poisson model is explained and fitted to a set of data on local-l
evel migration within the English county of Hereford and Worcester. Ho
wever, the sparse nature of the data set raises problems in assessing
goodness of fit because the deviance value is unusually low. This is t
ackled here with a simulation methodology.