|
| Introduction |
This example is concerned with individuals' migration histories within
Great Britain, where migration is a residential move between two localities.
Boundary choice is crucial in defining what is a migration move
(White and Meuser, 1988).
In this analysis migration is taken as an inter-county move. It is therefore
concerned with moves which involve breaking away from social and community
ties.
For a recent text on migration see for instance Boyle, Halfacree and Vaughan (1998).
|
| The data |
The data are derived from a large retrospective survey of life and work histories
carried out in 1986 under the Social Change and Economic Life Initiative (SCELI),
funded by the ESRC.
The data were therefore not specifically collected for the study
of migration, but were drawn from an existing data set which includes
information on where individuals had lived all their working lives.
The variables selected from the primary data set are those
which are suggested in the research literature as important for
explaining individual migration behaviour.
Temporary moves of a few months duration do not imply commitment to
a new area and are not regarded as migration. Migration data are therefore
recorded on an annual basis.
The respondents were aged 20 to 60 and lived in the travel-to-work area of
Rochdale, just to the north of Manchester.
(Rochdale was one of six localities chosen for the SCELI survey
for their contrasting experience of recent economic change.)
As the analysis is concerned with internal migration within Great Britain,
individuals who had lived abroad during their working lives are excluded
from the data set.
The information for 1986 is incomplete and is therefore not
included.
The data set contains the migration histories of 348 males
during their working, or potentially working lives, starting from the
completion of education up to 1985.
The data set is longitudinal, with one observation for each individual per
calendar year. There are a total of 6349 annual observations.
The start year for the collection of data for each individual is different,
but the final year is the same.
The response variable of interest is binary, indicating for each individual
and for each calendar year, whether or not there was a migration move.
The explanatory variables are age, calendar year, duration of stay at each
address, education, and information on marriage, children, employment and
occupational status for each year.
NEXT:The longitudinal data set |
Home page | Contents | Previous |