Modelling the duration of residence and plans for future residential relocation: a multilevel analysis.

Among the multitude of studies of factors that determine residential (im)mobility, relatively little attention has been paid to the length of time that people spend in a particular location and the importance of duration of stay for future relocation propensities. This study uses a large and detailed commercial survey sample of individuals in England and Wales and an appropriately tailored statistical approach to uncover new insights into the multilevel and spatially heterogeneous interactions that exist between residential duration, place attachment and plans for future residential relocation. We demonstrate how an individual's residential duration, as an essential ingredient for the accumulation of social capital and place-based attachment, is critical for informing plans for future (im)mobility. After controlling for a range of individual and contextual covariates, the predicted probability of planning a residential relocation is found to increase initially with duration of stay, to a peak after 4-5 years, and then to decline as the length of duration increases. However, there is evidence of strong geographical variation in this relationship, with some neighbourhoods being characterised by stable or even increasing propensities for movement with duration. The paper pays particular attention to the importance of wider neighbourhood dynamics (composition, selective sorting and population (in)stability), suggesting that they too play an important role in mediating duration-of-stay effects for individuals. The paper concludes by highlighting the need for researchers and policy practitioners interested in community dynamics, the development/accumulation of social capital and place attachment/rootedness, to give due consideration to multilevel durations of residence and, more broadly, the inherently spatial and temporal ties that bind individuals to place. [ABSTRACT FROM AUTHOR]