The proposed project aims to assess the geographic dynamics of the last three decade U.S. elderly migrations. The project will primarily be undertaken at the state level using the decennial census data of 1970 to 1990. The theoretical underpinning of the project is the decentralization perspective, which implies that elderly migration in the U.S. will shift from the pattern of focused origins to focus destinations (e.g., Northeast to Florida or Arizona) to the pattern of focused origins and dispersed destinations. To test this hypothesis and to examine factors and spatial patterns that are associated with-elderly migration over time, the proposed research develops around three specific aims: (1) to develop a set of constrained gravity models to test if the elderly migration streams to the traditionally popular migration destinations, such as Florida or Arizona, have declined while controlling origin and destination populations; (2) to assess migration propensities to all the interstate destinations to see if there is a decentralization trend similar to that in some European countries; (3) to model geographic factors that associate with the dynamics of the elderly migration system using geographic information systems and county-to-county migration data for the 1980 and 1990 censuses. This project is the first step toward understanding medium-and long-term elderly migration trends. The next step is to undertake corroborative studies with more recent data, and assess the impact of changes in elderly migration flows on elderly services from both public and private sectors. Methodologically, we view our research as a basis for more sophisticated modeling of the effects of individual reasons (e.g., living arrangements, return migrations) on aggregated migration outcomes.
