Abstract
This paper employs a household approach to elderly migration analyses and compares it with the traditional individual approach. The first part of the paper develops some concepts about household mobility and relates them to individual mobility. It then compares the two mobility measurements in a case study using the Public Use Microdata Sample (PUMS) from the 1990 Census. The results show that the mean household size for the elderly moving together tends to be smaller than that for elderly stayers. It also demonstrates the utility of the household approach on profiling elderly movers' living arrangement choices. The second part of the paper calibrates a set of discrete choice models based on the household and individual approaches. While in most cases the two approaches yield similar parameter estimates, new insights can still be gained through the household approach.
| Original language | English |
|---|---|
| Pages (from-to) | 285-300 |
| Number of pages | 16 |
| Journal | Papers in Regional Science |
| Volume | 76 |
| Issue number | 3 |
| DOIs | |
| Publication status | Published - Jul 1997 |
| Externally published | Yes |