I propose to model and forecast age- and duration specific marriage, divorce, widowhood, and remarriage rates in the United States. While the focus of past research on divorce and remarriage has been on the outcomes for children and working-age women, marriage and divorce patterns also have important consequences for the elderly. Nuptuality trends help determine the structure of the multi-generational kinship and support networks available to the elderly. Furthermore, forecasts of remarriage at older ages - following both divorce and widowhood - directly affect the living arrangements of the elderly and the companionship and support available to them. The lower mortality and better health observed among married people appears due not only to a selection effect of those who enter and stay married but to a protective effect from marriage. The methods I plan to use are based on the innovative forecasting techniques developed for mortality forecasting by Ronald Lee and Larry Carter. The project will (1) extend these statistical methods in order to work with the duration- and age- specific nature of divorce and remarriage, (2) forecast future nuptuality rates and their implications for the marital status of the elderly and their families, and (3) produce a consistent time series of marriage, divorce, and remarriage rates for use by the research public. Data will be from a combination of NCHS vital statistics and retrospective martial histories from the Current Population Survey. This pilot project is intended to lead to a subsequent request for longer-term support such as a FIRST grant or Independent Investigator Award to study further the consequences of divorce and remarriage on modern kinship networks.
| Goldstein, J R (1999) The leveling of divorce in the United States. Demography 36:409-14 |
