Affordable Access

A method for the simultaneous estimation of parallel processes in the human life course.

Authors
Type
Published Article
Journal
Studia demograficzne
Publication Date
Issue
105
Pages
111–124
Identifiers
PMID: 12286132
Source
Medline
Keywords
License
Unknown

Abstract

The period of time over which humans live their respective lives may be regarded as being comprised of a set of interdependent processes evolving simultaneously over time. Some parallel factors and processes are fertility, education, labor force participation, and migration. While researchers may be aware that these interrelated phenomena are taking place, it is very difficult to measure the effects of and from individual processes were they to operate independently. Longitudinal data must be gathered and employed in lieu of cross-sectional data to best measure these dynamic processes on a continuous basis. In so doing, classical methods of simultaneous estimation will be inadequate. This paper proposes alternative estimation methods which are illustrated in theoretical and empirical case studies. The empirical data are drawn from a 1984 Dutch event history survey of 1600 men and women aged 18-54 years. Histories of cohabitation, marriage, dissolution, and fertility were recorded. Semi-parametric estimates of partial and simultaneous effect coefficients as well as the size of the exposure populations on which they are based are presented in the paper. While additional studies and experimentation needs to be conducted on simultaneous estimation, results nonetheless suggest that planning to have children does not lower the odds that reproducing partnerships will eventually be abrogated. Reproduction strains partnerships and may worsen prospects for lasting relationships. Furthermore, the impact of conjugal union disruption on fertility is much stronger than the reverse effect.

Statistics

Seen <100 times