Multidimensional life-table analysis of the effect of child mortality on total fertility in India, 1992–93, 1998–99, 2005–06

Abstract

This paper presents a new method for estimating the effect of child mortality on the total fertility rate (TFR). The method is based on discrete-time survival models of parity progression that allow construction of a multivariate multidimensional life table of fertility with four dimensions: woman's age, parity, duration in parity, and number of previous child deaths. Additional socio-economic variables are included in the set of predictor variables in the underlying survival models of parity progression. The life table yields a replacement rate, which measures the effect of one additional child death on the TFR. The method is illustrated by applying it to three Indian National Family Health Surveys. Major findings are that dead children are incompletely replaced, and that the replacement rate rises as the TFR falls, reflecting women's increasing ability to control their fertility.

Keywords:

• fertility
• TFR
• child mortality
• multidimensional life table
• survival models
• replacement rate
• India

Notes

1. Hassan Eini-Zinab is at the programme on Population and Health Studies at the East–West Center, 1601 East–West Road, Honolulu, HI 96848-1601, USA. E-mail: [email protected] or [email protected]

2. Support for this research was provided by Grant 1R01HD057038 from the US National Institutes of Health. The author is grateful to Robert Retherford for helpful comments on earlier drafts of this paper. Computer programs (in SAS) are available from the author at the e-mail addresses above on request.