How similar are the determinants of mortality and fertility?

Abstract

Multivariate analyses offer an explanation of the effects of demographic, socioeconomic, and biological factors on children ever born and child mortality in Pakistan in 1990-91. Data are obtained from the 1990-91 Pakistan Demographic and Fertility Survey. Household durables (10 items) are used as a proxy for economic status. Sex preference is measured as the percentage of females in the household. Multicollinearity was not a problem. The results showed that both women's age and contraceptive use had a positive, significant effect on fertility and mortality. For all cohorts, as age increased the magnitude of the coefficient decreased. Marriage age was found to be negatively, significantly associated with fertility and mortality. For women aged 35 years and older, contraceptive use became negatively associated with child mortality. This study showed women's work status after marriage was inversely related to fertility and negatively related to mortality. Urbanization only had a significant, negative association with fertility and was unrelated to mortality. More girls in the family was related to higher fertility but not mortality. Standard of living was significantly, negatively related to fertility and mortality. Education of women aged 35 years and older had a stronger effect on fertility than husband's education. Husband's education beyond the primary level had a significant, negative effect on child mortality. Breast feeding longer than 12 months was significantly, negatively related to fertility and mortality. Breast feeding under 6 months was related to higher fertility than those who breast fed for 6-12 months. Immunization had a insignificant, positive relationship to fertility and was significant only for women under 25 years of age. There was a negative, significant effect on mortality.