Most Recent Birth Intervals in a Traditional Society: A Life Table and Hazards Regression Analysis

Abstract

Retrospective data is obtained from the 1988-89 Survey of the Effects of Socioeconomic Factors on Fertility among Scheduled Castes in Rural Karimganj District Assam India (1805 scheduled caste households in selected villages). Multivariate hazard regression models are used to estimate the net effect of explanatory variables on the most recent (closed) birth interval and the interaction of explanatory variables. Rindfuss et al. guidelines and Namboodiri and Suchindran life-table techniques are used in constructing median birth intervals and the proportion not conceiving within 97 months and measuring the tempo and amount of fertility. Birth intervals exclude the postpartum amenorrhea period which has a median of 9-13 months in Assam. Findings show that the median most recent birth (MRB) interval is 2.25 years. The shortest interval is an average of 1.5 years and followed an infant death. MRB interval increases with decreasing age at marriage and increased age parity and spousal age differences. MRB interval increases also with increased income. The Wilcoxon test shows statistically significant differences between groups for the following factors: marriage age maternal age parity spousal age difference survival status of the next-to-the-last birth and family income. Multivariate hazards regression models show that females married at 19 years of age or higher have a greater likelihood of having a long MRB interval. Longer birth intervals are also more likely among the youngest female cohort aged less than 25 years the highest parity of 5 or higher and the lower income households. The youngest birth cohort has a risk of having an MRB over 2 times higher than the oldest birth cohort. The risk is 1.69 times higher for women with a prior childs death than for women without this experience. Maternal age is highly significantly related to parity and education. The length of MRB interval is significantly more likely to increase with increased maternal age of literate mothers with lower parities. A younger age at marriage older cohorts lower parities lower income groups higher age differences between spouses and having a prior surviving child are related to a slower pace at which the last child was born.