Factors affecting fertility in India: a fresh exploration

We used District Level Household Survey round 3 conducted in India in 2007-08 to test the following hypotheses: 1) Child mortality is positively associated with fertility; 2) primary infertility is negatively associated with fertility; and 3) son preference is positively associated with fertility. For analysis purpose, we divided India into 85 natural regions based on the agro-climatic scheme proposed by the census of India. We estimated all the independent variables and the dependent variable for each of the 85 natural regions of India. Bivariate LISA, Moran’s I, ordinary least squares (OLS), and instrumental variable (IV) regression were used to test the hypotheses. Improved water and improved sanitation were used as instruments in the IV regression. Primary infertility, use of family planning methods, and female literacy were negatively associated with TFR in the OLS (child mortality was not included in OLS). Instrumental variable regression (also included child mortality) confirmed the findings of the OLS. But, the statistical significance of female literacy reduced slightly in the IV regression. Child mortality and son preference were not statistically associated with TFR in the IV regression.