Content area
Abstract
This dissertation uses Indian data from the National Family Health Survey, 1992–93 and the 1981 Census to examine how son preference affects two demographic outcomes: fertility behavior and childhood mortality. Chapter One explores how son preference expressed through fertility decisions shapes family structures, particularly family size and the sex composition of children. It postulates that couples with son preference practice differential stopping behavior (DSB), whereby those with sons cease having children, while those with only daughters continue bearing children in an attempt to have sons. One consequence of this behavior is that, as family size increases, the proportion of sons in the family decreases. Another consequence of DSB is that, controlling for family size, characteristics of couples who have the strongest son preference are the same as characteristics of couples who have the highest proportion of sons. These findings dispute the commonly held demographic assumption that boys and girls are randomly assigned to families irrespective of the couple's desire for sons. They also demonstrate that son preference has a profound and predictable effect on the sex composition of children in a family. Chapter Two uses district level data to map the geographic variation in the sex composition of children and several forms of female disadvantage. This exploratory work finds that the practice of DSB frequently coincides with, but is not coterminous with, low female literacy, low female labor force participation, and excess female childhood mortality. Chapter Three explores how parental preferences for boys and girls affect the likelihood that a child is vaccinated and the chances that a child survives to age five. It shows that children who are unwanted by their parents at the time of birth have 24% lower odds of being vaccinated and 38% higher odds of dying than wanted children. This chapter also shows that parental sex preferences indirectly affect the morbidity and mortality of a child through the composition of the child's siblings. The results show that having a high proportion of unwanted sisters is most beneficial, while having a high proportion of wanted brothers is most detrimental.
