| Birth spacing is the age interval among the children within one family. Birth spacing directly decides the investment of parents to their children and their careers, and then places an important role in the development of children. China has been applying one child policy for such a long time. With the changes in child policy in local government, having a second child may be possible under some circumstances. Motivated by this change, I investigate birth spacing and try to light some methodologies in birth spacing research in China. To find whether the birth spacing is the longer the better, I use NLSY to investigate the birth spacing effects on children's educational outcomes. Because birth spacing may be endogenous, I use instrumental variables to test the relationship between birth spacing and educational outcomes. This is one of the main contributions in the thesis. Specifically, in this thesis I use two instrumental variables: Twin and Membership in Rome Catholic. Moreover, polynomial regressions and semi-parametric estimations, comparing to OLS and 2SLS, give more information about birth spacing effects by picturing inverse U shape birth spacing effects curves. That is to say, the birth spacing effects do not increase linearly as birth spacing increases, but reach to a peak value then decrease. Through semi-parametric estimation using instrumental variables, we found that when birth spacing is around 29 months, the math, reading recognition and reading comprehension scores reach their peak value. This birth spacing is so called golden birth spacing. |