The secondary sex ratio, paternal age, maternal age and birth order in Japan.

The simultaneous effects of several variables on the secondary sex ratio have been examined using data from over 3.7 million births which occurred in Japan during 1975--6. A weak and negative association between sex ratio and birth order was observed but it was not significant in the statistical sense. A negative effect of paternal age--birth order interaction was obtained when maternal age was controlled. The quadratic model is much more powerful than the linear model in explaining the sex ratio variability.