A mixed effects model of birth spacing for pre-transition populations

It has often been argued that there are good theoretical and historical reasons to expect that deliberate birth spacing has played an important role in fertility patterns before the demographic transition. Yet, it has proved difficult to find hard empirical evidence. In this article, we propose a new model of the speed of parity progression that includes both fixed and random effects and that efficiently captures unobserved heterogeneity between couples in fecundability and postpartum amenorrhea. With this model, we demonstrate that pre-transition couples in the Netherlands indeed spaced their births during about the first ten years of marriage. In addition, we have found strong differentials in birth intervals by socio-economic position and religion. Finally, we also show how and why the model can be used with left-censored census data.