The 1979 National Longitudinal Survey of Youth (NLSY79) contains the most detailed data ever collected on fertility intentions and subsequent fertility behavior -- a large nationally representative sample of men and women born 1957-64 provided retrospective fertility histories and were asked their fertility intentions 16 times between 1979 and 2006. We use these remarkable data to build on prior work by Quesnel-Vallee and Morgan (2003) and Morgan and Rackin (2010) to describe the correspondence between intended family size (at various ages) and completed fertility. Earlier work shows that the mean error is negative - women and men have fewer children than intended -- but this error is modest in size. We also calculate the "gross error" (the absolute value of the individual differences between intentions measured at age 24 and behavior at age 45);this error is quite large approximating one birth for women and men. We propose to examine the correlates of "missing the target" both "low" and "high" over various ages (and years). These analyses correspond to and inform a "proximate determinants" model of low fertility proposed by Bongaarts (2001;also see Morgan 2003) and a theoretical framework that stresses that fertility intentions are highly contingent and highly constrained. We will also use these data to estimate dynamic models of interrelated fertility and intention change over the life course. Specifically, following Bollen and Curran (2004) we will estimate an "autoregressive latent trajectory model"- a model that subsumes the dominant approaches (the growth model and the autoregressive model) to modeling life course processes in longitudinal data.
The proposed research examines women and men's actual fertility behavior vis-`- vis their intentions for children. Unintended pregnancies impact the health and well-being of women and men as well as the children they bear. Likewise, unmet desire for children can have consequences for emotional well-being and life satisfaction.