A SOCIOECONOMIC ANALYSIS OF FERTILITY

点击翻译按钮获取中文摘要

This is a research project on the socioeconomic determinants of fertility in Iran. Our specific aims are (1) to extend our multivariate event history analysis of fertility so as to test for and estimate contextual effects and to specify the proximate determinants through which socioeconomic and contextual effects operate; and (2) to assess fertility trends during the period following the Islamic Revolution, 1979-1986. Aim (1) requires the continued development of new analytical methods, namely, the synthesis of event history modeling and multilevel analysis. This is facilitated by the Gibbs sampler, a recent innovation in statistics that allows models with many unobserved quantities to be easily specified. Here the unobserved quantities are the effects on fertility of unmeasured characteristics of the province, the local community, and the woman herself. To assess the effect of contraception, we have data only on "ever use", and we propose to translate that into estimates of current use at each time point by using external information on aggregate use in Iran. To implement aim (2), we will calculate demographic rates from the 1986 census and compare them with the IFS and the 1976 and 1966 censuses. To compare the 1986 census with the IFS data requires that we translate the event history model parameters estimated for the IFS into more familiar and interpretable demographic rates, and we propose to develop new methods for doing this.