PROJECTING KINSHIP RESOURCES FOR THE ELDERLY

人口比率的变化最终会导致人口数量和 老年人将拥有的生活亲属的类型，因此在他们的 潜在的基于亲属的资源。计算机微观模拟是一种用于 根据以下假设预测亲属关系资源的这些变化 人口比率的未来走向。这个项目将从一个 外部效度研究，据信是有史以来第一次对 亲属关系预测，比较20世纪80年代早期的亲属关系预测 与全国家庭和家庭调查的经验估计相反 家庭(NSFH)。接下来，采用不同建模的效果 技巧将通过比较两组中国人的亲属关系来评估 预报，一种是通过计算机微观模拟得出的，另一种是由 家庭状况生命表的替代技术。微观模拟 预测模型和美国历史汇率集将于 这些测试的基础。NSFH的KIN计数信息将是 使用模拟WHERE样本外推到80岁以上的人群 规模不足以做出稳定的估计。新的亲属关系预测 因为美国将在2050年之前做好准备，分别为黑人 对于白人和其他人，使用一系列替代预测 未来的人口比率。NSFH使评估 不同类型的亲属在亲属关系配置中的程度 不同的类型现在实际上提供了各种支持。这 将允许构建和投射基于亲属关系的资源 美国的指数超出了亲属的计算范围。特价 重点将放在对老年人亲属关系网络的评估上 “最年长的老人”。

Changing demographic rates entail eventual changes in the numbers and types of living kin that elderly persons will have and thus in their potential kin-based resources. Computer microsimulation is a tool for projection these changes in kinship resources based on assumptions about the future course of demographic rates. This project will begin with an external validity study, believed to be the first ever conducted for kinship projection, comparing earlier projections of 1980s kinship against empirical estimates from the National Survey of Families and Households (NSFH). Next the effect of employing different modelling techniques will be assessed by comparing two sets of Chinese kinship forecasts, one derived by computer microsimulation and the other by the alternative technique of family status life tables. The microsimulation projection model and U.S. historical rate sets will be revised on the basis of these tests. NSFH information of kin counts will be extrapolated to the over-80 population using simulation where sample sizes are insufficient for stable estimates. New kinship projections for the U.S. will then be prepared down to 2050, separately for blacks and for whites and others, using a range of alternative forecasts of future demographic rates. The NSFH makes it possible to assess the degree to which kin of different types within kinship congfigurations of different types actually do now provide support of various kinds. This will allow the construction and projection of kinship-based resource indices for the United States which go beyond kin counts. Special emphasis will be placed on assessment of kinship networks for the "oldest old".