POPULATION CHANGE FROM A FAMILY PERSPECTIVE

本研究的目的是从实证的角度来研究这些问题的原因 自1940年以来，美国家庭和家庭的变化， 整个成年人的生命周期。 使用公共使用样本微数据 根据1940年、1950年、1960年、1970年和1980年的人口普查， 审查了家庭和家庭的变化，如制度化， 老年人、独居、同居、女户主家庭和大家庭 家庭 要研究的变化的决定因素包括：1） 收入、年龄、婚姻状况、子女数量、种族、 性别、居住地和教育，2）上述变量的平均值变化， 3)住户和家庭模式的群组变化; 4） 个别变数的影响;及5）随时间的剩余变动 控制人口构成的变化。 此外，该研究还将分析 使用总数据计算的住户和家庭结构的剩余或净趋势 住房供应、消费者价格、社会 福利项目和性别比例 用于完成项目目标的方法包括分析 五个普查年的合并样本，其中添加剂和 时间的交互作用被明确地考虑。 多项logit 分析是用来获得独立的影响估计 住户/家庭/结构的名义计量变量，以及 多元分解技术被用来检查组成 家庭/家庭中随时间变化的过程性组成部分 结构措施 然后使用广义最小二乘来获得 时间序列模型的估计。 该项目的最终结果将不仅是推进我们的经验 了解家庭和住户的长期趋势和变化 结构，而且还通过明确地提高我们的理论知识 测试消费者需求、人口构成和文化品味 家庭和家庭变化的解释。

The objective of the proposed research is to empirically study the causes of household and family change in the U.S. since 1940 with consideration of the entire span of the adult life cycle. Using Public Use Sample microdata from the 1940, 1950, 1960, 1970, and 1980 Censuses of Population, the study examines household and family changes such as institutionalization of the elderly, living alone, cohabitation, female-headed families, and extended families. The determinants of the changes to be studied include: 1) variables such as income, age, marital status, number of children, race, sex, residence, and education, 2) changes in the means of above variables, 3) cohort shifts in household and family patterns; 4) changes in the effects of individual variables; and 5) residual changes over time controlling for changes in the composition of the population. Additionally, the study will analyze the structural determinants of the residual or net trends in household and family structures using aggregate timeseries measures of housing availability, consumer prices, social welfare programs, and sex ratios. The methods used to complete the goals of the project involve analysis of a combined sample from the five census years in which the additive and interactive effects of time are considered explicitly. Multinomial logit analysis is used to obtain estimates of the effects of the independent variables on the nominal measures of household/family/structure, and multivariate decomposition techniques are used to examine the compositional and processual components of changes over time in the household/family structure measure. Generalized least squares is then used to obtain estimates for the time-series models. The end result of the project will not only be to advance our empirical knowledge of the long-term trends and changes in family and household structure, but also to advance our theoretical knowledge by explicitly testing consumer demand, demographic composition, and cultural tastes explanations of family and household change.