Public Use Microdata Sample on the 1930 Census

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): This project will create a nationally representative one-in-100 public use microdata sample of the 1930 United States Census of Population. Preparation of the sample includes eight steps: 1) Data entry of information on approximately 1,232,026 individuals from the original enumerators' manuscripts; 2) Development of dictionaries to translate each alphabetic entry into numeric codes compatible with the existing series of microdata for other census years; 3) Evaluation of sample quality through consistency checks, random blind verification of approximately 123,200 cases and comparison with aggregate statistics in the published census volumes; 4) Editing, cleaning and allocation of missing, illegible and inconsistent data through logical rules and imputation procedures; 5) Construction of new variables on household composition, relationships within families, economic status and ethnicity; 6) Development of documentation, including full descriptions of the sampling and data processing methods, a procedural history of the 1930 census, a detailed analysis of comparability issues, and a user's guide; 7) Evaluation and editing of data dictionaries and data processing procedures used in the existing samples covering 1850 to 1920 to ensure consistency with the new 1930 sample; 8) Integration of the sample and documentation into the Integrated Public Use Microdata Series (IPUMS). The new sample represents the capstone of a twelve-year effort to create an integrated census database describing the American population from the mid-nineteenth century to the present. It will complete the IPUMS database, a nationally representative high-precision series of census microdata files spanning the period from 1850 through 2000. The IPTJMS includes historical samples created at Minnesota for every surviving census from 1850 through 1920. For the period from 1940 through 2000, the IPUMS incorporates samples that were originally prepared by the Census Bureau. The 1930 census is the missing link that will close the gap between the Minnesota samples and the Census Bureau samples. The completed data series will include large, high-precision compatible samples for fifteen census years between 1850 and 2000; taken together, these census files will constitute a resource of unprecedented power for the study of long-run demographic and economic change.