Big Data for Population Research

描述（由申请人提供）：该提案寻求资金，通过添加描述 1790 年至 1930 年美国整个统计人口的人口和地理数据来扩展综合公共使用微观数据系列 (IPUMS)。该项目将提供超过 6 亿人的特征数据，使可用于科学研究的美国人口普查微观数据量增加四倍。数据将覆盖完整地理区域的所有人口 详细信息，提供有关社区特征的背景信息，包括种族构成、人口行为和人口流动性。这些数据提供了有关关键社会和经济特征的最早信息，并将为长期人口和经济变化过程提供宝贵的见解。这些数据将为国家统计基础设施带来永久性和实质性的补充，并将对社会和行为科学研究产生深远的影响。该项目的实现得益于美国人口普查中两个主要家谱组织准备的大量高质量、经过验证的人口普查信息转录的捐赠。将大量原始数据转换为适合科学分析的格式需要执行以下任务：() 对地理位置进行分类和编码，以与已发布的人口普查报告中使用的类别兼容； (2) 评估数据转录的完整性和准确性； (3) 将字母字符串数据转换为随时间可比较的数字类别； (4) 采用新的数据清理软件来识别和纠正常见的计数和转录错误； (5) 制定文档，包括数据处理方法的完整描述、可比性问题的详细分析以及全面的机器可处理元数据； （6）将数据纳入IPUMS数据访问系统，免费向科学界传播； (7) 实施安全数据保护和保存政策。该项目将由经验丰富的研究人员团队执行，他们在大规模数据创建、集成和传播方面拥有出色的能力，并将利用尖端技术以合理的成本处理前所未有的数据量。该项目是明尼苏达人口中心与世界上最大的家谱数据生产商的合作项目，允许以具有成本效益的方式利用稀缺资源来开发人口和健康研究的共享基础设施。

DESCRIPTION (provided by applicant): This proposal seeks funding to expand the Integrated Public Use Micro-data Series (IPUMS) by adding demographic and geographic data describing the entire enumerated population of the U.S. from 1790 to 1930. The project will provide data on the characteristics of over 600 million persons, quadrupling the quantity of U.S. census micro-data available for scientific research. The data will cover entire populations with full geographic detail, providing contextual information on neighborhood characteristics, including ethnic composition, demographic behavior, and population mobility. These data offer the earliest information available on key social and economic characteristics, and they will provide invaluable insight into processes of long-run demographic and economic change. The data will make a permanent and substantial addition to the nation's statistical infrastructure and will have far-reaching implications for research across the social and behavioral sciences. The project is made possible by the donation of a massive high-quality verified transcription of information in the U.S. censuses, prepared by two major genealogical organizations. Converting this immense body of raw data into a format suitable for scientific analysis will require the following tasks: () classify and code geographic locations to be compatible with categories used in the published census returns; (2) assess completeness and accuracy of the data transcription; (3) convert alphabetic string data into numeric categories that are comparable over time; (4) employ new data cleaning software to identify and correct common enumeration and transcription errors; (5) develop documentation, including full descriptions of data processing methods, detailed analysis of comparability issues, and comprehensive machine-processable metadata; (6) incorporate the data into the IPUMS data access system for free dissemination to the scientific community; and (7) implement secure data protection and preservation policies. The project will be executed by a team of highly-experienced researchers with exceptional proficiency in large- scale data creation, integration, and dissemination and will leverage cutting-edge technology to process an unprecedented volume of data at reasonable cost. The project is a collaboration of the Minnesota Population Center with the world's largest producers of genealogical data, allowing cost-effective use of scarce resources to develop shared infrastructure for population and health research.