Data Science Training in Demography and Population Health

项目概要/摘要 华盛顿大学人口学与生态学研究中心 (CSDE) 提出了一项 培训计划，旨在培养具备以下技能的新一代人口科学家和人口统计学家： 使用先进的计算和统计方法进行数据密集型研究。人口科学 为理解社会、经济、政治和人口提供了一个重要的跨学科框架 影响疾病、死亡率、生育率和移民模式的动态。但要从噪声中提取信号 这些过程之间复杂的相互作用，下一代需要接受最先进方法的培训。的 该计划将培养学员成为涉及广泛复杂数据（包括数据）问题的领导者 来自多个来源、不用于研究的数据（例如管理数据）、动态或空间数据， 网络数据和“大数据”源。该计划的核心是 CSDE、 eScience Institute（华盛顿大学数据科学中心）以及统计和社会科学中心（CSSS）。 这三个研究所结合了人口统计方法、数据科学、计算机科学方面的广泛专业知识 和统计方法，并在研究和培训活动方面拥有长期、成功的合作记录， 包括人口研究大数据中的 BD2K 轨道。通过此次合作，学员将获得 以下领域的系列课程：高级统计方法和机器学习、编码、 数据库、数据可视化/通信和数据伦理。学员还将参加团队辅导 流程，他们与主要导师建立学徒关系，但也可以直接学习 通过合作项目和研讨会从其他教师和学员那里获得。导师将来自这三个人 中心以及各种其他部门，被选来提供最相关的专业知识， 包括来自生物统计学、统计学和计算机科学等高排名院系的多名教师 与工程。此外，学员将通过在校内和当地的轮换获得实践经验 合作伙伴。每位学员将轮流学习 eScience 的再现性和开放科学特殊兴趣 Group (SIG)，开发促进可重复研究的资源。培训计划将确保 学员获得必要的理论和应用经验来应对任何挑战和 接下来的数据革命带来的复杂性。

Project Summary/Abstract The Center for Studies in Demography and Ecology (CSDE) at the University of Washington proposes a training program to create a new generation of population scientists and demographers equipped with skills to conduct data intensive research with advanced computational and statistical methods. Population science provides a crucial interdisciplinary framework for understanding the social, economic, political, and demographic dynamics affecting patterns of disease, mortality, fertility, and migration. But to extract signal from noise in the complex interplay of these processes the next generation requires training in state-of-the-art methods. The program will prepare trainees to be leaders in problems involving a wide range of complex data, including data arising from multiple sources, data not intended for research (e.g.administrative data), dynamic or spatial data, network data, and “big data'' sources. At the core of this program is a collaboration between CSDE, the eScience Institute (UW's data science center), and the Center for Statistics and the Social Sciences (CSSS). These three institutes combine extensive expertise in demographic methods, data science, computer science and statistical methods and have a long, successful record of collaboration on research and training activities, including a BD2K track in Big Data for Demographic Research. Through this collaboration, trainees will take a series of coursework in the following areas: advanced statistical methods and machine learning, coding, databases, data visualization/communication, and data ethics. Trainees will also take part in team-mentoring processes, where they develop an apprenticeship relationship with a primary mentor but also learn directly from other faculty and trainees through collaborative projects and seminars. Mentors will come from all three centers, along with a variety of additional departments, selected to provide the most relevant expertise, including multiple faculty from the highly-ranked departments of Biostatistics, Statistics and Computer Science & Engineering. Further, trainees will receive hands-on experience through rotations with on-campus and local partners. Each trainee will do a rotation in eScience's Reproducibility and Open Science Special Interest Group (SIG), which develops resources that facilitate reproducible research. The training program will ensure trainees obtain the requisite theoretical and applied experience to take on whatever challenges and complexities the data revolution next provides.