Collaborative Research: Understanding drivers of innovation in the use of science in Federal NEPA decision-making

有效的公共决策需要循证方法，领先的科学家经常呼吁产生与政策相关的科学，但人们对政府官员如何在日常工作中使用科学知之甚少。该项目将通过对美国联邦政府多个级别上千项环境政策决策中的科学使用进行系统分析来填补这一空白。我们关注美国林务局（USFS）根据《国家环境政策法》编制的环境影响报告书（EIS）的内容。更详细地说明科学信息如何用于政府决策，将使政府机构能够改进纳入科学信息的程序。与此同时，提供公共决策过程中使用哪些类型的科学信息的详细信息将有助于科学家、大学和资助机构优先考虑与政策相关的科学。为了实现这些目标，该项目将发布一个索引，帮助科学家比较不同出版物的政策影响，并与环境规划者开展外联活动，帮助他们了解研究的意义。该项目将使用网络抓取和计算文本分析来收集和分析2006年至2018年期间产生的所有USFS EIS中的科学使用情况。这些环境影响报告记录了在各种情况下至少2，000项关于公共土地管理的决定的科学依据。该项目有两个目标：（1）了解是什么推动了创新的科学信息（数据，事实，证据）在USFS环境影响报告中的纳入;（2）了解是什么推动了USFS决策者采用创新的科学实践（操作程序，协议，规范）准备环境影响报告。使用计算文本分析和机器学习工具，我们将测量每个文档的科学内容的各个方面，例如引用，关键概念和分析实践。我们将联合收割机将这些信息与有关每份文件中科学信息的公开数据相结合，例如出版地点、制作文件的科学家的雇主、公众参与文件的程度？的准备，以及政府办公室使用的科学信息。这一分析将有助于理解科学信息的特征和政府部门的特征如何影响科学在公共决策过程中的应用。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估来支持。

Effective public decision-making requires an evidence-based approach, and leading scientists frequently call for producing policy-relevant science, yet little is known about how government officials use science in their day-to-day work. This project will fill this gap by conducting a systematic analysis of use of science in thousands of decisions about environmental policy made at multiple levels of the US federal government. We focus on the contents of Environmental Impact Statements (EISs) prepared by the US Forest Service (USFS) under the National Environmental Policy Act. Providing a more detailed picture of how scientific information is used in government decision-making will enable government agencies to improve the processes through which they incorporate scientific information. At the same time, providing details of what types of scientific information are utilized in public decision-making processes will help scientists, universities, and funding agencies prioritize science that is policy-relevant. To forward these goals, the project will publish an index that will assist scientists in comparing the policy impact of different publications and will conduct outreach activities with environmental planners to help them understand the implications of the research.This project will use web scraping and computational text analysis to collect and analyze the use of science in all USFS EISs produced between 2006 and 2018. These EISs record the scientific basis for at least 2,000 decisions made about public land management in a wide variety of contexts. The project has two objectives: (1) Understand what drives the inclusion of innovative scientific information (data, facts, evidence) in USFS EISs; and (2) Understand what drives the adoption of innovative scientific practices (operating procedures, protocols, norms) by USFS decision-makers preparing EISs. Using computational text analysis and machine learning tools, we will measure various aspects of the scientific content of each document such as citations, key concepts, and analytical practices. We will combine this information with publicly available data about the scientific information in each document, such as place of publication, employer of the scientists producing the document, extent of public participation in the document?s preparation, and the government office using the scientific information. This analysis will help in understanding how characteristics of scientific information and characteristics of government offices influence how science is used in public decision-making processes.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.