Collaborative Research: An Expanded Framework for Inferring Public Policy Diffusion Networks

州政府通常被称为“民主的实验室”，因为几十年来，他们一直在尝试新的政策解决社会上最重要的问题。公共政策学者和政治实践者经常向各州寻求可以被全国其他政府效仿的创新政策；称为策略扩散的过程。然而，状态之间的复杂关系（即策略实际上如何从一个状态扩散到另一个状态的模式、途径和机制）并没有得到很好的理解。在本研究中，pi开发了创新的工具和数据，以促进政策扩散的大规模计算和数据密集型研究。他们试图开发和利用最大的关于各州决策的数据库来采取政策，以更好地了解哪些州是政策领导者，哪些是政策追随者，以及为什么。这将为涵盖美国各州的政策扩散过程提供一个新颖、广泛和系统的视角。因此，政策制定者和其他感兴趣的利益相关者将有机会全面了解州政府在金融、环境、卫生、安全和其他社会问题领域的创新传播情况。技术方面本研究试图对公共政策扩散的研究做出三个主要贡献。首先，它将产生一个包含美国各州数百项政策的数据库，其规模至少是目前可用的最大数据库的三倍，并产生一个更能代表各州制定法律的政策范围的政策样本。它还将利用这些新数据来绘制和分析政策扩散的网络路径（即，在政策扩散中建立领导者和追随者的国家）。正如首席研究员最近发表的研究表明，尖端的计算方法可以应用于包含政府何时采取政策信息的大型数据库，以确定政策持续扩散的潜在网络。本研究的另一个关键贡献是改进了现有的方法，以最适合政策扩散研究（以及更广泛的社会科学）的方式推断网络联系，并在R统计环境中以用户友好的软件包实施。最后的贡献，旨在最大限度地利用这些数据和方法的潜在用户基础，将建立一个交互式的数据在线门户，完成可视化和自动分析。

General SummaryState governments are commonly referred to as "laboratories of democracy?"because for manydecades they have experimented with new policy solutions to society's most important problems.Public policy scholars and practitioners of politics often look to the states for innovative policiesthat can be emulated by other governments across the country; a process referred to as policydiffusion. However, the complex relationships between the states (that is, the patterns,pathways, and mechanisms underlying how policies actually diffuse from one state to another)are not well understood. In this research, the PIs develop innovative tools anddata to facilitate the large-scale computational and data-intensive study of policy diffusion. Theyseek to develop and utilize the largest database on states' decisions to adopt policies to betterunderstand which states are policy leaders, which are policy followers, and why. This willprovide a novel, expansive, and systematic view of the policy diffusion process that covers theAmerican states. As a result, policymakers and other interested stakeholders will have access to acomprehensive look at the spread of innovation among state governments regarding financial,environmental, health, security, and other social problem domains.Technical AspectsThis research seeks to make three central contributions to the study of public policy diffusion.First, it will produce a database on the adoption of hundreds of policies in the American states, atleast tripling the size of the largest database currently available and yielding a sample of policiesthat is more representative of the universe of policies for which states make laws. It will alsoleverage these new data to map and analyze the network pathways according to which policiesdiffuse (i.e., establish states that are leaders and followers in the diffusion of policies). Asrecently published research by the principal investigators shows, cutting-edge computationalmethods can be applied to large databases containing information on when policies were adoptedby which governments in order to identify underlying networks along which policies persistentlydiffuse. Another key contribution from this research is to improve upon the existing method forinferring network ties in ways that are most suitable for policy diffusion research (as well associal science more generally) and implement it in a user-friendly package in the R statisticalenvironment. A final contribution, which is intended to maximize the potential user-base of thesedata and methods, will be to build an interactive online portal to the data, complete withvisualizations and automated analytics.