Collaborative Research: EAGER: Cross-platform Election Advertising Transparency Initiative

合作研究：EAGER：跨平台选举广告透明度倡议

• 批准号: 2235008
• 负责人: Travis Ridout
• 金额: $ 1.99万
• 依托单位: Washington State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2235008&HistoricalAwards=false
• 关键词: Collaborative; Research; EAGER; Cross; platform

Nearly $2 billion was spent on digital advertising in the presidential general election in 2020, and online advertising is only expected to grow. Campaigns’ rapid shift to digital, however, has caused headaches for those who track political advertising. The data provided by online platforms, such as Facebook and Google, come in different formats, contain different collections that are not immediately comparable, and lack metadata necessary to find all ads related to a specific campaign for office, all ads featuring a specific candidate, or all ads within either set that discuss a specific issue. The sheer volume of content requires the use of computational methods to extract the information necessary to answer even basic descriptive questions such as how much money was spent online in a particular race or across all federal races. This project provides the infrastructure necessary to answer critical questions about the role of digital advertising in American democracy, including the extent to which “dark money” dominates campaigns, and the spread and reach of misinformation in campaigns.The project team is building expandable infrastructure to acquire, process, integrate, label, and distribute digital election advertising data from two large online platforms. The result is a centralized repository that provides robust documentation of all procedures and code such that they can be modified to apply to other platforms and contexts for expansion. Cross-platform integration and standardization of digital election advertising data through human and state-of-the-art computational methods reduces costs to individual researchers, provides parallel procedures already in place for analyzing TV advertising (and thus maximizing comparability to existing data), provides linkage information to other data sources, and produces accessible data to benefit the community. Accomplishing this aim is a high risk-high reward endeavor. For instance, new approaches or methods are needed to capture and label relevant advertising by focus, especially when the true sponsor is unclear or disguised.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.

2020年，在总统大选中，在数字广告上花费了近20亿美元，而在线广告只有预计才会增长。但是，竞选活动的迅速转移到数字化方面，已经为那些追踪政治广告的人造成了标题。在线平台提供的数据，例如Facebook和Google，采用不同的格式，包含不同的集合，这些集合无法立即可比，并且缺乏找到与特定办公室的所有广告系列相关的所有广告，所有具有特定候选人的广告或所有广告中的所有广告，或任何一组讨论特定问题的广告。内容量的数量需要使用计算方法来提取所需的信息，甚至可以回答基本的描述性问题，例如在特定种族或所有联邦种族中在线花费了多少钱。该项目提供了回答有关数字广告在美国民主国家中作用的关键问题所需的基础设施，包括“黑暗货币”主导竞选活动的程度，以及在广告系列中的失误和覆盖范围的传播和影响力。结果是一个集中存储库，可提供所有程序和代码的强大文档，以便将它们修改为应用于其他平台和上下文以进行扩展。通过人类和最先进的计算方法对数字选举广告数据进行的跨平台集成和标准化降低了对个别研究人员的成本，提供了已经制定的并行程序来分析电视广告（从而最大程度地提高了与现有数据的可比性），提供与其他数据源的链接信息，并为可访问的数据提供可访问的数据以使社区受益。实现这一目标是一项高风险高的奖励努力。例如，需要新的方法或方法来按重点捕获和标记相关的广告，尤其是当真正的赞助商不清楚或伪装时。该奖项反映了NSF的法定任务，并通过使用基金会的知识分子的智力优点和更广泛的影响来评估NSF的法定任务。