HCC: Small: Incorporating Procedural Fairness in Flagging Mechanisms on Social Media Sites

本研究将分析社交媒体平台用户对标记工具在标记生命周期的每个阶段的期望和互动。平台提供了标记，这是一种技术功能，使用户能够举报不适当的帖子或不良行为者，以减少网络伤害，如仇恨言论、未经同意的性照片分享等。然而，先前的研究表明，报告伤害可能是第二次受害，特别是当受害者认为缺乏程序正义时。标志在维护内容审核系统的可行性方面发挥着关键作用，这是识别需要审核员或自动化工具仔细审查的内容的第一步。因此，设计标记界面以减轻报告的负面体验是至关重要的。为了实现这一目标，我们必须了解用户如何理解标记，他们寻找什么信息，以及他们如何浏览标记界面。该项目将分为三个阶段：(1)对最近标记帖子的社交媒体用户进行访谈，以了解他们的动机，心理模式和标记之前的关注点。(2)使用交互原型的用户研究将检查在使用标记界面报告帖子时遇到的意识，导航和使用问题。(3)参与式设计讲习班将开发标记后信息和通信系统的原型。在每个阶段，研究成果将与边缘化社区的个人共同设计，他们面临着不成比例的网络虐待。分析将从程序正义及其声音和一致性概念的解释角度出发，调查用户如何发展他们对标记过程的公平看法。它将产生标记决策树的分类，提供一个设计维度的生成集，平台可以在此基础上改变未来的标记实现。该奖项反映了美国国家科学基金会的法定使命，并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，被认为值得支持。

This research will analyze expectations and interactions of social media platform users with flagging tools in each phase of the flagging lifecycle. Platforms offer flagging, a technical feature that empowers users to report inappropriate posts or bad actors, to reduce online harm, such as hate speech, nonconsensual sharing of sexual photos, etc. However, prior research shows that reporting harms can be experienced as secondary victimization, especially when victims perceive a lack of procedural justice. Flags play a critical role in maintaining the feasibility of content moderation systems, an initial step to identifying content that requires careful review by moderators or automated tools. It is, therefore, vital to design flagging interfaces in ways that ease the negative experiences of reporting. To accomplish this goal, we must understand how users make sense of flagging, what information they seek, and how they navigate flagging interfaces.There will be three phases in this project: (1) Interviews will be conducted with social media users who have recently flagged a post to understand their motivations, mental models, and concerns prior to flagging. (2) A user study with interactive prototypes will examine the awareness, navigation, and usage issues encountered while using flagging interfaces to report a post. (3) Participatory design workshop sessions will develop prototypes for post-flagging information and communication systems. In each phase, the research output will be codesigned with individuals from marginalized communities, who face disproportionate online abuse. Analysis will draw from the interpretive lens of procedural justice and its notions of voice and consistency to investigate how users develop their fairness perceptions of the flagging processes. It will produce a taxonomy of flagging-decision trees, offering a generative set of design dimensions upon which platforms may alter future flagging implementations.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.