The Ethics of Content Labeling
The Content Labeling Project operates within the Northeastern Ethics Institute. My primary objectives within the project were initially (a) cataloguing of examples and summaries of context for platform implementation; (b) creation of master spreadsheet and exemplar catalog; and (c) assist with literature review and other project areas as needed. In addition to content collection, I cultivated an in-depth policy review and content criteria evaluation of eight different online platforms - Facebook, Instagram, WhatsApp, Twitter, YouTube, Reddit, TikTok, and Snapchat.
Project Overview
The Ethics of Content Labeling - Project Description
Northeastern University Ethics Institute
Project Overview
As discussed with members of the Facebook Product/Policy Research and Stakeholder Engagement teams, the primary aim of this project from the Northeastern University Ethics Institute is to provide robust social and ethical analysis and evaluation of “candidate approaches” to online content labeling. By “approaches” we mean what the possible solutions are – e.g.,whether it is focused on users, content, or sources; whether it is positive or negative; or who takes on the responsibility in what respects. There will be a range of options considered, and clearly and systematically articulating that range is part of the initial stage of the project.The project will be relevant to content moderation and misinformation efforts by providing deeper and broader conceptual analysis and ethical evaluation than is currently available,drawing on and synthesizing rich intellectual resources across a variety of fields. The project’s central research question will be: How can online social platforms meet the needs and demands of society in terms of labeling informational content in an ethical and effective way? The project will draw on knowledge gained from current attempts at online content labeling in general, as well as from strong analogs in content labeling, such as in library sciences and food labeling. We will not be limited to the United States in our project scope, and we will incorporate international, cross-cultural perspectives and the views of global researchers into our work. The project will:
Overall, this project will provide a framework for identifying best practices in online content labeling, as well as for identifying considerations that need to be addressed in order for different approaches to be successful from both ethical and efficacy perspectives
Northeastern University Ethics Institute
Project Overview
As discussed with members of the Facebook Product/Policy Research and Stakeholder Engagement teams, the primary aim of this project from the Northeastern University Ethics Institute is to provide robust social and ethical analysis and evaluation of “candidate approaches” to online content labeling. By “approaches” we mean what the possible solutions are – e.g.,whether it is focused on users, content, or sources; whether it is positive or negative; or who takes on the responsibility in what respects. There will be a range of options considered, and clearly and systematically articulating that range is part of the initial stage of the project.The project will be relevant to content moderation and misinformation efforts by providing deeper and broader conceptual analysis and ethical evaluation than is currently available,drawing on and synthesizing rich intellectual resources across a variety of fields. The project’s central research question will be: How can online social platforms meet the needs and demands of society in terms of labeling informational content in an ethical and effective way? The project will draw on knowledge gained from current attempts at online content labeling in general, as well as from strong analogs in content labeling, such as in library sciences and food labeling. We will not be limited to the United States in our project scope, and we will incorporate international, cross-cultural perspectives and the views of global researchers into our work. The project will:
- Develop a typology of possible approaches that describes the solution space for online content labeling
- Create an assessment method/tool to evaluate candidate approaches to content labeling with respect to ethical, social, legal, technical, and effectiveness criteria
- Experimentally evaluate candidate online content labeling strategies to generate robust profiles of them that highlight their benefits and challenges. Rather than just doing observational studies, we will manipulate variables in our own online experiments (not requiring Facebook’s platform or data)
- Investigate the perceived benefits and challenges of various labeling methods paired with the efficacy of the labeling tools
- Investigate individual differences (such as age, gender and education) of labeling efficacy for candidate approaches to content labeling.
Overall, this project will provide a framework for identifying best practices in online content labeling, as well as for identifying considerations that need to be addressed in order for different approaches to be successful from both ethical and efficacy perspectives
The Evolution of Social Media Content Labeling: An Online Archive (2021)
A visual appendix of social media content examples, with photos and comprehensive policy analysis regarding label implementation. First, raw samples had to be collected from various social media platforms: Facebook, Instagram, Twitter, YouTube, Reddit, TikTok, WhatsApp, and Snapchat. I proceeded to conduct qualitative analysis on the privacy and content policies that affected content labeling. With both documented examples and insight on platform procedures, I developed a visual appendix of content labeling examples with a qualitative analysis of labeling efficacy, and developed these online tools to support future research on content labeling.
Authors: Jessica Montgomery Polny, Graduate Researcher; Prof. John P. Wihbey, Ethics Institute/College of Arts, Media and Design
Last Updated: June 3, 2021
Last Updated: June 3, 2021
The Emerging Science of Content Labeling:
Contextualizing Social Media Content Moderation (2020)
The product of the developing research for content labeling efficacy on social media platforms. My contributions to this paper are the raw examples and screen captures of labeled content, in addition to providing insights on a more organized taxonomy to refer to different formats of labeling across platforms.
Morrow, Garrett and Swire-Thompson, Briony and Polny, Jessica and Kopec, Matthew and Wihbey, John, The Emerging Science of Content Labeling: Contextualizing Social Media Content Moderation (December 3, 2020). Available at SSRN: https://ssrn.com/abstract=3742120 or http://dx.doi.org/10.2139/ssrn.3742120
Media Analysis
Public Opinion and the Effectiveness of Democracy (2015)
The Public Opinion course was offered in the Communications department at Salem State, and was taught by a Professor Mark Zaitchik,
who would also teach courses in Philosophy. This literary analysis assignment discusses the influences of public opinion and public advocacy on the effectiveness of democracy, and was the spark of my interest in media research with a philosophical and ethical approach.
who would also teach courses in Philosophy. This literary analysis assignment discusses the influences of public opinion and public advocacy on the effectiveness of democracy, and was the spark of my interest in media research with a philosophical and ethical approach.
Your browser does not support viewing this document. Click here to download the document.
