Measuring the Monetary Value of Online Volunteer Work
Not on the map. Why?
Abstract (arXiv)
Online volunteers are a crucial labor force that keeps many for-profit systems afloat (e.g. social media platforms and online review sites). Despite their substantial role in upholding highly valuable technological systems, online volunteers have no way of knowing the value of their work. This paper uses content moderation as a case study and measures its monetary value to make apparent volunteer labor's value. Using a novel dataset of private logs generated by moderators, we use linear mixed-effect regression and estimate that Reddit moderators worked a minimum of 466 hours per day in 2020. These hours amount to 3.4 million USD a year based on the median hourly wage for comparable content moderation services in the U.S. We discuss how this information may inform pathways to alleviate the one-sided relationship between technology companies and online volunteers.
Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). First posted on arXiv 28 May 2022.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2023: 1; 2024: 2; 2025: 1; 2026: 1.
5 papers.
- 2024SLM-Mod: Small Language Models Surpass LLMs at Content Moderationcited by 27
- 2025MoMoE: Mixture of Moderation Experts Framework for AI-Assisted Online Governancecited by 13
- 2023Toxicity Detection is NOT all you Need: Measuring the Gaps to Supporting Volunteer Content Moderatorscited by 10
- 2024Venire: A Machine Learning-Guided Panel Review System for Community Content Moderationcited by 7
- 2026PluRule: A Benchmark for Moderating Pluralistic Communities on Social Mediacited by 0
- 2026PluRule: A Benchmark for Moderating Pluralistic Communities on Social Media16 May 2026
- 2025MoMoE: Mixture of Moderation Experts Framework for AI-Assisted Online Governance20 May 2025
- 2024Venire: A Machine Learning-Guided Panel Review System for Community Content Moderation30 Oct 2024
- 2024SLM-Mod: Small Language Models Surpass LLMs at Content Moderation17 Oct 2024
- 2023Toxicity Detection is NOT all you Need: Measuring the Gaps to Supporting Volunteer Content Moderators14 Nov 2023