Bright and Dark Side of Social Media in the Marginalized Contexts
Permanent URI for this collection
Browse
Recent Submissions
Item Unveiling Gender Dynamics for Mental Health Posts in Social Media and Generative Artificial Intelligence(2025-01-07) Mattson, Tom; Weng, Qin; Ren, JieWe investigate the level of empathy mental health posts receive on social media and generative artificial intelligence (GenAI). Specifically, we examine gender effects to determine if posts authored by self-identified men, women, or unknown (no self-identified gender discloser) receive varying levels of empathy across different technical platforms. Using a sample of mental health posts from Reddit, we find that self-identified women receive more empathy relative to men across all platforms. We further find that Inflection Pi, a GenAI tool specifically designed to be empathetic, provides the most empathy, but it still favors self-identified women over men. Self-identified men attempting to receive empathy for their prolonged emotional distress are disadvantaged relative to self-identified women.Item How Does Crowdsourced Fact-Checking Approach Tackling Misinformation Affect Audience Engagement? Evidence from Twitter’s Community Notes Program(2025-01-07) Zhou, Yingxin; Hou, Jingbo; Gao, Yi; Chen, Pei-YuThe spread of online misinformation is a significant issue, prompting the use of crowdsourced fact-checking. While its impact on audience engagement with fact-checked content is well-studied, effects on audience engagement with fact-checked authors remain underexplored. Using Twitter’s Community Notes program as our research context and applying counterfactual estimation, we examine how crowdsourced fact-checking influences audience engagement with fact-checked authors. Our findings reveal that community notes received by fact-checked authors can increase audience engagement with these authors in the short term, but they decrease audience engagement in the long term. This study enhances our understanding of how crowdsourced fact-checking influences audience response.Item Sharing Lies: A Socio-Technical Review on False Information Dissemination(2025-01-07) Walke, Fabian; Nürnberger, ThaddäaThis paper provides a comprehensive review of false information dissemination, focusing on factors influencing its spread in the context of social media and information systems. The study synthesizes recent literature to identify and categorize 30 influence factors into eight main categories: demographic, personality-related, psychological, policy- and values-based, informational, media consumption-related, motivational, and preventive factors. Key findings indicate that low education, high extraversion, and conservative values significantly increase false information dissemination. Additionally, social media usage, emotional responses, and information overload play critical roles in promoting its dissemination. Preventive strategies, such as labeling content and training in false information recognition, are also examined. This review aims to improve understanding of the dynamics of false information dissemination and proposes strategies to mitigate its impact.Item Introduction to the Minitrack on Bright and Dark Side of Social Media in the Marginalized Contexts(2025-01-07) Mattson, Tom; Ren, Jie; Weng, QinItem Don’t Forget Your Mask! Impression Management in Open Source Communities(2025-01-07) Fitzgerald, Brian; Chen, Liwei; Daniel, SheraeOpen source is increasingly present in all software delivered today, including that which is entirely commercial in origin. Thus, ensuring sustainability of open source projects becomes increasingly important. The situation is not helped by that the fact that historically the level of female participation in open source projects has been abysmally low, as are challenges to participation that can be experienced by those with different ethnic backgrounds. It has been suggested that as the involvement of commercial organizations in open source has increased, the level of female participation and ethnic diversity has increased. We investigated this issue with ten male and female practitioners from different geographic locations who have extensive experience in open source communities. Drawing on concepts from digital impression management, we investigated the manner in which these impression management techniques were used to participate in open source communities, and the differences between usage by male and female participants.