@icsjournal.bsky.social #iCS

🙏 Huge thanks to @hannahgreber.bsky.social, @svenjaschäfer & @solecheler.bsky.social for invaluable feedback, and to @danielwiesner.bsky.social for the pretty graphs 🙂

5/
TL;DR: Users want deletion for insults, dialogue for misinformation — and trust humans more than AI when it comes to moderating online content.

4/
Key insights:
🗑️ For insults → deletion was preferred
💬 For misinformation → interactive strategies (guideline reminders, counterspeech) were favored
👥 Human moderators were trusted more than AI
⚖️ Individual values matter: opposition to censorship + right-leaning identity = lower acceptance

3/
I ran a factorial survey experiment with Austrians tested the acceptance of moderation depending on:
- type of comment (insults vs. misinformation)
- source (human vs. AI)
- transparency (explanation vs. none)
- moderation strategy (deletion, warning labels, guideline reminders, counterspeech)

2/
The paper asks: How should platforms moderate online comments — especially when they are insulting or misleading but still fall under free speech?

Thank you 😭

Tackling Online Incivility and Misinformation: A User-Centric Evaluation of Different Content Moderation Strategies
Andrea Stockinger @andiisto.bsky.social l
Communication, University of Vienna, Wien, Austria @ipk-univie.bsky.social #ica25

Congratulations on this well-deserved recognition! 👏🏼✨