Concept

“I Won the Election!”: An Empirical Analysis of Soft Moderation Interventions on Twitter Analysis of RQ3 - user interaction with tweets containing warning labels

  • Qualitative analysis of 3 kinds of tweets – tweets with warning labels that quote other tweets with warning labels; tweets with warning labels that quote other tweets without warning labels; tweets without warning labels that quote other tweets with warning labels
  • For group 1, tweets with warning labels that quote other tweets with warning labels, comments to quoted tweets generally had the purpose of correcting misinformation and providing accurate information
  • For group 2, tweets with warning labels because they tweet quotes without warning labels with questionable content, users were found to be confirming misinformation
  • Group 2 also demonstrated that Twitter is capable of updating and changing warning labels as they see fit, but these efforts are inevitably inconsistent
  • For group 3, tweets without warning labels quoting tweets with warning labels, tweets were discovered to be mocking the tweets and authors of tweets with warning labels, a small percentage of tweets actually tried to debunk misinformation, and an even smaller number of users tweeted to continue the spread of questionable content already labeled with warnings
  • 3 tweets in group 3 were about the need for hard moderation for tweets with warning labels and 1 tweet argued that Twitter was trying to dismiss election fraud through their use of warning labels

0

1

Updated 2021-05-18

Tags

CSCW (Computer-supported cooperative work)

Computing Sciences