How do online users who share rumor and conspiracy theory-related content operate across platforms?
Cross-platform activity of rumor repeat spreadersUnderstanding how content sharing strategies and messaging differ by prominent spreaders of rumors across Twitter and alt-tech platforms using lexical similarity, topic modeling, and qualitative analysis
Misleading visual media in an online rumoring campaign
Characterizing the types of content, narrative strategies, and communities that share visual media to support election-fraud rumors and conspiracy theories using computer vision, network analysis, and qualitative analysis
Online rumor and conspiracy theory narratives on and across alt-tech platforms
Examined how election narratives differ with respect to platform distribution, affective harms, audience engagement, social identity dynamics, and mobilization potential using natural language processing, hypothesis testing, and qualitative analysis
Election Integrity Partnership
Conducted short-suspense threat analysis of the online political information environment and election rumoring on Twitter using quantitative and qualitative analysis
How do prosocial online interactions impact user beliefs and behaviors?
Constructive conversations countering conspiracy theoriesExamining how pro-social language cues impact engagement with direct replies and additional conspiracy theory content using natural language processing and time series analysis
How do online platforms contribute to user engagement and disengagement with rumors and conspiracy theories?
Motivators of conspiracy theory engagement, disengagement, and recovery onlineCharacterized online-mediated radicalization and recovery, resulting harms, and platform solutions surfaced through interviews with former conspiracy theory believers
QAnon conspiracy theory user engagement, spread, and activities after community bans
Measured and characterized QAnon Reddit users’ activity patterns, participation, and production of harmful content following community deplatforming using quantitative and qualitative analysis