Exploring Conceptualizations of COVID-19 Risk in Ideologically Distinct Online Communities: A Computational Grounded...

Area of Science:

• Social Psychology
• Public Health Communication
• Computational Social Science

Background:

• COVID-19 pandemic highlighted the subjective nature of risk perception, influenced by social interpretations.
• Online communities played a significant role in sense-making during the crisis, offering insights into collective risk interpretation.
• Understanding how online groups construct risk is crucial for effective public health strategies.

Purpose of the Study:

• To explore how distinct online communities develop and reinforce divergent risk conceptualizations.
• To identify the role of group identity in shaping risk interpretation and communication within these communities.

Main Methods:

• Computational grounded theory approach analyzing ~500,000 comments from Reddit.
• Latent Dirichlet allocation (LDA) for topic modeling to identify key themes.
• Qualitative thematic analysis and Word2Vec word embedding for validation.

Main Results:

• Divergent risk concerns identified: Masks4All focused on health threats, while LockdownSkepticism prioritized economic and social risks.
• Group identity significantly shaped risk perceptions, with members reinforcing in-group positions and out-group condemnation.
• Both communities centered risk but interpreted its facets differently based on their ideologies.

Conclusions:

• Online communities exhibit distinct, ideologically driven risk perceptions of the same event (COVID-19).
• Group identity reinforces these divergent interpretations, necessitating tailored public health interventions.
• Engaging with diverse ideologies and value systems is essential for effective risk communication.