(WITHDRAWN) Framing Disaster Via Twitter: Utilizing Social Media for Social Work Research

Background/Purpose

This presentation focuses on an analysis of the disaster discourses generated by key social media user groups via Twitter in the aftermath of Hurricane Maria in Puerto Rico. Hurricane Maria, the third most destructive hurricane in American history, resulted in billions of dollars in damage and the loss of nearly 3,000 lives. The discourses that emerge on social media after disasters are significant in how they frame public narratives. The social construction of disaster points to the contested nature of these narratives as they vie for dominance in the public sphere, including social media communicative spaces. The literature suggests that there are numerous key interpretive communities and narratives present at any given time. This study focused on six of these groups (individuals, government, military, media, nonprofits, and others) and their corresponding disaster narratives. By utilizing a social constructionist/critical theoretical framework, the prevalent frames embedded in the disaster discourses were identified. The research questions that guided this study were: 1) what disaster and recovery narratives are constructed by key user groups (individuals, government, military, media, nonprofit, and “other”) via Twitter using the hashtag #HurricaneMaria; and 2) How do the post-Maria disaster and recovery narratives produced by the six key user groups converge and diverge?

Methodology

An instrumental case study design was utilized, bounded by timeframe (September 21, 2017 to September 21, 2018) and location (focusing on the impact of the hurricane in Puerto Rico). Purposeful sampling was used to collect tweets containing the hashtag #HurricaneMaria, accessed through the Twitter website. The final dataset included 1,296 tweets. Thematic content analysis was conducted, using qualitative, interpretive methods (Braun & Clarke, 2001) and quantitative methods including frequency calculations and sentiment analysis via the sentimentr package in R.

Results

Five prevalent frames or themes were identified in the corpus of tweets. These frames include the political frame, destruction frame, victim/hero frame, military/humanitarian aid frame, and counter-narratives. Different user groups were active in constructing these frames in the post-Maria discourse. Counter-narratives and survivor stories were notably scarce in the Twitter discourse. Sentiment analysis revealed differences in the tone of tweets (positive, negative, or neutral) posted by the six user groups over the one-year time frame. Different patterns of Twitter activity were also evident among the six user groups, including peaks during first week of the disaster and at the one-year anniversary.

Conclusions/Implications

In post-disaster situations, the frames produced by key actors can be influential in defining problems, causes, responsibility, and responses or solutions (Entman, 1993). In this study, the competing frames constructed by the six user groups told different stories of Hurricane Maria, emphasizing some aspects of the disaster and ignoring others. Social workers, by virtue of our positionality can promote alternative frames and advocate for the voices of marginalized communities to be included in policy actions and decision-making in the aftermath of disasters. This study also illustrates the potential of social media as a rich data source for social work research, policy practice, and advocacy.