Understanding your audience: The influence of social media user‐type on informational behaviors and hazard adjustments during Hurricane Dorian

• Published in: Meteorological applications 2023-09, Vol.30 (5), p.n/a
• Main Authors: Silver, Amber, Behlendorf, Brandon
• Format: Article
• Language: English
• Subjects: Atlantic Canada; behavioral response; Behavioural responses; Conduits; Evacuation; Hurricanes; information seeking; information sharing; protective action decision‐making; Questionnaires; risk perception; Social media; Storms
• ISSN: 1350-4827; 1469-8080
• DOI: 10.1002/met.2148

In 2019, Hurricane Dorian affected Atlantic Canada with widespread impacts across the region. In the days preceding landfall, there was a great deal of discussion about the storm and its potential impacts. This discussion also extended onto Twitter, which provided a platform for users to engage with storm‐related information. In this research, we disseminated a questionnaire to residents of Atlantic Canada from late September to late October through Qualtrics , an online survey provider. The questionnaire explored how Twitter influenced respondents' ( n  = 1218) self‐reported informational behaviors (i.e., searching, sharing, and processing) and behavioral responses before, during, and after the storm. The results demonstrate that users' informational needs and preferences were closely related to their online behaviors. For example, conduits (i.e., those who both searched for and shared information) were highly proactive users who disseminated information about evacuations, recommended protective actions, and other official guidance more so than others. Conduits were also the most likely to heed official guidance in terms of their own preparedness and response. Amplifiers (i.e., those who only share information) and consumers (i.e., those who only search for information) were also motivated to take action by information they saw online, albeit at lower rates than conduits. Lastly, the results demonstrate that users can be positively influenced by information they see online even if they do not actively engage with it. Taken together, the results of this study suggest that Twitter users may interact with storm‐related information in more nuanced and complex ways than previously understood.