DISASTER COMMUNICATION PATTERNS AND BEHAVIORS ON SOCIAL MEDIA: A STUDY SOCIAL NETWORK #BANJIR2020 ON TWITTER

Abstract

Purpose of the study: The purpose of this research is to analyze the disaster communication patterns and behaviors of Twitter users. Flood disaster in the Jabodetabek area became an unexpected event in early 2020. The flood inundated 23 areas in Bekasi, two regions in Bogor, and 17 areas in Jakarta. Information about floods became a trending topic on the 1st of January 2020. Methodology: The method used is social network analysis and text analysis #Banjir2020 on Twitter, using Netlytic and Gephi. The sample analyzed 1000 tweets from 304 users and 670 edges. The data was selected from the 10th to 13th of January 2020. Netlytic.org limits that we can only retrieve tweets data from Twitter for less than 2 weeks due to API limitations. Main Findings: The result shows that #Banjir2020 disaster communication patterns on Twitter formed five significant clusters on its network. The communication occurred as one-way communication. A low level of network density showed that the quiet intensity of communication and slow information to be able to spread throughout vast networks. Every twitter user involved can directly provide information to others. Judging from the messages conveyed, the most formed behavior is the behavior of information dissemination regarding this flood. The next significant response is the defense of DKI Jakarta Governor. Implications of this study: The disaster communication behaviors on #Banjir2020 is dominated by flood disaster information in some areas. Communication patterns form vast networks but still lack in terms of intensity, two-way communication, and slow information to move throughout the system. Novelty/Originality of this study: The research of #banjir2020 through Twitter using the analysis of SNA and disaster communication behavior has never been done by other researchers.