Crisis communication on social media: A comparative study of COVID-19 tweets by four international public health organizations and four news media organizations and the public responses to the tweets (Preprint)

Author:

Song TingORCID,Ke Jiang,Yu Haiyan,Yecies BrianORCID,Yu PingORCID

Abstract

BACKGROUND

Social media platforms are crucial for public health and news media organizations in crisis communication with the public, as seen during the COVID-19 pandemic. However, limited research on how these organizations use social media for communication may inhibit their public relations and community engagement in managing future crises.

OBJECTIVE

This study investigates early COVID-19 communication by four public health organizations and four news media organizations on Twitter and the public responses to these tweets.

METHODS

A mixed qualitative and quantitative data analysis was conducted. An unsupervised learning approach based on a convolutional neural network was used to classify the 15,711 tweets crawled from Twitter according to their meaning. The top 30 tweets in the validated 37 clusters were subjected to in-depth thematic analysis. Public responses to different content and communication strategies were then compared using descriptive statistics.

RESULTS

Six topics were identified: (1) case updates, (2) medical research and treatment information, (3) health instructions and suggestions, (4) impacts and consequences, (5) policies, methods and action, and (6) opinions and responses. Public health organizations tweeted most about "health instructions and suggestions" (52.6%) and least about "impacts and consequences" (2.7%). News media organizations tweeted most about "policies, methods and action" (28.1%) and least about "health instructions and suggestions" (3.6%). Public health organizations’ tweets about "medical research and treatment information" attracted the highest replies/tweet (24.4%), retweets/tweet (27.2%), and likes/tweet (27%). News media organizations’ tweets about "opinions and responses" attracted the highest replies/tweet (33.5%) and likes/tweet (23%) and "medical research and treatment information" attracted the highest retweets/tweet (21%). Both types of organizations received the fewest replies/tweet (10.6% vs 6.7%), retweets/tweet (9.4% vs 8.6%), and likes/tweet (8.8% vs 8.8%) about "impacts and consequences". The public was receptive "opinions and responses" for likes and replies for and "health instructions and suggestions" for retweets. They responded much less to "impacts and consequences" and "policies, methods and action". Five communication phases were identified: inception, awareness, panic, spreading, and co-habitant. For communication strategies, more hashtags used led to fewer public responses in the format of replies, retweets, and likes. Longer tweet text led to fewer retweets but no significant effect on replies or likes. An exception is that longer tweet text resulted in more retweets and likes about "medical research and treatment information", and more replies, retweets, and likes about "health instructions and suggestions".

CONCLUSIONS

Facing a public health crisis like COVID-19, the public engages more with social media content about pandemic updates and health instructions. To improve communication, more concise language and fewer hashtags may be used in when reporting factual pandemic information. Conversely, incorporating comprehensive details and a higher frequency of hashtags may be advantageous when providing health instructions.

Publisher

JMIR Publications Inc.

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3