Beyond organization and reputation: A case study of government responses and resulting online word-of-mouth by multiple stakeholders to post-Fukushima food imports

We examine the formation of thematic patterns embedded in online governmental crisis response strategies and the resulting online word-of-mouth over food imports after the Fukushima nuclear disaster. We ask: How did major themes embedded in each strategy and online word-of-mouth form, evolve, and reach closure? What are the semantic characteristics of the online word-of-mouth across social media publics, online news media, and search engines? Following Hallin and Mancini (2004), we compare results from Hong Kong, a “democratic corporatist” society, with those from Mainland China, a “polarized political” society, and Taiwan, a "liberal model.”

Using a mixed-method case study, we identified 1) themes underlying governmental crisis response strategies using thematic analysis; and 2) the network of associations between concepts expressed in texts online word-of-mouth through computer-assisted semantic network analysis focused on debates on online news, search-based contents and social media messaging among different stakeholder groups over imports of Japanese food. We compiled an organizational corpus which included 17 Centre for Food Safety press releases from Hong Kong, 21 from Mainland China’s National Center for Food Safety Risk Assessment, and 20 Taiwan Food and Drug Administration press releases. The word-of-mouth corpus included more than 10,000 online comments from each region.

By following Braun and Clarke (2014)’s guidelines to thematic analysis in health risk and wellbeing research based on grounded theory tradition (2006; p.87), we 1) transcribed and re-read data for initial ideas; 2) generated codes to identify key features; 3) searched for themes by collating into potential categories; 4) reviewed themes to align coded data with themes; and 5) defined and named themes. Semantic network analysis (SNA) is a form of content analysis that identifies the network of associations between concepts expressed in texts. By taking advantages of network analysis software UCINET and Gephi, SNA was conducted to provide a representational framework to identify the relational structure of specific associations between the expressed language.

This study extends SCCT (Coombs, 2007; 2016) to interpret constructions of meaning in crisis response strategies and by incorporating a multi-stakeholder perspective and suggesting SCCT outcome variables beyond reputation.