Testing agenda building theory in a social mediated

Telematics and Informatics 34 (2017) 765–773

Contents lists available at ScienceDirect

Telematics and Informatics journal homepage: www.elsevier.com/locate/tele

Public relations, media coverage, and public opinion in contemporary China: Testing agenda building theory in a social mediated crisis Yang Cheng a,⇑, Yi-Hui Christine Huang b, Ching Man Chan b a b

School of Journalism, University of Missouri-Columbia, United States NAH Room 202, Chinese University of Hong Kong, Shatin, Hong Kong

a r t i c l e

i n f o

Article history: Received 22 August 2015 Received in revised form 19 March 2016 Accepted 5 May 2016 Available online 21 June 2016 Keywords: Public relations Time series Social media Agenda building Crisis China

a b s t r a c t This study investigates a social mediated crisis triggered by the Guo Meimei incident that negatively affected the Red Cross Society of China (RCSC). RCSC’s public relations efforts, media coverage, and public opinion on micro-blogs are examined through the theoretical grounding of agenda building. Evidence shows that organizational public relations activities strongly influence media coverage at the first level, but exert no influence upon online public opinion. RCSC’s agenda neither forms issue salience of online public opinion, nor effectively influences public’s attitude on the issue. This study extends the theorization of the effects of agenda building by adding contextual factors about social media, political, and cultural characteristics in China. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction This paper uses agenda building theory to examine specific forms and effects of public relations activities adopted by the Red Cross Society of China (RCSC) in handling a social mediated crisis. On June 21, 2011, netizens discovered the micro-blog of a 20-year-old woman named Guo Meimei on Sina Weibo. On this webpage she exhibited her wealth by posting photos of luxury possessions, including dozens of Hermes bags and a Maserati car. Guo identified herself as the ‘‘General Manager of the Red Cross Society.” According to data from Sina Weibo, the number of registered users exceeds 300 million. The users collectively generate more than 100 million posts daily (Xinhua Net, 2012). Within two hours, Guo’s post was retweeted a thousand times; within 24 h, this number rose to 100,000. Netizens attacked Guo fiercely for flaunting her wealth, and speculation rose as to whether her affluence was funded by public goodwill. Infuriated Internet users began a massive search for Guo Meimei’s personal information and any possible links to the RCSC. The incident aroused enormous criticism of the RCSC for its lack of transparency and cast doubts of financial misappropriation upon its leadership. Media organizations followed up shortly with their own reporting. As a result, furious accusations from both the public and the media led to a critical crisis initiated on social media for the RCSC and shook public trust in China’s charity enterprises. As reported, overall charitable donations from the public dropped 80% within the first few months after this social mediated crisis (Ma, 2011). The RCSC was unable to regain its trust from the public in the following two to three years that the donation

⇑ Corresponding author. E-mail addresses: [email protected] (Y. Cheng), [email protected] (Y.-H.C. Huang), [email protected] (C.M. Chan). http://dx.doi.org/10.1016/j.tele.2016.05.012 0736-5853/Ó 2016 Elsevier Ltd. All rights reserved.

766

Y. Cheng et al. / Telematics and Informatics 34 (2017) 765–773

continued to drop since 2011 (Lu and Hunt, 2015). In the 2014 Yunnan earthquake, the RCSC only received $7.65 million in monetary donations, which was a relative small amount given the large scale of the earthquake (Tiezzi, 2014). This study is a theoretical reflection upon traditional agenda-building theory, which originated in Western countries. It examines a case study drawn from an online environment in a Chinese context. The relevant theoretical implications of this study are as follows. First, this study examines the impact of public relations activities on media and public agenda, a combined effect few scholars have studied by exploring the interactions between organizational crisis communication activities, media coverage, and public opinion (Kiousis et al., 2006; Kiousis and Strömbäck, 2010). Second, this study uses an online context to extend the research that has explicated aspects of agenda building theory as it relates to traditional news coverage and organizational structure (Kiousis et al., 2007; Ku et al., 2003; Lancendorfer and Lee, 2010). Third, this study proposes a different methodology from existing agenda-building research that has mainly depended on data derived from cross-sectional investigation at a certain point in time (Kim et al., 2015; Rogers et al., 1993). As suggested by Conway and Patterson (2008), the self-reported survey data from such studies can be too suggestive or otherwise prone to error. In contrast, we collected public relations activities, online public comments, and news coverage in a longitudinal timeframe, using Granger causality analysis to draw inferences about the agenda building effects. Finally, this paper engages with appeals for more cultural and context-sensitive perspectives in communication and public relations scholarship (Choi and Cameron, 2005; Corley et al., 2006; Miike, 2007). By examining the agenda building theory in a unique crisis setting with specific political, social, and cultural characteristics pertaining to China, this study should shed light to previous discussion on conditional factors in the agenda building research (Curtin, 1999; Zhang and Cameron, 2003). 2. Literature review 2.1. The concept of agenda building The concept of agenda-building develops from the theory of agenda setting, which hypothesizes that the objects or attributes of these objects in the media agenda can significantly influence those in the public agenda (McCombs, 2006; McCombs et al., 2000). Prior scholarship of agenda building (Cobb and Elder, 1972; Lang and Lang, 1981) examined how the agenda building takes place through the interaction between media and other institutions such as governments and advertising agents, and how the media filters information from various sources. Gandy (1982) further developed the concept of agenda building and suggested that we should look beyond the theory of agenda setting and focus instead on ‘‘who sets the media agenda, how and for what purpose it is set, and with what impact on the distribution of power and values in society” (p. 266). In other words, agenda building studies examine how the media and public agenda can be influenced by outside forces such as public relations campaign messages (Kiousis et al., 2007), political candidates (Lancendorfer and Lee, 2010), and foreign governments (Zhang and Cameron, 2003). There are two levels where agenda building occurs. The first-level agenda building suggests the transfer of issue salience from the organization to the media and public agenda. The second-level agenda building contends that the attributes of the issue emphasized by the organization affect the salience of those attributes in media coverage and public opinion. Specifically, the attributes can be examined from the substantive or affective dimension. Substantive attributes such as the personality and policy position of candidates refer to the cognitive characteristics of an object (Kim et al., 2015), while affective attributes refer to the emotional characteristics of an object such as the tonality of an issue (i.e., positive, negative or neutral tone) (Carroll and McCombs, 2003). Numerous studies (Curtin, 1999; Kiousis and Strömbäck, 2010; Kiousis et al., 2006, 2007; Ku et al., 2003) have found that information subsidies influence not just which topics are covered by the media (first-level), but also how these topics are described (second-level). For example, in the 2002 Florida Gubernatorial election, press releases were used by political candidates to successfully shape news story contents at the first- and second-level of agenda building (Kiousis et al., 2006). In the state of Michigan, journalists were influenced to use certain attributes to portray objects in their news articles (Lancendorfer and Lee, 2010). 2.2. Agenda building and public relations Agenda building in the public relations field ‘‘refers to the sources’ interactions with gatekeepers, a give-and-take process in which sources seek to get their information published and the press seeks to get that information from independent sources” (Ohl et al., 1995, p. 90). Practitioners in public relations fields usually play the role of information supplier to journalists through interviews, press conferences, and news release. Furthermore, these ‘‘journalists respect their official sources, reporting what these sources tell them” (Gans, 2003, p. 46). Thus, the salience of the media agenda is influenced by the information subsides provided by public relations practitioners. According to Sweetser and Brown (2008), up to 80 percent of news coverages was originated from news releases provided to journalists. Ohl et al. (1995) found that news releases have a substantial agenda-building impact on the media coverage of corporate takeovers. Kiousis et al.’s (2007) research of 28 U.S. companies supported the positive relationship between the public relations efforts and media coverages on corporate reputation. Previous studies have also focused on how public relations practices influence public opinion (Carroll and McCombs, 2003; Cho and Benoit, 2005). For example, Kim et al.’s study (2015) showed that there was a strong correlation between cor-


767

porate public relations messages and consumers’ messages online at the second-level. Lang and Lang (1981) found that in the Watergate scandal, policy elites played an important role in influencing public opinions. 2.3. Agenda building and social media The process of agenda building has been complicated with the emerging of social media such as Wikipedia, Twitter, and Facebook (Kreiss, 2014; Parmelee, 2013). On the one hand, information subsidies can apply both traditional and new media to influence the media and public agenda, while the agenda-building effects differ among types of information subsidies (Kiousis et al., 2009). For instance, Harmon and White (2001) found that video news releases can raise the salience in subsequent media coverage. Parmelee (2013) found that factors such as the credibility of information sources and the availability of other information influence the agenda-building effects when political leaders used tweets to influence the news reporting. On the other hand, the interactive social media increased the public’s access to information and the power to set the media and organizational agenda (Luo, 2014). Online activities such as crowdsourcing or collective intelligence generate possibilities for intensive crises, which can be initialized or transmitted through social media (Cheng and Cameron, 2016). With the above mentioned changes, we argue that the traditional agenda building model needs a revisiting in a social mediated crisis context in China. Such first-level and second-level agenda building theories are mostly applied in democratic societies where interest groups and policy makers reach the public and influence their opinions by building mass media agendas (Berger, 2001; Kim et al., 2015; Kleinnijenhuis et al., 2013). Democratic conditions for agenda building, however, seem not to be present in the RCSC case given the social and political situations in China. This calls into question the relevance of agenda-building paradigms to scholarship on the contemporary Chinese media landscape. On the one hand, the RCSC is a quasi-official organization, and it has potentially powerful agenda-building abilities insofar as the state-controlled Chinese media help to shape its media agenda. However, the online environment may weaken public relations practitioners’ power to influence public opinion via traditional mass media. On the other hand, the RCSC crisis goes deeper than credibility: it also involves issues of trust and accountability. Researchers such as Fukuyama (1995) and Wang and Yang (2012) have designated China a ‘‘low-trust society,” which may even worsen the trustworthiness of the RCSC and further weaken its agenda-building potential. Therefore, the RCSC crisis initiated on social media first poses a challenge to the theory of agenda building as it is currently conceived. Moreover, it allows for a test of established theoretical links between organizational crisis communication activities, news coverage, and online public opinion. To recap, this study aimed to understand what the RCSC’s agenda-building process was in a social mediated crisis in China. Specifically, the first level agenda-building effects among public relations activities, media coverage, and online public opinion were tested in H1a and H1b. H1a. The amount of RCSC’s public relations activities will be positively related to the amount of media coverage the RCSC receives.

H1b. The amount of RCSC’s public relations activities will be positively related to the amount of online public opinion. Because the relationship between the RCSC and state-owned media is more evident than that between the RCSC and nonstate-owned media, it makes sense to posit H1c. H1c. The amount of RCSC’s public relations activities will be more strongly correlated with the amount of state-owned media coverage than the amount of coverage by non-state-owned media. The following H2a and H2b were proposed based on the logic of the second-level agenda building among RCSC’s public relations activities, media coverage, and online public opinion. Then H2c asks whether the RCSC’s agenda was more strongly related to the state-owned media agenda than the non state-owned one regarding attribute salience. H2a. The overall tonality of public relations messages will be positively related to the tone of the same issue in media coverage.

H2b. The overall tonality of public relations messages will be positively related to the tone of the same issue in online public opinion. H2c. The RCSC’s public relations activities will be more strongly correlated with the state-owned media than the non-stateowned media regarding attribute salience.

3. Method A quantitative content analysis was conducted for public relations materials, media coverage from newspapers, and public comments on micro-blogs. Data were collected over a period of 6 weeks (42 days) ranging from June 22 to August 2, 2011.

768


Major databases for data collection included WiseNews, Sina Weibo, and the RCSC’s official website, Weibo account, and ifeng.com. 3.1. Data collection and measures 3.1.1. RCSC’s public relations materials Data on the RCSC’s responses were collected from its official website, Weibo account, and a ‘‘Guo Meimei incident” column on ifeng.com. The ifeng website is owned by Phoenix Television and ranks as one of the top five most influential websites in Mainland China (China websites ranking, 2011). RCSC’s public relations efforts are represented by press releases, public statements, interviews, press conferences, and other announcements. In the 42-day period, starting from the first announcement on the RCSC website denying any connection with Guo Meimei on June 22 until August 2, the RCSC issued seven statements, held two media conferences, and released forty posts on its official micro-blog, and the top officials attended several media interviews. This resulted in a total of 66 organization activities. 3.1.2. News coverage To collect the news articles, WiseNews is used as the database, which covers more than 335 representative full-text newspapers in mainland China (Cheng et al., 2015). According to people.com.cn, newspapers in Mainland fall into two categories: state-owned (dangbao) or non-state owned (fei dangbao). In order to limit variables, data on media type were collected solely for examining the correlation of media coverage and media type. Total media coverage per day for state-owned and non-state-owned media were collected separately by keyword searches of ‘‘Guo Meimei” or ‘‘Red Cross” from June 22 to August 2. Of the total number (2376) of reports collected, every three reports were systematically sampled out of the pool, meaning a total of 792 news articles were used for the content analysis. 3.1.3. Public opinion Sina Weibo, as the China’s most popular micro-blogging platform, served as the database for collecting online public opinion (BBC, 2016). By entering the search string ‘‘Guo Meimei and Red Cross,” micro-blog streams were collected each day from June 22 to August 2, 2011. ‘‘Top News” and ‘‘New Weekly” were selected as sources for coding the public comments because both of them are formal news media organizations and have immense follower bases of 2951,622 and 4836,218, respectively. As a result these two accounts provide ready access to a large quantity of public opinion about current RCSC’s crisis. From June 22 to August 2, 2011, 34,840 comments were collected from both micro-blogs. A sampling ratio was set at 5% so that 1742 public comments composed the sample for content analysis. 3.2. Data coding, inter-coder reliability, and data analysis 3.2.1. VAR for Granger causality In order to analyze the salience transmission of the RCSC’s agenda building, data were analyzed through a time series analysis using the vector autoregression (VAR) model (Freeman et al., 1989; Sims, 1980), which has been widely applied in time series analyses in agenda-building research (Cheng, 2016a; Kleinnijenhuis et al., 2013). VAR is actually a set of multivariate regression equations. Current value of each variable is regressed on the lagged values of all the other variables in the analysis. In this study, the amount of RCSC’s public relations activities is regressed against its lagged values as well as the lagged values of all other variables. The same is done with the amounts of media coverage and public opinion. Each measure is a dependent variable in one of the three simultaneous equations. Beginning with organizational agenda and ending with media content, a direct causal pathway should form as long as the media can be relied upon as faithful transmitters. If the media do not act as faithful transmitters (i.e., if they act independently), then this pattern will either dissolve or reverse itself. The rapid pace of information exchange in online communities has also influenced data collection for this study. Weekly printed news articles or public statements of the RCSC are not likely to influence online discussions, instead individuals more often than not discuss issues within just a few days of coverage in the news media (Roberts et al., 2002). Although agendasetting research suggests that the optimal time lag ranges from four to five days for national network news to eight weeks for news magazines (Wanta, 1997), our time series analysis traces the influence of news coverage for time lags ranging from 1 day to 1 week. We have focused on this range because the time lag for traditional news media’s effect upon online commentary should be relatively short (Roberts et al., 2002). 3.2.2. Content analysis Content analyses of public relations materials, media coverage, and online public comments were conducted to investigate the second-level agenda building. Relationships linking the attitudes of public opinion and tone of media coverage to the RCSC’s tonality were studied using quantitative database searches for comments from Sina Weibo and newspaper reports on dates after the RCSC’s responses. Three independent coders received well training and a codebook was used in introducing coding guidelines. When disagreements occurred among coders, we discussed the discrepancies with them until they reached a resolution for coding discrepant items. The evaluative tone of the RCSC adopts in portraying the issue are categorized as positive (coded as 1), neutral (coded as 2), and negative (coded as 3). Accordingly, media tone and type are decided for analyzing media responses to the same issue.

769


State-owned newspaper is coded as 1, while non-state-owned newspaper is coded as 2. The tonality of media is categorized as positive (coded as 1), neutral (coded as 2), or negative (coded as 3). Only those contents concerning ‘‘Red Cross” in the Guo Meimei incident are considered as applicable data; otherwise, the data was coded as non-applicable (9). Public opinion on this issue are examined through content analysis of Internet users’ comments on Sina Weibo. The public’s attitude in each thread on Weibo is categorized as either positive (coded as 1), neutral (coded as 2), or negative (coded as 3). Otherwise, the data was coded as non-applicable (9). Three coders first coded the tone of media coverage. A ratio of 5% from the coded 30% sample (792 in total) was selected for inter-coder reliability. The result of composite inter-coder reliability for media tone reached 0.95 by applying Holsti’s (1969) formula. Similarly, all the RCSC responses were also coded and the reliability is 0.91. For the coding of public’s attitude, coders are required to take a test of 435 pieces of comments (25% of the total 1742 comments) and the composite inter-coder reliability for the tone of public opinion is 0.92. 4. Findings The first set of hypothesis investigated the RCSC’s agenda building effects on the media (H1a) and public (H1b) agenda at the first level. The arrow in Table 1 indicates Granger causality from the coefficient block in the second column to the dependent variable in the first column. Two key findings emerged from this analysis to answer H1a and H1b. For H1a, the organizational crisis communication responses significantly predicted the amount of news coverage (v2 = 19.24, p < 0.01), even after controlling for the effect of the public agenda. H1b concerned the influence of organizational salience on the public’s agenda. Results showed that after controlling for the effect of media agenda, no significant relationship was found (v2 = 2.44, p > 0.05, see Table 1). In summary, the results involving H1a and H1b indicated that the salience of issues in the RCSC’s crisis responses indeed led to the salience of issues in the media agenda, but not those in the public agenda. For H1c, which asked whether the relationship between the RCSC and state-owned media agenda was more evident than that between the RCSC and non-state-owned media agenda, a further VAR Granger causality test was run to identify the respective effects of the RCSC’s agenda on non-state-owned and state-owned media agenda at the first-level (see Fig. 1 for detailed daily counts of non-state media, state media, and organization time series). Table 2 shows that both non-state-owned media (v2 = 18.05, p < 0.001) and state-owned media coverage (v2 = 7.88, p < 0.05) were significantly shaped by the organization’s public relations efforts. The salience of an issue in the RCSC’s crisis responses led to the salience of issues in both state-owned and non-state-owned media agenda. However, although both relationships were statistically significant, the Chi-square (18.05) for non-state-owned media was larger than that for state-owned media (v2 = 7.88). Moreover, a post hoc analysis also revealed mutual influence between non-state-owned and state-owned media coverage, with a stronger influence of non-state-owned on state-owned coverage than vice versa. After controlling for influence from the organization, non-state-owned newspaper coverage (v2 = 14.33, p < 0.001) had a more immediate, stronger, and persistent influence on state-owned media coverage (v2 = 6.06, p < 0.05) than vice versa. In other words, non-state-owned media seemed to have stronger agenda-setting power in terms of transmitting the issue salience to state-owned media than vice versa. Thus, H1c was not supported. H2a and H2b tested the relationship among RCSC’s public relations activities, media coverage, and online public opinion at the second level. By applying the Spearman correlation analysis, results showed that the overall tonality of public relations messages was significantly related to the tone of media coverage (r = 0.25, p < 0.01). However, contradicting with the expectation, we found that when the RCSC’s messages became positive, the media tone became even more negative. Thus, H2a was not supported. H2b then asked how the RCSC’s tonality in public relations messages was related to public’s attitude. Findings revealed there was no existence of a fixed pattern of relationship between the organization and public agenda regarding the attribute salience (r = 0.08, p > 0.10). This means that no matter which tonality the RCSC adopted, the public’s attitude towards this issue remained negatively oriented. Thus, H2b was not supported. Finally, to test H2c, we split the relationship between the RCSC and state-owned versus non-state-owned media. Results showed that H2c was supported but in a different manner. The Spearman correlation test showed there was no existence of a fixed pattern of relationship between the RCSC and state-owned media agenda (r = 0.40, p > 0.05). This means that no matter how the RCSC performed public relations, the tone of state-owned newspaper coverage remained unchanged. In Table 1 VAR Granger causality tests examining the organization-public and organization-media relationships. Dependent variable

Coefficient block

Chi-square

p-value

Media

Organization Public

19.24 12.54

0.00 0.01

Public

Organization Media

2.44 3.95

0.49 0.27

Note: Arrows indicate Granger causality from coefficient block to the dependent variable. Three lags of each independent variable are included in the model. VAR results satisfy stability test; diagnostic tests indicate no evidence of residual autocorrelations (up to four lags); residuals are normally distributed; model residuals are white noise. N = 42. Analyses were done using EViews 5.

770


Fig. 1. Daily counts for non-state-owned, state-owned media, and the RCSC time series (June 22–August 2, 2011).

Table 2 VAR Granger causality tests examining the non-state-owned media coverage, state-owned media coverage, and the RCSC’s public relations activities. Dependent variable

Coefficient block

Chi-square

p-value

Non-state-owned media

RCSC State-owned media

18.05 6.06

0.00 0.05

RCSC

Non-state-owned media State-owned media

2.75 5.79

0.25 0.06

State-owned media

Non-state-owned media RCSC

14.33 7.88

0.00 0.02

Note: Arrows indicate Granger causality from coefficient block to the dependent variable. Two lags of each independent variable are included in the model. VAR results satisfy stability test; diagnostic tests indicate no evidence of residual autocorrelations (up to five lags); residuals are normally distributed; model residuals are white noise. N = 42. Analyses were done using EViews 5.

contrast, a certain pattern of relationship between non-state-owned newspaper tone and uses of crisis responses was found (r = 0.22, p < 0.05), i.e., when the RCSC began to adopt a more positive and accommodative attitude, the tone of non-state-owned media became more negative. 5. Discussion and conclusion This study adopted a dynamic and process-specific method to explore the agenda building effects in a non-democratic context. Researchers discussed multiple relationships among a quasi-governmental organization, the media and public agenda over an extended timeframe. Theoretical implications of the results were presented below. 5.1. The effect on media agenda The RCSC’s public relations efforts demonstrated statistically significant relationship with the media coverage. Results revealed that the RCSC’s public relations campaign significantly predicted the transmission of issue salience in both the state-owned and non-state-owned news media. Moreover, the RCSC’s tonality also appeared to be significantly related to the tone of media coverage in general. Although RCSC’s agenda had effects on media agenda, the way of the influence demonstrated a different pattern from the propositions of agenda building theory. That was when the RCSC’s attribute towards the issue moved from negative to positive, the tone of media coverage actually became more negative. In essence, the RCSC lost its power to influence the tone of media coverage. State-owned media tone remained negative, while non-state-owned media tone became even more negative as the crisis evolved. State-owned newspapers did not mirror the tone of the RCSC and showed no significant change in media tone. For the case of non-state-owned newspapers, results also revealed that the RCSC’s crisis responses were ineffective, and the media tone was even more negative when the RCSC adopted an accommodative and positive attitude towards the issue.


771

5.2. Comparison between state-owned and non-state-owned media The RCSC’s influence on state-owned media was larger than the influence it had on non-state-owned media at both levels’ effects. The pattern of influence, however, was subtle. State-owned media tended to report comparatively fewer negative stories than the non-state-owned media. This phenomenon can be explained by using Huang’s (2012) claim that a costreduction paradigm (rather than a revenue-generating paradigm) dominates during a crisis event when a favorable prior relationship exists between an organization and its stakeholders. With respect to the present case study, this could be due to the fact that state-owned media in China are still controlled by the government as part of the stability maintenance apparatus (Hassid, 2012; Shirk, 2011). Non-state-owned media may be asserting their limited autonomy from this media environment. This may explain why non-state-owned media waded into the RCSC controversy: not to defend the organization or appease public opinion but to criticize the government by crying for reforms of state-controlled NGOs. Moreover, it is worth noting that there is also mutual influence involving the first-level effect and the influence of non-state-owned on state-owned media is more evident. 5.3. Lack of effect on public agenda This study also found that the RCSC failed in influencing online public opinion. The RCSC’s agenda building on issue salience did not form the issue salience of the public agenda, nor did the RCSC’s tonality effectively influence the public’s attitude on this issue. No matter what strategies the RCSC adopted, the public’s attitude remained negative. The RCSC credibility crisis lasted for 42 days, and social sentiment varied little online. Instead of provoking anti-government movement or evolving into public protests, the Internet facilitated the emotional catharsis. 5.4. Contextual factors to extend agenda building As seen in the RCSC incident, prolonged online dispute, competing media coverage, intensive public engagement, and a substantive drop in public donations all shed light on a deeper understanding of the Chinese context for communicative action. This study extended the theorization of the effects of agenda building by adding information about Chinese political and cultural characteristics. In essence, this study demonstrated that the nature of a crisis event, new media environments, a country’s unique media landscape, a country’s political system, and Internet users’ power of collective action all helped determine the effects of agenda building. The RCSC’s public relations efforts seemed to have limited success with respect to the transfer of issue salience and attributes onto the media, let alone the public, in a crisis situation. The first factor influencing the overall media response (including state-owned and non-state-owned media) in the RCSC case in China could be the nature of the crisis event. The RCSC is not directly connected to political governance but is more related to social and public interests associated with the non-profit sector. The crisis itself is more of a management issue than a political issue. Thus, it could be inferred that if the political stakes are low, Chinese media may have a certain degree of freedom to take a strong journalistic stance. In post-Mao China in particular, government authority and the ruling party’s political agenda have been challenged by the existence of online forums (Cheng and Chan, 2015; Luo, 2014). Through discussion and interaction in cyberspace, Chinese netizens demonstrated their ability to reverse agenda-building strategies of political and corporate institutions (Cheng, 2016b,c). The second factor is the uniqueness of Chinese netizens and online communication behaviors. Compared with Twitter in Western societies, Weibo in China (a clone of twitter) is more like a highly virtual online playground than a news-reporting platform (Yu et al., 2011). Using Weibo the public devotes itself to a quest for information outside the control of traditional state-run media and central government. In this way, Chinese citizens can engage in new forms of civic participation as users construct or archive personal information as well as tag or edit media content (Goode, 2009). In the future social media such as Weibo will coexist with traditional media (Yin and Liu, 2012), but will exert an increasing influence on it as well as on civic participation. The third factor concerns the proposition that China is a low-trust society. China’s designation as a ‘‘low trust society” includes institutional trust as well as individualized and interpersonal trust (Cheng, 2016c). The RCSC crisis materialized in a distinctive sociocultural context in which low trust levels are pervasive. Low trust that occurred during the crisis may be a key factor undermining the RCSC’s crisis communication. Distrust of state-affiliated organizations intensifies the Chinese public’s dependence on the Internet as a source of information. These findings also correspond with Zhou and Moy’s findings (2007) that due to low levels of trust in Chinese media outlets, netizens are making judgments in ways exactly opposite to the media’s arguments. 6. Limitation and directions for future research Future research may engage more in-depth with stakeholder perspectives (Turk, 1986). Because information subsidies may include news releases, press conferences, interviews, and forms of new media, we cannot assume that these subsidies all have an equal relationship with both the media and the public agenda (Kiousis and Strömbäck, 2010). Thus, future research may take into account diverse forms of information subsidies and compare the differences between traditional

772


and new media communication effects in the public relations field, which should ultimately deepen our conceptual understanding of the agenda-building theory. Moreover, as the Guo Meimei case still evolved until 2014 when Guo herself was arrested and sentenced to jail as reported by the official media in China, future research may extend the timeframe of data collection and examine the agenda-building effects longitudinally. Declaration of conflicting interests The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. References BBC. (2016, January 20). Sina Weibo ends 140-character limit ahead of Twitter. BBC. Retrieved from: http://www.bbc.com/news/technology-35361157. Berger, B.K., 2001. Private issues and public policy: locating the corporate agenda in agenda-setting theory. J. Public Relat. Res. 13 (2), 91–126. Carroll, C., McCombs, M., 2003. Agenda-setting effects of business news on the public’s images of major corporations. Corporate Reputation Rev. 6, 36–46. Cheng, Y., 2016a. Who is leading whom in the General Motors recall: understanding media impacts on public relations efforts, public awareness, and financial markets. Res. J. Inst. Public Relat., 1–25, Retrieved from: http://wwwinstituteforpr.org/leading-general-motors-recall-understanding-mediaimpact-public-relations-efforts-public-awareness-financial-markets/. Cheng, Y., 2016b. The third-level agenda-setting study: an examination of media, implicit, and explicit public agendas in China. Asian J. Commun. 26 (4), 319–332. http://dx.doi.org/10.1080/01292986.2015.1130159. Cheng, Y., 2016c. Social media keep buzzing! A test of contingency theory in China’s Red Cross credibility crisis. Int. J. Commun. 10, 3241–3260. http://ijoc. org/index.php/ijoc/article/view/4895/1705. Cheng, Y., Cameron, G., 2016. The status of social mediated crisis communication (SMCC) research: an analysis of published articles in 2002–2014. In: Austin, Lucinda, Jin, Yan (Eds.), Social Media and Crisis Communication. Routledge. Cheng, Y., Chan, C.M., 2015. The third level of agenda setting in contemporary China: tracking descriptions of moral and national education in media coverage and people’s minds. Int. J. Commun. 9, 1090–1107, DOI: 1932–8036/20150005. Cheng, Q., Chang, S.S., Guo, Y., Yip, P., 2015. Information accessibility of the charcoal burning suicide method in mainland China. PLoS ONE 10 (10), e0140686. http://dx.doi.org/10.1371/journal.pone.0140686. China websites ranking, 2011. The comprehensive ranking of Chinese websites. Retrieved from http://www.chinarank.org.cn/overview/Info.do?url= ifeng.com. (In Chinese). Cho, S., Benoit, W., 2005. Primary presidential election campaign messages in 2004: a functional analysis of candidates’ news releases. Public Relat. Rev. 31, 175–183. Choi, Y., Cameron, G.T., 2005. Overcoming ethnocentrisms: the role of identity in contingent practice of international public relations. J. Public Relat. Res. 17 (2), 171–189. Cobb, B.W., Elder, C.D., 1972. Participation in American Politics: The Dynamics of Agenda Building. Allyn & Bacon, Boston. Conway, M., Patterson, J.R., 2008. Today’s top story? An agenda-setting and recall experiment involving television and internet news. Southwestern Mass Commun. J. 24, 31–48. Corley, K.G., Harquail, C.V., Pratt, M.G., Glynn, M.A., Fiol, C.M., Hatch, M.J., 2006. Guiding organizational identity through aged adolescence. J. Manage. Inq. 15 (2), 85–99. Curtin, P.A., 1999. Reevaluating public relations information subsidies: market driven journalism and agenda-building theory and practice. J. Public Relat. Res. 11, 53–90. Freeman, J.R., Williams, J.T., Lin, T., 1989. Vector autoregression and the study of politics. Am. J. Political Sci. 33, 842–877. Fukuyama, F., 1995. Trust: The Social Virtues and the Creation of Prosperity. Free Press Paperbacks, New York. Gandy, O.H., 1982. Beyond Agenda Setting: Information Subsidies and Public Policy. Ablex, Norwood, NJ. Gans, H.J., 2003. Democracy and the News. Oxford University Press, New York. Goode, L., 2009. Social news, citizen journalism and democracy. New Media Soc. 11, 1287–1305. Harmon, M.D., White, C., 2001. How television news programs use video news releases. Public Relat. Rev. 27, 213–222. Hassid, J., 2012. Safety valve or pressure cooker? Blogs in Chinese political life. J. Commun. 62, 212–230. Holsti, O.R., 1969. Content Analysis for the Social Sciences and Humanities. Addison-Wesley, Reading, MA. Huang, Y.H., 2012. Gauging an integrated model of public relations value assessment (PRVA): scale development and cross-cultural studies. J. Public Relat. Res. 24 (3), 243–265. Kim, J.Y., Kiousis, S., Xiang, Z., 2015. Agenda building and agenda setting in business: corporate reputation attributes. Corporate Reputation Rev. 18, 25–36. http://dx.doi.org/10.1057/crr.2014.18. Kiousis, S., Strömbäck, J., 2010. The white house and public relations: examining the linkages between presidential communications and public opinion. Public Relations Rev. 36 (1), 7–14. Kiousis, S., Mitrook, M., Wu, X., Seltzer, T., 2006. First- and second-level agenda-building and agenda-setting effects: Exploring the linkages among candidate news releases, media coverage, and public opinion during the 2002 Florida Gubernatorial election. J. Public Relations Res. 18, 265–285. Kiousis, S., Popescu, C., Mitrook, M., 2007. Understanding influence on corporate reputation: an examination of public relations efforts, media coverage, public opinion, and financial performance from an agenda-building and agenda-setting perspective. J. Public Relations Res. 19 (2), 147–165. http://dx. doi.org/10.1080/10627260701290661. Kiousis, S., Kim, S., McDevitt, M., Ostrowski, A., 2009. Competing for attention: information subsidy influence in agenda building during election campaigns. J. Mass Commun. Q. 86 (3), 545–562. Kleinnijenhuis, J., Schultz, F., Utz, S., Oegema, D., 2013. The mediating role of the news in the BP oil spill crisis 2010: How US news is influenced by public relations and in turn influences public awareness, foreign news, and the share price. Commun. Res. 42 (3), 408–428. http://dx.doi.org/10.1177/ 0093650213510940. Kreiss, D., 2014. Seizing the moment: The presidential campaigns’ use of Twitter during the 2012 electoral cycle. New Media Soc. http://dx.doi.org/10.1177/ 1461444814562445. Advance online publication. Ku, G., Kaid, L., Pfau, M., 2003. The impact of web site campaigning on traditional news media and public information processing. J. Mass Commun. Q. 80 (3), 528–547. Lancendorfer, K.M., Lee, B., 2010. Who influences whom? The agenda building relationship between political candidates and the media in the 2002 Michigan governor’s race. J. Political Marketing 9, 186–206. http://dx.doi.org/10.1080/2f15377857.2010.497737. Lang, G.E., Lang, K., 1981. Watergate: an exploration of the agenda building process. Mass Commun. Rev. 2, 447–468. Lu, S., Hunt, S., 2015. Infamous Chinese Social Media Celebrity Guo Meimei Sentenced. CNN, Retrieved from: http://www.cnn.com/2015/09/11/asia/chinaguo-meimei- charge/. Luo, Y.J., 2014. The Internet and agenda setting in China: The influence of online public opinion on media coverage and government policy. Int. J. Commun. 8 (24), 1289–1312, Retrieved from http://ijoc.org/index.php/ijoc/article/view/2257.


773

Ma, Q. 2011. Charity orgs see donations plummet in wake of scandals. People’s Daily. Retrieved from: http://english.peopledaily.com.cn/90882/7581246. html. McCombs, M., 2006. Setting the Agenda: The Mass Media and Public Opinion. Polity Press, Malden, MA. McCombs, M., Lopez-Escobar, E., Llamas, J.P., 2000. Setting the agenda of attributes in the 1996 Spanish general election. J. Commun. 50 (2), 77–93. Miike, Y., 2007. An Asia centric reflection on Eurocentric bias in communication theory. Commun. Monogr. 74 (2), 272–278. Ohl, C.M., Pincus, J.D., Rimmer, T., Harrison, D., 1995. Agenda-building role of news releases in corporate takeovers. Public Relations Rev. 21, 89–101. Parmelee, J.H., 2013. The agenda-building function of political tweets. New Media Soc. 16, 434–450. http://dx.doi.org/10.1177/1461444813487955. People.com.cn. 2006. News of Communist Party of China. Retrieved from: http://cpc.people.com.cn/GB/64162/116900/index.html. (In Chinese). Roberts, M., Wanta, W., Dzwo, T.H., 2002. Agenda setting and issue salience online. Commun. Res. 29 (4), 452–465. Rogers, E.M., Dearing, J.W., Bregman, D., 1993. The anatomy of agenda setting research. J. Commun. 43, 68–84. Shirk, S.L., 2011. Changing Media, Changing China. Oxford University Press, New York. Sims, C.A., 1980. Macroeconomics and reality. Econometrica 48, 1–48. Sweetser, K.D., Brown, C.W., 2008. Information subsidies and agenda-building during the Israel-Lebanon crisis. Public Relations Rev. 34 (4), 359–366. Tiezzi, S. 2014. Charities and corruption in China. The Diplomat. Retrieved from: http://thediplomat.com/2014/08/charities-and-corruption-in-china/. Turk, J.V., 1986. Information subsidies and media content: a study of public relations influences on the news. J. Monogr. 100, 1–29. Wang, J.X., Yang, Y.Y., 2012. Bluebook of Social Mentality. Social Sciences Academic Press, Beijing. Wanta, W., 1997. The Public and the National Agenda: How People Learn About Important Issues. Lawrence Erlbaum Associates, Mahwah, NJ. Xinhua Net. 2012. Sina microblog users topped over 300 million Posting over 100 million posts every day. Retrieved from http://news.xinhuanet.com/tech/ 2012-02/29/c_122769084.htm. (In Chinese). Yin, Y.G., Liu, R.S., 2012. Annual Report on Development of New Media in China. Social Sciences Academic Press, Beijing, China. (In Chinese). Yu, L., Asur, S., Huberman, B.A., 2011. What Trends in Chinese Social Media. Paper Presented at the 5th SNA-KDD Workshop’ 11. San Diego, CA. Zhang, J., Cameron, G.T., 2003. China’s agenda building and image polishing in the U.S.: assessing an international public relations campaign. Public Relations Rev. 29 (1), 13–28. Zhou, Y.Q., Moy, P., 2007. Parsing framing processes: the interplay between online public opinion and media coverage. J. Commun. 57 (1), 79–98.