Contextual awareness on organizational crises: National context and crisis attribution

Abstract

This study examines the influences of national contexts on crisis attribution to address the recent appeal for a more contextually sensitive perspective in crisis communication research. Specifically, this study revisits the situational crisis communication theory (SCCT) in the Chinese context by taking an online discussion on Weibo about a high-profile homicide in a McDonald’s restaurant as the case. Built on the framing theory and categorization of national contexts, an inductive framing analysis of 100 top forwarded posts demonstrates a complex negotiation process of context-embedded frames and its significant impacts on crisis attribution. The model of SCCT is then refined by integrating national contexts into the concept of ‘modifier’. Implications for both theory and practice are also discussed.

Keywords

China context crisis attribution organizational crisis situational crisis communication theory social media

Crisis communication scholars have generated a substantial body of scholarship to investigate the influences of national context on crisis communication (e.g. Frandsen and Johansen, 2010; Kim et al., 2016; Meyers and Rozen, 2014). By viewing crises as social constructs that emerge in the social negotiation process (Bundy et al., 2017; Zhao et al., 2017), the contextually sensitive crisis communication research draws particular attention to cultural (e.g. Taylor, 2000), political (e.g. Dai and Hyun, 2010), technological (e.g. Falkheimer and Heide, 2009), and historical (e.g. Huang et al., 2016) contexts that determine specific crisis communication practices. Researches on the national context’ impacts on crisis communication have advanced the theoretical knowledge of crisis communication in the global context (Schwarz et al., 2016). Among them, crisis communication studies in non-Western contexts (e.g. Barker, 2011; Huang and Bedford, 2009; Meyers and Rozen, 2014) in particular investigate the external validity of crisis communication theories originated in the Western context. Although the non-Western perspective has contributed distinct value to crisis communication scholarship, most extant researches maintain their foci on crisis response strategies (e.g. An et al., 2010; Haruta and Hallahan, 2003; Jiang et al., 2015; Kim et al., 2016) and hence have exclusively enriched crisis response repertories that are applicable in non-Western contexts (Cheng, 2016a; Veil et al., 2011). Although fruitful, research testing the contextual influences on crisis attribution is limited. Crisis attribution is considered as a crucial factor in understanding the reputation threat posed by crisis situation (Claeys and Cauberghe, 2014; Ma and Zhan, 2016; Roh, 2017). A specific focus on the interplay of national contexts and crisis attribution would contribute to the accurate evaluation of crisis situation, as well as the success of crisis communication.

This study addresses the above issue by revisiting the concept of ‘modifier’ in situational crisis communication theory (SCCT) (Coombs, 2014) in the Chinese context. SCCT acknowledges the fundamental role of crisis attribution in determining both crisis situation and the selection of crisis response strategies. This study aims to refine the conceptional model of crisis attribution in SCCT by integrating the impacts of national contexts. The Chinese context is chosen as a case. A somehow unique regime type and the patterns of development as well as the profound transformation in various aspects of the Chinese society (Tang, 2016) make the Chinese context different from contexts elsewhere, including the exceptional economic development path, the unprecedented transition from an agricultural to a manufacturing country, and the prominent roles of historical, cultural, and traditional factors. Given such complexity, the Chinese context is a valuable object of study that calls into question many existing assumptions, conceptions, or models of crisis communication originated in the Western context (Huang et al., 2016; Huang and Kim, 2018).

In the first part, after situating this inquiry in the literature on SCCT and context-oriented crisis communication research, this study points out that both the concept of ‘modifier’ and its impacts on crisis attribution have restricted the interrogation to the organization domain. To address this gap, this study subsequently proposes a combined theoretical framework, which consists of Child’s (2009) categorization of national contexts and framing theory (Scheufele, 1999). Inductive framing analysis is adopted to investigate whether, how and to what extent national contexts influence the framing process of crisis. The findings from this study help fill a critical gap in discerning contextual influences on crisis attribution by integrating insights from a non-Western country to advance Western-originated crisis communication theory and by offering organizations practical insights that can be employed to infuse a contextual awareness of crisis attribution into their crisis communication plans, practices, and processes.

Theoretical framework

The determinant and change of crisis attribution

SCCT (Coombs, 2014) adopts an attribution-centered approach, which suggests attributed responsibility is the essential factor accounting for the reputation threat posed by crisis situation. Attribution theory (Weiner, 1986) is the root of SCCT, which offers the idea that people search for causes of unexpected and negative events. By applying the attribution theory–based idea to a broader array of crises, SCCT defines crisis attribution as crisis responsibility – the degree to which publics perceive that an organization should be responsible for a crisis (Coombs, 2007a). Coombs (1998) distinguishes two dimensions of crisis attribution: personal control and external control of crises. While the personal control of causality concerns with the extent that stakeholders believe organizational actions caused the crisis, the external control refers to the responsibility of external agents (Coombs, 1995). Based on crisis attribution, SCCT categories crises into three types: (1) victim type has low crisis attribution and the organization is viewed as a victim of the event; (2) accidental type has medium crisis attribution and the organization’s misdeed is viewed as unintentional or uncontrollable; and (3) intentional type has strong crisis attribution and the organization’s behavior is considered to be purposeful (Coombs and Holladay, 2002). Crisis managers could use the assessment of crisis attribution to decide crisis response strategies (i.e. deny, diminish, and rebuild).

Coombs (2006) further notices that crisis attribution is not fixed during crisis communication. He coins the concept of ‘modifiers’ to identify variables that ‘… can alter attribution generated by the crisis type’ (Coombs, 2006: 162). Modifier implies the unstable character of crisis attribution. Following his argument, existing studies identify performance history (e.g. Coombs, 2014; Coombs and Holladay, 2001; Griffin et al., 1991) and prior reputation (e.g. Coombs, 2007b; Coombs and Holladay, 2002) as modifiers. Although existing studies provide valuable insights into the concept, the research scope remains constrained to the organizational domain. The impact of external factors beyond the organization is largely overlooked.

National contexts and their influences on crisis communication

Most crisis communication studies focus on organizational contexts. For one thing, they explore how organizational contexts influence crisis response strategies (e.g. Cancel et al., 1997; Massey, 2001). For another, they investigate how crises influence the organizational contexts (e.g. Pang et al., 2006, 2010). Some researchers contend that contextual consideration in crisis communication should include not only the internal contexts but also the external contexts (e.g. Gilpin and Murphy, 2008; Schultz and Raupp, 2010). Frandsen and Johansen (2017) suggest the necessity of broadening the contexts to the macro-level, including politics, economics, media systems, and cultural traits. The extant literature demonstrates the contextual divergence and its impact on crisis communication. Countries and contexts that have been studied include Canada (Greenberg and Hier, 2001), Belgium (Johnson and Peppas, 2003), Sweden (Falkheimer and Heide, 2006), Denmark (Frandsen and Johansen, 2010), Israel (Meyers and Rozen, 2014), Japan (Tanifuji, 2000), Korea (Kim et al., 2008), and so on.

The uniqueness of Chinese contexts has drawn increasing attention from crisis communication scholars (e.g. Lyu, 2012b; Wu et al., 2016; Zhao et al., 2017). For example, Huang et al. (2016) reveal the following features of crisis communication in Chinese societies: collectivistic culture, nationalism, rationalism, face-giving/saving, striving for the ‘golden mean’, the preference for passive communicative strategies, and the avoidance of extreme strategies. Zhao et al. (2017) identify an authoritarian government structure and a digital transformation of society as the main features of the Chinese context. The existing literature explores the influences of these Chinese contexts on different aspects of crisis communication, such as roles of crisis communicators (e.g. Bowen and Heath, 2007; Chen, 2009; Lee, 2004), crisis response strategies (e.g. Huang and Bedford, 2009; Jiang et al., 2015; Kim et al., 2016; Lyu, 2012a), and effectiveness of crisis communication (e.g. Cheng, 2016b; Lee, 2004). Still, the relationship between the Chinese context and crisis attribution maintains unsolved.

Crisis frame as an intermediary between crisis attribution and context

Framing is an intrinsic part of SCCT (Coombs, 2007b), in which crisis communication is conceptualized as a process and various social actors utilize crisis frames to define the problems, interpret causality, and assign solutions. Moreover, framing has been recognized to take on a crucial role in the evolution of crisis situation (Liu and Kim, 2011), the effectiveness of crisis response strategies (Gerken et al., 2016), and the attribution of crisis responsibility (Schultz and Raupp, 2010). More importantly, crisis frame functions as an intermediary between context and crisis attribution. On one hand, the evaluation of crisis attribution is relied on ‘the frame used to interpret the crisis’ (Coombs, 2006: 203). Crisis frame hence determines crisis attribution. On the other hand, the generation, facilitation, and modification of frames are influenced by contextual factors (Benford and Snow, 2000). Accordingly, crisis frame can be viewed as an indicator of contextual influences.

Prior framing studies (e.g. Gans, 1979; Shoemaker and Reese, 1996; Tuchman, 1978) illustrate that social-structural contexts (e.g. political, cultural, social values) exert significant influences on frames. The contextual influences lead to the ‘unstable’ nature of frames (e.g. Van der Meer et al., 2014) and the multiple-frame condition (e.g. Zhao, 2017) in times of crisis. In this sense, the complex nexus of competing frames is inherent in crisis communication. Also, we need to consider the multiple-frame condition as a framing process (Scheufele, 1999) within which multiple frames are in contentious and confluent interaction with each other. Framing is contentious in the sense that it involves the generation of interpretive frames that not only differ from existing ones but also challenge them (e.g. Benford and Snow, 2000). It is confluent as framing involves power negotiation among multiple frames that result in the alignment of different frames (e.g. Van der Meer et al., 2014). The dynamic framing process reflects the contextual constraint and facilitation of crisis communication, where contextual factors influence, alter, or even formulate various frames (Benford and Snow, 2000).

The identification of national contexts

By considering the Chinese context as context-speciﬁc or context-bounded, Child (2009) develops a categorization to conceptualize and measure national contexts. According to Child (2009), the major components of national contexts are material and ideational systems. Both systems are external to organizations but have significant impacts on organizations. Child (2009) further divides the material system into economic and technological ones and categorizes the ideational system into cultural values and rationales, religious values, and political values. Following the categorization, this study further defines national contexts as external factors beyond the organizational domain that can be identified from a national material system and ideational system.

A combined theoretical framework is employed to investigate whether, how, and to what extent national contexts impact the crisis attribution (RQ). As discussed, this study adopts Child’s (2009) categorization of national contexts as the framework to identify the main components of contexts in which crisis communication evolved as the first step. Next step, framing theory (Entman, 1993; Gamson, 1992) is integrated into the theoretical framework. With a combination of these above two, the overall goal of the theoretical framework is to reveal how and to what extent the factors identified from Child’s categorization of national contexts influence the crisis attribution that is determined by the crisis frame. Accordingly, the following sub-questions guide this study:

RQ1. What are the different frames on social media in an organizational crisis?

RQ2. How does crisis attribution change during crisis communication?

RQ3. How do national contexts influence the crisis frame and further alter crisis attribution?

Methodology

Case selection

This study examines an online discussion on Sina Weibo (hereafter, referred to as Weibo) about a homicide case that occurred in Zhaoyuan, China on 28 May 2014 at a McDonald’s restaurant. According to the report (Liu, 2014b), the victim, a 35-year-old woman, was beaten to death while waiting for her 7-year-old son in the McDonald’s restaurant. While the victim was in the McDonald’s restaurant, six members of a religious group known as ‘The Church of Almighty God’ were trying to recruit new members. After the victim refused to give them her cell phone number, this group of people beat her to death with chairs and a metal mop handle. The crime was captured by both CCTV cameras and witnesses’ smartphones. The footage quickly went viral online and attracted significant attention from various social groups (Ramzy, 2014). The homicide occurred in commercial premises, a large amount of online discussion focused on why the McDonald’s staff did not intervene. Consequently, many people began to question McDonald’s responsibility for the murder, discussions that significantly threatened or deteriorated McDonald’s reputation. Crisis communication scholars define organizational crisis as a perception that an organization has committed some wrongdoing (cf. Coombs, 2014; Hearit and Courtright, 2003). As such, this case is an appropriate test for the SCCT model of organizational crisis.

Data collection

This study began gathering data at 10:00 on 6 June 2014 from the Weibo search engine (http://s.weibo.com/) by choosing ‘advanced search’. ‘招远’ (Zhaoyuan) and ‘麦当劳’ (McDonald’s) were used as key words as this combination of search terms refers to the location of the murder. According to the report released by Media Opinion Monitoring Office (Liu, 2014a), this study chooses between 21:00 on 28 May 2014 and 00:00 on 6 June 2014 as the timeframe, which covered the event from the outbreak to the end. The top forwarded 100 posts were collected from Weibo for coding.

Framing measures

To investigate the contextual impacts on crisis attribution, this study used inductive framing analysis that codes the framing elements (frame device and reasoning device) directly from the content and provides an outline of a frame package (Gamson, 1992; Van Gorp, 2005). A frame package is a cluster of logically organized devices that function as an identity kit for a frame. This integrated structure includes a core frame, framing devices, and reasoning devices.

Inductive framing analysis entails an explanatory procedure (e.g. Gamson, 1992; Van Gorp, 2005) that is appropriate for the inquiries of this study as it holds an open view to reveal the range of possible frames. Because the present study regards frames as dependent variables, each frame that emerges during the crisis communication encompasses a certain context. Therefore, detecting the context-embedded frames to the utmost ensures that researchers do not omit any possible impacts from the contexts. In addition, inductive framing analysis unfolds the negotiation process of crisis framing. As inductive framing analysis demonstrates when, how, and by whom an event has been framed, researchers can observe the complexity of crisis framing, including changes of crisis attribution, and can further scrutinize how context facilitates the process.

Analysis procedures

Two junior researchers conducted the coding processes in four sequential steps. There was not an explicit test of intercoder reliability in this study. As this study takes an inductive approach to framing analysis, we decide to keep the interpretation process as open as possible. Nevertheless, we did discuss the general principles as so two coders reached a broad agreement with regard to identifying the elements in the empirical materials. The analysis proceeded through the following five sequential steps: the first step was to identify framing devices, namely the demonstrable indicators of the frame (Gamson and Modigliani, 1989). Two coders applied an open coding system and did initial coding (Charmaz, 2006) separately. Two datasets (including citations and illustrations) were generated from the posts: one consisted of 751 citations and the other consisted of 763 citations. The independent but complementary coding process is to ensure that all possible framing devices are considered. Subsequently, the two datasets were combined to develop 100 key terms by merging synonyms, similar arguments, and so on. These key terms were used as a manifest of framing devices. To make sure the exhaustion of framing devices in the data, the analysts constantly scanned and compared the research material. Next, the two coders connected the reasoning devices to framing devices. They looked for the reasoning device as long as an element in the inventory suggested causal reasoning (Gamson and Modigliani, 1989). The operational elements of reasoning devices consist of problem definition, causal interpretation, moral judgments, and treatment recommendation (Matthes and Kohring, 2008), which is based on Entman’s (1993) definition of framing. Axial coding (Charmaz, 2006) was adopted to determine the patterns in the inventory. Two coders worked together to group the elements in the inventory by comparing similarities, differences, and contrasts. Then core frames were synthesized based on inherence and coherence among the framing devices and reasoning devices. Based on a previous study (Zhao, 2017), a matrix was constructed to present frame packages as the last step. In this revised matrix, the row represents the core frames and the column represents the framing devices and reasoning devices (Table 1).

Table 1.

Frame matrix for the online discussion on Weibo about a homicide case in McDonald’s.

Core frames	Sub-frames	Reasoning devices				Frame devices
Core frames	Sub-frames	Problem definition	Causal interpretation	Moral judgments	Treatment recommendation	Verbal devices	Visual devices
1. Problem identification frame	1.1 Ordinary criminal case	Criminal homicide	The murderers beat the victim to death	The brutal and bloody acts of the assaults defy all comprehension	Solving the crime and punishing the murders	Quarrel, molestation, died despite efforts, arrest, sub-judice	the murderers were beating the victim; police were investigating the crime scene
1. Problem identification frame	1.2 Religion-related criminal case	Criminal homicide	The members of cult beat the victim to death	The brutal and bloody acts of the cult’s members	Launching a special crackdown on cults	Religious cult, victim, suspects, crackdown	Television interview of the murderers
2. Attribution frame	2.1 Government	Access to public information; statutory construction	The reason the assailants were so unbridled is that the authorities take part with dignitaries	The untransparency and unfairness of the trial	Improve relevant laws and regulations	Government, police, duty, handle the case according to laws, official, nomocracy, death penalty	Screenshot from government microblog account
2. Attribution frame	2.2 Corporation	The obligations of corporations; the rights of consumers	The poor security and the nonfeasance of the staff led to the tragedy	The nonfeasance of McDonald’s staff	McDonald’s should improve security and compensate the victim’s family	Consumer, staff, security personnel, nonfeasance, McDonald’s, surveillance video, suspense of business	The picture of McDonald’s restaurant; McDonald’s logo
	2.3 Citizen	Criminal homicide	The victim was dead because no one lent a hand	The indifference of bystanders	Citizens need to pressure authorities by expressing their opinions	We, strangers, bystanders, silence, ignorance, assistance, obligation	the murderers were beating the victim; police were investigating the crime scene
	2.4 Media	Criminal homicide	The mass media have not released enough information to the public	The lack of media supervision	Improve the transparency of public information	Media, news, report, broadcast, interview, publicity, real name	*
	2.5 Reform	*	The violence is due to the social climate	Violence	Punish the murderers; poetic justice	Social problem, social climate, indifference, insecurity, scary	*
3. Human interest frame	3.1 People involved in the crisis	A tragedy of a family	The family disruption is due to the loss of family members			Son, mother, husband, wife, family love	Pictures of the victim
3. Human interest frame	3.2 Ordinary people	An opportunity to educate public	One needs to learn how to save oneself because of the insecurity of society	Insecurity of society	Learn how to protect oneself and save others	Help, guideline, skill, protect, help, what to do, how to do	*
4. Sentiment frame		Criminal homicide	*	*	*	Sadness, pathetic, indignation, petition, unfairness, Chinese idioms, sarcasm	the murderers were beating the victim; police were investigating the crime scene

Insignificant

Findings

Multiple crisis communicators and their diverse frames

To address RQ1, the inductive framing analysis shows that the crisis is jointly constructed by the participation of multiple actors through their Weibo accounts, be they individual Weibo users, media, government agencies, organizations, or non-governmental organization (NGO). In this study, several crisis communicators formulated four frames, including nine sub-frames (Table 1). The following paragraphs discuss these frames more fully.

Problem identification frame

This frame includes posts that describe the event, offer a definition of the event, or list a related issue regarding the event. In particular, this study differentiates two sub-frames: ordinary criminal case and religion-related criminal case. When framed as an ordinary criminal case, the definition of the homicide emphasized that this event was caused by a ‘quarrel’ and the suspects had confessed to their crime of ‘beating’ the victim (e.g. Zhaoyuan Municipal Public Security Bureau, 2014). The national police (Fighting the Four Blacks Eliminating the Four Harms, 2014) redefined this event by using the religion-related criminal case frame and highlighted the suspects of the homicide were members of a religious cult.

Attribution frame

The frame includes posts that address attribution, such as who or which organization should take responsibility for the event, including the cause or possible solutions. There are five variants of this frame with different viewpoints about the attribution of the event: government, corporation, citizens, media, or reform. For example, the posts in government attribution frame emphasized that the governments should take responsibility for the crisis. These posts mentioned the slow response of the police to the emergency call, the lack of transparency of public information, and the unsound legal system. For example, The Global Times (2014), an English-language Chinese newspaper, asked the police departments to respond to Internet users:

[…] There is a public outcry, but the press release from Zhaoyuan Municipal Public Security Bureau did not answer the questions from netizens; its superior @ Shandong Public Security remains silent. As a department of people’s governments, you have duties to respond to netizens as soon as possible […] (23:44, 30 May 2014)

The corporation attribution frame, also widely used during the online discussion, mainly focused on the obligation of a corporation, such as the responsibility of the staff and security personnel: ‘[…] one customer was beaten to death in McDonald’s, the staff and the security men just stayed with nonfeasance […]’ (Yellow Fish Veda, 2014). A feminist NGO mentioned the importance of violence prevention of staff training: ‘[…] McDonald’s should establish security measures to ensure safe diner places for customers’ (The Voice of Feminist, 2014).

Human interest frame

This frame lends a human face, feeling, and concern to the online discussion of the event. It includes posts that care about personal lives, emphasize how individuals and groups suffer from or are affected by the event, or offer advice on how to protect individuals or rescue others in a similar incident. This frame is divided into two sub-frames with different focuses: people involved in the event and ordinary people. The first sub-frame concerns people involved in the crisis. Such a frame personalizes or emotionalizes the crisis by paying more attention to the victims or her family members. The second sub-frame focuses on normal people and further provides knowledge of how to protect oneself if in a similar crisis.

Sentiment frame

The posts in this frame show intense emotional expressions that induce public anger, sadness, or sarcasm by using visual or video information. Internet users appealed to a sense of social injustice, such as sadness and sympathy. For example, Yu Jianrong (2014), a professor at the Chinese Academy of Social Science, expressed his sadness:

[…] a person has been beaten to death in broad daylight and in public […] no one on the scene lent a hand to the victim, and society is so indifferent that everyone is afraid of being deterred by the danger […].

The negotiation process of multiple frames and its influences on crisis attribution

To answer RQ2, the empirical analysis shows that crisis communication is a complex negotiation process in which multiple frames meet, compete, and collaborate. As a consequence, several dominant crisis frames emerge during the process that alter the crisis attribution. A public opinion analysis of this event considers the online discussion as four stages (Liu, 2014a). Following the analysis, this study identifies the dominant crisis frames in the four stages. By using time as the horizontal axis and corporation attribution (i.e. the McDonalds’ responsibility) as the vertical axis, Figure 1 presents the changes in dominant crisis frames and their influences on corporate crisis attribution.

Figure 1.

Changes in corporate attribution.

Stage 1: Ordinary criminal case sub-frame and medium attribution

Before this event caught wide attention, most of the online discussion in the first stage adopted the ordinary criminal case sub-frame. During this stage, the local police department and McDonald’s were the main contributors. The local police used the ordinary criminal case sub-frame and defined this event as an incident caused by a ‘quarrel’ (Zhaoyuan Municipal Public Security Bureau, 2014). McDonald’s (2014a), however, emphasized that the police department should be responsible for the investigation: ‘[…] The relevant departments are handling this case; We fully co-operate with the investigation’. This post reflects the initial attempt of McDonald’s to categorize this event as an ‘accident’, de-emphasizing its responsibility (Coombs, 2010: 39).

Stage 2: Attribution frame and high attribution

Empirical data show that attribution frame dominated the online discussion in the second stage. Ordinary Weibo users and NGOs became the main contributors in this stage of online discussion. To interpret this event, they mainly adopted the attribution frame. The posts in the attribution frame and, in particular, those using the corporation responsibility sub-frame, consequently appointed a high crisis responsibility to McDonald’s: ‘[…] I recommend the victim’s family sue McDonald’s and to claim compasation […]’ (Yellow Fish Veda, 2014) and ‘[…] McDonald’s should take responsibility for the failure of protecting the customers […]’ (The Voice of Feminist, 2014). I recommend the victim’s …’.] In this way, the neutral criminal case turned into a ‘preventable’ (Coombs, 2014) crisis for McDonald’s, which addressed its high level of responsibility in this event.

Stage 3: Religion-related criminal case frame and low responsibility

After the national police initiated the religion-related criminal case sub-frame (Fighting the Four Blacks Eliminating the Four Harms, 2014), crisis communication entered the third stage. By using this frame, the national police’s post stated that the event was caused by the members of a religious cult and had something to do with the ‘evil faith’ of the religious cult. Therefore, the atrocity of the homicide was in principle out of the control of any organization. Following the national police’s frame, McDonald’s (2014b) distributed its second post regarding this event, adding that one of their staff was injured when she tried to prevent the crime. In this post, McDonald’s emphasized that it was also a ‘victim’ (Coombs, 2015). In this way, attribution frames, in particular, corporation responsibility sub-frame, were dramatically de-emphasized. As a consequence, McDonald’s attribution was reduced to ‘minimal crisis responsibility’ (Coombs, 2015: 39).

Stage 4: Sentiment frame

Ordinary Weibo users made sentiment frame dominant in the last stage of online discussions. As shown in Yu Jianrong’s (2014) post, ordinary Weibo users expressed their discontent toward the dignitary and resignation of social reality by using sentiment frame. In this stage, the attribution, such as the cause or a solution, for the event was no longer the focus of the crisis communication. Consequently, the pressure on McDonald’s as the main responsible organization in this case was released.

Discussion

Identification and assessment of contextual impacts on crisis attribution

By using the categorization of national contexts (Child, 2009) as a framework, this study makes a precious identification, comparison, and assessment of the special features of the Chinese context in which the crisis was located and evolved. To answer RQ3, three main contextual factors, namely technology, political value, and cultural value, have been identified from the context-embedded frames that have particular relevance to the crisis communication. In the following parts, this study elaborates on the extent and manner to which the Chinese contexts is different from those of other countries and how the uniqueness of the Chinese context influences the crisis attribution.

Technology: The use of information and communication technologies

Social media have altered the structure of China’s traditional media system that had long consisted of centralized institutions serving as the ideological apparatus of the communist party-state (Zhao, 2004). Even ordinary people are able to make their voices heard ‘in pursuit of a range of selfish or altruistic goals’ (Holladay and Coombs, 2013: 131) in a decentralized space provided by social media. Thus, the generation of crisis frames and their corresponding crisis attributions are not the privileges of organizations that have access to the traditional media. Weibo in this case, provide a relatively open public space for alternative frames to emerge. As demonstrated in the empirical data, multiple social actors that get access to social media can frame the crisis from their own perspectives. In particular, in Stage 2, ordinary Weibo users, public opinion leaders, and NGOs generated their own frames, which led to the multiple-frame condition. It is of particular important for the evolvement of crisis where the emergence of multiple frames in Stage 2 altered the attribution in Stage 1.

Political value: State control

After the national police redefined this event (Fighting the Four Blacks Eliminating the Four Harms, 2014) as a criminal case by evil cult members, the homicide became politically sensitive in China. Because the Chinese government declared that the evil cult offers a rival ideology to state power and contests party legitimacy (Cheung, 2004), the discussion of evil cult-related issues should be guided and highly controlled by the state. Therefore, the high control of the state as a unique political context in China influences the crisis framing in Stage 3. The empirical data exemplified how the state controlled the online discussion by using the religion-related criminal case sub-frame: first, the state initiated a new frame and set ‘the size of the mouth’ (Zhao, 2004: 179) for the online discussion. The religion-related criminal case sub-frame dominated in the second stage, which narrowed down the discussion space of this event. Second, those who have a close relation to the state became the main contributors of the religion-related criminal case sub-frame, such as party-owned media (e.g. the People’s Daily and China Central Television) and other police departments (e.g. Zhaoyuan Municipal Public Security Bureau and Shandong Public Security). Most of the crisis communicators who were active in the previous stages withdrew from crisis definition frame, due to the politically sensitive nature of the event.

Cultural value: Anti-authority Chinese Internet culture

Chinese Internet culture, with its feature of anti-authority, works as a special context for crisis communication in Stage 4. Sentiment frame, the dominant frame in the last stage, was facilitated by the anti-authority cultural value on the Chinese Internet (Yang, 2009). The focuses on crisis attribution have been blurred by cynicism in the sentiment frame. For instance, the joke disseminated by individual Weibo user ‘Writer Cui Chenghao’ (2014) served the humor function and the expression of dissent:

While passing by McDonald’s and feeling hunger, I was thinking to drop in. Suddenly, I realized that there was a Cayenne outside the restaurant. That was a narrow escape! I rushed into a KFC nearby in a hurry […] (21:52, 31 May 2014)

This is mainly because passive emotions led the sentiment frame. Most of the posts using this frame serve as a channel for ordinary Weibo users to express their communal disappointment about the social climate, resentment against social inequalities, and disaffection with bureaucracy. Although the posts still have some relationship with the events, they did not devote themselves to rational discussion, such as attributing responsibility. When the public stop making attribution about crisis responsibility, there is no crisis for the organizations anymore (Coombs, 2015).

Implications for theory

Based on the above elaboration of how national contexts impact crisis attribution, this study continues the inquiry into what the theoretical implications are for the classic theory, SCCT, in the field of crisis communication research. This study contributes to the literature by offering empirical support to the argument that contexts beyond the organization exert significant influences on crisis attribution. Most studies on the changing of crisis attribution have examined the influences of internal contexts (Coombs, 2007b; Coombs and Holladay, 2001). This study demonstrates that the influences of national contexts, namely new technology, the high control from the state, and anti-authority Chinese Internet culture, altered the crisis attribution generated by crisis type. Given the fact that the Chinese context plays a relevant role in shaping crisis attribution, the concept of ‘modifier’ should integrate national contexts for advancing the model of SCCT. By consulting the categorization of national contexts (Child, 2009), this study considers national contexts in terms of both material and ideational systems in a national domain (see Figure 2). National contexts hereby refer to the external factors that can be identified from a material and ideational system and that exert impacts on crisis communication. Consequently, modifiers in crisis communication should entail both internal factors, that is, performance history and prior reputation, and national contexts, which are located in the material and ideational systems in a given situation.

Figure 2.

The integration of national contexts into the concept of ‘modifier’ in SCCT.

Implications for practice

Given the support that national contexts have shaped the way the crisis is being framed, this study further suggests that the assessment of crisis attribution should be treated as an ongoing process. The empirical results indicate that there is no fixed frame that multiple crisis communicators agree on to determine one crisis type during crisis communication. Rather, multiple frames emerge in different stages of the crisis, which meet, compete, collaborate, and negotiate with each other. After a complex negotiation process, limited dominant frames remain in certain stages, which offer us some references to identify crisis types. In light of the findings, organizations should pay more attention to the evaluation of crisis attribution and taking into account the influences of national contexts. It would be worthwhile for organizations to assess crisis type as a processual and contentious process as well as adopt the concept of ‘framing’ to evaluate crisis situation. Accordingly, the identification of the crisis type should be based on the analysis of crisis framing process and with reference to dominant crisis frames in different stages.

Conclusion

To address the recent appeal for a more contextually sensitive perspective in crisis communication research, this study analyzes the interplay between the Chinese context and crisis frame and how the interplay, in turn, might lead to the change of crisis attribution. The theoretical framework and empirical illustration enrich crisis communication theory in three ways: first, the results suggest several compelling reasons for extending mainstream crisis communication theories that this study found inapplicable to crisis communication in China. Second, the findings explicate the crucial role extra organizational contexts play in crisis communication. The empirical results demonstrate the complex negotiation process of multiple context-embedded frames and their influences on crisis attribution. Three main contextual factors (i.e. technology, political value, and cultural value) and their relevance to crisis communication have been identified and discussed. Third, this study proposes that a revision of Western-originated theories is needed if the findings of crisis communication are to be more widely applicable. To this end, this study refines the model of SCCT by integrating national contexts to the concept of modifier.

Despite the original findings and relevant implications, this study has some limitations. This study only examines the crisis that took place in the Chinese context. Future research could test the findings in different national contexts or take a comparative approach to provide valuable insights into contextual differences. Another limitation is that the data collection is restricted in the microblogging site, Weibo. Future research could include other types of social media, such as social networking, picture sharing, video sharing, and product/service review, to detect the impacts of national contexts in greater depth.

Footnotes

Acknowledgements

The author would like to thank anonymous reviewers for their valuable comments and suggestions.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Hui Zhao

Author biography

Hui Zhao is a PhD candidate in the Department of Strategic Communication at Lund University, Sweden. She received her MA from Hong Kong Baptist University. Her research interests include social media, social change, crisis communication, and e-governance. Some of her publications have appeared in International Journal of Strategic Communication and Journal of Communication Management, among others. She has also received several awards and scholarships such as the Top Student Paper Awards from the Organizational Communication Division and the Public Relations Division of International Communication Association (ICA) in 2017; Best Student Paper from 1st Annual Conference of the Public Relations Society of China and 9th International Forum on Public Relations & Advertising in 2016; Chinese Government Award for Outstanding Self-financed Students Abroad of 2018; Landshövding Per Westlings Fund in 2017; and Campus Vänner scholarship in 2016.

References

S-K

Park

D-J

Cho

, et al. (2010) A cross-cultural study of effective organizational crisis response strategy in the United States and South Korea. International Journal of Strategic Communication 4(4): 225–243.

Barker

(2011) Online crisis communication response. Communication 37(1): 118–136.

Benford

Snow

(2000) Framing processes and social movements. Annual Review of Sociology 26: 611–639.

Bowen

Heath

(2007) Narratives of the SARS epidemic and ethical implications for public health crises. International Journal of Strategic Communication 1(2): 73–91.

Bundy

Pfarrer

Short

, et al. (2017) Crises and crisis management. Journal of Management 43(6): 1661–1692.

Cancel

Cameron

Sallot

, et al. (1997) It depends. Journal of Public Relations Research 9(1): 31–63.

Charmaz

(2006) Constructing Grounded Theory. London: Sage.

Chen

(2009) Institutionalizing public relations. Public Relations Review 35(3): 187–198.

Cheung

(2004) In search of a theory of cult and freedom of religion in China: the case of Falun Gong. Pacific Rim Law & Policy Journal 13(1): 1–30.

10.

Cheng

(2016a) How social media is changing crisis communication strategies. Journal of Contingencies and Crisis Management 26(1): 58–68.

11.

Cheng

(2016b) Social media keep buzzing. International Journal of Communication 10: 3241–3260.

12.

Child

(2009) Context, comparison, and methodology in Chinese management research. Management and Organization Review 5(1): 57–73.

13.

Claeys

A-S

Cauberghe

(2014) What makes crisis response strategies work? Journal of Business Research 67(2): 182–189.

14.

Coombs

(1995) Choosing the right words the development of guidelines for the selection of the ‘appropriate’ crisis-response strategies. Management Communication Quarterly 8(4): 447–476.

15.

Coombs

(1998) An analytic framework for crisis situations. Journal of Public Relations Research 10(3): 177–191.

16.

Coombs

(2006) Crisis management. In: Botan

Hazleton

(eds) Public Relations Theory II. Mahwah, NJ: Routledge, pp.171–197.

17.

Coombs

(2007a) Attribution theory as a guide for post-crisis communication research. Public Relations Review 33(2): 135–139.

18.

Coombs

(2007b) Protecting organization reputations during a crisis. Corporate Reputation Review 10(3): 163–176.

19.

Coombs

(2010) Parameters for crisis communication. In: Coombs

Holladay

(eds) The Handbook of Crisis Communication, pp.17–53.

20.

Coombs

(2014) Ongoing Crisis Communication. Thousand Oaks, CA: Sage.

21.

Coombs

Holladay

(2001) An extended examination of the crisis situations. Journal of Public Relations Research 13(4): 321–340.

22.

Coombs

Holladay

(2002) Helping crisis managers protect reputational assets. Management Communication Quarterly 16(2): 165–186.

23.

Dai

Hyun

(2010) Global risk, domestic framing. Asian Journal of Communication 20(3): 299–317.

24.

Entman

(1993) Framing. Journal of Communication 43(4): 51–58.

25.

Falkheimer

Heide

(2006) Multicultural crisis communication. Journal of Contingencies and Crisis Management 14(4): 180–189.

26.

Falkheimer

Heide

(2009) Crisis communication in a new world. Nordicom Review 30(1): 55–65.

27.

Fighting the Four Blacks Eliminating the Four Harms (2014) Chinese police online, 31 May. Available at: https://www.weibo.com/2328516855/B6OvYpQkG?type=comment (accessed 1 November 2018).

28.

Frandsen

Johansen

(2010) Apologizing in a globalizing world. Corporate Communications 15(4): 350–364.

29.

Frandsen

Johansen

(2017) Organizational Crisis Communication. Los Angeles, CA: Sage.

30.

Gamson

(1992) Talking Politics. Cambridge: Cambridge University Press.

31.

Gamson

Modigliani

(1989) Media discourse and public opinion on nuclear power. American Journal of Sociology 95(1): 1–37.

32.

Gans

(1979) Deciding What’s News. New York: Pantheon.

33.

Gerken

Van der Land

Van der Meer

TGLA

(2016) Crisis in the air. Public Relations Review 42(5): 879–892.

34.

Gilpin

Murphy

(2008) Crisis Management in a Complex World. New York: Oxford University Press.

35.

Greenberg

Hier

(2001) Crisis, mobilization and collective problematization. Journalism Studies 2(4): 563–583.

36.

Griffin

Babin

Attaway

(1991) An empirical investigation of the impact of negative public publicity on consumer attitudes and intention. Advances in Consumer Research 18(1): 334–341.

37.

Haruta

Hallahan

(2003) Cultural issues in airline crisis communications. Asian Journal of Communication 13(2): 122–150.

38.

Hearit

Courtright

(2003) A social constructionist approach to crisis management. Communication Studies 54(1): 79–95.

39.

Holladay

Coombs

(2013) Public relations literacy. Public Relations Inquiry 2(2): 125–146.

40.

Huang

Y-HC

Bedford

(2009) The role of cross-cultural factors in integrative conflict resolution and crisis communication. American Behavioral Scientist 53(4): 565–578.

41.

Huang

Y-HC

Kim

(2018) Cultures of crisis response. Chinese Journal of Communication 11(1): 1–4.

42.

Huang

Y-HC

Cheng

(2016) Crisis communication in context. Public Relations Review 42(1): 201–213.

43.

Jiang

Huang

Y-HC

, et al. (2015) At the crossroads of inclusion and distance. Public Relations Review 41: 50–63.

44.

Jianrong

(2014) Yu Jianrong, 31 May. Available at: https://www.weibo.com/1827652007/B6N7F4Fh2?type=comment (accessed 1 November 2018).

45.

Johnson

Peppas

(2003) Crisis management in Belgium: The case of Coca-Cola. Corporate Communications 8(1): 18–22.

46.

Kim

Zhang

(2016) Self-mocking crisis strategy on social media. Public Relations Review 42(5): 903–912.

47.

Kim

Cha

Kim

(2008) Developing a crisis management index. Journal of Public Relations Research 20(3): 328–355.

48.

Lee

(2004) Audience-oriented approach to crisis communication. Communication Research 31(5): 600–618.

49.

Liu

Kim

(2011) How organizations framed the 2009 H1N1 pandemic via social and traditional media. Public Relations Review 37(3): 233–244.

50.

Liu

(2014a) Woman beaten to death at McDonald’s for refusing to give her phone number. Metro, 1 June. Available at: http://www.thenanfang.com/blog/refused-to-give-away-her-phone-number-shandong-woman-beaten-to-death/ (accessed 1 November 2018).

51.

Liu

(2014b) An analysis of public opinion toward ‘5.28’ intentional homicide in Zhaoyuan. Yuqing.People.com.cn, 10 June. Available at: http://yuqing.people.com.cn/n/2014/0610/c210114-25129414.html (accessed 1 November 2018).

52.

Lyu

(2012a) A comparative study of crisis communication strategies between Mainland China and Taiwan. Public Relations Review 38(5): 779–791.

53.

Lyu

(2012b) How young Chinese depend on the media during public health crises? Public Relations Review 38(5): 799–806.

54.

Zhan

(2016) Effects of attributed responsibility and response strategies on organizational reputation. Journal of Public Relations Research 28(2): 102–119.

55.

McDonald’s (2014a) McDonald’s, 30 May. Available at: https://www.weibo.com/1947211342/B6IqCvHEJ?type=comment (accessed 1 November 2018).

56.

McDonald’s (2014b) McDonald’s, 30 May. Available at: https://weibo.com/1947211342/B6PSA00gk?type=comment (accessed 1 November 2018).

57.

Massey

(2001) Managing organizational legitimacy. Journal of Business Communication 38(2): 153–182.

58.

Matthes

Kohring

(2008) The content analysis of media frames. Journal of Communication 58(2): 258–279.

59.

Meyers

Rozen

(2014) Framing the Flotilla. Communication Monographs 81(1): 108–132.

60.

Pang

Cropp

Cameron

(2006) Corporate crisis planning. Journal of Communication Management 10(4): 371–389.

61.

Pang

Jin

Cameron

(2010) Strategic management of communication. In: Heath

(ed.) The SAGE Handbook of Public Relations (2nd edition). Thousand Oaks, CA: Sage, pp.17–34.

62.

Ramzy

(2014) Anger in China after McDonald’s patron beaten to death. The New York Times, 2 June. Available at: http://sinosphere.blogs.nytimes.com/2014/06/02/anger-in-china-after-mcdonalds-patron-beaten-to-death/?_r=0 (accessed 1 November 2018).

63.

Roh

(2017) Examining the paracrisis online. Public Relations Review 43(3): 587–596.

64.

Scheufele

(1999) Framing as a theory of media effects. Journal of Communication 49(1): 103–122.

65.

Schultz

Raupp

(2010) The social construction of crises in governmental and corporate communications. Public Relations Review 36(2): 112–119.

66.

Schwarz

Seeger

Auer

(2016) The Handbook of International Crisis Communication Research. Hoboken, NJ: Wiley Blackwell.

67.

Shoemaker

Reese

(1996) Mediating the Message. New York: Longman.

68.

Tang

(2016) Populist Authoritarianism. Oxford: Oxford University Press.

69.

Tanifuji

(2000) Crisis awareness and organizational response capabilities in present Japanese local governments. Journal of Contingencies and Crisis Management 8(1): 30–41.

70.

Taylor

(2000) Cultural variance as a challenge to global public relations. Public Relations Review 26(3): 277–293.

71.

The Global Times (2014) The Global Times, 31 May. Available at: https://www.weibo.com/1974576991/B6RFQf3hK?type=comment#_rnd1513717408345 (accessed 1 November 2018).

72.

The Voice of Feminist (2014) The Voice of Feminist, 5 June. Available at: https://www.weibo.com/1740974192/B7zMAah2O?type=comment#_rnd1513717697881 (accessed 1 November 2018).

73.

Tuchman

(1978) Making News: A Study in the Construction of Reality. New York: Free Press.

74.

Van der Meer

TGLA

Verhoeven

Beentjes

, et al. (2014) When frames align. Public Relations Review 40(5): 751–761.

75.

Van Gorp

(2005) Where is the frame? European Journal of Communication 20(4): 484–507.

76.

Veil

Buehner

Palenchar

(2011) A work-in-process literature review. Journal of Contingencies and Crisis Management 19(2): 110–122.

77.

Weiner

(1986) An Attributional Theory of Motivation and Emotion. Berlin: Springer.

78.

Writer Cui Chenghao (2014) Writer Cui Chenghao, 30 May. Available at: https://www.weibo.com/2834256503/B6MZG6dyG?type=comment#_rnd1513718167437 (accessed 1 November 2018).

79.

Huang

Y-HC

Kao

(2016) East meets west. Asian Journal of Communication 26(4): 350–370.

80.

Yang

(2009) Of sympathy and play. The Chinese Journal of Communication and Society 9: 39–66.

81.

Yellow Fish Veda (2014) Yellow Fish Veda, 30 May. Available at: https://www.weibo.com/2177245391/B6JjTCsS7?type=comment (accessed 1 November 2018).

82.

Zhao

(2017) The impacts of contextual factors on social media crises. International Journal of Strategic Communication 11(1): 42–60.

83.

Zhao

Falkheimer

Heide

(2017) Revisiting a social constructionist approach to crisis communication. International Journal of Strategic Communication 11(5): 364–378.

84.

Zhao

(2004) The state, the market, and media control in China. In: Thomas

Nain

(eds) Who Owns the Media? London; New York: Zed Books, pp.179–212.

85.

Zhaoyuan Municipal Public Security Bureau (2014) Zhaoyuan Municipal Public Security Bureau, 29 May. Available at: https://www.weibo.com/2239424023/B6vtpxiba?type=comment (accessed 1 November 2018).