«13TH INTERNATIONAL PUBLIC RELATIONS RESEARCH CONFERENCE “Ethical Issues for Public Relations Practice in a Multicultural World” Holiday Inn ...»
Based on the above results, we examined the data-model fits of a model that excluded the non-significant paths between (a) crisis responsibility and relational trust (H4), (b) negative emotion and relational trust (H2a), and (c) positive emotion and information seeking intention (H3b). The model without non-significant paths had similar data-model fits to the proposed model (i.e., χ2/df = 2.79, CFI =.903, and RMSEA =.065), thus we did not revise the proposed model.
Mediation effects of emotions. Examining RQ 2 about the role of discrete emotions in the interpretation of outcome variables (i.e., relational trust and information seeking intention), mediation tests were used following Baron and Kenny Steps (1986). Sobel z scores were used for statistical decisions. In the proposed model, negative and positive emotions mediate the effects of crisis responsibility on relational trust and intention to seek information. Negative emotion partially mediated the relationship between crisis responsibility and information seeking intention (Sobel z statistic = 2.05, p =.04). Positive emotions fully mediated the effect of crisis responsibility on relational trust (Sobel z statistic = 2.44, p =.01). Table 5 shows the mediation test results.
Discussion This study investigates the role of both positive and negative emotions to a crisis in the context of an influenza outbreak on a college campus. We aimed to extend situation-based approaches to understanding crisis dynamics (e.g., Coombs & Holladay, 2002) by introducing emotion-driven interpretations of a crisis into a theoretical model. We report four major findings
that have both theoretical and practical implications in the field of public relations:
1. A variety of discrete emotions were experienced during an influenza outbreak.
2. Crisis responsibility was associated with both positive and negative emotions.
3. Positive and negative emotions mediated the relationships between crisis responsibility and among crisis outcome variables.
4. Relational trust was strongly and positively associated with intention to seek information.
First, students not only felt both negative and positive emotions during a flu pandemic, but they experienced some of positive emotions more frequently than negative emotions. That is, “interest, alert or curious” and “sympathy or compassion” were the most frequent emotional responses (followed by “anxious, worried or concerned”). It is noteworthy that similar emotions (i.e., sympathy, grateful, and interested) were primarily drawn by the September 11 attacks 433 (Fredrickson et al., 2003), but not by a corporate transgression case (Choi & Lin, 2009). This could be explained by the common characteristics between an influenza pandemic and a terrorism: both crises (a) create identifiable victims, (b) can directly impact anyone with random probability, and (c) receive relatively low organizational responsibility due to the lack of intentionality and controllability from the organization side (in case of terrorism, people attributed responsibility to an external entity who actually committed the crime). Intentionality and organizational involvement to an issue have been used to categorize crisis types (e.g., Coombs, 1995; Jin et al., 2007). It is possible that crisis emotions are closely related to the type of crisis because characteristics of the crisis itself play a major role in shaping attributions of responsibility (Coombs, 2004).
Discrete emotions were grouped into two categories based on previous theoretical models (e.g., Izard, 1977, Fredrickson et al., 2003) and empirical evidence provided by this study.
Consistent with previous categorizations used in psychology (Izard, 1977), anger, sadness, anxiety and fear were grouped as negative emotions and interested, love, grateful, hopeful and sympathy were grouped as positive emotions. For negative emotions, fear and anxiety were the primary indicators supporting Pang, Jin and Cameron’s (2009) notion that fear and anxiety dominates human emotion in crises, especially when uncertainty pervades in crisis situations. It has been argued that anxiety is the “default” emotion that stakeholders experience in any type of crises (e.g., Jin, 2009; Jin et al., 2007, 2008; Pang et al., 2009), while fright is predominantly induced by unpredictable and uncontrollable crises (Jin, 2009). Combined with these propositions, this study proposes anxiety and fear (or fright) as the primary negative emotions that should be addressed in facilitating emotional coping of stakeholders during an influenza outbreak.
Secondly, crisis responsibility was associated with both negative and positive emotions with a stronger association for negative emotion. These direct paths are consistent with Weiner’s (1985) notion that causal attributions affect emotions generated by such events. While negative emotional responses to corporate transgression cases have been found to be positively correlated with or influenced by the blame and responsibility directed toward an organization (e.g., Cho & Gower, 2006; Choi & Lin, 2009; Kim & Yang, 2009), no association was found for positive emotions. Alert, which is categorized as positive emotion in this study, was not associated with crisis responsibility in the Mattel product recall case either (Choi & Lin, 2009). For an influenza outbreak case, students might have associated crisis responsibility with the severity of a crisis and felt “grateful, appreciative or thankful” of not being infected by the flu. In previous research, crisis severity has been suggested to be a predictor of crisis responsibility (e.g., Coombs, 1998;
While alert was negatively associated with reputation in a corporate transgression case (Choi & Lin, 2009), positive emotions (including “interested, alert or curious”) were positively associated with relational trust in this study. The more students felt positive emotions, the more they trusted the college health center. Attribution Theory (Weiner, 1995) might predict that positive emotions would influence relational trust if the student health center was viewed as a provider of aid in response to a crisis with external causes. Combined with Choi and Lin’s (2009) study, the same emotion could have different influence on crisis dynamics depending on the type of a crisis (e.g., locus of crisis responsibility). Further research would be needed to 434 unpack the relationship between crisis responsibility and positive emotions in a variety of crisis contexts.
Third, both negative and positive emotions were significant mediators between crisis responsibility and crisis-related outcome variables. Specifically, negative emotion partially mediated the relationship between crisis responsibility and intention to seek crisis-relevant information. This finding reinforces Weiner’s (1995) proposition that emotions mediate between responsibility and behavioral intentions. This mediation path could indicate a desirable effect of negative emotions drawn by a crisis that stimulate stakeholders to search for more information about the issue. Providing sufficient information to stakeholders is especially important at an early stage of a crisis for two reasons (e.g., Coombs, 1999; Sturges, 1994): to (a) reduce fear and anxiety caused by the uncertainty of impact that a negative incident have on each stakeholder, and (b) prevent further spread of similar incidents (e.g., as for an influenza outbreak, relevant information would aid in reducing future influenza transmission). Although negative emotions have been mostly associated with negative impact on an organization in a crisis (e.g., Choi & Lim, 2009; Coombs et al., 2007), negative emotions could also provide an organization with chances to provide important information about a crisis and engage stakeholders in crisis communication.
In addition, positive emotion was a significant mediator between crisis responsibility and relational trust. Reinforcing Fredrickson and Joiner’s (2002) proposition that positive emotions work as a “breather” by enhancing broadminded coping, positive emotion was found to play a meaningful role in retaining public trust on an organization in a crisis. When mediated by positive emotions, the influence of negative evaluation about an organization (i.e., perceived crisis responsibility) on relational trust could be minimized by allowing stakeholders to be more flexible in interpreting crisis situations (Reed & Aspinwall, 1998; Trope & Pomerantz, 1998).
Fourth, a strong direct path was found between relational trust and intention to seek information. This suggests that trust is a valuable organizational asset that should be protected during a crisis (Coombs & Holladay, 2002). Supportive intention is believed to be a function of reputation (e.g., Coombs, 1998, 1999). Lee (2005) also found that participants' degree of trust in the organization has a significant direct effect on purchase intention. Addition to these behavioral intentions, we found a similar relationship also holds for stakeholder’s intention to seek crisisrelated information. Considering the strong path coefficient between relational trust and intention to seek information, a particular emphasis should be also paid to the role of positive emotion in shaping relational trust in crisis management.
Implications Combining the results from this study with others affords the opportunity to refine our understanding of emotion-based interpretations of a crisis. This study attempted to measure discrete emotions during an actual crisis event to collect data that best represent what stakeholders experienced in a real crisis. Study findings emphasize the importance of understanding individual emotional reactions to a crisis in effective crisis management. The theoretical implications of this study are threefold: (a) the identification of discrete emotions in the context of an influenza pandemic, (b) the categorization of crisis emotions based on both theoretical models and empirical evidence, and (c) the evaluation of differential roles of positive and negative emotions in crisis dynamics.
435 The results of this study have practical implications for dealing with health crises that involve disease transmission and share similar crisis profiles with an influenza outbreak, including lack of intentionality and limited organizational control. Crisis managers should factor stakeholders’ emotional reactions into their evaluation of the crisis situation and subsequent selection of a crisis response strategy. For instance, when “anger” is the dominant reaction, an organization might focus on explaining its position about the accusation to reduce the level of responsibility. Yet, for managing a crisis that primarily evokes “fear” or “anxiety,” providing relevant information and showing its commitment to reducing the impact of a crisis would be more important. A direct path from negative emotion (“fear” and “anxiety” were the primary indicators) to information seeking intention might indicate what types of emotions catalyze information seeking in this particular type of a crisis.
Limitations and Suggestions for Future Studies Caution should be used when interpreting relationships between study variables. Crosssectional survey data cannot provide definitive information about causal ordering between variables. The use of SEM does permit the test of a theoretical model that assumes a particular order of variables and, in doing so, assesses the plausibility of that causal order. Furthermore, surveys provide a useful methodology for studying emotions in a real crisis case due to the time sensitive nature of the topic. Future studies should consider longitudinal and/or experimental designs to confirm the temporal ordering of the variables studied here.
In the crisis communication literature, crisis responsibility has been found to be a significant predictor of organizational reputation or trust (e.g., Choi & Lin, 2009; Kim & Yang, 2009; Lee, 2005). However, this study didn’t support a direct relationship between these variables and only a small indirect path via positive emotion (indirect B =.29 *.19, or.06) was found. While the unique characteristics of the chosen crisis case could partly explain the lack of support (e.g., an organization’s role as a care provider than a crisis creator), it might be meaningful to further investigate the variables associated with the attribution of responsibility, including locus of attribution (personal vs. organizational), risk susceptibility or risk severity, to better understand the nature of the relationship between crisis responsibility and trust. Another topic for future work would involve the identification of factors that predict positive emotions in response to a crisis, considering its positive association with information seeking within crisis communication.
Conclusions There is a great deal to learn about conditions that govern the impact of emotions in the crisis context. It is our hope that this study’s focus on the role of positive and negative emotions in shaping crisis responsibility, relational trust and intentions to seek crisis-related information move us closer to this goal.
Arnold, M. B. (1960). Emotion and personality. New York: Columbia University Press.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.
Barton, L. (2001). Crisis in organizations II (2nd ed.). Cincinnati, OH: College Divisions SouthWestern.
Breckler, S. J. (1993). Emotion and attitude change. In M. Lewis & J. M. Haviland (Eds.), Handbook of emotions (pp. 461-473). New York: Guilford.
Broom, G. M, Casey, S., & Ritchey, J. (2000). Concepts and theory of organization- public relationships. In J. A. Ledingham and S. D. Bruning (Eds.), Public relations as relationship management: A relational approach to the study and practice of public relations (pp. 3-22). Mahwah, NJ: Lawrence Erlbaum Associates.
Bruning, S. D., & Ledingham, J. A. (1999). Relationships between organizations and publics:
Development of a multi-dimensional organization-public relationship scale. Public Relations Review, 25, 157-170.
Butler, J. K. (1991). Toward understanding and measuring conditions of trust: Evolution of conditions of trust inventory. Journal of Management, 17, 643-663.