HOW CRISIS COMMUNICATION
CAN COUNTER NEGATIVE
CONSUMER OUTCOMES AFTER A
SEVERE CSR-RELATED CRISIS IN
THE FASHION INDUSTRY
Name: Rosanna Cadeddu Student number: 11398809
Program: Business Administration Track: Marketing (MSc BA) Institution: University of Amsterdam
Faculty: Economics and Business Supervisor: Lars Moratis
Date of submission: January 26, 2018
Crisis type, crisis response strategy and message framing influence consumer emotions and perceived response credibility and corporate credibility
2 STATEMENT OF ORIGINALITY
This document is written by Rosanna Sofia Elaine Cadeddu, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 ABSTRACT
Aim - This research gives insights into the effects of CSR engagement and related crisis, crisis
response strategies and message framing on consumers’ emotions and judgements after a CSR-related crisis has occurred. Several significant effects are identified, which provide practical implications for future crisis communication.
Background - An increasing number of organizations engage in CSR activities and the expectations
of consumers regarding these activities are also on the rise. When CSR activities becomes a public expectation, perceptions of corporate irresponsible behaviour are easily created. Such reputational threat can lead to a severe corporate crisis.
Method - A scenario based 2 (embedded versus peripheral CSR and related crisis) x 2 (diminishing
versus rebuilding response strategy) x 2 (emotional versus rational message framing) experiment based on vignette is conducted by means of a fictive news article about a leading fashion company. In total, 313 respondents successfully participated in the study.
Findings - Main effects were found for all dependent variables. First, a crisis at an organization with
embedded CSR and a related crisis leads to higher levels of anger among consumers, compared to an organization with peripheral CSR and a related crisis. Second, a rebuilding strategy leads to higher levels of perceived corporate and response credibility than a diminishing strategy. Third, using a rebuilding strategy leads to a higher perceived corporate credibility and more feelings of sympathy, but only for organizations with peripheral CSR and related crisis.
Managerial Implications - This study provides practical implications and shows that a crisis
response strategy only leads to more positive consumer’ emotions and judgements when a firm experiences a CSR crisis that is related to its peripheral CSR activities and pursues a rebuilding strategy. When a firm is involved in a crisis related to its embedded CSR activities, the use of crisis response strategies did not counter any of the negative outcomes.
Keywords: CSR engagement; crisis response strategies; crisis type; consumer emotions; consumer judgements; consumer attitude
4 TABLE OF CONTENTS
ABSTRACT ... 3
1. INTRODUCTION ... 7
1.1. General Context ... 7
1.2 Research paradox & Literature gap ... 8
1.3 Research goal ... 9 1.4 Academic relevance ... 10 1.5 Managerial relevance ... 11 1.6 Overview ... 11 2. LITERATURE REVIEW ... 12 2.1 CSR ... 12
2.1.1 The need for CSR programs ... 13
2.1.2 CSR practices ... 14
2.2 CSIR and corporate crises ... 15
2.2.1 Corporate crisis ... 17
2.3 Crisis communication effects on consumers ... 18
2.3.1 Perceived Credibility ... 19
2.3.2 Consumer’s emotions ... 20
2.3.3 CSR engagement and related crisis and consumer responses ... 21
2.4 Organizational crisis communication ... 22
2.4.1 Crisis communication response strategies ... 23
2.4.2 Crisis response strategies and consumer responses ... 24
2.5 Message framing ... 25 2.6 Interaction effects ... 27 2.7 Conceptual model ... 28 3. METHOD ... 29 3.1. Design ... 29 3.2 Procedure ... 30 3.3 Stimulus material ... 31 3.4 Pre-test study ... 32 3.4.1 CSR activities ... 32
3.4.2 Crisis Response Strategy ... 33
3.4.3 Message Framing ... 34
5
3.6 Instruments ... 36
3.7 Manipulation checks main study ... 38
3.8 Control variables ... 39
3.8.1 Consumers’ crisis involvement ... 39
3.8.2 Sustainability orientation ... 39
4. RESULTS ... 41
4.1 Normality check ... 41
4.2 Correlations ... 41
4.3 Main effects of CSR engagement and related crisis ... 43
4.4 Main effects of crisis response strategies ... 44
4.5 Moderation effect of message framing ... 44
4.6 Moderation effect of crisis response strategy ... 47
4.7 Moderating Moderation effect of Message Framing on Total Model ... 50
5. DISCUSSION ... 54
5.1 Discussion of the results ... 54
5.1.1 Effect of CSR engagement and related crisis ... 54
5.1.2 Effect of crisis response strategies ... 56
5.1.3 Crisis response strategies and message framing ... 57
5.1.4 CSR engagement and related crisis and crisis response strategies ... 59
5.2 Managerial implications ... 60
5.3 Limitations and future research ... 61
5.4 Conclusion ... 62
6. LITERATURE ... 64
APPENDIX A - Questionnaire ... 70
APPENDIX B - Scenarios ... 75
APPENDIX C - Results ... 83
LIST OF TABLES & FIGURES Figure 1. Conceptual Model ... 29
Table 1. Manipulation Check of CSR-engagement Type ... 33
Table 2. Manipulation Check of Crisis Response Strategy ... 34
6
Table 4. Distribution Scenarios among Participants ... 35
Table 5. Demographics of Participants ... 36
Table 6. Manipulation Checks Main Study ... 38
Table 7. Means, Standard Deviations and Correlations ... 42
Table 8. Main Effects of CSR related Crisis to CSR engagement ... 43
Table 9. Main Effects of Crisis Response Strategies ... 44
Table 10. Moderation Effect of Message Framing on Crisis Response Strategy ... 46
Figure 2. Moderating Effect Crisis Strategy on CSR Crisis and Corporate Credibility ... 48
Figure 3. Moderating Effect Crisis Strategy on CSR Crisis and Sympathy ... 49
Table 11. Moderation Effect of Crisis Response Strategy on CSR engagement and related Crisis ... 49
Table 12. Moderation Effect of Message Framing on Total Model ... 52
7 1. INTRODUCTION
1.1. General Context
In November 2017, one of the world’s most successful fashion companies ZARA came in the news because shoppers in Istanbul found unexpected tags in the ZARA merchandise, declaring “I made this item you are going to buy, but I didn’t get paid for it.” Turkish employees owed three months’ loan and tried to get the attention of consumers by attaching the tags, in the hope that ZARA would finally pay them (Racked.com, 2017). Unfortunately, this is not the only incident ZARA has been accused of. The organization previously has been sued for child and slave labor, as well as exploiting Syrian refugees whereof the youngest was 15 years old (Racked.com, 2017). The organization reacted to such claims by promising that they would look into the incidents, however without taking responsibility for the incidents that happened in their third-party factories. As the labor and human rights website Equal Times wrote: “Zara is a company that would rather pay fines than rectify its bad labor practices.” Similar reactions are often used by fast-fashion retailers; it is a response that brands regularly use to elude responsibility (Racked.com, 2017).
This example is one of many, in which organizations are accused of corporate social misconduct. In recent years, there have been numerous reports of corporate misconduct in which companies did not lived up to their professed standards regarding their Corporate Social Responsibility (CSR) activities. Managers need to be extremely aware of the dangers of being perceived as socially irresponsible, especially since consumers tend to listen more to negative rather than positive information regarding CSR messages (Bhattacharya & Sen, 2007). Such incidents, as the example described above, led to consumer skepticism towards companies claiming to engage in CSR activities (Skarmeas & Leonidou, 2013). Consequently, consumers only believe 38% of the CSR messages organizations provide, while 72% of the consumers claim to be very interested in the CSR-topic (GlobeScan, 2012).
8
Skepticism is a serious issue, since it leads to negative effects on retailer equity, lower resistance to negative company information among consumers and an increase in negative worth-of-mouth (Skarmeas & Leonidou, 2013).
1.2 Research paradox & Literature gap
Incidents like the one described about fashion leader ZARA is just one of many. Such incidents raise many questions concerning the negative impacts it can have on organizations; how to minimize the negative impacts; whether an organization can influence these negative outcomes and even turn it to something positive; and if this effect is different for organizations with a high CSR reputation versus organizations with a low CSR reputation. The latest studies on CSR identified risks as well as the opportunities of having a CSR reputation (e.g. Brown & Dacin, 1997; Diehl, Terlutter & Mueller, 2015; Drumwright, 1994; Osterhus, 1997; Sen & Bhattacharya, 2001). Moreover, some studies also analysed the use of CSR as a way to cope with a severe corporate crisis (Vanhamme & Grobben, 2009; Klein & Dawar, 2004) or even defined crisis response strategies organizations could use after a crisis (Coombs, 2007). However, little is known about whether such crisis response strategies are also effective in countering the negative consumer outcomes when looking at different CSR engagement levels and if this effect differs among crises that relate to these engagement levels. Companies with a high CSR reputation may need to pursue different response strategies than companies that are less known for its CSR activities, especially when the crisis relates to its CSR engagement activities. Moreover, the current crisis response strategies do not take emotions into account that are evoked by consumers during a crisis (Claeys & Cauberghe, 2014). A tool that could serve as an additional response strategy that does look at emotions is the concept of message framing. Message framing is a widely effective and persuasive tool, often used in marketing communications (McKay-Nesbitt, Manchanda,
9
Smith, & Huhmann, 2011). Hence, it could serve as an additional crisis communication strategy that might influences the effects of the crisis response strategies identified by Coombs (2007).
1.3 Research goal
The primary goal and contribution of this research is to give more insights in the effects of CSR engagement and related crisis, crisis response strategies and message framing on consumer emotions and perceived response and corporate credibility. The research question is:
What is the effect of a CSR related crisis based on the CSR engagement type on consumers’ emotions, response credibility and organizational credibility; and how is this relationship moderated by crisis response strategy and message framing?
Several sub questions are formulated to answer the research question:
1. How do prior crisis CSR engagement and a crisis related to an organizations CSR activities affect consumers’ emotions and perceived response and corporate credibility?
2. How does the use of crisis response strategies affect consumers’ emotions and perceived response and corporate credibility?
3. How does the use of message framing influence the relationship between crisis response strategies and consumers’ emotions and perceived response and corporate credibility?
4. How does the implementation of a crisis response strategy influence the relationship between prior crisis CSR engagement and a crisis related to an organizations CSR activities on consumers’ emotions and perceived response and corporate credibility?
10 5. How does the interaction between CSR engagement and related crisis, crisis response strategies and message framing affect consumers’ emotions and perceived response and corporate credibility?
To answer the research questions, an experimental mixed design was conducted based on three main manipulations. First, the participants were shown to one of the eight conditions. The layout and length of the articles are the same, whereas differences in content reflected the manipulations. After reading the fictive article, which was an imitation of a NOS.nl article, the manipulation questions were represented towards the respondents. Then, respondents answered the questions related to the dependent variables.
1.4 Academic relevance
This research adds value to the existing literature of CSR in a few ways. First, this study looks at different CSR engagement levels, relates the crisis to the activities belonging to the CSR engagement and subsequently measures its effects on consumers’ emotions and perceived response credibility and corporate credibility. To my knowledge, this is the first study that has looked at different CSR crisis types relating to the CSR engagement of the organization on these dependent variables. Second, this research will explore the variable message framing as a potential moderator of the effects of crisis response strategies on consumers’ emotions and perceived response and corporate credibility. This effect only has been studied by Claeys and Cauberghe (2014); however, they looked at the purchase intention of consumers after a crisis and used a fictive organization with an expertise in water treatment installations, which is a rather specific organization type. This research will focus on the fashion industry, since CSR crises in this sector are unfortunately not rare and such company is relevant for most consumers. Besides, Claeys and Cauberghe (2014) examined
11
the effects on purchase intention, whereas this research shall focus on consumer emotions and perceived response and corporate credibility. Overall, this research will give new insights into crisis engagement and related crisis, crisis response strategies, message framing, consumers’ emotions and perceived response and corporate credibility after a crisis.
1.5 Managerial relevance
While an increasing number of corporations are dedicating substantial efforts to CSR related activities, the number of corporate social irresponsibility (CSIR) incidents continue to increase as well. When the perceived credibility is low and negative consumer emotions are created, a severe crisis is established and serious organizational harm is done. Therefore, it is extremely relevant for companies to know how to handle such CSR-related crisis correctly, to be still be perceived as credible and to be able to create positive consumer attitudes as these affect behavioural consumer responses. By investigating multiple variables that could influence consumer perceptions of organizations post-crisis, this research will provide new insights to develop a crisis response strategy after a crisis.
1.6 Overview
The structure of this paper will be as followed. First, existing literature on CSR, CSIR, crisis communication, message framing and consumer outcomes like emotions and perceived response and corporate credibility will be explored. Subsequently, hypotheses will be formulated and a conceptual framework will be provided. Following, an experimental study and its results will be presented, followed by a final discussion. Lastly, managerial implications will be provided.
12 2. LITERATURE REVIEW
This literature review gives insights into the academic literature on CSR, the effects of CSR and its risks, the responses of consumers towards CSR crises and the crisis communication strategies organizations could employ after a severe CSR-related crisis.
2.1 CSR
Nowadays, consumers expect companies to engage in CSR-related activities. Especially, due to extreme events like global warming and climate change. But, what does it mean for a company to engage in social responsible activities? Since the 1960s, many scholars have tried to create a comprehensive definition of the concept (Carroll, 1991). Different definitions of CSR are found among the existing literature; yet most definitions are similar and overlap. Carroll (1979) defines Corporate social responsibility (CSR) as “the social responsibility of business encompasses the economic, legal, ethical and discretionary expectations that society has of organizations at a given point in time’’ (p.500). He defined four basic social responsibility expectations (economic, legal, ethical and philanthropic) in a more comprehensive categorization. Within this categorization, each business responsibility can be scrutinized in relation to all stakeholders of the company (e.g. employees, customers, community, public at large and owners) (Mohr, Webb & Harris, 2001). The concept of CSR has evolved massively after the development of this definition. McWilliams and Siegel (2001) define CSR as ‘’situations where a company goes beyond compliance and engages in actions that appear to further some social good, beyond the interests of the firm and that which is required by law’’ (p. 112). Subsequently, Van Marrewijk (2003) defines CSR as “the company's’ activities – voluntary by definition – demonstrating the inclusion of social and environmental concerns in business operations and in interactions with stakeholders” (p. 102). Kotler and Lee (2005) then define CSR as “a commitment to improve community
well-13
being through discretionary business practices and contributions of corporate resources” (p.3). The differences between the definitions indicate the importance of CSR and that it is much more than just social restrictions imposed by law.
2.1.1 The need for CSR programs
More than 80% of the Fortune 5000 companies highlight their CSR activities on their websites (Bhattacharya & Sen, 2004). CSR has become an integrated part of the activities of modern companies which are dedicating substantial resources to all kinds of social initiatives like cause-related marketing, corporate philanthropy, minority support programs and socially responsible employment and manufacturing practices (Bhattacharya & Sen, 2004; Du, Bhattacharya, & Sen, 2010). Organizations are under a rising pressure to engage in social responsibility initiatives while maintaining profitable (Mohr, Webb & Harris, 2001), mainly due to the growing public standards for social performance (Pirsch et al, 2007; Juščius & Snieška, 2015). Consumers expect organizations to be in congruence with their social values (Maignan, Ferrell & Ferrell, 2005) and tend to prefer to be associated with companies that have good reputations (Maignan & Ferrell, 2001). Firms realize that engaging in CSR activities is not only “the right thing to do”, but it also leads to more positive outcomes from their key stakeholders (Bhattacharya & Sen, 2004), like positive reputations (Fombrun & Shanely, 1990). CSR can be used as a persuasive marketing tool, since it is built around the expectations of consumers. Consequences of CSR are that it leads to positive consumer awareness attitudes, sense of attachment, good-will among consumers, improves the company’s overall evaluation, reputation and corporate image in the long run (Hur, Kim & Woo, 2013; Chernev & Blair, 2015; Brown & Dacin, 1997).
14
2.1.2 CSR practices
Firms differ in their motives to engage in CSR. Existing literature found several distinctions between CSR practices and motives. The two CSR types that will be leading in this research is the distinction based on the work of Aguinis & Glavas (2013). They argue that companies engage in either embedded or peripheral CSR. Embedded CSR is built around the core competencies of a company and the CSR activities are integrated in the strategy, routines and operations of a firm. In contrast, peripheral CSR refers to CSR activities that are not integrated in a firm’s strategy, routines and operations (Aguinis & Glavas, 2013). Many examples can be found of firms that engage in CSR initiatives that are a part of its strategy or daily practices and routines, but not in both. In all these cases, the firm pursues peripheral CSR rather than embedded CSR (Aguinis & Glavas, 2013). This categorization of embedded versus peripheral CSR is based on previous literature on sustainability, but is different from other categorizations. Other categorizations focus on either the outcomes of CSR (e.g., substantive versus symbolic) or predictors of CSR (i.e., normative versus instrumental). For example, the predictor categorization of CSR can be both found in the categorization of embedded versus peripheral CSR. As Aguinis and Glavas (2013) state: “For example, consider the case of philanthropy and volunteering, which are examples of peripheral CSR. These activities can be based on normative (i.e., they are the right thing to do) or instrumental (i.e., there are financial benefits derived from them) motives (p.315).”
Previous research on CSR programs and consumers has created mixed findings. Frankental (2001), for example, argues that consumers find it difficult to differentiate between the true intentions of the organization and whether the CSR activities are just a form of greenwashing. However, Becker-Olsen, Cudmore, and Hill (2006) argue that greenwashing (employing CSR activities for self-interest) is not per se perceived as something negative. Beyond, they found evidence that if the CSR efforts are not aligned with
15
the organizational core business, consumers view it negatively. In other words, organizations that employ peripheral CSR will experience a more negative financial impact since consumers are less willing to buy from such organizations. Aguinis and Glavas (2013) argue that “consumers prefer to see a ‘‘win–win’’ approach in which products help both the company and society/environment at the same time” (p. 324). Concluding, by re-examining the empirical results derived from the marketing field, peripheral CSR leads to mixed and often even more negative outcomes than embedded CSR. When it comes to embedded CSR, consumers hold even higher quality standards of such organizations. Results from the work of Bhattacharya and Sen (2003) show that if the company is perceived of having its CSR values embedded and is perceived as being sincere about its intentions (embedded CSR rather than peripheral), the congruence between the values of the firm and consumer increases and will lead to higher identification levels with the organization. Overall, previous research suggests that companies that are engaged in CSR should built those activities around its core business and be transparent and sincere about their efforts.
2.2 CSIR and corporate crises
While more and more companies pursue CSR activities to create reputational and financial benefits, simultaneously the number of incidents of corporate socially irresponsible (CSIR) behaviour of firms increases substantially (Skarmeas & Leonidou, 2013; Wagner, Lutz, & Weitz, 2009; Jones, Bowd, & Tench, 2009). Corporate social irresponsibility (CSIR) refers to all corporate actions at the expense of employees, consumers, suppliers, the environment, other organizational stakeholders, and even society at large (Pearce & Manz, 2011). A recent case of CSIR is the emission fraud by the Volkswagen Group. Over the years, Volkswagen had developed a positive reputation due to their CSR activities. However, their CSR image became their liability when they were accused of implementing fraudulent
16
software in their diesel engines, which is extremely contradicting to the message they were claiming as being a climate friendly option. This software led the public believe that the emission of their cars was lower than it actually was, meaning that Volkswagen violated the Clean Air Act. Another recent case of social irresponsible behaviour is that of leading fashion company ZARA. In November 2017, the company came in the news because shoppers in Istanbul found unexpected tags in the ZARA merchandise, declaring “I made this item you are going to buy, but I didn’t get paid for it.” Unfortunately, this is not the only incident ZARA is associated with. The leading fashion brand already has been sued for child and slave labour, as well as exploiting Syrian refugees whereof the youngest was 15 years old (Racked.com, 2017). Other well-known examples of CSIR are the sweatshop and child labour activities at the Nike and Gap plants in Asia, the controversy around Shell Oil and the Niger delta and Nestlé's aggressive marketing of breastmilk substitutes in Africa.
While there are certainly cases in which consumers care more about their personal gain rather than ethical norms (Brunk & Blümelhuber, 2011), managers need to be extremely aware of irresponsible behaviour, since most consumers are more sensitive to negative rather than positive information (e.g., Sen & Bhattacharya 2001; Brunk & Blümelhuber 2011). By publicly engaging in CSR efforts, to enhance the organization’s reputation for example, managers also create new crisis risks. When CSR becomes integrated into the corporate reputation and thus becomes a public expectation, perceptions of social irresponsibility can become a reputational threat. As CSR has become the performance standard the company must achieve, stakeholder might perceive actions as socially irresponsible more easily when the firm cannot live up to their CSR expectations. Therefore, as Bhattacharya et al. (2007) state: “there is an asymmetric effect and “doing bad” hurts more than “doing good” helps (p.18). A serious consequence of being accused of irresponsible behaviour is that consumers start doubting the extent to which companies actually comply with their alleged standards.
17
This leads to an increase in consumer scepticism towards CSR activities (Skarmeas & Leonidou, 2013). Thus, even though CSR is used to build a positive reputation, it could also easily destroy one. Stakeholders that are dissatisfied with a corporation’s social performance can manufacture a reputational-based threat, and a potential crisis by challenging an organization’s commitment to responsible behaviour (Coombs & Holladay, 2015). Coombs et al. (2015) study this concept and define it as a CSR-based challenge. A CSR-based challenge transforms the positive outcomes of a CSR program from a potential strength to a potential weakness. A CSR program becomes a potential liability when stakeholders believe the challenge and thus question the commitment of the company to its CSR efforts. If not handled correctly by the company, a CSR-based challenge can turn into a crisis that has great reputational risks as consequence (Coombs & Holladay, 2015).
2.2.1 Corporate crisis
Coombs (2015) defines a crisis as “the perception of an unpredictable event that threatens important expectancies of stakeholders related to health, safety, environmental, and economic issues and can seriously impact an organization’s performance and generate negative outcomes” (p. 3). As reported by the Reputation Institute in 2012, 42 percent of an organization’s reputation is based upon perceptions of their CSR efforts. From a marketing communication perspective, crisis accusations are critical since they can have a deleterious impact on consumers’ perceptions of the incriminated company and its products. Years of work invested in building a strong company reputation can be virtually ruined overnight by an article published or broadcasted in the media. The negative information, as consequence of a crisis, has a large potential of destroying the reputation and corporate image of a firm. Brown & Dacin state that "negative CSR associations can have a detrimental effect on overall product evaluations, whereas positive CSR associations can enhance product evaluations” (1997, p. 80). This is mainly because people have the tendency emphasis more attention to
18
negative information rather than positive information when evaluating a company (Mizerski, 1982; Bhattacharya & Sen, 2007). Like Warren argued; “It takes 20 years to build a reputation and five minutes to ruin it. If you think about that, you'll do things differently”.
As reported by the Image Restoration Theory, the nature of the crisis should be considered in understanding image repair strategies. A crisis consists of two components: 1) the accused is held responsible for an action, and; 2) that act is considered offensive (Benoit & Dorries, 1996). In both cases, perceptions are more important than reality. According to Benoit & Dorries (1996): “The important point is not whether the business in fact is responsible for the offensive act, but whether the firm is thought to be responsible for it by the relevant audience” (p. 178).
2.3 Crisis communication effects on consumers
To examine the effects a crisis has on consumers, several attributes should be analysed. Ajzen & Fishbein (1980) developed the Theory of Reasoned Action (TRA, Ajzen & Fishbein, 1980) and argue that responses could be divided into beliefs and behaviour. Since perceptions leads to consumer’s behaviour, the beliefs of consumers are extremely important after a crisis. To test a consumer’s post-crisis attitude based on their beliefs, measuring emotions is commonly used. An interesting issue is to investigate how companies can overcome such CSR crisis. Since the company’s (CSR) reputation is damaged, responding to the crisis can become crucial in how they are perceived afterwards (Janney & Grove, 2011). Another issue that is highly relevant in the CSR literature is credibility. Corporate crises lead to scepticism among consumers and consequently leads to negative effects on consumer behaviours. To minimize scepticism among consumers, it is extremely important that they perceive the organization as credible.
19
2.3.1 Perceived Credibility
Credibility is an important issue in the CSR literature. Many studies stress that credibility perceptions should be included in studying crisis response (Fediuk, Pace & Botero, 2010; Kim & Cameron, 2011), since it is an essential persuasive feature in crisis communication strategies (Heath, 1997; Reynolds & Seeger, 2005). This study will make a distinction between response credibility as well as corporate credibility, since both are viewed as fundamental in the crisis communication literature.
Heath (1997) states that: “A company suffering a crisis must be able to tell a credible story, one that has factual fidelity that can withstand the scrutiny of reporters, governmental investigators, and concerned citizens” (p. 319). Belch et al. 1987 state that: “The term `source credibility' often is used to refer to attributes of the communicator including expertise, trustworthiness, attractiveness, and power (p. 60).'' Source credibility favourably improves the validity and/or believability of the message of the informer as perceived by the receiver (Ohanian, 1990).
Furthermore, previous studies have found that CSR has a positive effect on corporate credibility (Hur, Kim & Woo, 2014) and vice versa (Alcaniz, Caceres & Perez, 2010). When a brand is perceived as credible, it also has a positive effect on the CSR image of a company (Alcaniz, Caceres & Perez, 2010). Erdem, Swait & Louviere (2002) define perceived credibility as the credibility of the brand or a firm as a signal, which refers to the extent to which consumers believe that the company is trustworthy and has expertise. Trustworthiness indicates the level of trust the receiver has in the sender, whereas expertise includes the perceptions the receiver has about the source’s skills (Erdem et al., 2002). Newell & Goldsmith (2001) define corporate credibility as “the extent to which consumers feel that the firm has the knowledge or ability to fulfil its claims and whether the firm can be trusted to tell the truth or not” (p. 235). Thus, to be perceived as a credible organization or brand it needs to be perceived as “willing and able to deliver what is promised” (Erdem et al., p4). Generally,
20
credibility can be defined as the entire set of perceptions a receiver holds towards a certain source (Bettinghaus, 1969). As previously mentioned, a crisis could seriously harm the reputation of a company and its legitimacy. Reputation and legitimacy are often measured by credibility and trust (Coombs, 1998; Coombs & Holladay, 2001; Fombrun, 1996; Massey, 2001). In explaining the relationship between CSR and corporate reputation the variable credibility is found fundamental (Hur et al., 2014).
2.3.2 Consumer’s emotions
As consumer attitudes are the driver of their following behavioural responses, it plays an extremely important role in the crisis communication literature and research. To measure the attitude of consumers, the emotions post-crisis will be examined. Emotions are often investigated in the crisis response literature (Coombs & Holladay, 2005) as they are formed when interpreting an incident (Weiner, 1986). Anger and sympathy are the two emotions that consumer’s experience most frequently when exposed to a crisis situation (Coombs et al., 2005; Choi & Lin, 2009; McDonald, Sparks & Glendon, 2010). Coombs (2007a) argues that crisis response strategies influence consumers’ attribution of responsibility and the emotions they experience afterwards, which in turn create motivations for behaviour. When the attributions of responsibility are high, feelings of anger are evoked and consequently behavioural responses are negative (Coombs, 2007; Weiner, 2006). The higher the perceived responsibility, the stronger the feelings of anger and in some extreme cases “schadenfreude” can occur (when people get pleasure from the pain of others); while simultaneously feelings of sympathy reduce. Coombs (2007) states that: “Negative emotions can cause stakeholders to lash out at an organization (engage in negative word of mouth) or to severe interactions with the organization” (p. 169). When the attributions of responsibility are low, sympathy is evoked and the consequential behavioural responses are positive.
21
2.3.3 CSR engagement and related crisis and consumer responses
Having a CSR reputation could act as a buffer against a crisis, and may help a company to recover from a crisis. Research on this topic claim that consumers appear to provide greater support for companies that are socially and environmentally responsible (e.g. Pirsch et al., 2007; Creyer & Ross, 1997; Ellen, Mohr & Webb, 2000 and Sen et al., 2001), and have negative reactions to companies that are not (Barrett, 1996). However, others claim that companies are more heavily sanctioned on their CSIR activities, when they are known for their good CSR activities prior the crisis (Janney & Gove, 2011). Literature on corporate social irresponsibility argues that a perception of social irresponsibility becomes a threat for organizations with a CSR reputation (Bhattacharya et al., 2007; Coombs & Holladay, 2015). In such cases, CSR has become the performance standard it must achieve, and the organization cannot live up to the expectation of stakeholders their actions might be perceived as socially irresponsible behaviour. When the firm in crisis is known for being socially and environmentally responsible, a related crisis will evoke greater negative market reactions, compared to firms without a CSR reputation. Since there is a large discrepancy between their promised behaviour and actual behaviour, investors perceive such organizations as hypocrite (Tillman et al., 2009) or as decoupling (Zajac & Westphal, 2004). Organizations that are engaged in embedded CSR often have a CSR-based reputation, since the CSR activities are built around the core competencies of a company and are integrated in the strategy, routines and operations of a firm (Aguinis & Glavas, 2013). Therefore, it is expected that a crisis related to the CSR activities of a firm that has its CSR activities embedded will create more negative outcomes among consumers, since there is a greater discrepancy between their CSR program and organizational behaviour. In case of peripheral CSR, a firm is less known for its CSR activities. Here, the discrepancy between its CSR program and its behaviour is less large when experiencing a crisis. This leads to the following hypotheses:
22 H1: A crisis related to the embedded CSR activities of an organization leads to a (H1a) lower perceived response credibility, (H1b) lower overall corporate credibility, (H1c) more anger and (H1d) less sympathy towards the organization than when an organization experiences a crisis that is related to peripheral CSR activities.
2.4 Organizational crisis communication
Organizational crisis communication deals with the interrelationships between the situation of the crisis, the response strategies and the crisis perceptions (Schulz, Utz & Göritz, 2012). As previously mentioned, a corporate crisis is a major threat to the overall reputation of a company (Barton, 2001; Davies et al., 2003). A crisis gives people arguments to form negative associations about the organization, leading to a negative reputation (Coombs, 2007b). Since a corporate crisis can substantially impact the trust a consumer has in either the company or its products, organizations need to be aware of the importance of employing tactics to repair the trust and express their trustworthiness (Gillespie and Dietz, 2009; Xie and Peng, 2009). Therefore, scholars in both reputation management and crisis communication stress the importance of crisis communication, as it can reduce the potential reputational damage afterwards (Benoit, 1995; Coombs, 1995; Murray & White, 2005). Coombs (2010) describes crisis communication as “the collection, processing, and dissemination of information required to address a crisis situation” (p. 20).
Currently, there are two main theories that explain how a crisis influences the relationships between the stakeholders and the organization. Those theories are Weiner Attribution Theory (WAT, 1986) and Coombs and Holladay’s Situational Crisis Communication Theory (SCCT, 2002), which is revised by Coombs (2007b). Both theories argue that stakeholders make attributions about the cause of a negative event along several dimensions (McDonald et al., 2010). Such dimensions are formed by multiple sources, like the media and the company itself (Klein & Danwar, 2004). The SCCT examines how crisis
23
situations influence the attributions of crisis responsibility of the public, which subsequently influences their perceptions, affects and behavioural intentions regarding the organization (Coombs, 2007). The theory posits that the crisis manager should determine an appropriate crisis response based on the threat level of the crisis, especially since this response has a major influence on the attitude and behavioural and affective responses of the public afterwards (Coombs, 2010). Coombs developed a set of crisis response strategies (e.g., denial, diminish, rebuild, and reinforcing) that are proposed “to repair the reputation, to reduce negative affect and to prevent negative behavioural intentions” (p. 170). In analysing crisis communication, the concept of crisis response strategies is often used.
2.4.1 Crisis communication response strategies
The research done by Coombs (2007) established three main post-crisis strategies that organizations can apply. These main crisis response strategies are (1) deny, (2) diminish, and (3) rebuild strategies. Organizations that implement a deny strategy do not take responsibility for the crisis, or even claim that there is no crisis (Coombs, 2007b; Claeys et al, 2010). This strategy aims to remove any connection the organization has with the crisis. The diminish strategy intends to minimize the severity of the crisis and tries to lower the attributions of responsibility of the involved organization. Furthermore, the rebuilding strategy entails that the organization compensates the affected parties and releases an official statement of apology (Coombs, 2007b; Hegner, Beldad, & Kraesgenberg, 2016). Managers should implement a crisis response strategy that limits the harm and thus best serves to protect the organization. Therefore, many scholars stress the fact that it is extremely important that a manager bases its crisis response strategy on theoretically and empirically evidence (Coombs & Schmidt, 2000; Benson, 1988; Choi and Lin, 2009). The SCCT provides managers with several guidelines for choosing the best suited crisis communication response strategies (Coombs, 2006).
24
The SCCT respectively makes a distinction between three crisis types clusters: 1) victim cluster; 2) accidental cluster; and 3) preventable cluster. This is determined by the perceived responsibility attribution of consumers. If consumers attribute higher levels of crisis responsibility to the organization, it will be more damaging for the organizational reputation (Coombs, 2006, 2007b). This implies that preventable crisis, in which consumers fully attribute the responsibility at the organization, have the most negative impact on the reputation of the organization (Claeys et. al, 2010). Based on this knowledge, the higher the responsibility attributions the more negative the consequences are and the more relevant a crisis response strategy is (Hegner et. al, 2004).
2.4.2 Crisis response strategies and consumer responses
Hypotheses can be formulated by combining the crisis communication theory (SCCT), the crisis response strategies and the consumer responses. The SCCT argues that when consumers attribute high levels of responsibility of the crisis, they develop stronger feelings of anger. Consumers attribute stronger feelings of sympathy when they attribute low responsibility levels to the crisis. Claeys et al. (2010) argue that the post-crisis perception of a firm’s reputation among consumers does not differ between a matched and a mismatched crisis type and response strategy. However, this finding contradicts with the findings of Coombs & Holladay (1996). A rebuilding strategy results in the least negative effects on consumers’ trust, hence is the most suitable during a preventable crisis (Coombs, 2007; Claeys et al., 2010). This would imply that a rebuilding strategy will produce less negative emotions than other strategies. Therefore, a rebuilding strategy should lead to the most desired emotions and a restored organizational reputation. (Hegner, Beldad, & Kraesgenberg, 2016). Furthermore, Coombs and Holladay (2006) argue that the denial strategy is only a suitable strategy when the stakeholders genuinely think that the organization has no link to the crisis. As this study is only going to focus on crises that could be preventable, rather than
25
accidental, the focus will only lie on the diminishing and the rebuilding strategies for a comparison. Previous research claims that the rebuilding strategy is the most appropriate one to use when the organization is responsible for the crisis. However, this strategy is also the costliest option and could lead to a financial loss (Tyler, 1997). Accordingly, the question arises whether a rebuilding strategy is always necessary or if there are also situations in which the diminishing strategy is sufficient in fulfilling the needs of consumers in a crisis situation. Since previous research claims that the rebuilding strategy leads to the most desired emotions and higher levels of consumers’ trust, this must lead to a higher perceived credibility. Based on previous statements, the hypotheses on the effect of crisis response strategies on perceived credibility and consumer’s emotions can be defined:
H2: Respondents exposed to a rebuilding strategy leads to (H2a) a higher perceived response credibility, (H2b) higher overall corporate credibility, (H2c) less anger and (H2d) more sympathy towards the organization than respondents exposed to a diminishing strategy.
2.5 Message framing
Up until now, the focus of crisis research mainly focusses on studying the impact of attributions of responsibility on the emotions of consumers during such a crisis (e.g. sadness, anger) (Jin, 2009; Choi & Lin, 2009b; Coombs & Holladay, 2005). This is also in the case of the SCCT model. This model does not take the diverse emotions into account that are evoked in a crisis situation, while such emotions can influence the way people attribute responsibility (Kim & Cameron, 2011). In other words, the theory does not consider that people also make decisions based on both “emotions” and “logic” (e.g., Gordon & Arian, 2001). Choi and Lin (2009) state that: “It is important to examine how attribution independent emotions, along with attribution dependent emotions, can influence outcome variables in SCCT.” (p. 199). The theory emphasizes that the crisis type and/or situations shape the emotions of the public.
26
However, the theory did not look at the role of message framing in influencing emotional responses of the public in crisis situations. These emotions would in turn affect the following attributes, attitudes and behavioural intentions of the public. Since organizations often use emotional framing in their persuasive advertising communications (Cotte & Ritchie, 2005), it might also serve as an additional crisis communication strategy (Coombs, 2007). The way organizations frame their crisis information (rationally or emotionally) might influence consumer's’ willingness to pay attention to the content of the message and hence can be an effective additional crisis response strategy (McKay-Nesbitt et al., 2011). Previous studies found that consumers form attitudes differently towards rationally framed ads versus emotionally framed ads. Messages that are emotionally framed appeal to the emotions of individuals by using drama and subjective, evaluative arguments (Stafford & Day, 1995; Yoo & MacInnis, 2005) and lead to consumer responses like feelings (McKay-Nesbitt et al., 2011; Yoo & MacInnis, 2005). Content that is rationally framed appeal to the rationality of the individual by presenting content straightforward and objectively, which leads receivers to be more evaluative of the credibility of the content, since this appeals to the cognitions of individuals (MacInnis et al., 2002; McKay-Nesbitt et al., 2011; Yoo & MacInnis, 2005). Concluding, even though previous research suggests that a rebuilding crisis response strategy leads to the most desired consumer responses afterwards, this relationship might be influenced by the way the response is framed. As emotional framing appeals to the emotions of consumers, it is expected that this leads to more positive consumer emotions, whereas rational framing appeals more too cognitive cues thus leads to higher perceptions of credibility. To explore this relation, the following hypotheses are formulated:
H3: Message framing moderates the effects of crisis response strategy on (H3a) perceived response credibility, (H3b) overall corporate credibility and emotions like
27 (H3c) anger and (H3d), whereas a rational framing leads to a higher response and corporate credibility while emotional framing leads to less anger and more sympathy.
2.6 Interaction effects
Since the variables crisis response strategy, CSR engagement and message framing have not been studied together before in the fashion industry, investigating if there is a relation between these variables would be very interesting. To examine interaction effects between the several manipulations, an exploratory design should be used.
First, it would be interesting to analyse the relation between crisis response strategies (diminish vs. rebuild) and CSR engagement (embedded vs. peripheral). Although the previous literature argues that a rebuilding strategy is the strategy organizations should pursue during a preventable crisis, the combination with CSR engagement and a crisis that relates to this engagement may create different outcomes. Since the discrepancy between the CSR engagement and a CSR-related crisis is larger in the embedded condition, it is expected that only a rebuilding strategy could lead to more positive outcomes. Especially, since the crisis is related to the one of the core competencies of the organization. However, for the peripheral condition it is expected that both strategies would lead to more positive outcomes, since there is no physical harm (as in the case of a crisis related to embedded activity related crisis) and since the discrepancy between the reputation of the firm and its actions is less large. This leads to the following hypotheses:
H4: Crisis response strategy moderates the effect of prior crisis CSR engagement on (H4a) perceived response credibility, (H4b) overall corporate credibility and emotions like (H4c) anger and (H4d) sympathy, whereas a rebuilding strategy leads to more positive outcomes than a diminishing strategy in an embedded crisis situation and both strategies lead to more positive outcomes in a peripheral crisis situation.
28
Lastly, the interaction effects between all independent variables on the dependent variables will be examined. In the previous hypotheses, it is expected that only a rebuilding strategy will lead to positive outcomes in case of a crisis related to embedded CSR activities. However, the use of message framing as additional response strategy might influence this relationship. Since emotional framing appeals to the emotions of individuals, the combination with a diminishing strategy might be as effective as a rebuilding strategy. However, this option would be more cost effective for organizations, since a rebuilding strategy is the costliest option (Tyler, 1997). This leads to the following research question:
H5: To what extent does the effect of the prior CSR activities and a crisis related to these activities on (H5a) perceived response credibility, (H5b) overall corporate credibility and emotions like (H5c) anger and (H5d) sympathy depend on the interaction between the crisis response strategy and the framing of the message?
2.7 Conceptual model
The literature review on CSR engagement, crisis type, crisis communication and consumer attitudes and judgements leads to the identification of a literature gap regarding what the proper crisis response strategy is after an organization experiences a CSR-related crisis. In order to answer the research question and hypotheses, a quantitative approach was used. An experimental design was developed to analyse the interaction effects and differences between the independent variables on the dependent variables. This study looks at direct and moderating effects on the crisis communication after a CSR-related crisis. The relationships between the independent, dependent and moderating variables are visualized in the conceptual model below.
29
Figure 1. Conceptual Model
3. METHOD 3.1. Design
This study involves a 2 (Crisis at company with embedded versus peripheral CSR) x 2 (diminish versus rebuild response strategy) x 2 (emotional versus rational message framing) online experiment, based on a mixed design vignette study. A vignette study tries to examine the influence of different vignettes on individual responses via hypothetical situations (Atzmüller & Steiner, 2010). Atzmüller & Steiner (2010) describe a vignette as followed: “A vignette is a short, carefully constructed description of a person, object, or situation, representing a systematic combination of characteristics (p. 128).” This method allows to manipulate hypothetical situations and investigate causal effects on the dependent variables. In this research, the vignettes represent the CSR engagement and related crisis, crisis response strategy and message framing. A mixed design vignette was chosen, so that different groups of participants were exposed to various conditions (vignette sets) in which all respondents belonging to the same group reviewed the same condition (Birnbaum, 1999;
CSR engagement and related crisis
type A) Embedded B) Peripheral Crisis response strategy A) Diminish B) Rebuild Consumer’s emotions Anger Sympathy Consumer’s judgements Corporate credibility Response credibility Message framing A) Emotional B) Rational
30
Atzmüller & Steiner, 2010) via an online survey. When the described situation in the vignette is perceived as realistic, respondents tend to be more accurate in reflecting their own feelings, which leads to a high external validity (Finch, 1987).
3.2 Procedure
Three independent variables were applied in this study: corporate social responsibility program related crisis (embedded vs. peripheral related crisis), crisis response strategy (rebuilding vs. diminishing) and message framing (rational vs. emotional). This study used a quantitative research method, since a large data set had to be collected to measure various views and opinions. The respondents were selected through convenience sampling method, by personally contacting my network, mostly via social media and/or email. Through snowball sampling the survey was spread and 407 participants took part of the online questionnaire.
All participants were randomly assigned to one of the eight conditions via Qualtrics. The manipulated messages were evenly distributed across the participants. The online survey started with a general message that introduced the subject of the survey and included the notion that it was part of a master thesis. Then, the respondent got asked whether he or she would voluntarily like to participate to the study anonymously, also known as a notion of informed consent. This should minimize the chance that respondents would give socially desired answers (Saunders & Lewis, 2011). Subsequently, respondents read an introductory text about the company R&C that was engaged in either embedded or peripheral CSR activities. Then, they were exposed to one of the eight scenarios, followed by several manipulation checks and construct scales based on previous literature. Each condition presented an imitation of a NOS.nl article, in which the manipulation was done. The vignettes (conditions) were implemented via differences in content, while the layout and length of the articles remained the same. After reading the fictive article, respondents were exposed to the
31
manipulation questions. Lastly, respondents answered the questions that were related to the dependent variables. All questions could be answered on a 7-point Likert scale, varying from 1 (strongly disagree) to 7 (strongly agree).
3.3 Stimulus material
First, respondents read a text about the fictive fashion brand R&C. This text had two manipulations, namely a company with embedded versus peripheral CSR activities. In the embedded conditions, the fashion brand paid fair prices to its production workers in Asia and provided a good work environment; made environmentally friendly clothes; invested in its production workers by helping them develop themselves and setting up entrepreneurial activities; and donated money to various aid agencies. In the peripheral conditions, the organization only occasionally donated money to a certain aid agency that helps children in Africa. Subsequently, the respondents were exposed to a fictive article from NOS.nl, which reported a crisis for the fictive fashion brand R&C. The crisis type was manipulated in the beginning of the article by explaining what crisis occurred. The manipulation of the crisis response strategy was done via a reaction to the crisis that the CEO of R&C had given during a press conference. The crisis type manipulation provides two scenarios. The first is an embedded CSR-related crisis, whereby Amnesty International has found evidence that R&C exploits its workers and is involved with child labour. The crisis response manipulation was done via a reaction on the crisis by the CEO of the company. In the rebuild response strategy, the CEO apologizes and promises to change the working conditions, versus a diminishing response strategy in which the CEO nuances the crisis by saying that R&C employs the standard Asian working conditions just like other factories. In the manipulation of a peripheral CSR-related crisis, the organization was accused of being involved with the embezzlement of public donations together with aid agency “Future for Children”. In the rebuilding condition, the CEO apologizes for the crisis and promises to donate the same
32
amount of money again directly to children in need, versus the diminishing scenario where he says that there is always a risk with engaging with aid agencies and that R&C is not responsible. All scenarios could be found in Appendix B.
3.4 Pre-test study
To validate the effectiveness of the manipulations of the independent variables, a preliminary test was conducted. To investigate the effectiveness of these manipulations, a quantitative pre-test in the form of an online questionnaire was conducted among 30 respondents via Qualtrics. The respondents were randomly assigned to either the embedded CSR condition or the peripheral CSR condition. The organization that was used in the pre-test was a fictive fashion brand, called R&C. The respondents were randomly assigned to either the embedded CSR condition or the peripheral CSR condition. A distinction here was made based on the CSR activities of the organizations and a CSR crisis that was related to the described CSR activities. Subsequently, all respondents were exposed to the same questions.
3.4.1 CSR activities
A manipulation check for the embedded versus peripheral CSR activities of the organization was conducted to investigate whether individuals perceived that there was a significant difference between the CSR engagements of the organization. All respondents were exposed to a brand description and subsequently had to answer the following two questions on a 7-point Likert scale (1=strongly disagree, 7=strongly agree): “R&C has its
CSR activities integrated in its strategy, operations and daily routines” and “The CSR activities of R&C are part of its core competencies”. This self-developed scale was based on
Aguinis & Glavas (2013), who state that the distinction between embedded and peripheral CSR is based on the fact if the activities are part of the core competencies and whether it is integrated into the entire organizational strategy, operations and daily routines.
33
First, the reliability of this scale was measured. The Cronbach’s Alpha of this scale was 0.949, which is above the required minimum of 0.70. Therefore, this scale is confirmed to be reliable. Since the data is not normally distributed due to the small sample size, a Mann-Whitney Test was performed to check if the manipulation was significant. The test shows a p-value of p=0.001, thus the manipulation is confirmed. However, small changes will be made in the brand description to increase the perceived difference in the experimental design. For example, the peripheral condition in the pre-test included the following sentence: “From
their start nearly 20 years ago, the company has been donating money once a year to the aid agency “Future for Children”.” This will be changed into: “Occasionally, the company donates money to the aid agency “Future for Children”. Since small changes were made, to
increase its strength, this manipulation was included in the experimental survey.
Table 1
Manipulation Check of CSR-engagement Type.
CSR Activities Scale M SD N
Embedded CSR 5.75 0.876 16
Peripheral CSR 3.96 1.587 14
3.4.2 Crisis Response Strategy
A manipulation check was developed to investigate whether the respondents thought that the organization took responsibility for the crisis or not via its post-crisis response. The respondents were asked to rate one item on a 7-point Likert scale whether they felt if the organization took responsibility for the crisis (1=strongly disagree, 7=strongly agree). This item relates to the SCCT of Coombs (2007), which argues that the crisis response strategies differ in the level of responsibility companies take. A diminishing strategy implies that the organization lowers its responsibility, whereas in a rebuilding strategy the organization compensates the affected parties and releases an official statement of apology (Hegner et al., 2016). Since the scale only consists of one item, a reliability check is not possible. The mean
34
scores, however, are in line with the theory and confirm that the exposed texts have the intended effect.
Table 2
Manipulation Check of Crisis Response Strategy
Crisis response strategies M SD N
Diminishing Strategy 2.13 1.358 30
Rebuilding Strategy 5.23 1.654 30
3.4.3 Message Framing
The third manipulation check that was developed was to test whether the respondents thought the written text was emotional or rational. This manipulation was measured on four items, based on the research of Liu and Stout (1987), on a 7-point semantic differential scale (1=rational, 7=emotional). The reliability of this scale was measured and showed a Cronbach’s Alpha of 0.753, thus confirmed as reliable. To test the manipulation, a Mann-Whitney Test was performed, since the normality of data distribution is violated due to the small sample size. This tests shows a p-value of p=<0.000. Therefore, this manipulation is confirmed.
Table 3
Manipulation Check of Message Framing
Message Framing M SD N
Rational 3.02 0.986 30
Emotional 5.38 1.453 30
Lastly, the researcher extensively discussed the pre-test with 4 respondents after they had filled in all questions. The researcher asked the respondents what their argumentations behind their answers were. Subsequently, the researcher explained the purpose of the texts and which answers were desired. Then, together with the respondents, the researcher made small adjustments to the manipulated texts to increase the manipulation strength. Furthermore, the suitability and realism of the scenarios were verified by the participants.
35 3.5 Participants main study
In total, 645 participants started the online questionnaire via Qualtrics, of which 407 participants completed the survey. Most respondents were collected via the personal network of the researcher, the majority through social media and email. Furthermore, the personal network was asked to share the questionnaire among their network. To minimize the threats to the overall quality of the data, special attention was given to so-called “cheaters” and “speeders”. Smith, Roster, Golden & Albaum (2016) argue that respondents that intentionally provide wrong answers so that they can participate to the study and receive the following rewards can be classified as “cheaters”. They define “speeders” as respondents that do not read all questions extensively and minimizes his or her cognitive effort, so that they can receive their stimulus for minimal time effort (Smith et al., 2016). To decrease the probability of biases and increase the reliability of the data, a control question was included. Participants were asked whether the CEO apologized in his statement or not, which refers to the two crisis response strategies that were examined. In total, 33 respondents were removed from the data set since they answered this question wrong. Furthermore, a check of completion time was done. Before the online questionnaire was spread among the mass, a small sample of respondents was asked to fill in the survey as accurate and carefully as possible. This was asked to measure the minimum time a respondent needs to extensively read all texts and answer all questions as good as possible. Based on this small sample, the minimum time to answer all questions carefully was five minutes. Thus, all respondents that completed the survey in less than five minutes were removed from the data. After this data cleaning, 318 responses were left for analyses.
Table 4
Distribution Scenarios Among Participants
Message Framing
CSR program Emotional Rational Total
Embedded Crisis response
strategy Total Diminish Rebuild 33 46 79 41 41 82 74 87 161
36
Peripheral Crisis response strategy Total Diminish Rebuild 39 38 77 44 36 80 83 74 157
Total Crisis response
strategy Total Diminish Rebuild 72 84 156 85 77 162 157 161 318
Overall, the average age of all respondents was 30.67, of which the minimum age was 18 and the maximum age 72 years old. With regards to their sex, 148 participants were male and 170 were female. Furthermore, most participants were Dutch (85,9%).
Table 5 Demographics of Participants Frequency Percentage Gender Male Female 148 170 46,5% 53,5% Age M=30.67 Nationality Dutch Belgium German British Italian Portuguese Swiss Irish Russian Greek Bulgarian Brazilian Canadian Chinese Ghanaian Indian Indonesian Malaysian Mexican American Total 273 4 13 4 4 1 1 2 1 1 2 1 2 2 1 1 1 1 1 2 318 85,85% 1,26% 4,09% 1,26% 1,26% 0,31% 0,31% 0,63% 0,31% 0,31% 0,63% 0,31% 0,63% 0,63% 0,31% 0,31% 0,31% 0,31% 0,31% 0,63% 100% 3.6 Instruments
The instruments that were used in this questionnaire were mostly based on existing scales, retrieved from construct developing research. All questions could be answered through a 7-point Likert scale. All constructs in the study are reliable, based on Cronbach’s Alpha’s that lie between α = .721 and α = .918. The questionnaire with all constructs is to be found in Appendix A.
37
To measure the concept of sustainability orientation, respondents were asked to rate five statements on a 7-point Likert scale, based on the construct of Maignan (2001). The corresponding reliability coefficients were: 0.97, 0.92, and 0.96.
People attribute the responsibility level for a crisis based on experiencing an emotional reaction to it. The core emotions in the Attribution Theory are anger and sympathy. Emotions and the attribution of responsibility can evoke motivations for action. The construct by Lee (2004) was used to identify negative emotions towards the organization. The sympathy scale by McDonald (2010) was used to measure positive emotions towards the organization.
One item based on the Situational Crisis Communication Theory (Coombs, 2007) was included to this construct to measure the crisis response strategy as manipulation check, in which respondents were asked to rate the degree to which they perceived the organization took responsibility for the crisis on a 7-point Likert scale.
Message framing was measured on a four-item 7-point semantic differential scale based on the study of Liu and Stout (1987) (α = .83). Respondents were asked to rate the reaction of the organization on the crisis on each of the items (e.g. emotional versus rational; subjective versus objective).
The scale of Kopalle & Lehmann (2001) (α = .87) was used to measure consumers’ crisis involvement. This scale includes three items (e.g., “These events are very important to me”).
To measure response credibility, the newly developed scale of Van Riel (2017) was used (α = 0,86). This scale exists of eight items and is a combination of several items based on different other existing scales. Two items from the scale used by Coombs and Holladay (1996) were included (i.e., I think that the message in R&C’s response is clear). Also, one item from Turker (2009) was included (i.e., R&C is committed to take social responsible
