Sorry Seems To Be The Hardest Word
The interplay of crisis response strategy, emotional appeal, and medium on post-crisis reputation
Master's Thesis
Master: Communication Science Track: Corporate Communication
Faculty: Social and Behavioural Sciences Student: Esther van der Hulst
Student number: 10207449 Supervisor: Toni van der Meer
Abstract
Crisis communication plays a critical part in minimising possible reputational damage. It is therefore important for crisis communication managers to implement the most beneficial crisis response strategy. Previous research implies that this includes taking full responsibility for the situation, which may have legal consequences. To overcome this problem, this study examined whether a diminish strategy in combination with other communicative elements can produce similar outcomes to repair the post-crisis organisational reputation. Therefore, the study examines the effects of communicating emotions and medium choice in combination with crisis response strategy. Moreover, the moderating effects of perceived responsibility and trustworthiness were tested. The results of a 2x2x2 experimental design (N = 166), demonstrate a direct effect of crisis response strategy on the post-crisis organisational reputation. In addition, it was found that medium choice significantly influences the effect of crisis response strategy on the evaluation of the organisation. However, communicated emotion was not found to influence the post-crisis reputation in any way. Furthermore, the results reveal that perceived responsibility moderates the effects of crisis response strategy and medium on the post-crisis reputation. This indicates that when the perceived responsibility is low, the positive effects of crisis response strategy and medium are strengthened.
Introduction
Crisis communication is the heart of crisis management, meaning that when this is not effective neither will the crisis management effort be. Through crisis communication an
organisation can prevent a crisis from destroying an organisation by avoiding reputational damage (Coombs, 2014). A crisis can be referred to as a sudden and unexpected event occurs that can disrupt the operations of an organisation. In addition, the situation can affect a broad range of stakeholders who can be harmed physically, emotionally, or financially (Coombs, 2007). However, the crisis can also have negative effects for the organisation since the situation can unexpectedly shed light on any issue in the design and actions of the organisation (Carmeli & Schaubroeck, 2008). Because of this, a crisis can pose a reputational threat for the organisation (Coombs, 2007).
Even though crises have real causes, whether this is due to external or internal factors, they are shaped by the communicative interplay between several actors (Kleinnijnhuis, Schultz, & Oegema, 2015). Due to the importance of communication in times of crisis, organisational
communication can affect how the crisis plays out. Several scholars even argue that it is not the crisis itself that poses a reputational threat, but the way the organisation responds to the situation (Weick, 1995; Carmeli & Schaubroeck, 2008). A crisis thus presents managerial challenges, because a weak or unfitting response can damage the organisation. This study argues that crisis communication can contribute to minimising the reputational damage in two ways. Therefore this study aims to contribute to the knowledge of effective crisis communication, which is applicable for crisis communication managers to compose a strong and fitting response to positively influence the post-crisis reputation of the organisation.
First, Coombs (2007) proposes that organisations should use crisis response strategies to reduce negative effects, prevent negative behavioural intentions and repair the organisational reputation. Coombs (2004) distinguishes three response strategies ranging in the amount of focus on the victim. Organisations can choose a deny or scapegoat strategy to remove their involvement with the crisis. When their involvement in the crisis can no longer be called into question, the organisation will have to acknowledge their connection to the crisis.
crisis by making excuses or by justifying their actions (i.e. diminish strategy).The alternative is to accept responsibility by publicly apologising to the stakeholders or offer them compensation (i.e. rebuild strategy). Coombs (2004) argues that the greater the reputational threat, the more the organisation needs to address the victims to prevent reputational damage. However, taking full responsibility can have serious (legal) consequences for the organisation. Since the organisation publicly admits their wrong doing in a statement, this could be used against them in court which could destroy the organisation (Hargie, Stapleton, & Tourish, 2010). For this reason, it is interesting to examine whether a diminish strategy in combination with other communicative elements can produce similar outcomes as a rebuild strategy on the post crisis reputation.
Second, organisations can choose to frame their messages with a rational or an emotional appeal (Claeys & Cauberghe, 2014). An emotional appeal in the message of the organisation is found to have a positive influence on the public's interpretation of crisis situations (Choi & Lin, 2007).However, the effect of communicating emotions by organisations on their post-crisis reputation is understudied (Van der Meer & Verhoeven, 2014). Most research in crisis
communication have focused on the emotions of the stakeholders, but not as a way to frame the message of the organisation (Claeys & Cauberghe, 2014).
In addition to these strategies, the role of the medium should not be underestimated. Schultz, Utz and Göritz (2011) even found that the medium has a larger influence on the public than the content of the message when the organisation applied a crisis response strategy. Since a crisis can pose a direct risk to the public, the situation requires immediate communication.
Organisations can choose to inform their public by updating the website or social media accounts. However, in contrast to communication on the website, the use of social media allows for two-way communication between the organisation and the public (Taylor & Perry, 2005). By making use of their social media accounts, the organisation can quickly and directly inform stakeholders, which in turn signals their concern for the well-being of the stakeholders (Utz, Schultz, & Glocka, 2013).
This study looks at the effects of the combination of crisis response strategy, communication of emotions, and medium on the organisational reputation. Since crisis
influenced each other based on problems that practitioners struggled with (Coombs, 2014). Early crisis communication research has therefore been characterised with case studies to guide
practitioners (Carmeli & Schaubroeck, 2008; Coombs, 2014). Since Coombs (2004) introduced the Situational Crisis Communication Theory (SCCT), crisis communication research has focused on crisis response theories and framing the message content. However, the communciation of emotions by the organisations is new to the field (Claeys & Cauberghe, 2014).Furthermore, classical crisis communication theories focus mainly on the interplay between crisis type and crisis communication strategy and neglect the role of the medium (Utz et al, 2013).
More importantly, research has not yet examined the effects of the combination of these variables on the post-crisis reputation. The purpose of this study is thus to contribute to the crisis communication literature by experimentally examine the interplay between the above described variables in order to provide a more detailed and nuanced understanding of effective crisis communication. Moreover, the study provides implications for the public since it will inform them about the ways of organisations to influence the beliefs of the public after a crisis. Furthermore, this study will provide implications for communication managers to make well informed decisions when composing and communicating their statements with regard to the use of crisis response strategy, and/or emotions, and/or the medium. Therefore this study aims to answer the following question:
RQ: To what extent does crisis response strategy, communication of emotions, and medium
influence the post-crisis reputation of an organisation?
Theoretical framework Reputation
A crisis situation can pose a severe reputational threat. However, it is essential for an organisation that their reputation stays intact, since this is an important strategic resource (Carmeli & Tishler, 2005). Corporate reputations are of an evaluative nature, therefore an organisation can have a favourable or unfavourable reputation. A favourable reputation has several advantages for the organisation such as attracting customers and employee talent, generating positive media coverage, motivating organisational members and contributing to investment interest (Fombrun &
Van Riel, 2003; Coombs & Holladay, 2006). A favourable reputation can be established by positive interactions between the organisation and their stakeholders, while negative interactions lead to an unfavourable evaluation (Coombs & Holladay, 2006). In addition to the importance of positive interactions, the organisation has to try and live up to stakeholder expectations. Specifically, stakeholders hold certain expectations about how the organisation is supposed to operate. When the organisation fails to meet these expectations, which is the case in an organisational crisis, this can be problematic for the organisation (Coombs, 2007). This section will continue by explaining how framing of organisational responses and medium of communication can contribute to minimise the reputational damage after a crisis has occurred.
Content Communication Strategies
The current study argues that organisations can frame their messages by applying a crisis response strategy and by (the lack of) communicating emotions. The theory of framing suggests that the way the information is presented will influence the reaction of the receivers of that
information (Nabi, 2003). Entman (1993) further explains framing by arguing that certain elements of a perceived reality are highlighted while others are simply ignored or downplayed. Accordingly, a specific interpretation, solution, evaluation or treatment recommendation can be encouraged.This means that when the presented frame is accepted by the receivers, the way the individual views the problem or required solution might be influenced (Nabi, 2003).
Likewise, Coombs (2007) argues that when the organisation communicates about the crisis, the crisis manager can choose to highlight or downplay certain aspects of the crisis. When a crisis response strategy is applied, the crisis manager does this by communicating about whether the crisis was accidental or intentional and if the responsibility of the crisis lies with the organisation or with an external factor. In addition, the choice can be made to use an emotional or rational approach to the situation. Following this reasoning, crisis response strategies and (not)
Crisis Response Strategies
Crisis response strategies are defined as what the management says and does after a crisis has occurred, this way the organisation can try to lessen the negative effects of the crisis by communicating about it. Crisis response strategies are used by crisis managers to establish or reinforce a specific frame, with the aim to minimise reputational damage (Coombs, 2007). Coombs (2007) identifies three clusters of crisis response strategies and seven primary crisis response strategies. In the denial cluster the organisation denies the crisis or their involvement, hereby removing the connection between the organisation and the crisis. However, in some situations the connection between the organisation and the crisis is undeniable. When the organisation
acknowledges the crisis and their involvement, either a diminish or rebuild strategy can be used. By using diminish crisis response strategies, the organisation attempts to lessen the connection between the organisation and the crisis (Coombs, 2007). The organisation then communicates they had little influence over the situation or that the crisis is not as serious as the people believe. These strategies thus involve making excuses and providing justification for the crisis. Whenever the crisis presents a severe reputational threat, Coombs (2007) recommends a rebuild strategy, which seeks to improve the organisation's reputation. This means positive actions are taken by crisis managers to balance the crisis by means of offering a full apology or
compensation to the victims. Organisations use these two clusters to empathise that they are doing everything in their power to rectify the harm that has been done (Van der Meer & Verhoeven, 2014). This study is mainly interested in when anorganisation has a clear connection to the crisis, so a denial strategy cannot be used (Coombs, 2014), therefore only the effects of a diminish and rebuild strategy are studied.
Since Coombs (2007) published these recommendations, more research has been conducted to examine the effectiveness of crisis response strategies. For example, Claeys, Cauberghe, and Vyncke (2010) presented empirical evidence by confirming that crises in the preventable cluster pose the highest reputational threat and that a rebuild strategy is the best strategy to restore the organisational reputation. In addition, Sisco (2012) conducted an
experimental research which showed that participants were more likely to engage with the
organisation when a rebuild strategy was applied opposed to a diminish strategy. Moreover, Claeys and Cauberghe (2014) also found that rebuild strategy produced the most favourable outcome when the crisis was preventable. In line with previous research of Kim, Avery, and Lariscy (2009) they identified full apology as the best strategy to positively influence the organisational reputation. Based on this the following hypothesis can be stated:
H1: A rebuild strategy will have a more positive influence on the post crisis organisational
reputation than a diminish strategy.
Emotional vs. Rational Appeal
Building on the SCCT of Coombs (2007), Jin and colleagues (2007) argue that a perspective based on emotions should be integrated in the crisis response approach of the
organisation. Even though research on crisis communication and emotions has thus far focused on the emotions of stakeholders, organisations can also choose to communicate their emotions about the crisis (Claeys & Cauberghe, 2014). More specifically, organisations can choose for either an emotional or rational appeal to frame their messages. When the message is framed as emotional, this appeals to the emotions of the public by being subjective. In contrast, when organisations present information about the crisis in an objective manner, they appeal to the rationale of the public (Claeys & Cauberghe, 2014).
Yoo and MacInnis (2005) found that when an emotional approach was used in a
commercial, the attitude towards the brand was mainly driven by feelings. While it is not surprising that positive feelings created by the commercial resulted in a more positive evaluation of the advertisement and the brand, even some negative feelings caused the same effect. Although their research was of a more persuasive nature, this can also be applicable to corporate
communication. It can thus be reasoned that when an organisation communicates their emotions, this can influence the feelings of the stakeholders about the situation and the organisation. Crisis communication research has indeed found that the public relies on emotions for the interpretation
of the situation when a crisis occurs (Jin et al, 2007) and that emotions play a key role in the evaluation of the organisation by the public (Jin & Cameron, 2007).
In addition, Choi and Lin (2007) found that an emotional appeal in the message of the organisation is also found to have a positive influence on the public’s interpretation of a crisis situation. A possible explanation for this could be that when emotions are repeatedly paired with specific ideas or events, they will eventually create the approach of the individual to interpret and respond to the event (Nabi, 2003). Emotions can thus be used to frame the message of the organisation and hereby positively influence the attitude towards the crisis and the organisation. Based on this reasoning the following hypotheses can be formulated:
H2: Messages of the organisation with an emotional appeal will have a more positive influence on
the post-crisis organisational reputation than messages without an emotional appeal.
Medium Strategy
Classical crisis communication theories focus mainly on the interplay between crisis type and crisis communication strategy and neglect the role of the medium (Utz et al, 2013). However, the effect of the medium used to communicate the message should not be underestimated. Schultz and colleagues (2011) even found that the medium has a larger influence on the public than the content of the message when the organisation uses a response strategy. This can be explained by requirement of immediate communication by the organisation when a crisis occurs. The study of Taylor and Perry (2005) reveals that this especially holds when the crisis poses a direct risk to the public. An organisation can communicate about the crisis by updating the website or social media accounts of the organisation. This way, organisations choose a more proactive approach to communicating about the recalling of their products instead of waiting for traditional media to pick up on the crisis.
When an organisation tries to repair its image, it is found to be beneficial to respond quickly (Benoit, 1997). This might also explain why the social media usage of the public intensifies (Liu,
Austin, & Jin, 2011). Furthermore, in contrast to the organisational website, the social media accounts of the organisation would allow for two-way communication. Social media are therefore considered as more interactive and dialogic media compared to traditional media and other online media. In turn this could mean that they can be considered as quicker tools for the organisation to develop a relationship with the stakeholders (Kent & Taylor, 2002; Schultz et al, 2011). In addition, the case study of Utz and colleagues (2013) reveals that when the organisation communicated about the crisis through social media this resulted in a more favourable reputation. This effect particularly occurred when the messages was communicated through Facebook.
Thus, since a crisis often disrupts the relationship between the stakeholders and the organisation, and interactions are important in the process of building a relationship,
communicating through social media instead of the organisational website might be a faster way to rebuild this relationship (Schultz et al, 2011). Based on this the following hypothesis can be stated:
H3: Messages of the organisation that are communicated through social media will have a more
positive effect on the post crisis reputation of the organisation than messages that are
communicated through the organisational website.
Interaction Effects
While several scholars found that offering a full apology has the most positive influence on reputational restoration, this strategy was used the least (e.g. Kim et al, 2009). The infrequent use of a full apology can be explained since it requires the organisation to take full responsibility. The objective of this study is to find a strategy that will produce the same effect on the post-crisis reputation of the organisation as a rebuild strategy.
As argued previously, the communication of emotions is still understudied but can be considered as a type of frame (Nabi, 2003). Since crisis response strategy can also be seen as a type of frame, communicated emotion can complement this since both frames give the public cues about the interpretation of the crisis (Coombs, 2007; Jin et al, 2007). In addition, the frames of a crisis are found to affect the emotional responses of the public, which in turn can positively affect
the attitude towards the organisation (Kim and Cameron, 2011).
In addition, the consideration of the role of the medium in crisis communication is also understudied (Utz et al, 2013). However, it can be argued that the positive effects of the use of social media, as argued earlier can contribute to the effects of crisis response strategy. The crisis response strategy can be seen as content strategy whereas the choice of medium can be
considered as a medium strategy. This study argues that by combining a diminish strategy with either an emotional appeal or communicating the message through social media (or both), this can compare to the advantages of applying a rebuild strategy. An organisation can thus resemble the effect of a rebuild strategy on the organisational reputation without taking full responsibility and potential legal consequences. In order to examine these predictions, the interaction effects of the independent variables will be considered. Therefore the following hypotheses regarding the interaction effects are formulated:
H4: Messages that apply a diminish strategy in combination with an emotional appeal will have the
same effect on the post-crisis organisational reputation as messages that apply a rebuild strategy
without an emotional appeal.
H5: Messages of the organisation that apply a diminish strategy and are communicated through
social media will the effect on the post crisis reputation of the organisation as messages that apply
a rebuild strategy and that are communicated through the organisational website.
H6: Messages of the organisation that apply a diminish strategy in combination with an emotional
appeal and are communicated through social media will the effect on the post crisis reputation of
the organisation as messages that apply a rebuild strategy without an emotional appeal and that
are communicated through the organisational website.
Moderators
To further explain the predicted effects of crisis response strategy, emotional appeal, and medium the following section will discuss two moderators that have a possible influence on the relationship between these factors and the organisational reputation.
Moderating Role of Perceived Responsibility
According to the SCCT, the reputational threat of a crisis is influenced by three factors; crisis history, prior reputation and crisis responsibility (Coombs, 2007). Since the crisis and the
organisation in this study are both fictional, crisis history and prior reputation are not taken into account. However, crisis responsibility may affect the relationship between organisational response and the reputation. For example, Coombs (2006) found that stakeholders ascribe less blame to the organisation when the crisis involves a technical breakdown accident opposed to a human
breakdown accident.
The role of responsibility is explained by Benoit (1997) who argues that perceptions are more important than reality. The concern is thus not whether the organisation is responsible for its actions, but whether the public believes the organisation is responsible for the situation. However, if the organisation is not at fault this could be a crucial element of their response. Following the reasoning of Benoit (1997) and the example of Coombs (2006), stakeholders would attribute more responsibility to the organisation in case of a preventable crisis leading to the recommendation of the use of a rebuild strategy. However, if the perception of responsibility of the organisation is low, the organisation can try to minimise the reputational threat by using a diminish strategy. This would mean that when the perception of responsibility is low, the effects of the applied crisis response strategy on the organisational reputation would be strengthened since the perceived responsibility is an important contributor of a favourable reputation.
In addition, stakeholders engage in a number of coping strategies to help themselves deal with the stress caused by the crisis and to make sense of the crisis (Jin, 2010). More specifically, when stakeholders perceive the crisis as predictable and controllable, the preferred coping strategy is of cognitive nature. This means that the organisation should apply a rational appeal in their messages. However, when stakeholders perceive the crisis as unpredictable and uncontrollable, they would adopt an emotional coping strategy, and thus prefer an organisational response that applies an emotional appeal. Following the reasoning of Jin (2010), the stakeholders’ perception of the role of the organisation is of importance in the choice whether or not the organisation should
communicate their emotions in their messages. Thus, when the perceived responsibility of the organisation is low, this can contribute to the positive effect of communication of emotions on the organisational reputation.
The public can also be more forgiving of the organisation when the responsibility of the organisation for the crisis is perceived as low (Bradford & Garrett, 1995). This could also mean that the stakeholders are more willing to rebuild the relationship with the organisation. By
communicating their messages through social media, the organisation can have a more interactive and dialogic way of communicating with its stakeholders which contributes to the development of a relationship (Kent & Taylor, 2002; Schultz et al, 2011). The perceived responsibility thus
strengthens the effect of medium choice on the organisational reputation because when the perceived responsibility is low stakeholders will be more open to two-way communication with the organisation. Therefore, the following hypothesis can be stated:
H7: When perceived responsibility of the organisation for the crisis is low, this strengthens the
positive effect of (a) crisis response strategy, (b) communicated emotion, and (c) medium on
organisational reputation.
Moderating Role of Trustworthiness
Trust is characterised by steady and expected behaviour. From an organisational point of view, trust is described as the impression that the organisation is open, honest and will fulfil its
responsibilities. Trust is created slowly over a period of time and developed by interactions and shared experiences (DiStaso, Vafeiadis, & Amaral, 2015), which can be disrupted when an organisation is going through a crisis. Distrust can lead to uncertainty and stress among stakeholders, which can make them feel uncomfortable. In turn this will lead them to carefully watch the behaviour of the organisation and it is hard for the organisation to regain the trust of its stakeholders (Govier, 1992).
which can serve as evidence for its willingness to meet the expectations of the public (Jahansoozi, 2007). An organisation can for example achieve this by apologising to the stakeholders and hereby take full responsibility (i.e. apply a rebuild strategy). Since a rebuild strategy can increase the trustworthiness of the organisation, this can strengthen the effect of crisis response strategy on the organisational reputation. This could be due to that information coming from credible sources is able to influence the beliefs, attitudes or behaviour of the receivers (Belch & Belch, 2003).
An important factor of trust is integrity, which can be described as the extent to which someone complies with a set of principles that is found acceptable (Mayer, Davis, & Schoorman, 1995). By addressing the emotions of the stakeholders in the organisational response to the crisis, the organisation can signal selflessness to its stakeholders. In turn, this can contribute to their integrity, which can lead to more trust in the organisation. In addition, when the organisation does not communicate emotions this can be perceived by stakeholders as a lack of organisational involvement and sincerity (Van der Meer & Verhoeven, 2014). Communication of emotions can thus increase the trustworthiness of the organisation, which in turn can strengthen the effect of emotional appeal on the organisational reputation.
Another important factor of trust is openness. By communicating directly with their
stakeholders, the organisation can contribute to the perceived openness of the organisation, which in turn contributes to the trustworthiness (Mishra, 1996). By communicating through social media accounts, the organisation can achieve this. Furthermore, social media offers the possibility to engage in a dialogue between the stakeholders and the organisation, this can help the
organisation to build and maintain a relationship with the stakeholders (Kent & Taylor, 2002; Schultz et al, 2011). Communication though social media can thus increase the trustworthiness of the organisation, which in turn can strengthen the effect of medium on the organisational
reputation. Based on this the following hypothesis can be formulated:
H8: The trustworthiness of the organisation is high, this strengthens the positive effect of (a) crisis
response strategy, (b) communicated emotion, and (c) medium on organisational reputation.
Figure 1. Conceptual framework of hypotheses.
Method
The assumption of causal relationships in the hypotheses explains the choice for an experiment in this study. Furthermore, an experiment offers the possibility of control over the independent variables through manipulation.
Sampling
The experimental study was conducted online and consisted of a convenience sample of 180 participants. The participants were recruited through the personal network of the researcher using email and social media. However, 14 participants did not fully complete the questionnaire and were thus removed from the sample. The final sample thus consisted of 166 respondents of which 64.50% were female and 35.50% male. The age of the participants ranged between 18 and 64 years, with an average of 29.5 (SD = 11.5). Moreover, 41.00% of the respondents completed a master's degree, 34.90% a bachelor's degree, 9.60% intermediate-level education, 11.40% had
below intermediate-level education and the remaining 3.00% indicated that they had completed another level of education than any of the described above.
Experimental Design
The experimental design was a two between-subjects (crisis response strategy: diminish vs. rebuild) x two (communicated emotion: emotion vs. no emotion) x two (medium of
communication: website vs. Facebook) factorial design. This resulted in eight experimental conditions.
Procedure
The participants received a URL through email or a direct message which enabled them to fill out the online questionnaire. The first page contained general information about the experiment which informed the participants about the procedure. After agreeing to the terms and conditions, the participants were briefed about the fictional crisis by a source that was not related to the crisis. The crisis concerned a fictitious soda company (Soda&co) that was involved in a crisis because of distributing contaminated soda bottles which resulted in 32 people to become sick. After
discovering this fact, they recalled the bottles so no further harm could be inflicted.
Four response messages were composed for the experiment, which were published on either the (mock-up) website or Facebook page of the organisation. This resulted in eight experimental conditions, to which the participants were randomly assigned. Before reading the message the participants were asked to treat the crisis and company as if they were real, even though the crisis and organisation were fictional. Next, they were asked to read the response message of the organisation about the crisis. Each participant was exposed to one of the
messages for 30 seconds to ensure the message was read carefully. After, they were asked to fill out a questionnaire that measured the dependent and moderating variables. In addition, the participants answered several questions to confirm the manipulation and regarding their
demographics. Finally, the participants were thanked and debriefed by emphasising that the crisis and the company (Soda&co) were both fictional.
Independent Variables
The crisis response strategy was manipulated by either using a diminish or rebuild strategy. The messages in the diminish cluster used strategies of excuse and justification. This was
expressed by emphasising that the contaminated soda did not pose any serious health risks. This was expressed by the following sentences: “We would like to emphasise that even though the soda can make you sick, it does not pose any serious health risks” and “We ensure you we caught the problem in time and that the distribution of contaminated bottles was limited”. Furthermore, the messages stated that the situation was out of the control of the organisation by the following
sentence: “There was nothing more we could have done to control the situation since the machines are cleaned daily and checked weekly”. The messages in the rebuild cluster used strategies of compensation and apology. Apologies were expressed by for example stating: “We offer our sincere apologies to all who suffered from sickness after drinking our soda's”. In addition,
compensation was offered with the following sentence: “As compensation for the inconvenience, we would like to offer you a coupon for a free beverage”.
The communicated emotion was manipulated by changing the tone of voice of the message. In the emotions cluster the organisation used a more conversational human voice, opposed to the messages in the no emotions cluster. This was expressed by using more personal pronouns (Kelleher, 2009). More specifically, the conditions that had no emotional appeal were partly written from third person perspective; “Soda&co has made the precautionary decision to
voluntarily recall all possible affected bottles and we advise you to return recently purchased
soda's”. Moreover, the messages without the emotional appeal avoid using personal pronouns, by for example stating: “To compensate for the inconvenience, we offer a coupon for a free beverage” instead of “As compensation for the inconvenience, we would like to offer you a coupon for a free beverage”.Furthermore, the messages in the emotions cluster contained the following sentence to
address the emotions of the public: “We understand that you might be angry since this is not the service you expect from us”.
Finally, the medium of communication was manipulated by adjusting the setting in which the messages were displayed. The four messages as described above were shown to the respondents as if they were either published on the website or on the Facebook page of the organisation. In order to simulate the Facebook conditions as authentic as possible, an actual Facebook page of Soda&co was created where the messages were posted and a screenshot was shown. In addition, it was implied that the person reading the message 'liked' the Facebook page of Soda&co.
Moreover, the image showed that over 35,000 people liked the page, including 12 of the reader's friends. Furthermore, it was suggested that the messages of the organisation was posted very recently. In the website conditions the respondents were shown the home page of the website of Soda&co. The message of the organisation headlined 'caution' in order to draw attention. The address field at the top of the image displayed the domain of Soda&co. Furthermore, the image also provided navigation bars to the online shop, locations, about us, and contact pages. Moreover, the Twitter feed of Soda&co was shown. The manipulated messages can be found in the Appendix.
Measures
The variables that were used in this study are listed below. Respondents were asked to what extent they (dis)agreed with the presented statements. Their answers were recorded on a 7-point Likert scale, ranging from completely disagree (1) to completely agree (7), unless stated differently. After recoding reversed items when necessary, scale scores were calculated by taking the mean of the item scores.
Moderating variables
The perceived responsibility was assessed by using a three item scale from Griffin, Babin, and Darden (1992). The three items were: (a) “Circumstances, not the organisation, are
responsible for the crisis,” (b) “The blame for the crisis lies with the organisation,” and (c) “The blame for the crisis lies in the circumstances, not the organisation”. A factor analysis showed that the items loaded on one factor (eigenvalue = 2.29; explaining 76.40% of the variance). A reliability test showed that the internal consistency of the scale was high (Cronbach’s α = .84).
To determine the perceived trustworthiness, a four item scale from Sirdeshmukh, Singh and Sabol (2002) was used. The participants indicated to what extent they felt the organisation was dependable, competent, of integrity and responsive to customers. An example of such a statement was “I think this organisation is competent”. A factor analysis showed that the items loaded on one factor (eigenvalue = 2.57; explaining 64.20% of the variance). A reliability test showed that the internal consistency of the scale was high (Cronbach’s α = .81).
Dependent Variable
The reputation of the organisation was measured using The Organisational Reputation Scale of Coombs and Holladay (2002) which contained five items. The items used in the study were: (a) “The organisation is concerned with the well-being of its publics,” (b) “The organisation is basically dishonest,” (c) “I do not trust the organisation to tell the truth about the incident,” (d) “Under most circumstances, I would be likely to believe what the organisation says,” and (e) “The organisation is not concerned with the well-being of its publics”. A factor analysis showed that the items loaded on one factor (eigenvalue = 2.91; explaining 56.16% of the variance). A reliability the test showed that the internal consistency of the scale was high/satisfactory (Cronbach’s α = .82). The means, standard deviations, and internal consistency of the variables can be found in Table 1.
Table 1
Means, standard deviations, and internal consistency of moderators and dependent variable
M SD α
Perceived responsibility 4.36 1.83 0.84
Trustworthiness 4.21 1.00 0.81
Reputation 4.39 1.08 0.82
Pre-test
A pre-test was conducted to check whether the manipulation of the messages succeeded (N = 15). In the pre-test, the respondents were asked to evaluate the four messages on how much responsibility the organisation took for its actions, to what extent the organisation apologised for its actions, to what extent they felt the organisation approached the situation with an emotional appeal and to what extent they felt the organisation expressed their emotions. The results of four one sample t-tests showed a significant difference in how the messages were perceived (see below) and could thus be used in the main study. The respondents of the pre-test were excluded from the main study.
Looking at the messages in the emotions condition (M1&M2), the respondents thought that the organisation took less responsibility in the message that adopted a diminish strategy (M = 3.72,
SD = 2.05) compared to the message that adopted a rebuild strategy (M = 5.67, SD = 1.05, p
<.001). In a similar way, looking at the messages in the no emotions condition (M3&M4), the
respondents thought that the organisation took less responsibility in the message that adopted a diminish strategy (M = 3.07, SD = 1.62) compared to the message that adopted a rebuild strategy (M = 5.73, SD = 1.03, p <.001).
Looking at the messages in the emotions condition (M1&M2) the respondents felt the organisation apologised less for the situation in the message that used a diminish strategy (M = 4.87, SD = 1.25) compared to the message that used a rebuild strategy (M = 5.80, SD = 1.48, p
<.001). Participants felt that the organisation apologised less for the situation in the message that
used a diminish strategy (M = 3.20, SD = 1.57) compared to the rebuild strategy (M = 6.07, SD = 0.80, p <.001) with respect to the messages in the no emotions condition (M3&M4) as well.
Looking at the messages in the diminish condition (M1&M3) the respondents felt that the organisation approached the situation more emotional in the message that adopted an emotional appeal (M = 3.67, SD = 1.59) compared to the message that did not adopt an emotional appeal (M
= 3.13, SD = 1.55, p <.001). Likewise, the respondents felt that the organisation approached the
compared to the message that did not adopt an emotional appeal (M = 5.27, SD = 1.58, p <.001) looking at the messages in the rebuild condition (M2&M4).
Looking at the messages in the diminish condition (M1&M3) the respondents thought that the organisation expressed their emotions more in the message that adopted an emotional appeal (M = 4.13, SD = 1.41) compared to the message that did not adopt an emotional appeal (M = 3.13,
SD = 1.25, p <.001). In a similar way, looking at the messages in the rebuild condition (M2&M4),
the respondents thought that the organisation expressed their emotions more in the message that adopted an emotional appeal (M = 5.53, SD = 1.30) compared to the message that did not adopt an emotional appeal (M = 5.20, SD = 1.57, p <.001).
Manipulation Checks
To confirm whether the manipulations of the materials used in the experiment were
successful, the experiment controlled for whether the participants perceived the manipulations as intended. The participants were asked if they could recall where they had read the message, either on the organisational website, Facebook page, or Twitter page. In addition, the fourth option was “I don't know/other”. Out of the participants who were exposed to the messages that were published on the organisation's website, 76.00% correctly remembered seeing the message on the
organisational website. Of the participants who were exposed to the messages that were published on the organisation's Facebook account, 94.00% correctly remembered reading the message on Facebook. A Fisher's-exact test showed this difference was significant (p <.001), indicating the manipulation of medium was successful.
For the manipulation check of crisis response strategy, two independent sample t-tests were conducted. On a 7-point Likert scale participants had to indicate if they thought the organisation apologised for its actions and how much responsibility the organisation took. The results revealed that respondents significantly perceived the organisation to be more apologising in the rebuild conditions (M = 5.39, SD = 1.37) than in the diminish conditions (M = 4.00, SD = 1.84), t (164) = -5.49, p <.001. In addition, participants thought that the organisation significantly took more responsibility for its actions in the rebuild conditions (M = 4.99, SD = 1.68) than in the diminish
conditions (M = 4.17, SD = 1.79), t (164) = -3.02, p <.01. These results indicate that the manipulation of crisis response strategy was successful.
To check if the communicated emotion was perceived as intended, two independent sample t-tests were conducted. Participants had to indicate on a 7-point Likert scale to what extent they felt the organisation expressed emotions and to what extent they believed the organisation
approached the situation from an emotional point of view. The results demonstrated that
respondents in the emotions conditions felt the organisation expressed significantly more emotions (M = 4.49, SD = 1.38) than in the no emotions conditions (M = 3.92, SD = 1.61), t (164) = -2.47, p
<.05. In addition, participants significantly felt that the organisation approached the situation from a
more emotional point of view in the emotions conditions (M = 4.36, SD = 1.60) than in the no emotions conditions (M = 3.68, SD = 1.58), t (164) = -2.74, p <.01. These results indicate that the manipulation of communicated emotion was successful.
Results Data Analyses
After the data was gathered, first several basic analyses were performed to check for missing data and to label the values. Second, the scale variables for the moderating and dependent variables were created, as described in the previous section. Next, the means and standard deviations for the scale scores for the moderators and dependent variable were determined for each experimental condition (see Table 2.1 and 2.2). In order to perform the statistical analyses to test the hypotheses, the independent variables were dummy coded*. In addition, the correlations between the independent, moderating, and dependent variables were calculated (Table 3). Out of these variables, only crisis response strategy (marginally), r(164) = .14,
p < .1, and trustworthiness, r(164) = .54, p < .001, significantly correlated with reputation. This
indicates that only crisis response strategy and trustworthiness have a direct effect the
organisational reputation. More specifically, when the crisis response strategy increases by 1 (i.e. rebuild strategy opposed to diminish strategy), the organisation reputation will increase with 0.14.
Moreover, for each additional point on the scale of trustworthiness, which runs from 1 (completely not trustworthy) to 7 (completely trustworthy) the organisational reputation will increase with 0.54.
*crisis response strategy: 0 = diminish, 1 = rebuild, emotional appeal: 0 = no emotions, 1 = emotions, medium: 0 =
website, 1 = Facebook.
Main Effects
Hypotheses 1, 2, and 3 presumed a direct effect on the post-crisis reputation of the organisation. To test these hypotheses, a two-way ANOVA was performed with crisis response strategy, emotional appeal, and medium as independent variables and the organisational
reputation as dependent variable. The results indicated that the crisis response strategy (rebuild vs. diminish) had a marginal significant positive effect on the post-crisis reputation of the
Table 3
Correlations of variables
Correlations
Crisis response strategy Communicated emotion Medium Perceived responsibility Trustworthiness Reputation
Crisis response strategy 1 -0.061 -0.020 0.044 0.049 .143*
Communicated emotion 1 0.026 -0.089 -0.025 -0.067 Medium 1 -0.025 -0.033 -0.057 Perceived responsibility 1 -0.093 -0.121 Trustworthiness 1 Reputation 1 .539*** Notes: N = 166 for all variables. The Pearson product-moment correlation coefficients are shown.
* p < .1, ** p <.05,*** p <.001. Table 2.2
Means and standard deviations of the Facebook conditions
Facebook x Diminish x Emotions Facebook x Rebuild x Emotions Facebook x Diminish x No emotions Facebook x Rebuild x No emotions
M SD M SD M SD M SD
Responsibility 3.65 1.92 4.53 2.21 4.42 1.48 4.79 1.76
Trustworthiness 4.29 1.09 3.99 1.06 4.05 0.73 4.30 1.09
Reputation 4.38 1.18 4.14 1.37 4.27 0.95 4.52 1.07
Note: FxDxE n = 25, FxRxE n = 20, FxDxNE n = 23, FxRxNE n = 22 Table 2.1
Means and standard deviations of the website conditions
Website x Diminish x Emotions Website x Rebuild x Emotions Website x Diminish x No emotions Website x Rebuild x No emotions
M SD M SD M SD M SD
Responsibility 4.68 1.74 4.02 1.88 4.54 1.75 4.33 1.86
Trustworthiness 4.03 1.17 4.47 0.94 4.28 0.93 4.26 1.00
Reputation 4.11 1.08 4.69 0.79 4.17 1.11 4.87 0.88
organisation. This implies that applying a rebuild strategy (M = 4.55, SD = 0.12) would lead to a more positive reputation than a diminish strategy (M = 4.23, SD = 0.12), F (1, 158) = 3.68, p <.1. Therefore, hypothesis 1 can be accepted. However, neither the use of an emotional appeal (no emotions vs. emotions), nor the medium (organisational website vs. Facebook) had a significant effect of the post-crisis reputation of the organisation. More specifically, the messages that communicated emotions (M = 4.46, SD = 0.12) did not lead to a more positive reputation than messages that did not communicate emotions (M = 4.33, SD = 0.12), F (1, 158) = 0.58, n.s.
Moreover, the reputation did not increase when the message was published on the Facebook page (M = 4.33, SD = 0.11) opposed to when the message was published on the organisational website (M = 4.46, SD = 0.12), F (1, 158) = 0.62, n.s. Accordingly, hypotheses 2 and 3 were rejected.
Hypotheses 4, 5 and 6 assumed interaction effects between the independent variables. The analysis showed a significant marginal interaction effect for the interaction between crisis response strategy and medium on the organisational reputation, F (1, 158) = 3.53, p < .1, see figure 2. More specifically, when the messages were published on the website a rebuild strategy (M = 4.78, SD = 0.18) had a more positive effect on the organisational reputation than a diminish strategy (M = 4.14, SD = 0.17). However, when the messages were posted on the Facebook page of the
organisation, there was only a minor difference between the effect of a rebuild strategy (M = 4.33,
SD = 0.17) and a diminish strategy (M = 4.32, SD = 0.16) on the organisational reputation.
Nevertheless, since the message that applied a rebuild strategy and was posted on the website had the most positive effect on the post-crisis reputation, hypothesis 5 cannot be accepted.
Moreover, there were no significant effects found for the interaction between crisis response strategy and emotional appeal. Opposed to the assumption of hypothesis 4, the message that applied a diminish strategy and communicated emotion (M = 4.24, SD = 0.16), did not have a more positive effect on the post-crisis reputation than the message that applied a rebuild strategy and did not communicate emotions (M = 4.70, SD = 0.16), F (1, 158) = 0.81, n.s. This implies that hypothesis 4 must be rejected. In addition, no significant effect was found for the interaction between the three factors. This means that contrary to the expectation of hypothesis 6 that the message that applied a diminish strategy in combination with an emotional appeal and was
communicated through social media (M = 4.38, SD = 0.22) did not have a more positive effect on the post crisis reputation of the organisation than the message that applied a rebuild strategy without an emotional appeal and that are communicated through the organisational website (M = 4.87, SD = 0.23), F (1, 158) = 0.29, n.s. Hence, hypothesis 6 is rejected.
Figure 2. Effect of the interaction between crisis response strategy and medium on organisational reputation.
Moderating Effects
To assess the influence of perceived responsibility and trustworthiness on the relationship between crisis response strategy, communicated emotion, and medium choice and post-crisis reputation a multiple regression analysis was performed (see Table 4). The multiple regression analysis allowed to simultaneously test the main effects of crisis response strategy, emotional appeal, and medium on organisational reputation and the potential moderating effects of perceived responsibility and trustworthiness. Before this was done the assumptions for conducting a
regression analysis were tested. First, an analysis of standard residuals was carried out, which showed that the data contained no outliers. Second, the histogram of standardised residuals showed that the variables were normally distributed and the scatterplots showed that the data met the assumptions of homogeneity of variance and linearity.
computed. The multiple regression analysis was conducted with the standardised versions of crisis response strategy, communicated emotion, medium, perceived responsibility, and trustworthiness as independent variables and (the unstandardized version) reputation as dependent variable. Using the enter method, results showed that the independent and moderating variables explain a significant amount of the variance in the organisational reputation (R² = .36, F (11, 154) = 7.96, p <.001). .
In line with the results of the two-way ANOVA, the results showed a significant direct effect of crisis response strategy (b* = 0.13, t (154) = 2.01, p <.05). However, no significant effect was found for communicated emotion (b* = -0.06, t (154) = -0.98, n.s.) nor the medium (b* = -0.02, t (154) = -0.30, n.s.) on the organisational reputation. This means that the crisis response strategy significantly predicts the organisational reputation. More specifically, when the a rebuild strategy is applied, the organisational reputation will increase by 0.13.
Furthermore, the analysis showed no significant effect of perceived responsibility (b* = -0.08, t (154) = -1.27, n.s.) on the organisational reputation. However, a significant positive effect was found for the influence of trustworthiness (b* = 0.53, t (154) = 8.10, p <.001) on the
organisational reputation. This means that the organisational reputation can be predicted by the perceived trustworthiness of the organisation. More specifically, for each additional point on the scale of trustworthiness, which runs from 1 (completely not trustworthy) to 7 (completely
trustworthy), the organisational reputation will increase with 0.53. It must be noted that for this effect other independent variables are assumed to be held constant.
Looking at the moderating effects of perceived responsibility and trustworthiness, the analysis showed no significant effects of the interaction between crisis response strategy and trustworthiness (b* = -0.08, t (154) = -1.26, n.s.) on organisational reputation. In other words, trustworthiness had no significant effect on the relationship between crisis response strategy and organisational reputation. However, the analysis did show a significant negative effect of the interaction between crisis response strategy and perceived responsibility (b* = 0.12, t (154) = -1.77, p <.1). Put differently, perceived responsibility had a negative influence on the relationship between crisis response strategy and organisational reputation (see Figure 3). Since the results
also showed a significant effect of crisis response strategy on the organisational reputation, the effect of crisis response strategy is not dependent on the value of perceived responsibility. However, when ‘more’ crisis response strategy is applied (i.e. a rebuild strategy over a diminish strategy) and the perceived responsibility is low, the organisational reputation increases. This means that a low perception of crisis responsibility strengthens the effect of a rebuild strategy on the organisational reputation.
Figure 3. Effect of the interaction between crisis response strategy and perceived responsibility on organisational reputation.
The analysis furthermore showed that there were no significant effects between the
interaction between communicated emotion and perceived responsibility (b* = 0.06, t (154) = 0.91, n.s.) and communicated emotion and trustworthiness (b* = -0.08, t (154) = -1.11, n.s.) on the reputation of the organisation, In other words, perceived responsibility and trustworthiness have no influence on the relationship between communicated emotion and organisational reputation.
The analysis showed no significant effect for the interaction between medium and
trustworthiness on the organisational reputation (b* = -0.05, t (154) = -0.82, n.s.).This implies that trustworthiness has no influence on the relationship between medium choice and organisational reputation. However, a significant negative effect was found for the interaction between medium and perceived responsibility on the organisational reputation (b* = -0.11, t (154) = -2.16, p <.05).
This indicates that the perceived responsibility has a negative influence on the relationship between medium choice and organisational reputation (see Figure 4). Since the results did not reveal a significant effect of medium on the organisational reputation, the effect of medium is dependent on the value of perceived responsibility. The graph illustrates that the effect is strongest when medium is high (i.e. Facebook instead of organisational website). More specifically, if the perceived responsibility is low and the message is communicated through Facebook, the organisational reputation increases. However, when the perceived responsibility is high and the message is communicated through Facebook, the organisational reputation decreases. This means that a low perception of crisis responsibility strengthens the effect of communication through Facebook on the organisational reputation.
Figure 4. Effect of the interaction between medium and perceived responsibility on organisational reputation.
Based on these results, it becomes clear that hypothesis 7b must be rejected since there was no significant effect found of perceived responsibility on the relationship between communicated emotion and organisational reputation. However, since the results did reveal an increased effect of crisis response strategy on the organisational reputation when the perceived responsibility was low, hypothesis 7a can be supported. In addition, the results showed an effects of medium on the organisational reputation increased when the perceived responsibility was low. Hence, hypothesis 7c can be supported. Moreover, no significant effects were found for the moderating role of
trustworthiness on the relationship of any of the independent variables and the organisational reputation. Hypotheses 8a, 8b and 8c must therefore be rejected. The results are presented in Table 4.
Conclusion
The aim of this study was to give insight to the interplay of crisis response strategies, communication of emotions and medium on the post-crisis reputation of an organisation. This could be useful for crisis managers to use alternative strategies than a rebuild strategy in order to achieve the same effect on the organisational reputation. In addition, this research contributes to the crisis communication literature by being the first to look at the interplay of these three variables. Although no main effects of the communication of emotions and the choice of medium on the organisational reputation were found, crisis response strategy did seem to have a positive effect on the organisational reputation. More specifically, applying a rebuild strategy leads to a more
favourable organisational reputation than a diminish strategy. This is in line with the expectations based on the recommendations of Coombs (2007) and previous research which reveals that offering a full apology in case of a preventable crisis had the most positive outcome on the
evaluation of the organisational reputation (Claeys et al, 2010; Claeys & Cauberghe, 2014; Kim et al, 2009; Sisco, 2012).
Furthermore, perceived responsibility proved to have a negative effect on the relationship between crisis response strategy and organisational reputation. This implies that when the organisation applies a crisis response strategy, the organisational reputation becomes more favourable especially when the perceived responsibility of the organisation is low. However, this was only true when a rebuild strategy was applied, suggesting that high perceived responsibility does not influence the effect of crisis response strategy on organisational reputation. This finding is in not line with Coombs (2007), who suggested a diminish strategy would lead to less a favourable post-crisis reputation when the perceived responsibility of the organisation for the crisis is high. Based on the results of this study, it can be concluded that when the perception of responsibility is high, the organisation could apply either a diminish or a rebuild strategy. However, when the
perceived responsibility is low a rebuild strategy is more effective. This could be due to the fact that when stakeholders perceive the organisational responsibility has high, they already decided upon the role of the organisation in the crisis. Therefore a crisis response strategy would not change their evaluation of the reputation of the crisis. However, when the perception of responsibility is low, a rebuild strategy would contribute to a more favourable reputation (Coombs & Holladay, 2008). However, in practice it is hard to determine the level of responsibility stakeholders attribute to the organisation first since the organisation also has to respond to the crisis quickly (Coombs, 2007; Taylor & Perry, 2005).
No direct effect of medium on the organisational reputation was found which is not in line with the findings of Utz and colleagues (2013) and Schultz and colleagues (2011). This could be explained by the fact that their research compared social media strategies to (online) traditional media. The direct effect might thus not be due to the incorporation of social media, but due to the fact that the organisation was the communicator. However, the results did demonstrate that a combination of crisis response strategy and medium has a significant positive effect on the organisational reputation. Even though the results illustrate that the evaluation of the reputation remained more favourable when the organisation applied a rebuild strategy, the results also indicate that when the organisation published the message on Facebook (vs. the organisational website) the participants evaluated the organisation equally. The message that applied a rebuild
strategy thus only had more positive effect on the organisational reputation when it was published on the organisational website. Based on the results of this research it can thus be concluded that when the message is published on Facebook, the more favourable effect of a rebuild strategy disappears.
Moreover, medium choice especially seemed to have an effect on the organisational reputation when the perceived responsibility was high. This implies that the organisation should choose to communicate their message through the organisational website when the perceived responsibility of the organisation for the crisis is high. Again, it might be that stakeholders already decided upon the role of the organisation in the crisis. Therefore, they would not be open to rebuild the relationship with the organisation (Schultz et al, 2011), but purely want information from the organisation. This would also explain the finding that a using a rebuild strategy and publishing the message on the website had the most positive outcome for the organisational reputation. However, a low perception of crisis responsibility strengthens the effect of communication through Facebook on the organisational reputation. Which could mean that stakeholders want to engage with the organisation to rebuild the relationship.
Contrary to the findings of previous research (Claeys & Cauberghe, 2014; Van der Meer & Verhoeven, 2014)the use of an emotional appeal in the messages of the organisation did not contribute to a more favourable evaluation of the post-crisis organisational reputation in any way. This might be explained based on the crisis type used in this study. As discussed previously, Jin (2010) suggested that individual prefer a cognitive coping strategy when they perceive the crisis as predictable and controllable. However, the absence of the role of perceived responsibility cannot be explained based on the argumentation of Jin (2010). This effect might be beyond the scope of this research and should be explored further in future research. It might also be possible that communicated emotion did not add to the effects of crisis response strategy because these strategies already carry emotional factors (Van der Meer & Verhoeven, 2014). In particular, a rebuild strategy communicates more emotional components than a diminish because it includes an apology.
have any moderating effects on the relationship between crisis response strategy, communication of emotions, and medium and the post-crisis reputation. However, the results demonstrate that trustworthiness of the organisation predicts the evaluation of reputation. Trust is thus an important predictor of the reputation, which is in line with the findings of DiStaso and colleagues (2015).
To conclude, the research question of this study: “To what extent does crisis response
strategy, communication of emotions, and medium influence the post-crisis reputation of an organisation?” can be answered. Based on the results there seems to exists a noticeable
relationship between the interplay of crisis response strategy and medium and the post-crisis reputation of the organisation. However, the use of an emotional appeal did not contribute to a more favourable evaluation of the post-crisis organisational reputation.
Discussion Limitations
There are several limitations to this research, which can explain the deviant results of the study. First, there were some issues with the sample of the study. Namely, in the study women were overly represented, in addition to young and highly educated people. The sample was
therefore not representative for the population. Moreover, the size of the groups that were exposed to the experimental conditions were rather small.
Second, as the reputation of an organisation develops over time (Gray & Balmer, 1998) it is therefore difficult to assess by a single exposure to a fictional organisation. In addition, the study measures a post-crisis attitude towards the organisation, without measuring the pre-crisis attitude towards the organisation. It is therefore hard to examine the impact of the crisis and the response strategies on the organisational reputation. Future research should therefore also consider
including a control group that would not receive a crisis response strategy. However, the design of this study did not include a control group because the main objective of the study was to examine under which conditions an organisation could use a less expensive response strategy.
Third, the study focused on a preventable crisis which is the crisis type with the most severe reputational threat (Coombs, 2007). It is possible that another crisis type would produce
different results. In addition, only one organisation was examined, which was quite specific (food/beverages). Due to these limitations, the generalisability of the results is limited as well. Moreover, the possibility exists that a study that focuses on a different organisation or an organisation in another sector, would produce contrasting results.
Fourth, the questions were mainly answered somewhat higher than the centre of the scale. Greenleaf (1992) argued that this could be because of moderacy bias. This implies that
respondents have a preference for midpoint responses on the scales in the questionnaire. Within this research, the respondents had to answer the questions on a 7-point scale, by creating a 6-point scale this problem could be solved. This would mean that respondents are forced to answer either in favour or against a statement.
Lastly, to address the deviant results in the effects of the use of an emotional appeal in the organisational response, the study did not take the personal preferences of information processing into account. Verbeke (2005) argues that the preference of how information should be presented might be based on the personality of the receiver. Future research could take the influence of personal preferences into account and the emotions the crisis caused with the stakeholders. Moreover, since crisis type determines the preferred coping strategy (Jin, 2010), future research should examine whether a different crisis type would produce contrasting effects.
In addition, Jin et al (2007) argued that in order for crisis managers to make optimal communicative choices and address the specific needs of the stakeholders, they need to understand how stakeholders feel, evaluate and experience the crisis situation. However, this study did not take the emotions of the public into account. In addition, Coombs and Holladay (2005) also pointed out that crisis managers can make a more informed choice of response strategy when they understand the emotions of their stakeholders. Future research should thus look at the emotions of stakeholders that are evoked by the crisis and take these into account. Furthermore, this study looked at the communication of emotions in general. Future research should consider making a distinction between different emotions that are communicated based on the emotions that the crisis evoked among the stakeholders.
Implications
This research contributes to the crisis communication literature by experimentally studying the interplay of crisis response strategies, the communication of an emotions, and medium on the post-crisis organisational reputation. The findings of this study also contribute to further development of the SCCT. Moreover, they offer several managerial implications.
First, the role of perceived responsibility of the organisation should not be underestimated when choosing a crisis response strategy. As the results demonstrate, when the perceived
responsibility is high the organisation can choose to apply either a diminish or rebuild strategy and achieve an equal effect. Choosing a diminish strategy would be more beneficial for the
organisation since this does not include taking full responsibility. However, as mentioned
previously, this would be hard in practice becaue it is hard to determine the level of responsibility stakeholders attribute to the organisation before communicating an organisational response since the organisation has to respond to the crisis quickly (Coombs, 2007; Taylor & Perry, 2005).
Second, the study answered to the call of Utz and colleagues (2013) for more research on the role of medium in crisis communication. This produced an important insight for communication managers. Namely that even though a rebuild strategy proved to be the most beneficial strategy with regard to the post-crisis reputation, a diminish strategy can produce a similar effect when the message is posted on the Facebook account of the organisation. Since the differences in
evaluation of reputation are very small, organisations can best choose to apply a diminish strategy and publish their message on Facebook. However, when the perceived responsibility of the
organisation is high, the organisation should avoid communicating through Facebook.
On a final note, it is important to mention that even though these strategies can contribute to minimising the damage of a crisis, crisis management is more than having a plan and carrying this out during the crisis. Every organisation should check for potential crises on a daily basis and undertake action to prevent them from happening (Coombs, 2014).
