This is getting personal
_____
The relation between tone of voice and organizational
reputation in online crisis communication
Anne Bindels Dr. P. Verhoeven
10711449 26 juni 2015
Masterthesis
Abstract
As the field of crisis communication is dominated by case studies, the purpose of this study was to get a better empirical understanding of how tone of voice in social media relates to organizational reputation in the context of a crisis. With the use of an online survey and a 2 (tone of voice: conversational human voice and corporate voice) x 2 (crisis response
strategies: accommodative and defensive) experimental design, results show no differences between the two voices or between the two crisis responses. It was also found that with a personal, friendly, online message, the organization is seen as more social present and its account is more easily accepted. Moreover, social presence and account acceptance correlate positively but moderately with reputation. No differences in scores were found between the accommodative and defensive crisis response strategy.
Keywords: crisis communication, conversational human voice, social presence, crisis response strategy, Facebook, reputation
Introduction
“Update Geldermalsen: no trains Houten Castellum-Zaltbommel, no intercity’s Utrecht-Den Bosch. Travel via Arnhem or Rotterdam.” Clear, plain information about trains that do not run, a recurring and bothersome problem. This is an ordinary tweet from NS, the biggest railway organization of the Netherlands. NS is a commonly known organization that uses social media to inform the public real-time about problems that could affect travellers. The tweet is straightforward, but is it enough? When a crisis like this occurs, what are the factors that determine the reputational outcomes?
Since the rise of social media, consumers changed from passive receptors of information into active participants of the online marketplace (Hennig-Thurau, Malthouse, Friege, Gensler, Lobschat, Rangaswamy & Skiera, 2010). They gained control, because social media allowed them to share and create instead of just consume (Macnamara, 2010). Organizations are trying to tap into the online communication trend of getting close and familiar with their consumers. By trial and error they discover which strategies work and which do not, using case studies as an example. Theory, however, stays behind. Communication professionals often work with case studies to guide them in their online activities, because relevant literature barely exists yet. Scientific knowledge of online communication strategies is lacking, but much needed (Coombs & Holladay, 2009).
Communication strategies get attention to increasing extent. The world is largely an online environment, so organizations that want to communicate with the publics, need to be present in that environment. The shift from traditional media to social media is a difficult one for organizations because of the limitless possibilities. Twitter and Facebook seem to have taken the stage as platforms for organizations to communicate quick and real-time with their customers. The characteristics of these platforms (speed, interactivity, sociability) make them pre-eminently the tool for communication, especially crisis communication (Schultz, Utz & Göritz, 2011). When a crisis occurs, people search for reliable information that is up to
date. They view the Internet as ideal source for news because of its real-time possibilities and interactivity, so it is not unexpected that social media usage increases during a crisis (Liu, Austin & Jin, 2011). Social media allows information to spread rapidly and easily (Smith, 2010) and people have the possibility to share it and react on it. And because of this full speed distribution of information, it is of great importance that the crisis communication strategy is executed perfectly.
The right execution of communication concerning a crisis is important for several reasons with the main goal of protecting the company’s reputation. But what factors in crisis communication influence the consumers’ perception of the organization? Social presence and perceived interactivity are found to be relevant elements in online communication, which could ultimately affect reputation (Park & Cameron, 2014). The same counts for the crisis response strategy. According to existing literature on crisis communication, it is assumed that when a company sends out a crisis message, people are more willing to accept it when the company apologizes and admits its mistakes than when no excuses are made (Coombs, 2007; Schultz et al., 2011). In this study, an attempt is made to identify effective and ineffective crisis communication strategies. The purpose is to answer the following research question: what is the relation between of tone of voice and reputation in online crisis communication?
Theoretical background
Literature on crisis communication is not abundant. Most of it is based on case studies, while evidence-based theory is lacking (Coombs, 2007; 2015). Park and Cameron (2014) have addressed this issue in a complete and thorough manner by exploring the influence of tone of voice, source and crisis response strategy on reputation. They developed an online survey and fictitious blog messages concerning a crisis and used a 2 (tone of voice) x 2 (source) x 2 (crisis response strategy) mixed experimental design. The results showed that
interactivity. Furthermore, it was found that a high perception of social presence and interactivity led to positive post-crisis outcomes such as high reputation. The current
research builds on the findings of Park and Cameron (2014), their study therefore served as a guide. In this section, literature on crisis communication, social media and tone of voice will be reviewed. After that, hypotheses are formulated in accordance with the discussed literature.
Crisis communication
A crisis can be described as the perception of an unpredictable happening that causes a threat to stakeholders’ expectancies, the performance of the organization and its reputation (Coombs, 2007). In the field of corporate communication, literature on crises is mostly organization-centred. Crises are viewed as a threat to the reputation of the organization (Coombs, 2007; Coombs & Holladay, 2002; 2009; Verhoeven, Van Hoof, Ter Keurs, Van Vuuren, 2011). Although tangible damage like financial or property loss does not always occur, organizational perception of stakeholders may change nevertheless (Sohn & Lariscy, 2012; Stephens & Malone, 2010). Sensemaking and reducing uncertainty are important responsibilities of the organization, because without explanation, apology or elaboration, it is more likely the crisis will get out of control (Coombs, 2007; Smith, 2010: Verhoeven et al., 2011; Weick, 1988). This outcome can be prevented by well thought-out crisis
communication. A good crisis communication strategy will not only protect the company’s reputation, it might even improve it (Coombs, 2007; Sohn & Lariscy, 2012).
A great contributor to literature on crisis communication is Coombs. His extensive research on crisis communication and strategies led to the development of the situational crisis communication theory (SCCT), a framework that provides insights in how to protect the organizational reputation (Coombs, 2007). With SCCT, stakeholders’ perception of the organization’s reputation can be identified, which in turn will provide guidance for post-crisis
communication. Its roots are in the Attribution Theory, which states that people tend to seek for the cause of an event, especially an unfortunate or negative one (Weiner, 1985). People often will experience a core emotion of the Attribution Theory in reaction to the situation, anger or sympathy. SCCT builds on this theory by predicting the reputational threat and with that, suggesting crisis response strategies to prevent or revert harm to the reputation of the organization (Coombs, 2007). A crisis manager is more able to decide which crisis response strategy is best suitable to protect the organization’s reputation when he understands the crisis situation.
Coombs (2007) distinguishes SCCT crisis types and SCCT crisis response strategies. The crisis types are clustered in three groups: the victim cluster, in which the organization is also victim of the crisis, the accidental cluster, in which the actions of the organization were unintentional but led to a crisis, and the preventable cluster, were the organization took harmful actions, knowing the risks. These clusters respectively cause a mild, moderate and severe reputational threat. Crisis history also plays a role in determining the reputational threat, whereas crisis history or an unfavourable previous reputation
increases the threat (Coombs, 2007).
The crisis response strategies developed by Coombs are designed to fit the perceived acceptance of responsibility for the crisis (Coombs, 2007; 2015). The more accommodative the response, the more responsibility an organization takes. Words and actions of an organization shape people’s perception (Coombs, 2007), so response strategy has to be determined with care. The primary crisis response strategies can be grouped in three categories, based on acceptance of responsibility: denial, diminish and rebuild (Coombs, 2006). With the deny strategy the organization seeks to remove the connection between itself and the crisis situation. This could be done by denying the existence of the crisis or by attacking the accuser. The diminish strategy attempts to convince people that the situation is not so bad as they think, or that it is out of control of the organization. An
example of this is claiming there is minimal damage. An organization uses the rebuild strategy to generate positivity towards its reputation, for instance with an apology. It is the task of the organization to determine the best crisis response. But, as Coombs recognises, organizations often do not have the financial resources to go for the best crisis response. The best option is then to determine what strategy is the second best fit, but has less financial consequences.
In earlier times, research assumed traditional media are the tools for communicating with, or actually to the public, also when it comes to crises. Now people more and more rely on Internet as their source of news (Liu, Austin & Jin, 2011), it is time to bridge the gap in time. A tremendous amount of time is spent online, so online has automatically become the place where people get their information. Since information about organizations mostly derives from the media, and as discussed before, more and more from online media, there is little surprise in a need for an online crisis strategy (Coombs, 2007). Organizations should use this knowledge in their advance and embed social media in their crisis communication strategy.
Organizational use of social media
The use of social media provides various opportunities for organizations. With traditional media, there was a situation with mostly one-way communication (Grunig & Grunig, 1992). Organizations focused on how to get information to consumers hoping to gain attention. With the rise of new media and later social media, consumers changed from passive receptors into actives that share and create (Kaplan & Haenlein, 2010). For scholars do not yet agree on one definition of social media, a combination of several studies lead to the following description. Social media are regarded as the digital tools and applications build on the technical ideal of Web 2.0, that facilitate interactive communication, creation and exchange of information among and between the public and organizations (Liu et al., 2011;
Hennig-Thurau et al., 2010). Consumers are getting more and more control of what happens with online and offline content (Macnamara, 2010), becoming active market players, which in turn leads to organizations realising the need of extensive two-way communication. Two-way communication is defined as symmetrical communication in which the organization both seeks and gives information, actively pursuing the dialogue and engagement (Grunig & Grunig, 1992; Grunig, 2001). By being present and active on social network sites like Facebook and Twitter, an organization is able to get closer to its customers. With social media embedded in its communication plan, providing information is not the main issue anymore. More value lies in getting a conversation started, asking for feedback and ideas, listening to and monitoring consumers (Schultz et al., 2011, Schultz & Wehmeier, 2010). Social media offers the opportunity to engage.
Since use of social media in organizational context is relatively new, it is not yet fully embedded in the work of communication professionals (Grunig, 2009). As stated, social media offer the organization close contact with its customers, inviting them to get into a dialogue. This conversation can lead to a better understanding of the market (Muntinga, 2011; Kaplan & Haenlein, 2010), more brand love (Batra, Ahuvia & Bagazzi, 2012) and many more benefits.
On the other hand, because of social media being a relatively new element to the corporate communication, organizations often view them as a possible threat. The dialogue that arises is open and uncontrollable which causes that the organization can only make an attempt to manage content. The unpredictability of what happens with content once it is served up in the online environment is called the pinball effect (Hennig-Thurau et al., 2010). This metaphor illustrates how content, the pinball, is put in the marketplace and how
communication professionals can only try to steer what happens to it by using the flippers. In the chaos of the environment, however, it is difficult if not impossible to control what
Social media in a crisis
Social media have several characteristics that make them an obvious choice to disseminate information through when an organization is facing a crisis (Schultz et al., 2011). Social media can spread information at very high speed, and therefore the consumer can be informed within a very short period of time. Whether the crisis is internal or external,
consumers want to get information in a clear and quick way. The organization can create a message on Twitter or Facebook explaining what is happening, giving consumers the chance to understand. Also, those consumers can inform others. The rapid distribution of content is possible by consumers having the possibility to share it and react on it. Even when consumers do not share or react on it online, they probably will do offline, which is another argument for online crisis communication (Liu et al., 2011). People have the tendency to seek human contact in insecure situations, thus crisis information will also be spread offline (Liu, Fraustino & Jin, 2015; Sohn & Lariscy, 2012). The use of social media is therefore effective and efficient in crisis management.
Tone of voice
The tone of voice is a relevant element in the organization’s communication (Van Noort, Willemsen, Kerkhof & Verhoeven, 2014). According to research, content written in first-person narrative style and with a first-personal touch is more likely to be accepted and
appreciated than a message with a corporate tone of voice (Huibers & Verhoeven, 2014). In a context where people are in constant interpersonal contact, they expect the
communication to be open, transparent and authentic (Xifra & Huertas, 2008). Kelleher (2009) was the first research to explore the conversational human voice (CHV) and its effects on public’s response. He described CHV as “an engaging and natural style of organizational communication as perceived by an organization’s publics based on interactions between individuals in the organization and individuals in publics.” (Kelleher,
2009, p. 177). Characteristics of this communication style are messages written in first-person narrative style, using ‘I’ or ‘we’, speaking directly to the public, using humour, emotions and admitting mistakes (Kelleher & Miller, 2006).
On the other hand, an organization can communicate in a formal and distant way, with a corporate voice. This entails no use of first-person narrative style, no use of emotions and just providing straightforward information (Park & Cameron, 2014). With this way of communicating there is no intention to imply personal contact, a real conversation or dialogue. It is argued that communication with a corporate voice is less interesting to online publics, because it seems like the organization shows little interest in them. It rather sounds like a non-human entity than like a real person (Searls & Weinberger, 2001).
It is stated that the use of CHV has a positive effect on the response of the public because of its transparency and openness (Kelleher, 2009). According to previous
communication studies, it is plausible that this tone of voice influences attitude towards the organization in a positive way, as well as the intention to use word-of-mouth (WOM) communication (Park & Cameron, 2014). This positive effect may possibly extend to the reputation of the company. The main question in this research therefore is: what influence does conversational human voice have on perceived reputation?
Social presence
Literature states that the positive effects of CHV could be verified through the social
presence theory of Short, Williams and Christie (Park & Cameron, 2014). Their theory stems from the 1970’s and assumes that an important element of effective communication is that the counterparty is perceived as real and authentic. Social presence is the degree to which someone or something is seen as a real person. The higher the level and amount of interpersonal interactions, the higher social presence will be (Park & Cameron, 2014). People tend to get the feeling they engage in human interaction when the online
communication has features that imply a personal dimension (Kelleher, 2009). A personal dimension can be achieved by for example using the first-person narrative style, writing in an informal way or using humour. Therefore, the following hypothesis is:
H1: Perceived social presence is higher when a crisis management message is expressed with a conversational human voice rather than with a corporate tone of voice.
Interactivity
Another important aspect in online communication is interactivity. Interactivity is seen as the essence of new media in contrast to traditional media. Liu and Shrum defined it as “the degree to which two or more communicating parties can act on each other, on the
communication medium, and on the messages and the degree to which such influences are synchronized” (2002, p. 54). It implies two-way communication, giving the user the
opportunity to respond, give feedback or ask questions, like it would be in an offline conversation.
As far as research goes on interactivity, it showed that there is a positive influence of interactivity on attitudes towards the organization (Van Noort, Voorveld & Van Reijmersdal, 2012; Liu & Shrum, 2009). However, a distinction needs to be made between actual and perceived interactivity. Actual interactivity is objective, measurable by counting interactive elements. Perceived interactivity is subjective, and does not have to be equal with actual interactivity. It was found that the number of interactive functions does not correlate with perceived interactivity (Voorveld, Neijen & Smit, 2013). The incongruence between these two types of interactivity suggest that some features have the ability to really make a difference in the perception of consumers, while some are barely noticed and therefore do not result in giving an interactive feel to the website (Voorveld, Neijens & Smit, 2013).
and it is more effective in persuasion (Liu & Shrum, 2013). Wu, Hu and Wu (2010) also found that perceived interactivity has a positive effect on online trust. These findings make it arguable that the use of conversational human voice will increase the perceived interactivity. A corporate tone of voice suggests distance will probably not increase perceived
interactivity. The next hypothesis will thus be:
H2: Perceived interactivity is higher when a crisis management message is expressed with a conversational human voice rather than with a corporate tone of voice.
Crisis response strategy
Organizations should be aware of possible threats to their reputation because it enables the organization to respond to crisis situations in a quick and prepared way (Coombs, 2007). Awareness makes anticipation in case of a crisis easier, especially if anticipation exists in the shape of a crisis communication strategy. Whenever a crisis occurs, internally or
externally, a crisis response strategy is needed. A message from crisis management is more likely to be accepted when conveyed through conversational human voice than through an organizational voice, because the people prefer interpersonal communication over
impersonal communication (Liu et al., 2015). Again, a corporate, formal tone of voice is not engaging and creates distance between the organization and the public. This may be an indication for the following hypothesis:
H3: A crisis response is more acceptable when expressed with a conversational human voice rather than with a corporate voice.
Not only tone of voice might have an impact on whether a crisis response will be accepted. Also the actual content of the crisis message could make a difference. According to Coombs
(2007), an organization can choose from different options and combinations when setting up a strategy. There is the denial, in which the organization does not take any responsibility or even denies the existence of the crisis, the defensive strategy. Another option is the
accommodative strategy, where the organization takes responsibility for its actions and makes a sincere apology (Coombs, 2007; Park & Cameron, 2014; Lee, 2011). There is logic in assuming that a message with an apology is more acceptable than a message that just contains plain information and no excuse at all, but then it has to be perceived as sincere (Van Noort et al., 2014). However, Liu et al. (2011) found that people are willing to accept a defensive message if it comes directly from the organization experiencing the crisis. Schultz et al. (2011) also found that the medium matters more than the actual
message. But when source is left out as a variable, would a defensive strategy still work? By testing the following hypothesis, an attempt to answer to this question is made:
H4: A crisis response is more acceptable when expressed following an accommodative strategy in contrast to a defensive strategy.
The verification or falsification of this is hypothesis is relevant, because acceptation of a crisis message could lead to a higher reputation (Coombs, 2007; Schultz et al., 2011). This leads to the next hypothesis:
H5: The more a crisis message is accepted, the higher reputation will be.
Reputation
Even though it is commonly accepted that corporate communication does have a signicant impact on reputation, it is not yet fully understood how online communication strategies, in particular the crisis strategies, are linked to reputation (Floreddu, Cabiddu & Evaristo, 2014).
The reputation is one of the most valuable assets of the organization, and it is the one thing that distinguishes it from other organizations. It may even serve as buffer in times of a crisis (Sohn & Lariscy, 2012). Although it is not an easy task and there is no consensus over the construct reputation (Martin & Burke, 2012), scholars have attempted to create a definition of reputation (Gotsi & Wilson, 2001; Coombs, 2007, Fombrun, Gardberg & Sever, 2000). Fombrun et al. (2000) describe corporate reputation as a collectively made construct that portrays the perceptions of various stakeholders about the organization’s performance.
Since the use of conversational human voice in online communication is said to have a positive effect on attitude (Kelleher, 2009), it is expected that this effect may also be visible on reputation. As this is a study in the context of corporate communication, reputation is viewed as organizational reputation. Accordingly, the final hypothesis and with that the main question of this research is as follows.
H6: Reputation is higher when a crisis response message is expressed with a conversational human voice rather than with a corporate voice.
Method
In this study, a 2 (tone of voice: conversational human voice versus corporate voice) x 2 (crisis response: defensive versus accommodative) experimental design is used. Every subject was exposed to one of the four conditions, so both tone of voice and crisis response served as between-subject factors. The texts and questions in the online survey were all written in Dutch.
Sample and procedure
The participants of this study were 170 people in the network of the researcher. They were approached offline and online via e-mail and Facebook, which resulted in friends, family,
fellow students and acquaintances filling out the online survey. This is called a snowball sample and works like a chain referral where subjects ask other people to participate in the study.
Participants received a message in which they were asked to fill out an online survey with a reference to the survey link. No previous information about the purpose of the study was given to prevent biased answers. By clicking on the link in the message or e-mail, they were directed to the survey. For ethical reasons, participants had to approve of participating in the online survey. They read the following text and checked the box if they wished to continue, and by that agreed to the terms of ASCoR, Amsterdam School of Communication Research:
I hereby declare that I have been informed in a clear manner about the nature and method of the research, as described in the email invitation for this study.
I agree, fully and voluntarily, to participate in this research study. With this, I retain the right to withdraw my consent, without having to give a reason for doing so. I am aware that I may halt my participation in the experiment at any time.
If my research results are used in scientific publications or are made public in another way, this will be done such a way that my anonymity is completely safeguarded. My personal data will not be passed on to third parties without my express permission.
If I wish to receive more information about the research, either now or in future, I can
contact Anne Bindels at annebindels@hotmail.com. Should I have any complaints about this research, I can contact the designated member of the Ethics Committee representing the ASCoR, at the following address: ASCoR secretariat, Ethics Committee, University of Amsterdam, Postbus 15793, 1001 NG Amsterdam; 020 525 3680; ascorsecrfmg@uva.nl. √ I understand the text presented above, and I agree to participate in the research study.
Material
After agreeing on participating in the study, the participants were randomly assigned to one of the four conditions. They were exposed to a fictitious message, from the Dutch Railway Organization, in Dutch called Nederlandse Spoorwegen (NS). Four different versions of one crisis story were created as a Facebook message. NS is known for its presence on social media on Twitter and Facebook. Schultz et al. (2011) found that a message on Twitter would
evoke less negative crisis reactions than a blog message. Since Park and Cameron (2014) already used blog messages in their research, research with Facebook messages might bring new insights.
NS is very present in Dutch media because of two reasons: it is the main public transportation organization, and NS experiences a lot of technical problems that cause inconvenience for travellers. These conditions make NS susceptible for crises and therefore this organization makes a useful subject of study.
All four messages were about a technical problem that occurred at Utrecht Central Station that caused a delay of approximately one hour. Participants were randomly exposed to one of the four versions displayed in the online survey. Each version contained either a conversational voice or an organizational voice, and a defensive or accommodative response strategy. No explanation about the organization was provided, because it is commonly known in the Netherlands and detailed information was not necessary.
Independent variables
Tone of voice
One of the independent variables is tone of voice. Tone of voice is manipulated by using different types of voices in the crisis story. Based on the definition by Kelleher (2009), a conversational human voice evokes a feeling of personal contact. The messages with a human voice are written in first person narrative, with emotional en understanding elements. An example of a sentence in which the conversational human voice is clearly visible is the following: “Beste reiziger, helaas moeten we mededelen dat er een seinstoring rond Utrecht Centraal is opgetreden.” With “Beste reiziger”, the public is personally addressed, while the word “helaas” brings emotion in the message. The message is written in first person
narrative, which brings the organization closer to the reader and resembles an offline conversation.
The other two messages are written with a corporate voice and provided necessary information only. They are written in a formal, distant style. This means no use of a first person narrative and no emotional tone with the main purpose of offering clear information. In comparison to the previous shown sentence, there is less personality and human-like elements: “Er is een seinstoring rond Utrecht Centraal opgetreden.” A salutation is missing, and words that bring out emotions have been deleted.
Crisis response strategy
Following the theory of Coombs, the crisis response strategy continuum ranges from defensive to accommodative. The different levels represent the degree to which the organization takes responsibility for its actions and apologizes for them. In the
accommodative response strategy, the organization apologizes for its mistake and tries to turn matters around by showing what positive actions it undertakes. This strategy is shown by the organization apologizing for its actions and saying its employees will try to solve the problems. These elements are visible in sentences like “We (...) zullen er alles aan doen om de problemen zo snel mogelijk op te lossen. Onze welgemeende excuses voor het
ongemak.” As Coombs (2007) described, apologizing means taking full responsibility. NS also says they are working to solve the problems as fast as they can.
By taking the defensive response strategy, the organization does not take
responsibility for or even denies mistakes. The organization does not hold itself accountable for any problem. In the stories, this shows by the organization not apologizing and not taking the blame. Instead, those elements are completely left out. NS does not deny that there is a crisis, because posting the message is a sign of acknowledging it. But it does imply that the organization does not take any responsibility for the situation, it only provides information. A complete overview of the Facebook messages created can be found in Appendix 1.
Dependent variables
Social presence
To assess perceived social presence, participants of the study were asked how they felt about the message. Following Park & Cameron (2014), four items on a 7-point bipolar scale of Short, Williams and Christie (1976) were used to measure this variable. The
measurement consisted of the following items: item 1 ranged from personal to impersonal, item 2 ranged from warm to cold, item 3 ranged from unsociable to sociable and item 4 ranged from sensitive to insensitive. To ensure participants would randomly fill in the scales, item 3 was set up from negative to positive. After reversing the other three items, a reliability analysis was performed. The analysis showed that the four items together form one
construct and thus a reliable scale ( = .83)
Perceived interactivity
Perceived interactivity was measured by questions on how participants viewed the
interactivity of the message. Here, eight 7-point Likert scale items from McMillan and Hwang (2002) were used: the medium makes two-way communication possible, the medium is interactive, the medium works at the right speed, the medium enables personal
communication, the medium offers varied content, the medium gets and holds my attention, I can easily find my way within the medium, I can easily get answers to my question with the medium. Together these items form one scale, measuring one construct ( = .90).
Account acceptance
The level of acceptance of the organization’s crisis response strategy was also assessed. This was done by using a single-item measure from Jin (2010), following Park and Cameron (2014). The following question was asked: “How acceptable do you think the actions taken
by the company are?” with a scale ranging from 1 (totally acceptable) to 7 (not acceptable at all). Before performing analyses with this variable, the scale was reversed.
Organizational reputation
To measure how participants viewed the organization and therefore to measure
organizational reputation, the Harris-Fombrun Reputation Quotient (1976) was adapted. Nine 7-point Likert scale items were used: I have a good feeling about this organization, this company provides excellent value to the consumer, I trust this company, This company sells high quality products and services, this company is powerful, the products and services this company sells are important to society, this company never lives up to its promises, this company contributes very little to the economy and I am very familiar with the products and services of the company. Also here two items were in the opposite direction from the other seven items to ensure participants would pay attention filling out the survey. These items were reversed before analysis. The nine items constitute one reliable construct ( = .82).
A complete overview of the items about social presence, perceived interactivity, account acceptance and reputation can be found in Appendix 2.
Analysis
To test the hypotheses, several analyses were performed. For H1, H2, H3 and H6, independent sample T-tests were performed in which tone of voice was the grouping
variable and either social presence, perceived interactivity, account acceptance or reputation were the test variable. To test H3 and H4, an independent sample T-test was done with crisis response strategy as grouping variable and account acceptance as test variable. To test H5, a correlation analysis was performed.
Resultaten
The study contains data from 170 participants, of which 67,2% (N = 115) is female and 32,4% (N= 55) is male. The participants largely come from the network of the researcher, whether directly or indirectly. Only one person says he never travels with NS, which means 99,4% (N = 169) of participants do travel with NS, whether regularly or occasionally. The participants are randomly placed in one of the four conditions: they either read the message written with human voice and accommodative response strategy (N = 47), with human voice and defensive response strategy (N = 43), with corporate voice and accommodative strategy (N = 42) or with corporate voice and defensive response strategy (N = 38). Table 1 presents means and standard deviations for variables social presence, perceived interactivity,
account acceptance and reputation.
Table 1. Means and standard deviations of scores on social presence, perceived interactivity, account acceptance and reputation expressed with either conversational human voice or corporate voice.
Mean (Standard deviation)
Tone of voice Social
presence Perceived Interactivity Account acceptance Reputation Human 4.10 (.99) 4.18 (1.35) 4.72 ( .16) 4.84 ( .09) Corporate 3.10 (1.04) 4.15 (1.31) 4.14 ( .16) 4.82 ( .10)
Human voice and corporate voice
To examine the differences between the use of conversational human voice and corporate voice, T-tests are performed with the use of SPSS Statistics 21. In table 1, means and standard deviations of tone of voice in different variables are presented.
An independent sample T-test was conducted to test H1, which states that social presence is higher when a message concerning a crisis is expressed with a conversational
significant difference between conversational human voice (M = 4.10, SD = .99) and corporate voice (M = 3.10, SD = 1.04), t(168)= 6.47, p < .001, 95% CI [.70, 1.31]. The participants who read the crisis message written with a human voice rate social presence higher than the participants who read the message with a corporate voice. H1 was therefore supported.
It was hypothesized that perceived interactivity would be higher when a message is written with a human voice than with a corporate voice. However, no significant difference is found concerning perceived interactivity. There is no difference between human voice (M = 4.18, SD = 1.35) and corporate tone of voice (M = 4.15, SD = 1.31), t(168)= .13, p = .90, 95% CI [-.38, .43]. H2 has to be rejected.
The influence of tone of voice on account acceptance is also examined, concerning H3. The expectation was that a conversational human voice would have a more positive effect than the corporate voice on whether the respondent accepted the message of NS. The analysis shows a significant difference between a human voice (M = 4.72, SD = .16) and corporate voice (M = 4.14, SD = .16), t(168)= 2.60, p = .01, 95% CI [.15, 1.03]. People are more likely to accept the account of the organization when a message is expressed with a human voice. H3 is therefore accepted.
Following the main research question of this study, the effect of tone of voice on reputation was tested. The analysis shows no significant difference between conversational human voice (M = 4.84, SD = .09) and corporate tone of voice (M = 4.82, SD = .10), t(168)= .13, p = .90, 95% CI [-.24, .28]. This means that the human voice does not differ from corporate voice in terms of scores reputation, which results in the rejection of H6.
Table 2. Means and standard deviations of scores on account acceptance and reputation with defensive or accommodative crisis response strategy.
Mean (Standard deviation)
Crisis response
strategy Account acceptance Reputation
Defensive 3.50 ( .16) 4.85 ( .77)
Accommodative 3.61 ( .16) 4.82 ( .94)
Defensive response and accommodative response
Now it is clear that people are more likely to accept the organization’s account when they read a message written with a conversational human voice. Since crisis response strategy was also manipulated, it is interesting to see if there is a difference between the two types of crisis response strategy on account acceptance as shown in table 2. An independent sample T-test demonstrates no significant difference between the accommodative response (M = 3.61, SD = .16) and the defensive response (M = 3.50, SD = .16), t(168)= .49, p = .62, 95% CI [-.34, .56]. H4 thus has to be rejected.
Performing another independent sample T-test, the differences between the two types of crisis response regarding the scores on reputation were examined. It presents no significant differences between the defensive (M = 4.85, SD = .77) and the accommodative strategy (M = 4.82, SD = .94), t(168) = -.26, p = .80, 95% CI [-.29, .23]. These results mean that for reputation, it does not make a difference whether a defensive crisis response or an accommodative crisis response is used.
Since there is a significant difference between human voice and corporate voice as for social presence and account acceptance, it is be interesting to investigate the
relationship of those dependent variables with the variable reputation. This is done by looking and the Pearson’s Correlation. The results show that a significant, weak positive correlation exists between social presence and reputation (r = .30, p < .001). Apparently, the
more people view a crisis response as friendly and sociable, the more positive their opinion of the organization will be. The results also show that there is a significant, reasonably strong, positive correlation between account acceptance and reputation (r = .47, p < .001). The more acceptable the crisis message is, the higher the score on reputation will be, which means H5 is supported.
Discussion
This section is saved for conclusions drawn from the result section. A short summary will be followed by the results, linked to the literature discussed. After that, limitations will be
touched on and directions for future research will be given.
Conclusion
The research question of this study was what influence tone of voice has on perceived reputation. This study builds on the one of Park & Cameron (2014) that was used as an example, guide and inspiration. An attempt to answer the question was made by the creation of a crisis communication message, with tone of voice and crisis response strategy
manipulated. The Dutch Railway organization NS was taken as subject, because it is a well-known and crisis prone company in the Netherlands. The message was a fictitious
Facebook message. Using a 2 (tone of voice: human or corporate) x 2 (crisis response strategy: defensive or accommodative) experimental design, 170 participants were asked to fill out an online survey in which they got to see one of the four fictitious Facebook
messages from NS.
In order to answer the research question, several hypotheses were considered. Following Park and Cameron (2014), social presence, perceived interactivity and account acceptance were used as variables. The first hypothesis stated that social presence would be higher when a social media crisis message was written with a conversational human
voice than with a corporate voice. The results show that there is a significant difference between those two tones of voice, suggesting that when an organization uses a human voice in its messages, the public perceives them as more social present and therefore more ‘real’ (Park & Cameron, 2014; Short, Williams & Christie, 1976). This is in line with the findings of Park and Cameron. Since people prefer to human contact over all other types of communication (Liu et al., 2015), this is an important pursuit for organizations in their online communication. Also, it was found that social presence correlates with reputation. A high score on social presence could mean a high score on reputation, a useful finding for
organizations to keep in mind when constructing a crisis communication message. However, this correlation was weak and so a strong conclusion cannot be drawn.
Does this also count for perceived interactivity? It was hypothesized that perceived interactivity is higher when a human voice is used, rather than a corporate voice. A high score on perceived interactivity does not necessarily mean a high number of actual interactive features (Voorveld, Neijen & Smit, 2013). As a message written with a human voice is perceived as more real than written with a corporate voice, this effect may also be visible in the variable perceived interactivity. After all, when somebody evaluates the
medium as highly interactive, he might evaluate the communication as personal and similar to an offline conversation (Liu et al., 2015). However, the results do not support any
difference between the conversational human voice and corporate voice in perceived interactivity. This means that it doesn’t matter which tone of voice is used, perceived interactivity is equal with both tones of voice. Interactivity may not be in a voice, but it may be in content (Schultz et al., 2011). The implication of this finding is when organizations strive for high perceived interactivity, their focus should not be on tone of voice. This is not in line with the findings of Park and Cameron, as they found that the use of conversational human voice led to a higher score on perceived interactivity. An explanation for this can be found in the limitations section.
Another variable tested was account acceptance. It is reasonable to assume that when a message is perceived as being real and human, somebody will more easily accept the organization’s account than when the message is formal and distant. Indeed, the results show that when expressed with a human voice, the crisis message from NS is more
accepted than when expressed with a formal, corporate voice. An important finding, because one of the goals of post-crisis communication is getting stakeholders to accept what the organization has to say (Sohn & Lariscy, 2012) and this can only be done when
stakeholders perceive the apology as sincere (Van Noort et al., 2014). Apparently this goal is better reached with the use of a human voice than with a corporate voice. Making a post or Tweet in a personal style should be taken into account, especially when it concerns a crisis message.
Another expectation was that a message written with a human voice would result in a higher score on reputation than a message written with a corporate voice. Literature
suggests that the conversational human voice has a positive influence on attitude towards the organization (Kelleher & Miller, 2006; Kelleher, 2009; Park & Cameron, 2014; Van Noort et al., 2014). And, since it is the most valuable asset of an organization (Sohn & Lariscy, 2012), we need to know whatever we can do to create a strong and positive reputation. This is a very relevant issue for NS, because it regularly has to cope with reputational setbacks. Surprisingly, there was no difference in reputation between human voice and corporate voice. Apparently it does not matter, in terms of reputation, if a message is expressed in personal or formal style. This is in line with the findings of Park and Cameron (2014). What they did find, is that although tone of voice does not make a difference for reputation, it does for intention to buy or willingness to communicate. Maybe the effect of conversational human voice is just not visible in the items that question reputation, which is a wide concept, but it is visible in more specific concepts like brand attitude (Van Noort & Willemsen, 2012).
The findings on the differences between a conversational human voice and a corporate voice resulted in the model in figure 1. It shows that a message with the use of a human voice is rated significantly more social present than a message with a corporate voice. It also shows the positive correlation between social presence and reputation. Next to that, the figure shows that the scores on account acceptance were significantly higher when the participants read a message with a human voice than when they read one with a
corporate voice. Also, the correlation between account acceptance and reputation is presented.
Next to exploration of the influence of tone of voice, crisis response strategy was the other variable to explore. Following Park and Cameron (2014), crisis response strategy was manipulated into two extreme levels of the defensive – accommodative continuum. It was hypothesized that an accommodative strategy would lead to more acceptation than a defensive strategy. Surprisingly, this was not the case. Apparently consumers do not care about apologies and taking responsibility when they are informed about a crisis. Or, as Van Noort et al. (2014) highlighted, the apology was not perceived as genuine and did not fit the type of crisis (Coombs, 2007) Again, this is not in line with the results from the study from
r = .30 ** Account acceptance Social presence Tone of voice (CHV and CV) Organizational reputation t = 6.47 ** t = 2.60 * r = .47 **
Figure 1. Model of differences of tone of voice on social presence and account acceptance, and correlations between social presence and reputation, and account acceptance and reputation.
* Significant at the 0.01 level (2-tailed) ** Significant at the 0.001 level (2-tailed)
Park and Cameron. This could be an indication for too little distinction between the two types of responses in text. I will continue on this point under the limitations section.
The defensive and accommodative crisis response strategy also did not differ on the reputation scores. This is in line with the study of Schultz et al. (2011), where the medium seemed to be more important than the actual content of the message. It seems a little
rigorous to advice not to focus on crisis response strategy in crisis communication, but in this case one crisis response is not better than the other regarding reputation, similar as to what Park & Cameron (2014) found. The fictitious messages were created as Facebook posts, online messages, and thus should be placed in a social media context. It does seem
counterintuitive to not apology for an issue, so it has to be taken into account that the results of this study do not suggest to not publicly apologize in traditional media or offline situations.
Limitations and future research
This study has interesting implications, although some limitations have to be taken into account. One issue that should be addressed is the image of NS. Because of the newsworthiness and crisis susceptibility, NS was chosen as subject of study. In the
Netherlands, most people travel or have ever travelled with NS and it can be assumed that everybody knows of its existence. This is seen in the results that show that only 1 of 170 participants does not travel with NS. Next to that, the company is often in the news because of organizational or technical problem. It results in the fact that most people already have a rather complete image of NS and that raises the question how it affects the research. It would be interesting to compare the results of this study to the ones of a study in which a fictitious company is the subject.
An implication of the participants’ existing knowledge about NS is that it might block the effects variables may have on reputation. When there is no flexibility in reputation because of prior knowledge, chances are that scores on reputation do not come from independent variables in the different conditions but from the opinion the participant already
had. A suggestion to partly resolve this issue would be to ask questions about reputation before and after exposing the participants to one of the four crisis responses.
Also, the sample is an element that could have a significant influence on the results. The participants were found in the network of the researcher, which means most of them were students. People under the age of 40 are significantly more experienced with new technologies, among which social media. The manipulated Facebook message is much more familiar to them than to the older generation and that might have resulted in giving less attention to it or giving different scores on interactivity because they are familiar with the possibilities of Facebook. In future research, it would be useful to have more demographic information of participants to rule out possible influences or differences.
An explication for the findings on perceived interactivity, account acceptance and reputation may lie in the operationalization of the variables tone of voice and crisis response strategy. Some of them were not in line with previous studies and literature, and surprising because they are counterintuitive. For example, it seems strange that a difference in crisis response strategy does not have any effect on account acceptance. It was assumed that when an organization makes an apology and takes responsibility for the mistake,
participants would be more likely to accept its account. This was not the case, and the reason for that might be in the content of the Facebook message. Maybe the difference between the defensive and accommodative strategy was not clear enough for people to distinguish them from one another, or the cause of the findings lies in the unclear fit between crisis situation and response (Coombs, 2007; Coombs & Holladay, 2009; Van Noort et al., 2014). The same counts for the conversational human voice and corporate voice. In future research, this could be fixed by adding a manipulation check in which participants are asked about how defensive or personal the message was.
The implications and limitations offer numerous opportunities for future research. The focus can be on another operationalization, a more specific study on a particular variable, a
path analysis, causal relations and so on. As the purpose of this study was to explore the influence of voice in online crisis communication messages, a new building block was placed in the field of corporate communication.
References
Batra, R., Ahuvia, A. & Bagozzi R.P. (2012). Brand love. Journal of Marketing, 76 (2), 1-16 Coombs, W.T. (2007). Protecting Organization Reputations During a Crisis: The
Development and Application of Situational Crisis Communication Theory. Corporate Reputation Review 10 (3), 163-176
Coombs, W. T., & Holladay, S. J. (2002). Helping Crisis Managers Protect Reputational Assets: Initial Tests of the Situational Crisis Communication Theory.
Management
Communication Quarterly, 16 (2), 165-86Coombs, W. T., & Holladay, S. J. (2009). Further explorations of post-crisis communication: Effects of media and response strategies on perceptions and intentions. Public Relations Review, 35(1), 1–6
Coombs, W.T. (2015). The value of communication during a crisis: Insights from strategic communication research. Business Horizons, 58 (2), 141-148
Floreddu, P.B., Cabiddu, F. & Evaristo, R. (2014) Inside your social media ring: How to optimize online corporate reputation. Business Horizons, 57 (6) 737-745
Fombrun, C.J., Gardberg, N.A & Sever, J.M. (2000). The reputation quotient: a
multi-stakeholder measure of corporate reputation. Journal of Brand Management, 7 (4), 241-256 Gotsi, M. & Wilson, A.M. (2001) Corporate reputation: seeking a definition. Corporate Communications: An International Journal, 6 (1), 24-30
Grunig, J. E., & Grunig, L. A. (1992). Models of public relations and communication. In J.E. Grunig (Ed.), Excellence in public relations and communication management (285-325). New York: Routledge
Grunig, J.E. (2009). Paradigm of global public relations in an age of digitalism. PRism, blog.iain-tulungagung.ac.id, consulted on 10-06-2015
Hennig-Thurau, T., Malthouse, E. C., Friege, C., Gensler, S., Lobschat, L., Rangaswamy, A., & Skiera, B. (2010). The impact of new media on customer relationships. Journal of Service Research, 13 (3), 311-330.
J. Huibers & J. Verhoeven (2014). Webcare als online reputatiemanagement: het gebruik van webcarestrategieën en conversational human voice in Nederland, en de effecten hiervan op de corporate reputatie. Tijdschrift voor Communicatiewetenschap, 42 (2), 165-189
Kaplan, A.M. & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business Horizons, 53 (1), 59-68.
Kelleher, T. (2009) Conversational voice, communicated commitment, and public relations outcomes in interactive online communication. Journal of Communication, 59 (1), 172-188 Kelleher, T. & Miller, B. (2006) Organizational blogs and the human voice: relational strategies and relational outcomes. Journal of Computer‐Mediated Communication, 11 (2), 395-414
Lee, M. (2011). Configuration of external influences. The combined effects of institutions and stakeholders on corporate social responsibility Strategies. Journal of Business Ethics, 102 (2), 281-298
Liu, B.F., Austin, L. & Jin, Y. (2011). How publics respond to crisis communication
strategies: The interplay of information form and source. Public Relations Review, 37 (4), 345-353
Liu, B.F., Fraustino, J.D. & Jin, Y. (2015). How disaster information form, source, type, and prior disaster exposure affect public outcomes. Jumping on the social media bandwagon? Journal of Applied Communication Research, 43 (1), 44-65
Advertising, 38 (2), 53-68
Macnamara, J. (2010). Emergent media and public communication. Understanding the changing landscape. Public communication review 1 (2), 3-17
Macnamara, J. & Zerfass, A. (2012). Social media communication in organizations. The challenges of balancing openness, strategy and management. International Journal of Strategic Communication, 6 (4), 287-308
Martin, M. G., & Burke, R. J. (2012). Corporate reputation: Managing opportunities and threats. Gower Publishing, Ltd
McMillan, S.J. & Hwang, J. (2002). Measures of perceived interactivity. An exploration of the role of direction of communication, user control and time in shaping perceptions of
interactivity. Journal of Advertising, 31 (3), 29-42
Muntinga, D.G., Moorman, M. & Smit, E.G. (2011) Introducing COBRAs. Exploring
motivations for brand-related social media use. International Journal of Advertising 30 (1), 13-46
Park, H. & Cameron, G.T. (2014). Keeping it real: exploring the role of conversational human voice and source credibility in crisis communication via blogs. Journalism & Mass
Communication Quarterly, 91 (3), 487-507
Schultz, F. & Wehmeier S. (2010). Institutionalization of corporate social responsibility within corporate communications; Combining institutional, sensemaking and communication perspectives. Corporate Communications: An International Journal, 15 (1), 9-29
Schultz, F., Utz, S. & Göritz, A. (2011). Is the medium the message? Perceptions of and reactions to crisis communication via twitter, blogs and traditional media. Public Relations Review 37 (1), 20–27
Searls, D., & Weinberger, D. (2000). Markets are conversations. In R. Levine, C. Locke, D. Searls, & D. Weinberger (Eds.), The cluetrain manifesto: The end of business as usual (pp. 75–114). New York: Perseus.
Smith, B.G (2010)Socially distributing public relations: Twitter, Haiti, and interactivity in social media. Public Relations Review, 36 (4), 329-335
Sohn, Y.J. & Lariscy, R.W. (2012). A buffer or boomerang? The role of corporate reputation in bad times. Communication Research 42 (2), 237-259
Stephens, K. & Malone, P. (2010). New media for crisis communication. Opportunities for technical translation, dialogue and stakeholder response. In: Handbook of crisis
communication. Edited by Coombs, W.T. & Holladay, S.J. Malden: Wiley-Blackwell, 381-395 Verhoeven, J.W.M, Van Hoof, J.J., Ter Keurs, H. & Van Vuuren, M. (2012). Effects of
apologies and crisis responsibility on corporate and spokesperson reputation. Public Relations Review 38 (3), 501–504
Van Noort, G., Voorveld, H.A.M. & Van Reijmersdal, E.A. (2012). Interactivity in brand web sites. Cognitive, affective, and behavioral responses explained by consumers' online flow experience. Journal of Interactive Marketing, 26 (4), 223-234
Van Noort, G., & Willemsen, L. M. (2012). Online damage control: The effects of proactive versus reactive webcare interventions in consumer-generated and brand-generated Van Noort, G. Willemsen, Kerkhof & Verhoeven (2014). Webcare as an integrative tool for customer care, reputation management, and online marketing: a literature review. In Kitchen, P.J. & Uzungolu, E. (Eds.) Integrated communications in the postmodern era (77-99).London: Macmillan Publishers Ltd.
Voorveld, H.M., Neijens, P. & Smit, E. (2011). The relation between actual and perceived interactivity. Journal of Advertising 40 (2), 77-92
Weick, K. E. (1988). Enacted sensemaking in crisis situations. Journal of management studies (25) 4, 305-317
Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychology Review 92 (4), 548-573
and disposition to trust on initial online trust. Journal of Computer‐Mediated Communication, 16 (1), 1-26
Xifra, J. & Huertas, A. (2008). Blogging PR. An exploratory analysis of public relations weblogs. Public Relations Review, 34 (3), 269-275
Appendix 1
Four constructed Facebook messages of NS in original language (Dutch), all fictitious crisis responses about a technical problem, with either a conversational human voice or corporate voice, and in a accommodative strategy or defensive strategy.
1: CHV/A (Conversational human voice – accommodative strategy)
“Beste reiziger, helaas moeten we mededelen dat er een seinstoring rond Utrecht Centraal is opgetreden. Hierdoor rijden er geen treinen van en naar Utrecht. Dit zal extra reistijd opleveren van gemiddeld 2 uur. We begrijpen dat dit heel vervelend is en zullen er alles aan doen om de problemen zo snel mogelijk op te lossen. Onze welgemeende excuses voor het ongemak.”
2: CHV/D (Conversational human voice – defensive strategy)
“Beste reiziger, helaas moeten we mededelen dat er een seinstoring rond Utrecht Centraal is opgetreden. Hierdoor rijden er geen treinen van en naar Utrecht. Dit zal extra reistijd opleveren van gemiddeld 2 uur. We begrijpen dat dit heel vervelend is. Er wordt hard gewerkt om de problemen zo snel mogelijk op te lossen.”
3: CV/A (Corporate voice – accommodative strategy)
“Er is een seinstoring rond Utrecht Centraal opgetreden. Hierdoor rijden er geen treinen van en naar Utrecht. Dit zal extra reistijd opleveren van gemiddeld 2 uur. Onze excuses voor het ongemak.”
4: CV/D (Corporate voice – defensive strategy)
Appendix 2
Items of variables social presence, perceived interactivity, account acceptance and reputation in original language Dutch.
1. Social presence:
1. Persoonlijk – onpersoonlijk 2. Vriendelijk – onvriendelijk 3. Koel – warm
4. Fijngevoelig – niet fijngevoelig
2. Perceived interactiviy:
1. 1Het medium maakt tweezijdige communicatie mogelijk 2. Het medium is interactief
3. Het medium maakt persoonlijke communicatie mogelijk 4. Het medium werkt op goede snelheid
5. Het medium biedt gevarieerde content
6. Mijn aandacht wordt getrokken en vastgehouden door het medium 7. Ik kan gemakkelijk mijn weg vinden binnen het medium
8. Ik kan snel antwoord krijgen op vragen via dit medium
3. Account acceptance:
Hoe acceptabel vind je de acties die NS onderneemt?
4. Reputation:
1. Deze organisatie is van waarde voor de consument. 2. Ik heb een goed gevoel over deze organisatie.
3. Ik heb vertrouwen in deze organisatie.
4. De service en producten van deze organisatie zijn van hoge kwaliteit. 5. Dit is een machtige organisatie.
6. De producten en service van deze organisatie zijn van belang voor de samenleving. 7. Deze organisatie houdt zich nooit aan zijn beloften.
8. Deze organisatie draagt weinig bij aan de samenleving.
