1
Master’s Thesis:
Enhancing Deliberation through Online Crisis
Communication
Graduate School of Communication
Master’s Programme Communication Science
Line Frier, Student number: 11104600
Supervisor:
Dr. James Slevin
2 Abstract
The new communicative features of Facebook offer airlines various opportunities to reach news medias and other audiences quickly and to provide constantly updated information regarding accidents with their planes. However, once the story breaks on Facebook, the opportunity to provide factual information as well as the capability to influence the developing nature of the story is reduced to minutes. This can directly influence peoples’ perception of the accident and their ability to make proper informed decisions. Hence, airlines may benefit from fostering more deliberative debates on their Facebook. Deliberation involves many kind of interactions among individuals to reinforce relationships and decision-making. In online crisis communication, deliberation thus allows people to consent to specific decisions that affect them, and thereby reach truly just decisions rather than decisions based on rumors or misinformation provided by other Facebook users. For this purpose, 25 Facebook posts and 875 comments published in the timeframe 2014-2016 were analyzed and assessed in terms of indicators established to evaluate the degree of deliberation derived from the work of several deliberation scholars. Drawing on Coombs’ (1999) online crisis communication theory, which emphasizes the importance of a high amount of information, this study predicted that a timely and proactive online crisis approach giving instructing information would lead to deliberative debates on Facebook. It was further assumed that the direct involvement from the organization in the comment section would encourage people to deliberate more. The findings indicate that the specific elements from Coombs’ (1999) theory has a limited effect on deliberation. Only Facebook posts communicating about organizational actions that will directly influence people can increase the degree of deliberation, and thereby the likelihood of individuals making better, informed judgements during airline crises.
3 Introduction
When an accident happens with a plane from an international airline it will be within minutes on Facebook (FB) including photos and videos often taken by survivors or eyewitnesses (Stritesky, Stránska & Drábik, 2015). Once the story breaks on FB, it is very difficult for airlines to control the developing nature of the story and the ability to provide factual information is reduced to minutes (Chadwick, 2008). This is particularly due to the decentralization of communication on FB, which enables a one-to-many interaction (Graham, Avery & Park, 2015; Mayfield, 2006), and thus increases the probability of misinformation and rumors (Stritesky, 2015). This limits individuals’ possibility to make informed judgements relating to that specific situation (Graham et al., 2015).
In this context, internet advocates claim that FB posts posit great opportunities to improve deliberative practices based on the information flow and diversity of opinions (Halpern & Gibbs, 2013; Noveck, 2009). This is especially valuable for the airline industry, as they are continuously accused of “unpredictable events”, which force them to be in a continual pre-crisis mode and thus the capability to provide factual information is crucial (Ray, 1999). The transparent, explicit reasoning typical for deliberative processes would allow people to consent to certain decisions that affect them, and thereby reach truly just decisions (Fleck, 1991, Smith & Wales, 2000). Hence, this study addresses this area by exploring the extent to which FB posts can foster deliberation during airline crises.
Accordingly, Habermas (1989) defines deliberation as a notion of interchange of reciprocal arguments among a group of individuals that is triggered by a public or mutual problem, in which the main focus of discussion is to find a solution acceptable for all who have a stake in the issue. Several scholars further argue that deliberative debates are beneficial to organizational crises, as they can function as a steering force in balancing information and possible decisions as well as consider alternative courses of actions, which are just for the
4 comment interest in that specific context (Braunack et al., 2010; Wendling, Radisch & Jacobzone, 2013). However, the decentralization of communication on FB often leads to comments containing misleading and irrelevant content making it difficult for the discourse to live up to the deliberative ideals (Stritesky et al., 2015; Stroud, Scacco, Muddiman, & Curry, 2015). This has crucial impact on how people make legitimate, informed decisions (Stroud et al., 2015).
At this stage, research has yet to conclude, how online crisis communication can affect the degree of deliberation on FB (Baum, Jacobsen & Goold, 2009; Braunack-Mayer et al., 2010; Halpern & Gibbs, 2013). This study will thus examine, how airlines through an open and timely online crisis approach containing instructing information can foster deliberative debates on FB. This will be done by examining organizational FB posts and comments. Therefore, this study aims to contribute to the scientific discussion about deliberation as well as exploring new possibilities to foster deliberation on FB.
Accordingly, the following research question will be examined:
To what extent can deliberation in the comment section of organizational Facebook posts be increased through online crisis communication and direct involvement from the organization?
Theoretical Framework
The decentralization of communication on FB permits everybody to report about airline accidents, making it difficult for airlines to provide individuals with correct and factual information (Stritesky et al., 2015). This can lead to misinformation circling, which people base their judgements on (Graham et al., 2015). Not only does this hinder peoples’ ability to make legitimate decisions, it can also affect their perception of the accident and the airline (Stritesky et al., 2015; Wendling et al., 2013). In this context, more deliberative debates on FB
5 would provide individuals with factual and detailed information, which they can draw upon in their decision making (Braunack-Mayer et al., 2010).
Defining Deliberation
Habermas (1989) argues that the main idea behind deliberation is that the behavior and interaction of participants are in terms with the principles of political equality and egalitarian reciprocity. It further increases tolerance and allows individuals to be involved as well as engaged in public life (Walsh, 2004). In this context, Thompson (1995) further claims that individuals are free of external and internal coercion when considering and weighing the arguments proposed by others in the deliberative communication process (p. 255). Hence, a common notion to deliberation is the right to equal respect (Burkhalter, Gastil, & Kelshaw, 2002).
Research also found positive consequences related to deliberation such as increased familiarity and understanding of opposing views and thus the creation of stronger links between knowledge and attitudes (Cappella, Price, & Nir, 2002; Stroud et al., 2015). As a result, deliberation produces reflected public opinions, and therefore constitutes a so-called public sphere (Zimmermann, 2015), where people can freely discuss and identify societal problems (Habermas, 1989). This is of interest for organizations, since it flourishes more personal relationships with their publics (Thompson, 2005). Particularly in online crisis communication, this can help individuals make better informed judgements (Wendling et al., 2013).
Accordingly, for the purposes of this thesis, deliberation will be conceptualized within the notion of what Thompson (1995) call **the deliberative conception of democracy**, in which decisions are based on formed judgements (p. 255). It further empathizes three aspects of quality deliberation proposed by Camaj and Santana (2015): (1) a common agenda between participants, (2) individuals’ openness towards different opinions expressed by others, and (3)
6 the deliberation is determined by reason rather than coercion. Following Thompson (1995), the legitimation thus stems from the outcome of a generalized deliberation rather than predetermined will or set of preferences (p. 257).
The Role of Deliberation in Online Crisis Communication
The principles of deliberation exemplify an attempt to counteract the deficits of a representative democracy, which is particularly crucial in terms of legitimacy (Bohman, 1996; Escobar, 2009). Deliberation can gauge the views of the community, and thus enable organizations and publics to generate well-informed debates in order to avoid future issues (Braunack-Mayer et al., 2010). However, according to Tannen (1999), today’s social reality is rather conflictive, as **the best way to discuss an idea is to set up a debate** (p. 5). The conflicts are often dealt with rather counterproductive and self-perpetuating (Tannen, 1999), since it does not allow a thorough clarification of the issues under discussion (Escobar, 2009). This influences the discourse to be more misleading and irrelevant rather than contributing to more meaningful content (Coe, Kanski, & Rains, 2014). This is a common problem on FB (Stritesky et al., 2015; Stroud et al., 2015). According to Isaac (1999), these types of discourse also somewhat reinforce dramatic antagonism, which thus result in embedded monologues rather than dialogic conversations (Escobar, 2009; Isaac, 1999) leaving people unable to make informed judgements (Stroud et al., 2015). Hence, deliberation is thus a concept that has been developed from a crisis in dialogue (Escobar, 2009; Hyde & Bineham, 2000).
Davis (1999) argues that the misleading and irrelevant content is often based on individuals participating in online discussions that agree with their own attitudes, and therefore reinforce already pre-existing views rather than acknowledging different perspectives of a certain event. This hinders the possibility of allowing people to be well-informed in the specific situation (Burkhalter et al., 2002).
7 Consequently, Robertson, Vatrapu and Media (2010) highlight the potential of fostering deliberation on FB because it **most closely enable public sphere discourse for those who choose to enter the online “salons** (p. 29). In this regard, researchers highlight the benefits of deliberation to address difficult, value-laden organizational issues (Dahlberg, 2004; Baum et al., 2009), since it permits the transparency of diverse opinions (Thompson, 1995). This allows individuals to make better, informed judgements during organizational crises (Thompson, 1995). Therefore, the present study examines specific online crisis communication aspects and their potential to foster more deliberative debates on FB.
Online Crisis Communication and Deliberation
Several crisis communication scholars suggest that effective online crisis communication requires customized and personalized content-based messages needed and desired by customers (Greer & Moreland, 2003; Sturges, 1994). Hence, the success of a crisis communication is profoundly affected by what an organization says and does (Barker, 2011; Benoit, 1997; Coombs, 2004). This is supported by Lynette, Sparks and Glendon (2010) who found that a “no comment” online crisis response is more likely to have a negative influence on people.
Several approaches have been offered to categorize crisis strategies at the contextual level in social media. Most crisis communication theories build on Grunig and Grunig’s (2000) excellence model, which stresses the two-way symmetrical model. However, the new communication practices on FB has led to a skepticism towards the “idealism” of two-way communication (Barker, 2011). Fjeld and Molesworth (2006) argues that the skepticism is mainly due to the two-way complexity, and that the communication in a virtual environment is mostly grounded in one-to-many and many-to-many. In this context, the communicative features of the virtual environment allow organizations to reach their publics in a direct manner
8 without messages being filtered (Barker, 2011; Fjeld & Molesworth, 2006). This further means that organizations are able to provide their publics with factual information, and thus influence their perception of the situation (Gonzáles-Herrero & Smith, 2008).
Subsequently, Coombs’ (1999) online crisis communication theory is one way of facilitating FB posts during a crisis, and hence forms the basis for this study. The theory focuses on the potentials of communicative actions in a virtual environment. The notion is to be open and in candor with the media and publics as well as disseminate information in a timely and proactive manner (Coombs, 1999; Greer & Moreland, 2003). The time and openness of the posts are essential components, since they facilitate the characteristics of a real-time conversation between the organization and its publics (Greer & Moreland, 2003). Coombs’ (1999) theory further stresses the importance of providing individuals with instructing information, such as objective information about the accident, possible reservations to consider, and organizational actions done to solve the issue (Coombs, 1999). Hence, a central idea is to share as much information as possible during a crisis in order to keep the publics well-informed and involved, and thereby avoid misinformation and rumors (Barker, 2011; Coombs, 1999; Wendling et al., 2013).
In addition, Nonaka and Takeuchi (1995) argue that a high amount of information will lead to more transfer, sharing and processing of knowledge among participants allowing them to gain a deeper understanding of the situation (Braunuack-Mayer et al., 2010). Particularly information is a central tenet to deliberation, since it forms the basis for equal opportunities for participation (Braunuack-Mayer et al., 2010) and reduce existing social power structures, which allows participants to collaborate as equals (Elster, 1998; Manin et al., 1987). As such, individuals are able to deliberate on an equal basis (Barker, 2011). The sharing of information and knowledge furthermore gathers people with similar or dissimilar perspectives on the issue,
9 which increases the likelihood of a more deliberative debates in the comment section (Lynette et al., 2010), since deliberation thrives on the clash of competing views (Thompson, 1995).
It is therefore possible to assume that airlines addressing a crisis in a timely and open manner while giving instructing information can positively affect the degree of deliberation on FB. This is based on individuals’ permission to participate, share and process knowledge on FB, and thus create a dynamic context, in which information flows both ways. Therefore, the first hypotheses are formulated as followed:
Hypothesis 1: Timely and open information about the crisis has a positive influence on the
degree of deliberation.
Hypothesis 1a: Objective information about the crisis has a positive influence on the degree
of deliberation
Hypothesis 1b: Information about reservations has a positive influence on the degree of
deliberation
Hypothesis 1c: Organizational actions to correct the problem has a positive influence on the
degree of deliberation.
Moderation Effect of Direct Involvement from the Airline
To understand the influence of online crisis communication on deliberation, it is important to consider possible moderators. Research found that the direct involvement from an organization in the FB comment section can have a direct influence on the deliberative norms in the interaction (Stroud et al., 2015). Aarts and Dijksterhuis (2003) define norms as **situational or mental representations of appropriate behavior that guide behavior in certain situations and environment** (p. 18). Thus, the second part of the study aims to examine the
10 direct involvement from the airline as a possible moderator in the relationship between online crisis communication on deliberation.
Following a research conducted by Wise, Hamman, and Thorson (2006), it was found that site visitors were more likely to participate in online discussion, when components of moderation were present. Likewise, Ruiz et al. (2011) also found that organizational involvement via moderation can affect the comments to contain more relevant and thorough content. These findings suggest that direct organizational involvement can positively influence the discourse to contain more meaningful and thoughtful content. According to Baum et al. (2009), the aspects of moderation portrays the organization as somewhat open towards diverse points of views, value the opinions of their publics, and thereby designates the right to equal respect. These are central characteristics when fostering deliberation.
Accordingly, when an organization engages directly with the commenters, it cues essential group characteristics (Hogg & Reid, 2006; Stroud et al., 2015). This is particularly important in online crisis communication, as people often seek more human connections when faced with difficult decision-making (Wendling et al., 2013). According to Aarts and Dijksterhuis (2003), organizations embodying important group characteristics tend to be more trusted, and are therefore more likely to influence the norms in the interaction.
Thus, looking at the above listed findings of research, it could be assumed that the airline engaging directly with participants in the comment section can moderate the relationship between online crisis communication response and deliberation. To test whether this holds true or not, the following hypothesis was formulated:
Hypothesis 2: The direct involvement from an airline in the comment section positively
11 The hypotheses put together result in a conceptual model as displayed in Figure 1.
Figure 1: Hypotheses and variables in the conceptual model
Methodology
To test the proposed hypotheses, a quantitative content analysis was done to investigate the potential of fostering deliberation on FB through Coombs’ (1999) online crisis communication theory. The present study used this particular research methodology, since it provided a set of useful tools, which made it possible to compare various user-generated comments on FB as well as the option to examine whether these contained discourse in terms with this study’s conceptualization of deliberation (Fields, Swan, & Kloos, 2010; Neuendorf, 2011). Hence, the quantitative content analysis gave this study the opportunity to systematically and objectively identify essential characteristics in the comments to answer the research question (Holsti, 1969).
Research Design
For this purpose, the present study quantitatively evaluated FB posts and comments published by five international airlines that had experienced an organizational crisis in the
12 selected timeframe from 2014-2016. The selective approach of FB posts published by the airlines provided six information-rich cases, in which specific characteristics were uncovered. All posts published during the time of the crises were systematically collected. Additionally, comments linked to the posts were also examined to determine the degree of deliberation fostered on FB. Since the range of comments varied between four and 75, a stratified sampling analysis was used. The objective with a stratified sampling analysis was to select a sample from the FB comments, which would function as a “well-defined” estimate of the overall amount of comments (Tipton, 2013). A total amount of 35 comments were systematically chosen in order to avoid overestimations, and thereby maintain a uniform number of comments for each post (Halpern & Gibbs, 2013; Holsti, 1969). The stratified sampling analysis and the systematic selection of posts and comments were made to ensure that the sampling would contain a comparable measurement of the collected data.
Sample and Data Collection
The FB posts and comments were retrieved from Flydubai, Asiana Airlines, Dynamic Airways, Air Canada and Malaysia Airlines’ official FB pages. The airline industry was chosen for this study based on its continual pre-crisis mode forcing the airlines to provide factual information in order to avoid misinformation about accidents (Ray, 1999). Subsequently, the communicative features of FB also contribute to the visibility of airline accidents, which creates an even higher level of complexity for airlines’ online crisis communication (Greer & Moreland, 2003; Ray, 1999). It is thus crucial that airlines provide promptly and accurately information in order to keep the publics well-informed about a situation (Ray, 1999; Romanucci & Blandin, 2013).
Accordingly, Facebook was chosen based on individuals’ increased use during organizational crises (Wendling et al., 2013). The interchangeable nature of FB moreover
13 permits individuals to view the organizational posts without further admission from the organization, thereby creating an interconnected society between the organization and its publics (De Vries & Carlson, 2014; Kim & Liu, 2012).
During the sampling procedure, posts from the five airlines’ official FB page published in the selected timeframe were collected. Data collection of FB posts would thus start from the date the accident happened, and the data collection would end when the airline no longer posted about it. Thus, a total amount of 25 FB posts from the five airlines were collected along with 35 comments linked to each post. A total amount of 875 comments were coded. Hence, the overall sampling consisted of 900 items.
Variables and Measurements
A codebook was established to capture the variables under investigation (See Appendix 1). A total amount of 22 variables were conceptualized with definitions and explanations. The present study was divided into two dimensions: post attributes and comment attributes. The first dimension dealt with the content in the FB posts, whereas the latter examined the comments, and hence the degree of deliberation. All the variables were coded in terms of categorical variables, in which 1 represented ‘aspects included’, 2 meant ‘no aspect included’, and 99 indicated that the variable or item was not applicable to the post or comment. These types of variables were chosen, since this study then would obtain comparisons that are most logical in testing the mentioned hypotheses.
Post Attributes
The 25 posts were examined in terms of the main thrusts of the theoretical discussion of Coombs’ (1999) online crisis communication theory. An essential feature to the theory is to ensure direct real-time interactions by proactively addressing the situation. This was conceptualized as Time & Openness in the codebook (Coombs, 1999; Elliott, 2009). The
14
Objective Information referred to the amount/nationality of victims, the date of the accident,
and thereby the transfer and sharing of knowledge from the organization to the publics. In terms of Reservations to be taken in accordance to the situation, airlines would explain, how individuals should act or not act in the specific context. Lastly, organizational Actions indicated the procedures the airline was communicating about the actions taken to ease the crisis.
Comment Attributes
Next, the 875 comments were coded. Online deliberation research offers a choice of effective coding schemes (Camaj & Santana, 2015; Carter, 1998; Halpern & Gibbs, 2013; Janssen & Kies, 2005; Santana, 2014; Schmidt, 1998; Schneider, 1997; Willhelm, 1999), and thus this study operationalized basic deliberative indicators from previous studies and in terms with this study’s conceptualization of deliberation. Hence, each comment was coded in terms of five deliberative categories: Rationality, Reciprocity, Politeness, Civility and Opinion
Diversity.
Adopting Wilhelm’s (1999) instrument, three items were used to examine the discourse
Rationality, which would determine if the legitimation stemmed from deliberation or set of
preferences. Hence, the comments were coded in terms of (1) explicit assertion with internal validation, (2) external arguments with a rationale, or (3) an argument with validation in the form of the source of information. Similar to previous studies (Janssen & Kies, 2005; Willhelm, 1999), the present study measured Reciprocity by exploring the interaction among participants. The comments were thus coded as, (1) the commenter reply to previous comments, (2) the commenter added additional information to the discussion, or (3) an off-topic comment referring to comments did not relate to the initial post nor previous comments.
An essential feature of deliberation is the right to equal respect (Carter, 1998; Schmidt, 1998; Thompson, 1995). Hence, the variables Politeness and Civility were both included to
15 explore whether harsh language, verbal attacks and name-calling were present, as it can negatively affect agreements among participants (Halpern & Gibbs, 2013; Santana, 2014). To measure Opinion Diversity, the comments were measured for whether they indicated support or opposition towards the airline (Camaj & Santana, 2015). Subsequently, to investigate the moderation effect of the direct involvement from the organization, each comment was identified as either a Facebook user or the organization.
Plan of Analysis
Before analyzing the proposed hypotheses, it was important to have the statistical foundation in order. Hence, an overview of the pre-tests.
Inter-coder Reliability Test
To establish inter-coder reliability, the percent agreement of Krippendorff’s alpha was used for three variables coded: the degree of Deliberation, the mentioning of Crisis and Coombs’ (1999) Online Crisis Communication theory1. In order for the coding of a variable to
be perceived reliable, it was required that the Krippendorff’s alpha was .70 or higher (Lombard, Snyder-Duch & Bracken, 2002). The Kalpha values for all three variables are above .70, meaning that the inter-coder reliability was considered as valid (See Appendix 2, Table 7).
Measures and Normality Analysis
For the analysis of the constructs Rationality and Reciprocity in the codebook scales were constructed with three items. To see if the combination of the items measured the latent features of Rationality and Reciprocity, the reliability of the scale was tested. The Cronbach’s
1 This consists of the independent variables: Time & Openness, objective Information,
16 Alpha for the scale for Rationality was 1.000 and for Reciprocity it was also 1.000, indicating a strong reliability of the items. Thus, the scales for Rationality and Reciprocity were constructed as originally intended with three items.
Normality was also tested, since it would give certain assumptions about the data and which statistical tests were appropriate to use (Moore & MacCabe, 2007). The normality analysis showed that the data for this study was not normally distributed, since there was no evidence of a normal distribution shape in the histogram. A reason for this can be due to the variables, as they only had three answer possibilities, and therefore not a lot of diverse values. Due to the lack of normality, the Spearman rank correlation has been used to test the correlation in between the variables instead of the Pearson correlation.
Control Variables
Besides the independent variables, other variables could have an influence on the dependent variable: Deliberation as well. Control variables were thus considered to make sure that the results of the following analysis were without interference of confounding factors. Thus, the present study controlled for the number of Words, the type of Crisis and the
Frequency of posts.
Words was measured in terms of the number of words used in the comments. According
to Halpern and Gibbs (2013), longer comments are assumed to contain more complex sentences and thus more meaningful and thoughtful discourse. This could positively influence the degree of Deliberation. The type of Crisis was coded in terms of the amount of times it was mentioned that the airline perceived it as controllable. According to Lynette et al. (2010), controllable crises usually evoke emotions such as anger, frustrations as well as the need to complain among individuals. This could negatively affect Deliberation. The Frequency of posts was measured the amount of published posts during the crisis, and would thus facilitate real-time interactions
17 as predicted by Coombs (1999). A high amount of posts should thus positively affect
Deliberation. A possible relation between the control variables, the independent variables, and
dependent variable was presumed.
Results
The analysis of the data will be reported in three sections: regression analysis of the control variables and hypotheses 1, 1a, 1b and 1c, followed by a separate moderation analysis to test the moderation effect of Direct Involvement (Hypothesis 2).
Spearman Rank Correlation and Mean Score
The Spearman rank correlation show a high amount of significant correlations (See Appendix 3). However, the independent variables, the moderating variable as well as the control variables are all negatively correlated with the dependent variable: Deliberation. This is of interest, since the hypotheses of this research have assumed a positive correlation. In addition, the mean score was also calculated to explore the degree of deliberation present in the 875 comments. The mean score was 15,76, indicating a low presence of deliberation (See Appendix 2, Table 8). This is essential to consider in terms of the results from the regression analyses.
Effects of Control Variables on Deliberation
The control variables; amount of Words, mentioning of Crisis and Frequency of posts are used to explore how the various variables influences Deliberation. A regression analysis was done with the dependent variable and control variables.
18 Table 1
Regression of Control Variables on Deliberation
The control variable Words is not significant (p<.653). However, the control variable
Crisis did have a significant negative (Beta-Value = -.782; p<.000) effect on Deliberation,
indicating that the more the occurrence of the crisis is mentioned, the lower the degree of deliberation will be in the comments. The same goes for the control variable Frequency, which had a significant negative (Beta-Value = -.218; p<.000) impact on Deliberation, meaning that the regularity of posts published by an airline would not increase deliberation in the comment section.
Effects of Coombs’ (1999) Online Crisis Communication Theory on Deliberation
Hypothesis 1 of this study predicts that a timely and open approach in the communication of an organizational crisis on FB will have a positive effect on Deliberation. However, the hypothesis is not supported.
The conducted analysis (R-Square = 1.000; see table 3) showed a significant negative (Beta-Value = -1.059; p<.000) effect of Time & Openness on Deliberation. This indicates that the elements of a timely and open approach do not affect the degree of deliberation in the comment sections. The control variable Crisis was significant and had a positive (Beta-Value = 0.063) influence on Deliberation, meaning that the more a post contains information about the crisis being controllable, the higher the deliberation will be in the comment section. The
19 control variable Words was also significant but had a negative (Beta-Value = -.006; p<.000) impact on Deliberation. This means that the higher the amount of words is in the comment, the more the degree of deliberation will decrease. The control variable Frequency was also significant and a negative (Beta-Value = -.008; p<.000) influence on Deliberation, meaning that the frequency of published posts from the airline will not increase the degree of deliberation in the comment section.
Table 2
Regression of Time & Openness on Deliberation
Furthermore, this study presumed that objective information in the organizational post would have a positive effect on Deliberation (Hypothesis 1a). The hypothesis was not supported. As visible in Table 4, the regression (R-Square = 1.000) showed a significant negative effect of objective Information on Deliberation (Beta-Value = -1.108; p<.000). This mean that the more an organizational post contain objective information about the organizational crisis, the less deliberation will be present in the comment section.
20 Table 3
Regression of Info on Deliberation
The control variable Crisis, which measured the mentioning of the occurrence of the crisis, had a significant positive (Beta-Value = 0.160; p<.000) effect on Deliberation. This indicates that participants are more likely to deliberate, when there is more information about the occurrence of the accident. The control variable Words was also significant but had again a negative (Beta-Value = -.006; p<.000) influence on Deliberation. This further means that a high amount of words in the comments do not lead to more deliberation among participants. The same goes with the third control variable Frequency, which also had a negative (Beta-Value = -0.55; p<.000) impact on Deliberation. This stresses that several posts about an accident from an airline does not increase the likelihood of deliberative debates in the comment sections.
Further, for hypothesis 1b, it was predicted that reservations to be taken in accordance to the crisis provided by the airline would positively affect Deliberation. This hypothesis was not supported. There was a high correlation between the variables (R-Square = 1.000) applicable in Table 5. However, Reservations to be taken is significantly negatively (Beta-Value = -1.050; p<.000) influencing Deliberation. This leads to the notion of reservations already provided by the airline will not lead to more deliberation among the participants in the comment sections.
21 The control variable Frequency was significant and had a positive (Beta-Value = .074; p<.000) influence on Deliberation. This means that the more often an airline inform people about the reservations they should take in relation to the crisis, the more deliberation will be fostered among participants in the comments. Again, the control variable Words was significant but had a negative (Beta-Value = -.007; p<.000) effect on Deliberation, meaning that a higher amount of words do not increase deliberation in the comment section. The control variable
Crisis was not significant.
Table 4
Regression of Reservations on Deliberation
Moreover, this study also predicted that airlines communicating about the actions they are taking in order to ease the harm or solve the crisis would positively influence Deliberation (Hypothesis 1c). This hypothesis was supported. The conducted regression analysis (R-Square = 1.000; see Table 6) showed a significant positive (Beta-Value = .028; p<.000) effect of
Actions on Deliberation. This means that organizational actions being communicated in FB
22 Table 5
Regression of Actions on Deliberation
With regards to the control variables, the amount of Words (Beta-Value = -.001; p<.000), the mentioning of the occurrence of the Crisis (Beta-Value = -.805; p<.000) and the
Frequency of posts published (Beta-Value = -.223; p<.000) all had a significant negative
influence on Deliberation. This indicates that a high amount of words and information do not increase the degree of Deliberation. Posts often published with information about the organizational crisis will neither increase Deliberation.
With the results from the regression analysis Hypotheses 1, 1a and 1b are not supported. This means that posts containing Time & Openness concerning information, objective
Information and information about the Reservations to be taken will negatively affect the
degree of Deliberation in the comment section. Hypothesis 1c is supported, indicating that organizational posts communicating about the organizational actions to solve the issue will lead to more Deliberation.
Moderation of Coombs’ (1999) Online Crisis Communication Theory and Deliberation by Direct Involvement
Hypothesis 2 predicted a positive influence of a Direct Involvement from the airline in the comment section on Deliberation. The hypothesis was not supported.
In order to carry out a moderation analysis, an interaction of the independent variable:
23 term between was added into the regression to analyze the moderation effect for Hypothesis 2 (R-Square = 1.000; see Table 7).
Table 6
Regression of Interaction Online Crisis Communication and Direct Involvement on Deliberation
The main effect of Online Crisis Communication (Beta-Value = 0.080; p<.000) is significantly and positively influencing Deliberation. This means that the elements constituting
Online Crisis Communication do increase the degree of Deliberation in the comment section. Direct Involvement from the organization (Beta-Value = -.001; p<.647) is not significantly
influencing Deliberation, indicating that the presence of the airline in the comment section does not increase deliberation. When analyzing the moderation effect through the interaction between Online Crisis Communication and Direct Involvement (Beta-Value = .001; p<.602), it is evident that the interaction is not significantly affecting Deliberation. This means that the
Direct Involvement from the airline does not moderate the relationship between Online Crisis Communication and Deliberation.
24 Conclusion
The airline industry is continual accused of “unpredictable events”, which force airlines to provide timely, factual information to people in order to avoid misinformation and rumors (Ray, 1999; Stritesky, 2015). However, the decentralization of communication on FB often challenges international airlines to do so, since each FB user is empowered to add their side of the story (Chadwick, 2008; Stritesky et al., 2015). Thus, a common problem on FB is the rather irrelevant and misleading content, which hinders the transparency of diverse opinions, and thereby individuals’ ability to make informed judgements. In this context, scholars have highlighted the potential for organizations to foster deliberation, since it can function as a steering force in providing individuals with valuable information, which thereby can help them reach truly just decisions (Braunack et al., 2010; Wendling et al., 2010). Hence, the present study explored the potential of fostering more deliberative debates on FB through elements from Coombs’ (1999) online crisis communication theory. For this purpose, a quantitative content analysis was carried out examining 25 FB posts and 875 comments, in which a codebook of 22 variables was developed. The choice of methodology helped this research to thoroughly investigate the content of the FB posts and whether the comments were in terms of this study’s conceptualization of deliberation.
As mentioned, Coombs’ (1999) theory formed the basis of this study, and four hypotheses were developed hereof. First, hypothesis 1 assumed that organizational posts published in a timely manner and addressing the issue proactively would positively affect the degree of deliberation. However, the results showed that this would merely decrease the degree of deliberation. It can thus be discussed that the approach rather increases the propinquity and thus constitute an ongoing conversation, as with Grunig and Grunig’s (2000) two-way symmetrical model. Nonetheless, the results did show that the mentioning of the crisis being controllable had a positive influence on deliberation. This finding is in line with literature by
25 Lynette et al. (2010), who argued that the repetition of the occurrence of the crisis encourage individuals to participate in debates, in which reasoned and informed discourse is used. It can therefore be concluded that organizations would benefit more from addressing the occurrence of the crisis rather than focusing on approaching the FB post in a timely and open manner in the aim of fostering more deliberation.
For the second hypothesis 1a, a high amount of objective information in the form of number of victims, their nationality and time of accident was presumed to lead to more deliberative debates. However, the assumption was not likely to foster deliberation. Accordingly, it can be argued that the transfer and sharing of knowledge does not encourage individuals to engage in deliberative discourse, but rather channels a way to tell their own stories (Nonaka & Takeuchi, 1995). However, the results again emphasize the occurrence of the crisis to be a factor that fosters more deliberation, and should thus be something for organizations to consider.
The third hypothesis 1b neither proved to increase deliberation in the comments. It was expected that organizational posts communicating about which reservations people should take in accordance with the situation would lead to deliberative debates. An explanation of the result may be that reservations are merely perceived as presented solution, which thus limits the possibility for individuals to participate in the interchange of reciprocal arguments (Stroud et al., 2015; Thompson, 1995). In this context, organizations could potentially foster more deliberation, if they showed more openness towards people contributing to which reservations are the most appropriate in the situation, since they are the ones also affected by the accident. As stated earlier, it was only hypothesis 1c, which showed an increase in deliberation. It was expected that the organizational actions taken in order to solve the problem would lead to more deliberative debates. The result can be explained through the literature by Braunuack-Mayer et al. (2010), who argues that a specific component of deliberation is for individuals to
26 identify the importance of preparing others in advance. The organizational actions can have a direct influence on, how individuals judge the situation, and their future associations with the airline. Hence, deliberative debates will help individuals and the organization reach decisions acceptable for all who have a stake in the issue (Braunuack-Mayer et al., 2010; Thompson, 1995).
Subsequently, the present study also investigated a possible moderator in terms of direct involvement in the comment section. In this context, hypothesis 2 assumed that the direct organizational involvement would affect the norms in the interaction to contain more thoughtful content. However, the results showed that this was not the case. Accordingly, it can be argued that the organization did not embody essential group characteristics, and thus had no influence on affecting the norms among the participant to be more deliberative (Aarts & Dijksterhuis, 2003). According to Stroud et al. (2015), it is crucial that an organization acts as an opinion leader in interactions in order to influence individuals’ norms. Hence, before engaging directly with the participants, it is crucial for the organization to acknowledge, which important characteristics are essential to embody in order to influence the commenters.
The evidence gathered from this study’s hypotheses suggest that some components of online crisis communication are better suited for deliberation than others, particularly those that allow for greater affordances of identifiability such as the occurrence of the crisis and organizational actions. These factors are easily relatable and influence how individuals perceive the airline and their further judgements of the situation. Hence, airlines communicating about tangible actions that directly affect people in their sharing and processing knowledge can ensure higher levels of meaningful and thoughtful discourse and thereby deliberative debates.
The pre-crisis mode the airline industry is forced to be in makes it crucial for airlines to help people be well-informed about the specifics of an accident. For this reason, deliberation
27 will be beneficial for airlines. It gathers people around a common agenda encouraging them to share and process information and knowledge. The openness and acceptance of various opinions as a consequence of deliberation will help people become better informed about a specific plane accident. This will contribute to better, legitimate decisions based on reason rather than coercion. The decisions will thus be acceptable for all who have a stake in the issue. Hence, the fostering of deliberation on FB will not only benefit airlines by limiting the possible damage of their reputation, it will also help individuals draw upon factual information rather than misinformation in their decision-making and thereby reach truly just decisions.
Discussion
Based on the previous sections, some aspects deserve mention, since they are perceived as limitations for this study. To start with, this study’s results showed limited deliberation fostered through online crisis communication on FB. Hence, it can be argued that some social media channels are better suited for deliberation than others. In this regard, the decentralization of communication on FB can to some extent limit the possibility for individuals to be equally well-informed, since each FB user is empowered to add their side of the story to the discussion. Thus, even though international airlines provide their publics with factual information in the form of the occurrence of the crisis and organizational actions taken, it can still be difficult to reach out to each and every one affected by the situation, and provide them with essential information that should enable them to make better, informed decisions Chadwick, 2008; Stritesky et al., 2015. Therefore, it can be argued whether FB is the right online venue to aim at fostering deliberation? Due to social media channels increasingly becoming more incorporated into our daily lives, it is important for future research to explore various channels and their potential to foster deliberation.
28 Moreover, although the present study benefitted from the quantitative content analysis, it is problematic to capture written discourse. In this context, the chosen methodology limited the effects of including normative dimensions of deliberative discourse. This may have had crucial effect on the inherited nature of the discourse taken place. It can thus be argued that a combination of a quantitative and qualitative approach in the examination of a progressive level of the public sphere would have benefitted the study. In so doing, it would have been possible to capture essential features such as reflexivity and discourse equality, which are crucial components of deliberation (Burkhalter et al., 2002). The combination of methodology would further have given a more appropriate evaluation of the degree of deliberation present in the comments (Dahlberg, 2004).
Likewise, the results of the study might have been affected by the chosen industry. The present study chose to only investigate one single industry: the airline industry, and hereby five international airlines and their representatives. It is thus not possible to know whether the results are generalizable to other industries or not. To only examine one industry further precludes generalization about the overall deliberative patterns in social media channels managed by other industries. Hence, in order to thoroughly investigate the effects of Coombs’ (1999) online crisis communication theory, this study would have benefitted from examining various industries and thus compare the results. This would have given it a more correct understanding of, how well the theory is able to foster deliberation on FB, and is thus something for future research to explore.
Besides the above mentioned aspects, also some minor technical features of this study are worth mentioning. As declared in the theoretical framework, there have been offered several approaches to categorize crisis strategies in social media channels. This study chose for Coombs’ (1999) theory. However, based on the results, it can be argued that the specific elements of the theory are not successful at fostering deliberation. Hence, it is worth
29 investigating several other online crisis communication theories, and explore the potential of their elements to foster more deliberative debates on FB. Similarly, some technical characteristics could also have affected the results of this study. It was only comments posted during the organizational crisis that were analyzed. This indicates that the conclusions are based on limited information. In this regard, it could have been essential to examine comments a couple of weeks before and after the crisis in order to examine the difference in the degree of deliberation, and thereby truly conclude the effects of the theory.
Additionally, some of the measures in this study have been limited in their measurement. The moderation effect of a direct involvement in the comment section was restricted due to low participation of organizational representatives. Hence, future research should consider examining to what extent the organization is present in the comment section in order to thoroughly investigate, whether the organization does have an influence on the norms in the interaction.
References
Aarts, H., & Dijksterhuis, A. (2003). The Silence of the Library: Environment, Situational Norm, and Social Behavior. Journal of Personality and Social Psychology, 84(1), 18-28.
Barker, R. (2011). Online Crisis Communication Response: A Case Study of Fraudulent Banking Transactions in South Africa. Communicatio, 37(1), 118-136.
Baum, N. M., Jacobsen, P. D., & Goold, S. D. (2009). “Listen to the People”: Public Deliberation About Social Distancing Measures in a Pandemic. The American Journal
30 Benoit, W. L. (1997). Image Repair Discourse and Crisis Communication. Public Relations
Review, 23(2), 177-186.
Braunack-Mayer, A., Street, J. M., Rogers, W. A., Givney, R., Moss, J. R., & Hiller, J. E. (2010). Including the Public in Pandemic Planning: A Deliberative Approach. BMC
Public Health, 10(501), 1-9.
Burkhalter, S., Gastil, J., & Kelshaw, T. (2002). The Self-Reinforcing Model of Public Deliberation. Communication Theory, 12, 398-422.
Camaj, L., & Santana, A. D. (2015). Political Deliberation on Facebook during Electoral Campaigns: Exploring the Relevance of Moderator’s Technical Role and Political Ideology. Journal of Information Technology & Politics, 12, 325-341.
Cappella, J. N., Price, V., & Nir, L. (2002). Argument Repertoire as a Reliable and Valid Measure of Opinion Quality: Electronic Dialogue During Campaign 2000. Political
Communication, 19(1), 73-93.
Carter, S. L. (1998). Civility: Manners, Moral, and the Etiquette of Democracy. New York: Basic Books.
Chadwich, A. (2008). Web 2.0: New Challenges for the Study of E-Democracy in an Era of Informational Exuberance. I/S: A Journal of Law and Policy for the Information
Society, 5(1), 9-41.
Coe, K., Kenski, K., & Rains, S. A. (2014). Online and Uncivil? Patterns and Determinants of Incivility in Newspaper Website Comments. Journal of Communication, 64(4), 658-679.
Coombs, W. T. (1999). Ongoing Crisis Communication: Planning, Managing, and Responding. Sage, Thousand Oaks, CA.
31 Dahlberg, L. (2004). Net-Public Sphere Research Beyond the First Phase. Javnost-The Public,
11(1), 27-44.
Davis, R. (1999). The Web of Politics: The Internet’s Impact on the American Political System. Oxford: Oxford University Press.
De Vries, N. J., & Carlson, J. (2014). Examining the Drivers and Brand Performance Implications of Customer Engagement with Brands in the Social Media Environment.
Journal of Brand Management, 21(6), 495-515.
Elster, J. (1998). Deliberative Democracy. Cambridge, MA: Cambridge University Press. Escobar, O. (2009). The Dialogic Turn: Dialogue for Deliberation. Inspire Journal of Law,
Politics and Societies, 4(2), 42-70.
Fields, A. M., Swan, S., & Kloos, B. (2010). What it means to be a Woman: Ambivalent Sexism in Female College Students’ Experiences and Attitudes. Sex Roles, 62, 554-567. Fjeld, K., & Moleworth, M. (2006). PR Practitioners’ Experiences of and Attitudes Towards
the Internet’s Contribution to External Crisis Communication. Corporate
Communications: An International Journal, 11(4), 391-405.
Fleck, L. (1991). Just Health Care Rationing: A Democratic Decision-making Approach.
University of Pennsylvania Law Reviews, 140, 1597-1636.
Gonzáles-Herrero, A., & Smith, S. (2008). Crisis Communication Management on the Web: How Internet-Based Technologies are Changing the Way Public Relations Professionals Handle Business Crises. Journal of Contingencies and Crisis
Management, 16(3), 143-153.
Graham, M W., Avery, E. J., & Park, S. (2015). The Role of Social Media in Local Government Crisis Communications. Public Relations Review, 41, 386-394.
32 Greer, C. F., & Moreland, K. D. (2003). United Airlines’ and American Airlines’ Online
Crisis Communication Following the September 11 Terrorist Attacks. Public
Relations Review, 29, 427-441.
Grunig, J. E. (2000). Collectivism, Collaboration, and Societal Corporatism as Core
Professional Values in Public Relations. Journal of Public Relations Research, 12(1), 23-48.
Habermas, J. (1989). The Structural Transformation of the Public Sphere: An Inquiry into a Category of Bourgeois Society. Cambridge, MA: MIT Press
Habermas, J. (2006). Political Communication in Media Society: Does Democracy Still Enjoy an Epistemic Dimension? The Impact of Normative Theory on Empirical Research. Communication Theory, 16, 411-426.
Halpern, D., & Gibbs, J. (2013). Social Media as a Catalyst for Online Deliberation? Exploring the Affordances of Facebook and Youtube for Political Expression. Computers in
Human Behavior, 29, 1159-1168.
Hogg, M. A., & Reid, S. A. (2006). Social Identity, Self-Categorization, and the Communication of Group Norms. Communication Theory, 16(1), 7-30.
Holsti, O. R. (1969). Content Analysis for the Social Sciences and Humanities. Reading, MA: Addison-Wesley.
Hyde, B., & Bineham, J. L. (2000). From Debate to Dialogue: Toward a Pedagogy of Non-Polarized Public Discourse. Southern Communication Journal, 65, 208-223.
Isaac, W. N. (1999). Dialogue and the Art of Thinking Together. A Pioneering Approach to Communicating in Business and in Life. New York: Currency.
33 Janssen, D., & Kies, R. (2005). Online Forum and Deliberative Democracy. Acta Politica, 40,
317-335.
Kim, S., & Liu, B. F. (2012). Are All Crises Opportunities? A Comparison of How Corporate and Government Organizations Responded to the 2009 Flu Pandemic. Journal of Public
Relations Research, 24(1), 69-85.
Lombard, M., Snyder-Dutch, J. & Bracken, C. C. (2002). Content Analysis in Mass Communication. Human Communication Research, 28(4), 586-604.
Lynette, M. M., Sparks, B., & Glendon, A. I. (2010). Stakeholder Reactions to Company Crisis Communication and Causes. Public Relations Review, 36, 263-271.
Manin, B. (1987). On Legitimacy and Political Deliberation. Political Theory, trans. E. Stein, and J. Mansbridge, 15(3), 338-368.
Mayfield, A. (2006). What is Social Media? Retrieved from
http://www.spannerworks.com/fileadmin/uploads/eBooks/What is Social Media.pdf
Moore, D. S., & MacCabe, G. P. (2007). Statistiek in de Praktijk. Schoonhoven: Academic Service, 5. ed.
Neuendorf, K. A. (2011). Content Analysis – A Methodological Primer for Gender Research.
Sex Roles, 64, 276-289.
Nonaka, I., & Takeuchi, H. (1995). The knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. New York: Oxford University Press. Noveck, B., S. (2009). Wiki Government: How Technology Can Make Government Better,
Democracy Stronger, and Citizens More Powerful. Washington, DC: Brookings Institution Press.
34 Ray, S. J. (1999). Strategic Communication in Crisis Management: Lessons From the Airline
Industry. Quorum Books, Westport, CT.
Robertson, S., P., Vatrapu, R., K., & Medina, R. (2010). Off the Wall Political Discourse: Facebook Use in the 2008 U.S. Presidential Elections. Information Policy, 15(1/2), 11-31.
Romanucci, & Blandin. (2013). 5 Most Common Causes of Plane Crashes. Retrieved from
http://rblaw.net/5-most-common-causes-of-plane-crashes/
Ruiz, C., Domingo, D., Mico, J. L., Diaz-Noci, J., Koldo, M., & Masip, P. (2011). Public Sphere 2.0? The Democratic Qualities of Citizen Debates in Online Newspapers. The
International Journal of Press/Politics, 16(4), 463-487.
Santana, A. D. (2014). Virtuous or Vitriolic: The Effect of Anonymity on Civility in Online Newspaper Reader Comment Boards. Journalism Practice, 8(1), 18-33.
Smith, G., & Wales, C. (2000). Citizens’ Juries and Deliberative Democracy. Political Studies,
48, 51-65,
Stritesky, V., Stránska, A., & Drábik, P. (2015). Crisis Communication on Facebook. Studia
Commercialla Bratislavensia, 8(29), 103-111.
Stroud, N. J., Scacco, J. M., Muddiman, A., & Curry, A. L. (2015). Changing Deliberative Norms on News Organization’s Facebook Sites. Journal of Computer-Mediated
Communication, 20, 188-203.
Tannen, D. (1999). The Argument Culture. Changing the Way We Argue. London: Virago Press.
Thompson, J, B. (1995). The Media and Modernity: A Social Theory of the Media. Cambridge: Polity Press.
35 Thompson, J. B. (2005). The New Visibility. Theory, Culture & Society, 22(6), 31-51.
Tipton, E. (2013). Stratified Sampling Using Cluster Analysis: A Sample Selection Strategy for Improved Generalizations from Experiments. Evaluation Review, 37(2), 109-139. Walsh, K. C. (2004). Talking about Politics: Informal Groups and Social Identity in American
Life. Chicago: University of Chicago Press.
Wendling, C., Radisch, J., & Jacobzone, S. (2013). The Use of Social Media in Risk and Crisis Communication. OECD Working Papers on Public Governance, 24, OECD Publishing.
Wise, K., Hamman, B., & Thorson, K. (2006). Moderation, Response Rate, and Message Interactivity: Features of Online Communities and Their Effects on Intent to Participate.
36 Appendix 1
Codebook
Technical Variables
Variable Name Variable Label Values of Variable
CODER Coder 1 Frier Line
2 Stieglmeier Elisa
The coder who coded the data
ORG Name of organization 1 Air Canada
2 Asiana Airlines 3 Dynamic International 4 Fly Dubai
5 Malaysia Airlines
The name of the organization
POCO Post or comment 1 Post
2 Comment
The post published by the organization or comment from a participant
IDENTITY Identity of the commenter 1 The Organization 2 Facebook User
The identity of the one leaving the post or comment.
DATE Date mmddyyyy
The date when the post/comment1 was given.
WORDS Number of words Xxxx
37 Post Attributes
Variable Name Variable Label Values of Variable
CRISIS Mentioning of crisis Xxxx
99 Not applicable
The amount of times the reason for the accident is mentioned.
If at least one of the following aspects is mentioned, Type of crisis is elaborated on: Technical-error accident, human-Technical-error accident.
TIME Time of the post 1 On the same day
2 Days after 99 Not applicable
The time of the post published about the organizational crisis
FREQUENCY The number of posts Xxxx
99 Not applicable
The total number of posts published about the organizational crisis
OPEN The openness of the crisis 1 Post addresses the accident proactively
2 No information about the accident is mentioned 99 Not applicable
In the post, the organization acknowledges the crisis by addressing it proactively.
Examples for 1: “The incident occurred at approximately 00:43 AT”; “We are aware of an
incident involving our flight”; “We regret to confirm that Flight FZ981 crashed on landing”; “Deeply sorry to confirm the following information in relation to the tragic accident”.
INFO Objective Information 1 Post includes information about the accident
38 2 No information about the
accident is mentioned 99 Not applicable
In the post, the organization includes hard and objective facts about the crisis.
If at least one of the following aspects is mentioned, Objective information is coded 01: Flight number, number of victims, nationality of victims, time of accident.
Examples for 1: “Flight FZ981 from Dubai to Rostov On Don”; “The accident occurred at
approximately 00:50 GMT”; “Carrying 133 passengers and give crew, was involved”; “44 Russian, 8 Ukrainian, 2 Indian, 1 Uzbekistani”; “Lost contact with flight MH17”.
RESERV Reservations 1 Post includes information about
the reservations
2 No information about the reservation is mentioned 99 Not applicable
In the post, the organization address those affected or unaffected by the crisis and present methods and procedures to link individuals to possible preventative or corrective actions.
Examples for 1: “Family members who seek information about passengers on Flight AC624
may telephone”; “Emergency contact line for passengers and their relatives are open”; “The investigation is ongoing; relatives must therefore be patient”.
ACTIONS Organizational action 1 Post includes organizational action
2 No information about the organizational action is mentioned
99 Not applicable
In the post, the organization writes about which actions are made to ease/solve the accident. If at least one of the following aspects is mentioned, Organizational action is coded 01:
39
Contacting family members, further investigation, change fees for passengers, alternative routes.
Examples for 1: “We are investigating further details”; “We are currently contacting relatives
of the passengers and crew”; “With immediate effect, all European flights will be taking alternative routes”; “We will be waiving any change fees for passengers who wish”.
Comment Attributes – Rationality
Variable Name Variable Label Values of Variable
ASSERTION Explicit
assertion/allegation
1 Comment includes explicit assertion/allegation
2 No explicit assertion/allegation is mentioned
99 Not applicable
In the comment, the commenter includes an opinion with either a rationale or personal emotions, prejudice, or judgments.
Examples for 1: “I think every flight company have different strategy”; “This is important to me
because my ultimate fear is dying in a plane crash”; “I kinda agree, instead of wasting 2 hours like that, they could have diverted to another airport”.
REASON Reasoned argument
(External validation)
1 Comment includes reasoned argument
2 No reasoned argument is mentioned
99 Not applicable
In the comment, the commenter includes an argument supported by a rationale.
Examples for 1: “If you feel intimidated by knowledge, you should lock yourself away in a dark
room rather than cry”; “Do the landing on another airport for temporary if there was a poor visibility or any obstacles”; “Fate is fate but taking a bas risk will higher your chances of
40
death”.
EVIDENCE Argument supported with
evidence
(External validation)
1 Comment includes argument supported with evidence 2 No argument supported with
evidence is mentioned 99 Not applicable
In the comment, the commenter includes an argument with evidence in the form of the source of information.
Examples for 1: “I am appalled by a report aired on national television by Good Morning
America”; “It is obvious that the Ukrainian army shot down this plane. See the photo”; “I do not think people should dwell on the route as it was open and the fight path approved”.
Comment Attributes – Reciprocity
Variable Name Variable Label Values of Variable
REPLY Reply to previous
comments posted by other participants.
1 Comment includes reply to previous comments 2 No reply is given 99 Not applicable
In the comment, the commenter replies to previous comments posted by other participants.
Examples for 1: “Marina Novikova, you have to wait to find out”; “Dear Marian Miller, I
would like to say that your immediate response to this tragedy is no different”; “Marina, just a question, do you work for aviation?”; “Dear Julia, while the emotional reaction is
understandable, the “blame culture” is actually unsafe”. “I agree with Maria and Thomas”.
ADDITIONAL Additional information to the discussion generated by the initial post.
1 Comment includes additional information
41 given
99 Not applicable
In the comment, the commenter replies with additional information to the discussion generated by the initial post.
Examples for 1: “It was not the airlines fault. The ACT told them that the weather is getting
better”; “It does not work like that. Plus, fuel only lasts so long”; “Really sad news, and I do not know why Fly Dubai plane 773-800 is not supporting fusion technology”; “Typical mistake of the pilots. Exactly the same situation like Tatarstan Flight 3636”.
OFF-TOPIC Comments is neither
related previous
comments nor initial post
1 Comment includes Off-topic 2 No off-topic comment is given 99 Not applicable
In the comment, the commenter provides a reply that is neither related to the initial post nor to previous comments among users.
Examples for 1: “If you are interested to know more about the aircraft visit this website”.
Comment Attributes – Politeness and Civility
Variable Name Variable Label Values of Variable
POLITE Impolite behaviors 1 Comment includes impoliteness
2 No aspects about impoliteness 99 Not applicable
In the comment, the commenter shows either polite or impolite behaviors in his/her response. If at least one of the following aspects is mentioned, Politeness is elaborated on: Curses, insults, words that indicates judgmental speak.
Examples for 1: “How stupid of you to assume that”; “You obviously do not know what you are
talking about!”; “HOW DARE YOU?!”; “You are timid with racism with your most stupid comments”.
