From Dialogue to Deliberation:
A Deliberative Perspective on Organizational Communication on
Social Media with Publics
Stephanie Balorian (10812814) Master’s Thesis
Graduate School of Communication Master’s programme Communication Science
Supervision by Dr. J.M. Slevin 29/06/2018
2 Abstract
It is important for organizations to maintain the organization-public relationship, because it is essential to their survival. Dialogue has been proposed as an ideal way to engage with publics, because it promotes dialogue. However, due to the anonymity and massiveness of social media platforms, dialogue on social media is impractical for communication
professionals. The massiveness of social media platforms make it difficult for communication professionals to keep up with the influx of replies sustain dialogue. Thus, communication professionals could benefit from adopting a different perspective on engagement with publics on social media, that is more in line with the characteristics of social media. This study investigates to what extent a different theory, in this case deliberation, can be used to engage with publics on social media. By examining to what extent a certain type of communication management can prompt deliberation, and further how deliberation by publics can be enhanced through moderation and time, this study offers communication professionals a different perspective to aid them in the development of future online corporate
communication strategies. A content analysis was conducted on 722 Facebook posts and comments from 10 companies on the 2018 Fortune 500 list. The main result of this study is that the voluntary disclosure of information by the organization in a Facebook post, in combination with moderation by the organization in the comment section, increases the degree of deliberation by publics on the given information.
3 From Dialogue to Deliberation: A Deliberative Perspective on Organizational Communication
on Social Media with Publics
The popularity of social media platforms has continued to rise over the years, with Facebook and Youtube leading the way (Smith & Anderson, 2018). In an effort to reach publics on these platforms, organizations have made sure to also be present on such platforms. Of the organizations on the 2016 Fortune 500 list, 88% percent have a Facebook account, and 79% have a Twitter account (Ganim Barnes & Daubitz, n.d.). Social media is stated as the most effective way to build brand awareness, and form relationships with publics (Ganim Barnes & Daubitz, n.d.). It is necessary to maintain and improve the organization-public relationship, because it is considered essential to the survival of an organization (Ledingham, 2003; Suchman, 1995; Sweetser, English & Fernandes, 2015). Improving the organization-public relationship online is possible due to the interactive capabilities of social media. These capabilities allow users to interact with one another, without having to be in the same space or time (Albu & Etter, 2016). And to disseminate information to many others at once (Gamboa & Goncalves, 2014). Social media can be defined as “internet-based channels that allow users
to opportunistically interact and selectively self-present, either in real-time or asynchronously, with both broad and narrow audiences who derive value from user-generated content and the perception of interaction with others” (Carr & Hayes, 2015, p. 50).
The two-way symmetrical model of communication is argued to be the most ethical way for an organization to interact with publics, because it promotes dialogue (Botan, 1997; Grunig & Grunig, 1992). However, engaging with publics on social media to form
relationships with publics, through dialogue is not practical for communication professionals. Dialogue can be described as both a precondition to, and a product of, a relationship that is based on mutual understanding and respect (Kent & Theunissen, 2016). The concept of dialogue cannot be scaled up to meet the demands of massive social media platforms. Social
4 media posts can become very popular, and receive more than a thousand replies (Gamboa & Goncalves, 2014). The goal of dialogue is to reach a level of mutual understanding between communication partners (Kent & Taylor, 2002). To reach mutual understanding,
communication partners must be committed to the dialogue. This commitment leads to sustained communication over time (Theunissen & Wan Noordin, 2012). Due to the
massiveness of social media platforms, the attempt of communication professionals to engage in dialogue is limited to one-time attempts, because of a lack of commitment from
organization and publics participant (Lane, 2014). Considering the theory behind dialogue, one-time attempts of initiating dialogue by an organization with individual users on social media does not constitute dialogue (Kent & Theunissen, 2016). Further, the anonymity of social media makes it also questionable whether it is the appropriate space for dialogic engagement between organization and publics. Reason for this questioning is that anonymity hinders the formation of relationships, which is necessary for dialogue to occur (Black, Mezzina & Thompson, 2016; Dwyer, 2007).
The concept of dialogue is very prominent in public relations research (Bortree & Seltzer, 2009; Kim, Kim & Nam, 2014; Rybalko & Seltzer, 2010; Kim, Chun, Kwak & Nam, 2014; Heath et al., 2006). However, communication professionals might benefit from
adopting a different perspective, that is more appropriate in the context of social media. In this study deliberation is proposed as that other perspective, because it is better equipped to deal the characteristics of social media that currently clash with dialogue theory. Deliberation is characterized by thorough group discussions, and the formation of well-informed judgements (Halpern & Gibbs, 2013). The aim of this study is to offer a deliberation as a more practical solution to engaging with publics on social media, and form relationships. The results of this study are relevant for communication professionals, who have struggled with improving their organization-public relationship by engaging publics through dialogue. This study is further
5 scientifically relevant as it contributes to the discussion on the role of dialogue on social media. Further, it contributes to the body of knowledge on fostering deliberation on social media, by investigating deliberation in the context of public relations.
To examine the extent to which a deliberative perspective on the engagement between organizations and publics on social media is appropriate, and practical in use for
communication professionals, the following research question is proposed:
To what extent can deliberation be prompted and enhance in Facebook comment of organizational Facebook posts from Fortune 500 companies, through reflective communication management, moderation by the organization and time?
Theoretical Framework
Clash Between Dialogue Theory and Public Relations on Social Media
Dialogue can be defined in various ways. One of the early influences on dialogue theory in public relations comes from philosopher Martin Buber. Buber (1958) introduced the notion of “I-It” and “I-Thou” interactions, which describes how humans interact with the world around them and with other humans. Following the perspective of Buber (1958) Cissna and Anderson (1998) describe dialogue as “a quality of relationship that arises, however briefly, between two or more people and a way of thinking about human affairs that highlights their dialogic qualities” (Cissna & Anderson, 1998, p. 65). Kent and Taylor define dialogue as “any negotiated exchange of ideas and opinions”(Kent & Taylor, 1998, p. 325).
In this study, the perspective of Kent and Theunissen (2016) on dialogue is followed. They describe dialogue as a precondition to, and product of a relationship based on mutual understanding and respect. This definition was selected, because it describes the concept of dialogue in its totality. Dialogue as a precondition refers to the communicative competence of communication partners, which is necessary for mutual understanding to reached (Habermas,
6 1970; Kent & Theunissen, 2016). Communicative competence describes the universal
understanding of grammar rules concerning the personal pronoun, the interrogative, the imperative and so forth (Habermas, 1970). Dialogue as a product refers to the accumulation of repeated interactions and exchanges between individuals (Kent & Theunissen, 2016).
Due to the interactive capabilities of the internet and social media, these spaces were thought to be ideal for facilitating dialogue between organizations and publics (Kent & Taylor, 1998; Valentini, 2015). Social media can be defined as “internet-based channels that allow users to opportunistically interact and selectively self-present, either in real-time or asynchronously, with both broad and narrow audiences who derive value from user-generated content and the perception of interaction with others” (Carr & Hayes, 2015, p. 50). Online organizations could reach new publics and directly communicate with them, and publics could were now no longer just recipients of corporate communication (Kent & Taylor, 1998; Kent, 2013). By creating their own content and sharing this with an audience online, they could voice their thoughts and opinions on what an organization was doing (Kietzmann, Hermkens, McCarthy & Silvestre, 2011; Pfeffer, Zorbach & Carley, 2014).
Contrary to the assumption of social media being a space for dialogue, not all social media are suitable. Facebook and YouTube are the most popular social media platforms in North-America (Smith & Anderson, 2018). However, Facebook is more appropriate than YouTube, because anonymity is not encouraged on Facebook (Zhao, Grasmuck & Martin, 2008). Anonymity can be a hindrance to the process of dialogue, because communication partners are unsure of whom they are communicating with (Black, Mezzina & Thompson, 2016; Dwyer, 2007). A certain degree of self-disclosure from either communication partner is necessary to establish a relationship, which is a necessary precondition to dialogue (Kent & Theunissen, 2016). On Facebook individual users share identifying information such as their real name, and who their family and friends are (Zhao, Grasmuck & Martin, 2008).
7 Considering the relationship between dialogue and anonymity, it could be argued that the anonymity would also have an impact on the likelihood of dialogue occurring between organizations and publics. Organizations often communicate as a collective (Christensen & Cornelissen, 2013). Depending on the organization, individual employees communicating with publics on behalf of their organization do not have to reveal their own identity (Rybalko & Seltzer, 2010). This causes an imbalance in the organization-publics relationship, which clashes with the notion of communication partners in dialogue being able to view each other equals (Kent and Taylor, 2002).
The massiveness of social media platforms also forms an issue. Social media
facilitates mass personal communication. An individual is able to speak to many others at an instance, and the many others can also reply at an instance (O’Sullivan & Carr, 2017). An outcome of this is that a single post by an organization on Facebook could receive more than a thousand replies. In a situation like this, trying to engage individual commenters in dialogue becomes impractical. Communication professionals have limited power in establishing dialogue as described by the principles of dialogue (Kent & Taylor, 2002). The first principle is mutuality which refers to the acknowledgement of the interdependency between
organizations and publics. The second principle is propinquity which refers to organizations consulting publics on matters that affect them, and publics being able and willing to voice their opinion on those matters. The third principle is empathy which describes that
supportiveness and trust should be present for dialogue to be fruitful. The fourth principle refers to the acceptance of risks that are associated with dialogue due to its open-ended character. And lastly, the fifth principle commitment refers to a commitment from both communication partners to reach mutual understanding on an issue.
PR practitioners can initiate dialogue, but publics might not want to stay committed to the dialogue (Lane, 2014). Thus, attempts at dialogue are often one-off attempts. It is
8 impossible to enact all the principles of dialogue in such short engagements with publics. Also, PR practitioners have to meet certain deadlines and requirements, set by their
organization, to reach organizational goals. Some PR practitioners note that such requirements hinder their capability to be more empathetic or committed to dialogue (Lane & Bartlett, 2016).
Thus, certain characteristics of social media clash with dialogue theory, making it questionable whether dialogue theory is the right solution for engaging publics on social media. The anonymity of social media platforms creates an imbalance in the organization-publics relationship, and the massiveness of the platforms limits the practical use of dialogue on social media. Considering this, PR practitioners might benefit from adopting a different perspective: deliberation.
Reflective Communication Management and Deliberation
Deliberation has roots in political theory (Edwards, 2015; Thompson, 1995). The concept can be described as “a process through which participants address their conflicts, share information, exchange arguments and make decisions” (Palazzo & Scherer, 2006, p. 80). It comprises four dimensions: rationality, reciprocity, respect and constructiveness (Esau, Friess and Eilders, 2017). Deliberation shares some of the same ideals as dialogue such equality and respect for one another. However, promoting deliberation rather than dialogue on social media could be a more realistic goal for public practitioners, because it is better
equipped to deal with the massiveness of social media platforms. Participants do not have to engage one another in dialogue to receive benefits from the interaction. Benefits such as being more well-informed on an issue, and having a higher perceived legitimacy of decisions
resulting from the deliberation, can also be received by simply observing the process of deliberation (Grönlund & Himmelroos, 2009; Stromer-Galley & Muhlberger, 2009; Thompson, 1995).
9 Online deliberation can be influenced by numerous aspects. In this study the influence of a communication strategy in a Facebook post on the degree of deliberation in Facebook comments is examined. The idealization of dialogue has made it seem as though other types of communication strategies are less ideal. Grunig and Grunig (1992) argued that the two-way symmetrical model of communication would be the most ethical, yet Kent and Taylor (2002) question this claim and note that there had been no evidence to support that claim. Theunissen and Noordin (2012) point out that both one-way and two-way communication are relevant, as a one-way strategy can promote awareness of an issue, and awareness is necessary before understanding can be reached. Van Ruler and Vercic (2004) proposed a reflective approach to communication management, which is described as a way of using different communication strategies interchangeably to solve managerial problems. The goal of approach is of
addressing public legitimacy of an organization. There are five models of communication in this approach: information, persuasion, relationship, dialogue and reflective. The first four communication models are similar to those by Grunig and Grunig (1992), but in the reflective approach they function as interchangeable strategies. The reflective model acts as an
umbrella. Considering the argument made in this study concerning the status of dialogue on social media, this strategy is considered as a Dialogue-In-Name-Only strategy, as coined by Kent and Theunissen (2016).
The way information is conveyed can have an influence on the degree of deliberation. Thompson (1995) noted that there is no basis for the assumption that dialogue would be more useful in prompting deliberation than a one-way form of communication. Edwards (2015) also noted that other types of discourse are relevant to deliberation, as different communication styles can “express the identities and positions of different groups involved in deliberation” (Edwards, 2015, p. 68). Gudowksy and Bechtold (2013) examined how different
10 type of information influenced the capacity of participants to deliberate. Thus, following the statements by Thompson (1995) and Edwards (2015), and considering the study by
Gudowsky and Bechtold (2013), it can be assumed that not one particular communication strategy would be better in prompting deliberation than another. To test this assumption the following hypothesis is formulated:
H1: Reflective communication management by a Fortune 500 company on Facebook has a positive influence on the degree of deliberation in Facebook comment section
Enhancing the Degree of Deliberation Through Moderation and Time
Studies have indicated that moderation and time could also affect the degree of deliberation (Janssen and Kies, 2005; Medaglia and Zhu, 2017; Wise, Hamman and Thorson. 2006). Medaglia and Zhu (2017) examined public deliberation on Chinese government-managed social media. Results indicated that users were more likely to interact with other users that had the same beliefs as them. This led to an effect of their existing opinions being more reinforced than pre-deliberation. Based on this Medaglia and Zhu (2017) proposed that moderation could improve the degree of deliberation, as a moderator could interact with users to improve opinion diversity. Janssen and Kies (2005) described two roles for a moderator: censor and promoter of deliberation. A censor would for example remove comments that are not in line with the theme of discussion (Janssen and Kies, 2005). A promoter of deliberation would for example give background information with the goal of promoting the deliberation.
Aside from promoting opinion diversity, the presence of a moderator could promote equal participation in the discussion. Wise, Hamman and Thorson (2006) examined through an experiment the effect of moderation on an individual’s intent to participate in discussion in online communities. The result indicated that participants were more likely to participate in the moderated community than the unmoderated community. The reasoning behind this is that
11 harmful messages can be a hindrance to participation, and moderation could reduce the
amount of harmful messages (Nonnecke, Preece, Andrews & Voutour, 2004). Moderation should however be done carefully. Disagreements should not be considered as harmful messages, as conflicting views in a discussion can contribute to an individual’s process of opinion formation (Price, Capella & Nir, 2002).
It could be assumed that moderation to promote deliberation could positively influence the relationship between the communication strategy and the degree of deliberation, by
promoting opinion diversity through interaction with commenters, and improving equal participation through actions such as monitoring of harmful messages. To test this assumption the following hypothesis was formulated:
H2: Moderation by the organization positively moderates the relationship between reflective communication management and the degree of deliberation in Facebook comments on Facebook posts from Fortune 500 companies
Time is another aspect to consider in deliberation. Janssen and Kies (2005) note that asynchronous discussion spaces are more suitable for deliberation than synchronous spaces, because it allows users to reflect on and justify their opinion before sharing it (Janssen & Kies, 2005). Synchronous spaces led to more straying away from the topic of discussion (Stromer-Galley & Martinson, 2009). Wise, Hamman and Thorson (2006) also found that the time that had passed between posts had a positive influence on the intent to participate in discussion. Further, the position of the post on a webpage might also have an influence. Schultz (2017) examined to what extent brand post characteristics would influence consumer engagement. Results from their study indicated that the longer a post stayed at the top of a Facebook page, the more likes, comments and shares the post received. The reasoning behind this being that the post received more exposure at the top of page, allowing users to see the post more easily and comment. In the context of deliberation, it could be assumed that the
12 length of time that a post is positioned at the top of a Facebook page positively moderates the relationship between reflective communication management and the degree of deliberation, as it allows users to see the post, and there would be more opportunity to participate as a
consequence. To test this assumption, the following hypothesis was formulated:
H3: The length of time that a post is positioned at the top of a Facebook page positively moderates the relationship between reflective communication management and the degree of deliberation in Facebook comments on Facebook posts from Fortune 500 companies
Figure 1. Conceptual model of present study
Methodology Research method
A quantitative content analysis was selected to examine the relationship between the independent variables Reflective Communication Management, Moderation by Organization, Time and the dependent variable Degree of Deliberation. This method was deemed
13 to determine the degree of deliberation in Facebook comments in an objective and systematic manner, by identifying specified characteristics of Facebook posts and comments (Boeije & 't Hart, 2009).
Sample
The social media platform Facebook was selected because it is indicated to be a suitable space for deliberation (Halpern and Gibbs, 2013). A systematic random sampling of organizations was drawn from the 2018 Fortune 500 list. This sampling technique was selected to benefit the external validity of this research (Boeije & 't Hart, 2009). The 2018 Fortune 500 list was selected, because the rankings are based on the yearly financial success of a company. In this study it presumed that online communication played a role in their achievement of financial success. Thus, an objective of this study is to examine their
communication strategy on social media with publics, and to what extent deliberation played a role. To be selected the companies had to have a verified Facebook page, and the Facebook posts had to have a minimum of five comments in response. If a page was not verified then the legitimacy of the page was determined by using logical reasoning. Following this criteria, the following 10 companies were selected: Walmart, Berkshire Hathaway, AT&T, Ford Motor Company, Costco, Kroger, Home Depot, The Boeing Company, Comcast and Target. Data collection
The last 10 Facebook posts per organization were collected on the 28th of May 2018 through the Netvizz API on Facebook. This collection resulted in 42 posts and 722 comments from 10 companies between the 10th of February 2018 and the 27th of May 2018. The total sample of coded items is 764. The registration unit included a Facebook post and a maximum of the first 20 comments in response to that post. The unit of analysis was a Facebook
14 possible for Facebook posts to have more than a thousand comments. Due to time constraints it would not have been feasible to analyse all of them. A disadvantage of this decision is that the sampling of comments might not be representative of the whole comment thread.
However, it does allow for the assessment of the degree of deliberation in the comment section to a certain extent.
Control variable
Amount of Facebook followers. The amount of Facebook followers on a page could influence the level of interaction on that page, and consequently the degree of deliberation in Facebook comments (Sabate, Berbegal-Mirabent, Cañabate & Lebherz, 2014). Therefore, this variable was added to analysis as a control variable to exclude it as a potential influence on the dependent variable Degree of Deliberation. The amount of Facebook followers per organizations was measured as a ratio variable.
Time. This ratio variable measured the timespan between Facebook posts. The operationalization of the variable is derived from the study by Schultz (2017). Time was measured by recording the time between Facebook posts in total amount of minutes.
Reflective Communication Management. This nominal variable measured the
communication strategy present in a Facebook post. The operationalization is based on the theory by Van Ruler and Vercic (2004). The variable is measured on a nominal level with four categories ranging from 1 to 4: (1) informational, (2) persuasive, (3) relational and (4) dialogic. Please see appendix A for a full overview of indicators per strategy.
Moderation by Organization. This ratio variable measures how often the organization interacted with users under a Facebook post with the goal of promoting deliberation.
Indicators for this variable are based on the conceptualization of moderation by Janssen and Kies (2005). Indicators of this variable are: addressing complaints and criticism by
15 commenters, giving out a warning for harmful content being posted and informing the
commenter of the possibility that their comment might be deleted, encouraging alternate viewpoints related to the topic of the post by asking questions, elaborating on the topic of a post by providing extra information, and asking general questions related to the topic of the post to commenters with the objective of promoting discussion.
Degree of Deliberation. In this study the operationalization of the concept deliberation by Esau, Friess and Eilders (2017) was selected, because of it its recency and its application to the social media platform Facebook, which is also being examined in this study. These factors benefit the internal validity of this research. The degree of deliberation was measured by seven individual items, and was later in the analysis computed into a scale from 0 to 7. The seven items are: topic relevance, whether or not the content of the comment is related to the topic of the post; reasoning, whether or not the comment contains justifications of a statement; general engagement, whether the comment addresses content in another comment;
argumentative engagement, whether or not the comment addresses an argument in another comment and critical engagement, whether or not the comment is critical of another
comment; respect is measured by the absence of aggressive and offensive language. Lastly, constructiveness is measured by whether or not a comment contains constructive elements such as suggestions for solutions. Each of these items is measured on a nominal level with dichotomous coding ranging from 0 (not present) to 1 (present).
Inter-coder reliability
Reliability was tested for the variables Reflective Communication Management, Moderation, Time and the individual variables comprising the variable Degree of Deliberation. A complete overview of the reliability scores is provided in appendix B. Reliability scores from .60 or .67 and up are deemed as acceptable (De Swert, 2012). The following three inter-coder reliability tests were applied on 12% of the total sample Facebook
16 posts and comments (N = 724): percentage agreement, Cohen’s Kappa and Krippendorf’s Alpha.
The reliability scores Cohen’s Kappa and KAlpha for the variable Critical Engagement could not be calculated, because the sample did not vary enough. Its percentage agreement was however a 100%, meaning that both coders agreed that it was absent in the sample. This result is however in line with the study from which the operationalization was derived from (Esau, Friess and Eilders, 2017). The lack of variety in the sample also applied to the variable Reflective Communication Management (% agreement = 100%, κ = 1.000, KAlpha = 1.000). In this case both coders agreed that only one case varied in the sample and the rest did not.
The variable Argumentative Engagement (% agreement = 96,6%%, κ = .651, KAlpha = .6509) and Respect (% agreement = 98.9%, κ = .662. KAlpha = .6629) received acceptable scores but should be discussed. A possible explanation for the difference between the average reliability scores and high percentage agreement could be due to the binary coding of the variables. The rarity of the variable being present in the comments could have led to a low reliability score, when in fact very few mistakes were made (De Swert, 2012).
Plan of analysis
To test hypothesis 1 through 3, the following plan of analysis was followed. First, the influence of the control variable Amount of Facebook followers on the dependent variable Degree of Deliberation was assessed through a linear regression analysis. This analysis was selected, because it allowed for the assessment of the degree to which the amount of
Facebook followers could predict the degree of deliberation in Facebook comments. A hierarchical multiple regression analysis was conducted to assess the influence of the independent variable, and the two moderators on the dependent variable Degree of Deliberation. This test was deemed the most appropriate statistical test, because the two
17 moderator variables and the dependent variable are measured on a ratio level or could be accepted as interval level.
The nominal independent variable Reflective Communication Management was recoded into three dummy variables and added to the regression analysis: RCM_persuasion (M = .10, SD = .30), RCM_relationship (M = .09, SD = .08) and RCM_dialogue (M = .08, SD = .28). Tests for multicollinearity proved that there was moderate collinearity between the variables. However, this was not of great concern, as all the values for VIF are below 5 and thus acceptable (Field, 2009). As the assumptions for a hierarchical multiple regression analysis were met, the test was performed with the use of the PROCESS macro version 3 by Hayes (2012) in SPSS 24.
Results Control variable
A linear regression analysis was conducted to assess the potential influence of the control variable Amount of Facebook followers on the dependent variable Degree of
Deliberation (Sabate, Berbegal-Mirabent, Cañabate & Lebherz, 2014). The regression model with the Degree of Deliberation (M = 6.85, SD = 21.69) as a dependent variable and the Amount of Facebook followers (M = 8027540.73, SD = 10391991) as independent variable was not significant, F(1,762) = .53, p = .465. Therefore it could not be used to predict the Degree of Deliberation. This result was confirmed by lack of significant effect of the variable Amount of Facebook Followers on the Degree of Deliberation, b* = -.026, t = -.73, p = .465, 95% CI [5.34, 9.24]. This result indicates that an increase in the amount of Facebook
followers was not related to an increase in activity in the comment section of a Facebook post. As an organization attracts more followers on Facebook, the degree to which publics would deliberate in the comment section of Facebook posts remained unaffected. Thus, it was
18 concluded that in this study that the amount of followers of an organization on Facebook, had no influence on the Degree of Deliberation.
Effect of Reflective Communication Management on Deliberation
A hierarchical multiple regression analysis was conducted to assess the influence of the independent variable Reflective Communication Management on the dependent variable Degree of Deliberation, and the moderators Time and Moderation by the Organization. The regression model with the Degree of Deliberation as a dependent variable and Reflective Communication Management, Moderation by Organization and Time as independent variables was significant, F(11, 752) = 2.27, p < .010. The regression model could therefore be used to predict the Degree of Deliberation, but the strength of the prediction was very weak: .03% of the variation in the Degree of Deliberation in Facebook comments could be predicted on the basis of Reflective Communication Management, Moderation by the Organization and Time (R2 = .03).
No support was found for hypothesis 1. Hypothesis 1 assumed that Reflective Communication Management would positively influence the Degree of Deliberation in Facebook comments. Significant results were found for the strategies Dialogue and
Information. Dialogue, b* = 9.20, t = -1.99, p = .046, 95% CI [-36.40,-.26] and Information, b* = 4.56, t = -2.39, p = .017, 95% CI [-19.85, -1.93] had a significant but weak association with the Degree of Deliberation. On average the use of a Dialogue strategy led to a decrease in the degree of deliberation by 18.33, and the use of an Information strategy led to a decrease of 10.88. For these effects, the other independent variables are assumed to be held constant.
The communication strategy in a Facebook post was coded as Dialogue when it contained a question, and Information when it contained neither of the elements of a dialogic, persuasive or relational strategy. The beforementioned results indicates that the addition of a
19 question to a Facebook post, in comparison to a purely objective post, would lead to a
significant decrease in the degree to which publics would deliberate in the comment section of that post.
Moderation analysis
To test the moderation effects of the variables Moderation by Organization and Time, six interaction terms were made between each individual moderator, and the three dummy variables from the variable Reflective Communication Management. Partial support was found for hypothesis 2. Hypothesis 2 assumed that Moderation by the Organization would positively moderate the relationship between Reflective Communication Management and the Degree of Deliberation. No significant main effect was found for the category Relationship on the Degree of Deliberation, b* = 19.12, t = .36, p = .718, 95% CI [-30.62, 44.44]. This
indicates that solely the voluntary disclosure of information by the organization about their involvement in societal issues was not enough to enhance deliberation by publics in the comment section.
However, the interaction between the strategy Relationship and Moderation by Organization did prove to be significant, b* = 1.00, t = 3.42, p <.001, 95% CI [1.46, 5.39]. Partial support was found for the conclusion that Moderation by the Organization has a moderating effect on the relationship between the Reflective Communication Management and the Degree of Deliberation. Figure 2 shows that on average Facebook posts with a Relationship strategy had a higher degree of deliberation in the comment section when
moderation by organization in the comment section was present, than posts that did not have a Relationship strategy nor moderation by the organization in the comment section. This
indicates that when the organization would voluntary disclose information about for example their involvement in environmental issues, and consequently increased their moderation of the
20 comment section by asking questions or addressing criticisms, publics would deliberate more in the comment section.
Figure 2. Interaction effect between independent variable Relationship and Degree of Deliberation and moderator Moderation by the Organization
Lastly, hypothesis 3 assumed that the Time that a Facebook post was positioned at the top of the Facebook page would positively moderate the relationship between Reflective Communication Management and the Degree of Deliberation in comments. Contrary to the prediction no support was found for this hypothesis, b* = -.001, t = -.932, p = .352, 95% CI [-.00, .00]. This indicates that regardless of the type of communication strategy present in the Facebook post, longer exposure to that Facebook post at the top of a Facebook page would not motivate users to deliberate more on the information in the Facebook post. See table 2 in appendix B for a complete overview of the effect sizes for all variables and interaction terms included in the hierarchical multiple regression analysis.
Conclusion 0 5 10 15 20 25 30
Low RCM_Relationship High RCM_Relationship
De p en d en t var iab le Low Moderation High Moderation
21 The research problem of this study was that dialogue had been proposed as an ideal way for organizations to engage with publics, and improve the organization-publics
relationships. However, considering the theoretical foundation of the concept dialogue, the application of the dialogue theory on social media was argued to be impractical, due to the anonymity and massiveness of social media platforms. The popularity of Facebook posts make it impractical for communication professionals to build a relationship and consistently engage in dialogue with individual users. The use of dialogue on social media was further argued to be questionable, due to the anonymity that social media provides. Organizations often do not require their employees to identify themselves online. Thus, the relationship is skewed in favour of the organization, and impedes the ability of communication partners to view each other as equals in dialogue. To provide communication professionals a more practical solution to engagement with publics on social media, this study examined the communication between organizations and publics on Facebook from a deliberative perspective.
The results of this study indicate that the degree of deliberation in Facebook comments could not be prompted by solely applying the theory of reflective communication
management. Nor could the degree of deliberation by publics be enhanced through longer exposure to the Facebook post at the top of a Facebook page. Moderation by the organization with the objective of promoting deliberation, was able to motivate publics to deliberate more in the Facebook comments. However, increasing the deliberation by publics on Facebook is only possible when the Facebook post contains voluntary disclosure of information by the organization.
Hypothesis 1 assumed that interchangeable communication strategies would each positively influence the degree of deliberation in Facebook comments. Contrary to the prediction made in this study, none of the interchangeable communication strategies of the
22 Reflective Communication Management approach were effective in prompting deliberation. Neither a relational nor a persuasive communication strategy was able to prompt deliberation. The use of an informational and dialogic communication strategy actually lead to a decrease in the degree of deliberation by publics. This result is contrast with the claim by Grunig and Grunig (1992) that a two-way symmetrical communication would be the most ideal way of engaging. The dialogue and relational strategy, as described by Van Ruler and Vercic (2004), can both be defined as two-way symmetrical communication models, because they both promote negotiation and discourse. A practical implication of this result is that
communication professionals, depending on the goal of their communication effort, may have to be more critical in choosing which communication strategy to apply on Facebook, as the use of a dialogic or informational communication strategy may be not as effective for engaging with publics as previously assumed.
Hypothesis 2 assumed that moderation with the objective of promoting deliberation would positively influence the degree of deliberation by publics in the comment section. Only the use of a relational communication strategy, in combination with moderation by the
organization in the comment section, was able to enhance the degree of deliberation. This result provides evidence for the proposal by Medaglia and Zhu (2017) to implement moderation in the process of deliberation. A practical implication of this result is that for communication professionals aiming to increase trust in their organization among publics, the use of a relational communication strategy on social media may be most effective. This also indicates that the moderation by organizations on such Facebook posts will have to increase, as the effect of the relationship strategy may not occur otherwise.
Hypothesis 3 assumed that the duration of exposure to a Facebook post at the top of the page would be able to increase the degree of deliberation in the Facebook comments. However, the results indicate that the duration of exposure to a Facebook post at the top of an
23 organizational Facebook page plays no role in enhancing the degree of deliberation in
Facebook comments. This result is in contrast with the arguments by both Janssen and Kies (2005), and Wise, Hamman and Thorson (2006) that asynchronous spaces would be more suitable for dialogue than synchronous spaces. The space wherein the process of deliberation in this study was investigated is Facebook. Facebook provides users asynchronous interaction. A possible explanation for the difference in result, is that Janssen and Kies (2005), and Wise, Hamman and Thorson (2006) investigated the role of time on online forums instead of social media. Online forums differ from social media in technical features. These may have affected the outcome of the results. Knowing that there is a difference between online forums and social media, the results from the study of Schultz (2017) were used in this study to place the role of time in the context of social media. However, this seemed to have no influence. A practical implication of the beforementioned result is that communication professionals, do not have to mind the duration that users are exposed to a Facebook post, when trying to promote online deliberation.
Discussion
This study has some limitations. The results of this study may have been affected by the quality of operationalization of the concepts. In particular of the concepts reflective communication management and moderation by the organization. The operationalization of the concepts were based on existing theories, or were modifications of existing measures. Further, the reliability of these concept had been confirmed through inter-coder reliability testing. However, in the case of the concept reflective communication management, there was not a pre-existing measure. Thus, some nuances in the operationalization of the variable may have been overlooked, leading to a less accurate measurement of communication strategies in Facebook posts. To increase the internal validity of this research it is recommended that future research uses another pre-existing measure of communication strategies to examine the
24 relationship between online deliberation and communication strategies, and compare the results.
Further, it is recommended to investigate to what extent different topics could have an influence on the outcome of this study. Studies on online deliberation have examined its role in a political context (Medaglia & Zhu, 2017; Halpern & Gibbs, 2013; Janssen & Kies, 2005; Stromer-Galley & Martinson, 2009) or in a news context (Esau, Friess & Eilders, 2017). The topic of the post in the beforementioned contexts is often more clearly defined than in the communication efforts by an organization on social media. In other words, it could be that the intention of the sender is perceived more clearly in the beforementioned contexts, than when an organization posts something on social media.
The population of this study is Fortune 500 companies, thus the generalizability of the results is currently limited to this population. To increase the external validity of this research it is recommended to investigate to what extent the type organization could have an influence on the outcome of the results. Further, a limitation of this study is that the perspective of the public was not explored in this study. Publics might not want to seek the same level of engagement as organizations do. To investigate how online publics view the organization-publics relationship, it is recommended to include their perspective via either qualitative methods such as interviews, or quantitative methods such as surveys. The addition of this type of data to the results of this study, can provide a more in depth understanding of the
25
References
Albu, O. B., & Etter, M. (2016). Hypertextuality and social media: A study of the constitutive and paradoxical implications of organizational Twitter use. Management
Communication Quarterly, 30(1), 5-31.
Black, E. W., Mezzina, K., & Thompson, L. A. (2016). Anonymous social media– Understanding the content and context of Yik Yak. Computers in Human Behavior, 57, 17-22.
Boeije, H., & 't Hart, H. (2009). In Hox J. (Ed.), Onderzoeksmethoden (8th ed.). Den Haag: Boom Lemma Uitgevers.
Bortree, D. S., & Seltzer, T. (2009). Dialogic strategies and outcomes: An analysis of environmental advocacy groups’ Facebook profiles. Public Relations Review, 35(3), 317-319.
Botan, C. (1997). Ethics in strategic communication campaigns: The case for a new approach to public relations. The Journal of Business Communication (1973), 34(2), 188-202. Carr, C. T., & Hayes, R. A. (2015). Social media: Defining, developing, and divining. Atlantic
Journal of Communication, 23(1), 46-65.
Christensen, L. T., & Cornelissen, J. (2013). Bridging corporate and organizational communication: Review, development and a look to the future.
In Organisationskommunikation und Public Relations (pp. 43-72). Springer VS, Wiesbaden.
Cissna, K. N., & Anderson, R. (1998). Theorizing about dialogic moments: The Buber‐Rogers position and postmodern themes. Communication Theory, 8(1), 63-104.
26 De Swert, K. (2012). Calculating inter-coder reliability in media content analysis using
Krippendorff’s Alpha. Center for Politics and Communication, 1-15.
Dwyer, C. (2007, January). Digital relationships in the" myspace" generation: Results from a qualitative study. In System Sciences, 2007. HICSS 2007. 40th Annual Hawaii
International Conference on (pp. 19-19). IEEE.
Edwards, L. (2015). The role of public relations in deliberative systems. Journal of Communication, 66(1), 60-81.
Esau, K., Friess, D., & Eilders, C. (2017). Design Matters! An Empirical Analysis of Online Deliberation on Different News Platforms. Policy & Internet, 9(3), 321-342.
Field, A. (2009). Discovering statistics using SPSS. Sage publications.
Gamboa, A. M., & Goncalves, H. M. (2014). Customer loyalty through social networks: Lessons from Zara on Facebook. Business Horizons, 57(6), 709-717.
Ganim Barnes, N., & Daubitz, C. (n.d.). Time for Reevaluation? Social Media and the 2016 Inc. 500. Retrieved June 29, 2018, from
https://www.umassd.edu/cmr/socialmediaresearch/2017inc500/
Grönlund, K., & Himmelroos, S. (2009). The challenge of deliberative democracy online–A comparison of face-to-face and virtual experiments in citizen deliberation. Information Polity, 14(3), 187-201.
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 (pp. 28 5‐325). Hillsdale, NJ: LEA
Gudowsky, N., & Bechtold, U. (2013). The role of information in public participation. Journal of Public Deliberation, 9(1).
27 Habermas, J. (1970). Towards a theory of communicative competence. Inquiry, 13(1-4),
360-375.
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(3), 1159-1168.
Hayes, A. F. (2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling [White paper]. Retrieved from http://www.afhayes.com/ public/process2012.pdf
Heath, R. L., Pearce, W. B., Shotter, J., Taylor, J. R., Kersten, A., Zorn, T., ... & Deetz, S. (2006). The processes of dialogue: Participation and legitimation. Management Communication Quarterly, 19(3), 341-375.
Janssen, D., & Kies, R. (2005). Online forums and deliberative democracy. Acta política, 40(3), 317-335.
Ledingham, J. A. (2003). Explicating relationship management as a general theory of public relations. Journal of public relations research, 15(2), 181-198.
Kent, M. L., & Taylor, M. (1998). Building dialogic relationships through the World Wide Web. Public relations review, 24(3), 321-334.
Kent, M. L., & Taylor, M. (2002). Toward a dialogic theory of public relations. Public relations review, 28(1), 21-37.
Kent, M. L., & Theunissen, P. (2016). Discussion, Dialogue, Discourse| Elegy for Mediated Dialogue: Shiva the Destroyer and Reclaiming Our First Principles. International Journal of Communication, 10, 15.
28 Kietzmann, J. H., Hermkens, K., McCarthy, I. P., & Silvestre, B. S. (2011). Social media? Get
serious! Understanding the functional building blocks of social media. Business horizons, 54(3), 241-251.
Kim, D., Chun, H., Kwak, Y., & Nam, Y. (2014). The employment of dialogic principles in website, Facebook, and Twitter platforms of environmental nonprofit
organizations. Social science computer review, 32(5), 590-605.
Kim, D., Kim, J. H., & Nam, Y. (2014). How does industry use social networking sites? An analysis of corporate dialogic uses of Facebook, Twitter, YouTube, and LinkedIn by industry type. Quality & Quantity, 48(5), 2605-2614.
Lane, A. (2014). Toward understanding the (lack of?) significance of dialogue to the practice of public relations. Asia Pacific Public Relations Journal, 15(1), 123-142.
Lane, A. B., & Bartlett, J. (2016). Why dialogic principles don’t make it in practice–And what we can do about it. International Journal of Communication, 10, 4074-4094.
Medaglia, R., & Zhu, D. (2017). Public deliberation on government-managed social media: A study on Weibo users in China. Government Information Quarterly, 34(3), 533-544. Nonnecke, B., Preece, J., Andrews, D., & Voutour, R. (2004). Online lurkers tell why. AMCIS
2004 Proceedings, 321
O’Sullivan, P. B., & Carr, C. T. (2017). Masspersonal communication: A model bridging the mass-interpersonal divide. New Media & Society, 1461444816686104.
Palazzo, G., & Scherer, A. G. (2006). Corporate legitimacy as deliberation: A communicative framework. Journal of Business Ethics, 66(1), 71-88.
29
Pfeffer, J., Zorbach, T., & Carley, K. M. (2014). Understanding online firestorms: Negative word-of-mouth dynamics in social media networks. Journal of Marketing
Communications, 20(1-2), 117-128.
Price, V., Cappella, J. N., & Nir, L. (2002). Does disagreement contribute to more deliberative opinion?. Political communication, 19(1), 95-112.
30 Appendix A
Codebook Introduction
Dialogue is considered to be the most ethical way to communicate on social media between organizations and publics. However, due to its theoretical foundations the
appropriateness of the theory on social media is questionable. Thus, in this study a different perspective on the communication between organizations and publics on social media is proposed: deliberation. The goal of this study is determine the extent to which deliberation could be an appropriate perspective on the communication between organizations and publics on the social media Facebook. This goal is achieved through assessing the influence of the independent variables Reflective Communication Management, Moderation by the
Organization and Time on the dependent variable Degree of Deliberation in Facebook comments.
Sample
Name Facebook followers
1. Walmart 34 million
2. Berkshire Hathaway 59.081 3. AT&T 6.5 million 4. Ford Motor Company 8.7 million
5. Costco 1.6 million
6. Kroger 1.5 million
7. Home Depot 4.5 million 8. The Boeing Company 1.2 million
9. Comcast 85.282
10. Target 23 million
Note. This is the amount of Facebook followers at the time of the sampling on the of sampling 28th of May 2018.
Registration Unit and Unit of Analysis
The unit of analysis is a Facebook comment. The registration unit includes the Facebook post and first 20 comments in response to that post.
31 Categories
A01 Coder ID State your coder ID. 1 Stephanie
2 Wilfred
A02 Organization ID
Indicate the organization that you are coding. Name
1. Walmart
2. Berkshire Hathaway 3. AT&T
4. Ford Motor Company
5. Costco
6. Kroger
7. Home Depot
8. The Boeing Company
9. Comcast
10. Target
A04 Amount of Facebook followers
Indicate the amount of followers per company
Name Facebook followers
1. Walmart 34 million
2. Berkshire Hathaway 59.081 3. AT&T 6.5 million 4. Ford Motor Company 8.7 million
5. Costco 1.6 million
32 7. Home Depot 4.5 million
8. The Boeing Company 1.2 million
9. Comcast 85.282
10. Target 23 million
A02 Date of the post
State the date of the Facebook post as yyyy/mm/dd.
Reflective Communication Management.
The description of the different communication strategies are based on the theory of reflective communication management by Van Ruler and Vercic (2004). The registration unit is the Facebook post.
V01 Indicate the type of communication strategy present in the Facebook post. Make use of the examples given below per strategy to aid your assessment.
1. Persuasion
The communication strategy is considered to be persuasive when the Facebook post contains one of the following:
The use of a celebrity to endorse a product created by the company through mentioning them using the product, service or being present at an event hosted by them. For example, “See Brad Pitt strut his stuff in our new line of casual wear”.
Emotional appeals are the use of emotion to motivate an individual in a preferred direction. Examples are:
o self-esteem, “Feel absolutely gorgeous in our new heels!”.
o social fears, “Don’t be left out of the watercooler chat. Watch Apple-pie now on demand.”
o authority or experts, “Alexandra Steele, dentist of 35 years, recommends this brand of toothpaste!”
Rational appeals make use of factual data to motivate individuals. For example, “80% of customers said yes to the dress!”.
33 2. Relationship
The communication strategy is considered to be relationship when the Facebook post contains voluntary disclosure of information (Bonson & Ratkai, 2013). Examples of this are:
o the descriptions of the organizations involvement and/or support concerning environmental issues, societal issues and diversity in human resources and career.
o disclosure of financial reporting by for example mentioning that they have a new report out
3. Dialogue
The intent of the communication strategy is considered to be dialogue when the Facebook post contains a question to the public. For example, “How do you guys feel about our new approach?”.
4. Information
The intent of the communication strategy is considered to be informational when the
Facebook post is absent of elements indicating of a persuasive, relational or dialogic strategy. 99. Unable to discern
The communication strategy in a Facebook post is considered Unable to Discern when it is not possible to place the strategy in any of the previous categories.
Time
V02 Determine the time span between Facebook postings.
The registration unit is a Facebook post. To determine the time, first subtract the minutes from the current post that you are coding with the previous post in the sample. Then convert the total amount of hours into minutes, by using the following formula: total amount of minutes = (a – b + (24 x amount of days)) x 60. a = current post and b = previous post.
34 Example, post A was placed at 13:45 on 25/05/2018 and post B was placed at 12:01 on
26/05/2018. First subtract the minutes, 1 – 45 = 44 minutes. Then convert the amount of hours into minutes by using the formula (12 – 13 + (24x1)) = 1380 minutes. And lastly, determine the total amount of minutes, 1380 + 44 = 1424 minutes.
Moderation
V03 Determine whether or not moderation with the objective of promoting deliberation is present in the comment section.
The registration unit is a Facebook comment. These instructions apply to the first 20
comments in the comment section. The moderator in the comment section is the organization. Example, you are coding the company H&M. If the comment carries the username H&M, then that comment is to be analysed for signs of moderation. Indicators of moderation with the objective of promoting discussion are the following:
- Moderator addresses complaints and criticism by commenters
- Moderator gives out a warning to a commenter for harmful content being posted and informs them of possibility to be deleted
- Moderator encourages alternate viewpoint related to the topic of the post - Moderator elaborates on topic of post by providing extra information - Moderator asks questions to commenters to promote discussion
Code (0) if no indicators per comment are present. Code (1) if one or more indicators per comment is present.
0 Not present 1 Present Deliberation
Determine the degree of deliberation in the comment section.
The registration unit is an individual Facebook comment. This applies to the first 20 Facebook comments. Choose Not Applicable when the comment contains only a hyperlink, emoji's or pictures.
35 Is the content of the comment related to the topic of the post?
0 Content of the comment is not related to the topic of the post 1 Content of comment is related to the topic of the post
99 Not applicable
D02 Reasoning
Does the comment contain a statement with one or more justifications?
Examples of justifications are the use of reasons, facts or explanations to justify their statements.
0 Comment does not contain a statement with one or more justifications 1 Comment contains a statement with one or more justifications
99 Not applicable
D03 General Engagement
Does the comment address general content in another comment?
General engagement is free of arguments and/or criticisms. Example, comment 1 says, "I like apple sauce". Comment 2 says, "What do you like about applesauce?". Comment 2 is an example of general engagement because it addresses the content in comment 1 and is free of arguments and criticism.
0 Comment does not address another commenter 1 Comment addresses an argument commenter 99 Not applicable
D04 Argumentative Engagement
36 An example of argumentative engagement is the following, “You’re calling for more diversity in the workforce, but wouldn’t it be better if we were all just colour-blind?!”
0 Comment does not address an argument in another comment 1 Comment addresses an argument in another comment 99 Not applicable
D05 Critical Engagement
Does the comment contain criticism of content in another comment?
An example of criticism is the following, “I do not agree with your assessment that H&M is solely to blame for the horrid work conditions in the fashion industry.”
0 Comment does not contain criticism of content in another comment 1 Comment contains criticism of content in another comment
99 Not relevant
D06 Respect
Is the comment free of aggressive and offensive language?
Examples of aggressive and offensive language is the use of curse words and descriptions of inflicting physical harm towards another individual.
0 Aggressive and/or offensive language is absent in comment
1 Aggressive and/or offensive language is present in comment addresses 99 Not relevant
D07 Constructiveness
Does the comment contain constructive elements?
The following is a description of constructiveness in a comment. For example, the topic of the post is about the canoes. Commenters have started to complain about the quality of the
37 canoes, and thus one commenter proposes a solution for their complaints. Example, “You should try to contact customer service and see if they can help you out.”
0 Comment does not contain constructive elements 1 Comment contains constructive elements
38 Appendix B
Tabel 1 Inter-coder Reliability Values
% agreement Cohen’s Kappa KAlpha Reflective Communication Management 100% 1.000 1.000 Moderation 88.9% .727 .7385 Topic Relevance 88.7% .753 .7539 Reasoning 85.5% .690 .6901 General Engagement 88.7% .740 .7409 Argumentative Engagement 96.6% .651 .6509 Critical Engagement 100% - - Respect 98.9% .662 .6629 Constructiveness 96.7% .709 .7109 Table 2
Multicollinearity Test Results of the Linear Regression Analysis (N = 764) Collinearity Statistics Tolerance VIF Time .92 1.09 Moderation by Organization .66 1.52 Relationship strategy .82 1.21 Dialogue strategy .87 1.15 Information strategy .82 1.23
39 Table 3
Hierarchical Multiple Regression Analyses Predicting the Degree of Deliberation (N = 764)
Note. * p <.05. ** p <.01. *** p <.001. Degree of Deliberation in Facebook Comments b* Constant 4.35*** Relationship strategy 19.12 Dialogue strategy 9.20* Information strategy 4.56* Moderation by Organization .55 Relationship strategy * Moderation by Organization 1.00*** Dialogue strategy * Moderation by Organization 1.06 Information strategy * Moderation by Organization .66 Time .0005
Relationship strategy * Time .01
Dialogue strategy *Time .0018
Information strategy * Time .0005 R2
F
.03
