News/Discussion Values and Interactivity in Corporate Social Responsibility Communication via Social Media

Abstract

Social media have great potential to facilitate corporate social responsibility (CSR) dialogue between companies and their stakeholders, but two fundamental issues remain: how to encourage greater participation/dialogue and how to avoid the development of echo chambers, whereby participants merely reinforce their previously held views, potentially increasing the polarization of stakeholders. The problem of participation is grounded in people’s reluctance to comment on social media, and concerns with echo chambers arise when social media comments merely reinforce company statements. This research hypothesized that people’s willingness to comment increases when company replies are perceived to be contingent on past comments and when there is uncertainty, rather than negativity, in the comments. Results supported only the latter. Additionally, the study investigated the valence of comments and responses, exploring whether valenced comments engender potentially opinion-reinforcing echo chambers. Results showed that uncertainty tended to foster more interaction and questions, and that negativity inspired more negative comments.

Keywords

corporate social responsibility social media echo chambers interactivity

Within the contemporary corporate social responsibility (CSR) landscape, dialogue between corporations and their stakeholders is positioned at the very core of strategic and effective CSR communication. Scholars have argued that dialogue is fundamental to how CSR is perceived by stakeholders (Morsing, 2017) and others have gone even further to argue that dialogue is constitutive of CSR (Schoeneborn et al., 2020). Although dialogue can take on many forms, CSR dialogue involves the co-construction of meaning between companies and stakeholders, seeks out the “resolution of conflict” via mutually acceptable solutions, and is “inherently action-oriented” (Golob & Podnar, 2014, pp. 249–250). CSR dialogue is more than a corporate communication tool to improve corporate reputation. It is an ethical practice that enables organizations and their publics to build relationships and develop social capital (Kent & Taylor, 2016). Ethically and managerially, dialogue is a necessity for organizations to engage stakeholders in CSR activities (Bialkova & Te Paske, 2021), and to keep organizations accountable in terms of their human rights records (Stohl & Stohl, 2017).

Among various communication channels, few hold more promise to fulfill the ideal of genuine CSR dialogue between companies and stakeholders than social media (Xu, 2023). The capacity for people to send and receive messages in an open, ongoing fashion makes social media suitable platforms through which companies and stakeholders can engage in the sort of two-way symmetrical communication normatively desirable for CSR dialogue (Capriotti, 2017). That is, stakeholders can voice concerns about CSR issues, gaining an active role in co-creating CSR.

However, CSR dialogue on social media can be unstructured (not enacted with concrete goals; Acuti et al., 2024). Furthermore, the fact that people have considerable liberty to speak their minds on social media, makes it challenging to anticipate how dialogue will pan out. There have been concerns that social media interactions may facilitate the formation of echo chambers—a phenomenon whereby “social media algorithms combine with [people’s] tendencies to interact with like-minded others… create an environment that predominantly exposes users to congenial, opinion-reinforcing content to the exclusion of more diverse, opinion-challenging content” (Kitchens et al., 2020, p. 1619). Put differently, open communication may be inhibited, and self-reinforcing thought silos may form—to the detriment of any genuine dialogue (Buhmann et al., 2021; Kitchens et al., 2020). Thus, people may become increasingly polarized by selectively exposing themselves to information that confirms, rather than disconfirms, their own beliefs (Kitchens et al., 2020; Sunstein, 2017). Toubiana and Zietsma (2017) found that Facebook acted as an emotional echo chamber rather than a dialogic medium, where negative¹ emotions about an organization were reinforced, amplified and led to increased adverse organizational perceptions.

Therefore, it is critical for CSR that social media not become echo chambers, but rather become spaces where people are willing to comment openly on CSR issues (Lew & C. Stohl, 2023). It is a farrago of affirmations, questions, probes, and disagreements that research suggests contributes to an open dialogue and diversity and quality of opinions (Price et al., 2002). However, a challenge remains: Dialogue cannot happen if people are unwilling to participate, and most people are lurkers who do not actively comment (Sun et al., 2014).

Given the limited participation rate of CSR dialogue on social media, little is known about the “black box” of dialogue—what features can (a) promote or sustain CSR dialogue while (b) reducing the formation of echo chambers (Acuti et al., 2024; Illia et al., 2017)? With this in mind, the present research examined two relevant outcomes: willingness to comment and comment valence/type. People’s willingness to comment on a CSR post is a key prerequisite to social media-based CSR dialogue (Lew & C. Stohl, 2023). Furthermore, comment valence/type provides insight into the “black box” of dialogue—with greater variety in comment valence/type (e.g., positive comments, negative comments, questions, etc.), there is a greater diversity of opinions, and therefore echo chambers should be less likely to form.

Several concepts borrowed from other fields were used to predict those two outcomes. The concept of interactivity (from interpersonal communication, referring to a semantically relevant response by a later message to an earlier message within a message chain [Rafaeli, 1988]) was used to motivate the line of questioning as to whether the presence/absence of a company’s reply to comments on its own Instagram page engenders greater willingness to comment among stakeholders. Specifically, the feature being studied is the presence of a company’s reply, which can foster greater perceived contingency (a sense that messages are interconnected, coherent, and meaningful; Sundar et al., 2016), which may then lead to greater willingness to comment, an integral foundation of CSR dialogue.

News/discussion values (from journalism), namely uncertainty and negativity, were studied for their potential to interact with the presence/absence of a company’s reply to jointly influence willingness to comment. Uncertainty refers to ambiguity as to what the facts are, and can be brought about via questions, and negativity refers to disagreeing in a negative (e.g., angry, dismissive, pessimistic) tone (Ziegele et al., 2014). Uncertainty may enhance stakeholders’ willingness to comment, especially if there is a question-and-answer scenario in which stakeholders ask questions and companies reply. But negativity—even if it is desirable for its potential to expand the diversity and improve the quality of opinions (Price et al., 2002)—may dampen willingness to comment if stakeholders suspect the dialogue will degenerate into a mud-slinging session.

In sum, interactivity and news/discussion values were predicted to influence two key outcomes. The first outcome, willingness to comment, is important for its antecedent role in encouraging CSR dialogue. The second outcome, comment valence/type, is important for its implications for diversity in dialogue and by extension, the reduction of echo chambers. Taken together, this research theoretically examines how to promote greater CSR dialogue and how to minimize echo chambers. Practitioners may also benefit from an empirical test of specific features that they can use to engender dialogue.

CSR Dialogue on Social Media

Two-way symmetrical communication has been, for some time, considered the gold standard for CSR communication (Grunig, 2001; Morsing & Schultz, 2006). Its core principles of co-constructing CSR efforts via ongoing dialogue with stakeholders and the lack of management-led agenda promise a more inclusive and legitimate version of CSR (see Palazzo & Scherer, 2006). Social media, with their easy accessibility and capacity for user-generated content (e.g., comments) and reactions (e.g., likes), have often been described as the vehicles that possess the strongest potential to meet this gold standard, surpassing traditional mass communication platforms that offer one-way communication (Capriotti, 2017). Social media affordances also enable dialogic processes which “contribute to stakeholder empowerment and facilitate CSR co-creation” (Xu, 2023, p. 155).

There are several ways to enact two-way symmetrical communication across social media. A common approach is for companies to set social media guidelines compatible with a two-way symmetrical communication orientation. Tracing the evolution of a multinational pharmaceutical corporation’s CSR communication on social media as a case study, Castelló et al. (2016) identified several factors crucial to what they call a networked strategy. For example, top management came up with social media guidelines that explicitly encouraged employees to participate in dialogue with stakeholders, instructed employees to interact with stakeholders on social media as they would to people in everyday professional situations, and urged employees to foster more comments from stakeholders.

Another method is to exploit the capabilities of social media to create suitable spaces for two-way symmetrical communication. Through interviews and content analyses, Illia et al. (2017) identified several such dialogue spaces. For example, microsites allow people to provide opinions or ask questions so that stakeholders can learn about a company’s CSR projects, online forums allow stakeholders to develop a collective agenda on future CSR issues, and voting systems with text-based feedback capabilities allow stakeholders to co-decide with companies what CSR policies should be.

A third approach is to use two-way symmetrical communication to form virtual communities on various social media platforms (Chen et al., 2020). Within these online communities, dialogic processes may support the development of a communal identity, shared values, and trust—factors essential to CSR (Korschun & Du, 2013; Xu, 2023).

Yet, two-way symmetrical communication is uncommon in practice. One content analysis showed that only a third of the world’s most valuable companies had social media guidelines that adopted a CSR co-creation frame (Stohl et al., 2017). The majority of companies in that study treated social media as corporate communication tools and desired to protect themselves from the risks of social media (Stohl et al., 2017). Additionally, many stakeholders seem skeptical of companies’ attempts to engage in dialogue. For example, interviewees in a study mentioned that stakeholders’ comments on social media are perceived as more trustworthy than messages that originate from a company’s official social media accounts (Dunn & Harness, 2019). Separately, a content analysis showed that CSR posts on social media received fewer comments, likes, and shares than non-CSR posts, demonstrating on one hand the informational—rather than dialogical—nature of CSR posts, and on the other hand stakeholders’ unwillingness to engage in dialogue (Cho et al., 2017).

The results due to a lack of two-way symmetrical communication are clear: Companies that moderate a dialogue by dominating the social media thread or by prematurely terminating the dialogue accrue more unfavorable attitudes and lower levels of trust than companies that do not moderate the dialogue (Buhmann et al., 2021).

Interactivity

Amid the somewhat pessimistic outlook vis-à-vis CSR dialogue on social media, one approach—centered on the concept of interactivity—seems to hold some promise. Within the CSR literature, Capriotti et al. (2021) proposed a theoretical framework containing several important dimensions of interactivity pertaining to dialogue. According to their framework, CSR dialogue can be broadly classified into two categories: the predisposition to interaction and effective interaction. The framework indicates that quality dialogue can be achieved if a company establishes the right conditions to foster greater predisposition to interaction among stakeholders (e.g., by maintaining an active presence on social media, by uploading content intended to trigger interactions, or by using technological functions that facilitate interactions such as links and hashtags) or if a company communicates effectively (e.g., responding frequently, reciprocating appropriately). As a whole, this conceptualization of interactivity is fairly macroscopic, with interactivity being theorized in a broad, typological fashion.

To advance the overall theorizing of interactivity in CSR dialogue, a more fine-grained conceptualization of interactivity (alongside its associated constructs such as perceived contingency) can be borrowed from the interpersonal communication literature and be commensurately applied to a CSR context. From this perspective, interactivity² is a message attribute indicating whether a later message in a message chain semantically refers to earlier messages (Rafaeli, 1988). The following example illustrates how Rafaeli’s (1988) argument translates to CSR. If a company makes an original CSR post on Instagram, the post is considered non-interactive because it is the first message in a message chain and therefore it cannot possibly refer to earlier messages. But as more stakeholders comment on the post or reply to existing comments in a semantically relevant way, these subsequent comments/replies are potentially interactive. It should be noted that messages are not automatically considered interactive because they are nested under earlier messages on social media interfaces (e.g., on Instagram and Facebook, comments are nested under posts, and replies are nested under comments). Non-sequitur comments or replies that do not semantically refer to earlier messages are not considered interactive even though they give the appearance of stakeholder engagement. In other words, Rafaeli’s (1988) conceptualization of interactivity strips dialogue down to its uncompromisable essence at a message attribute level, providing one way through which CSR scholars can study the “black box” of dialogue (see also the construct of double interacts in Weick, 1979).

A number of experiments have adopted Rafaeli’s (1988) approach to interactivity in CSR. Lee and Park (2013) found that participants indicated greater trust in companies that replied interactively to stakeholders who commented on their CSR videos than in companies that did not reply. Similarly, Go and Bortree (2017) found that when companies replied interactively to stakeholders’ comments on the company’s Facebook CSR post, they were perceived as having greater openness to dialogic communication, and as a consequence, greater credibility, than companies that did not reply. Lew and C. Stohl (2023) demonstrated that participants perceived greater message contingency—that is, greater semantic interrelatedness among messages (Sundar et al., 2016)—when companies replied interactively to stakeholders’ comments on the company’s Instagram CSR post than when companies replied non-interactively in a semantically tangential manner using boilerplate language. This greater perceived contingency then led participants to be more willing to comment on the CSR post and to have more favorable perceptions of the company’s CSR efforts.

Considering the normative need for diverse stakeholder participation in dialogue for CSR policies to be legitimized (Castelló et al., 2016; Glozer et al., 2019), willingness to comment is an indispensable part of CSR, above and beyond the spreading of positive electronic word-of-mouth in corporate communication contexts. When companies respond to stakeholders’ comments with interactive replies (interactivity as an attribute of individual messages), audience members can develop a sense of perceived contingency (a global perception of interrelatedness between several messages) vis-à-vis the message chain, in that they perceive that the messages in the message chain are interrelated (Sundar et al., 2016). Perceived contingency, on its part, fosters greater willingness to comment because it makes the overall communication more coherent, thus giving audience members the impression that they can contribute in a meaningful way (Lew & C. Stohl, 2023). In sum, greater interactivity leads to greater perceived contingency, which in turn leads to greater willingness to comment on a CSR post.³ Therefore, in the context of a company making a CSR post on its own social media page:

H1: People are more willing to comment on a social media post if a company replies to comments than if a company does not reply.

H2: The relationship between the presence/absence of a company’s reply and willingness to comment (as specified in H1) is mediated by perceived contingency.

But beyond willingness to comment, the valence/type of comments that people leave on CSR posts are equally crucial to understanding CSR dialogue. If a company’s CSR post on its own social media page attracts mostly (or only) positive comments, then there is a risk that the company will end up in a self-affirming echo chamber (see Kitchens et al., 2020; Sunstein, 2017). Unlike dialogue in corporate communication, where positive comments are valued for their utility in building company-stakeholder relationships (see Pang et al., 2018), dialogue in CSR should value negative comments for ethical reasons. CSR dialogue—and indeed the present study—places a premium on negative comments, which should ideally motivate companies to engage their stakeholders in a manner that relinquishes setting the CSR agenda or moderating the dialogue on their own social media pages (see Buhmann et al., 2021). Negative comments may be valuable not only for their role in minimizing the formation of echo chambers but they also add to the diversity of opinions (Price et al., 2002) and provide valid points of contestation regarding a company’s CSR communication (see Ganesh & Zoller, 2012). One study found that when a company’s CSR efforts were met with negative comments by people on social media, the company can mitigate the deleterious effects of those negative comments if they give two-sided responses that reaffirmed their altruistic intentions but also openly acknowledged their self-serving intentions (Rim & Song, 2016). Negative comments, thus, may even incentivize companies to be more open regarding their CSR activities in their CSR communication.

Relatedly, asking questions is an essential part of practical dialogue between the company and its stakeholders (Illia et al., 2015), and may engender greater participation and dialogic dynamics. Asking questions can also elicit uncertainty in social media users, which makes them more likely to contribute to the discussion (Ziegele et al., 2014; see next section for details).

Therefore, several types of comments on a company’s social media post are of particular interest in this context: (a) positively valenced comments, (b) negatively valenced comments, and (c) questions. If a company replies to stakeholders’ comments on its CSR post, are people’s subsequent comments more likely to be positively valenced towards the company, negatively valanced towards the company, neutral, or are they more likely to question the company?

RQ1: Do people comment differently (i.e., positively, neutrally, negatively, or ask questions) on a social media post if a company replies than if a company does not reply?

News and Discussion Values

Although interactivity and perceived contingency are useful concepts with which to analyze dialogue, they pertain to the semantic interrelatedness of messages and are silent on the thematic content of those messages (Sundar et al., 2016). Similarly, Capriotti et al.’s (2021) framework of interactivity in CSR dialogue focuses on how companies can make use of technological features or be oriented towards interacting with stakeholders on social media (among other things) but does not speak to the thematic content within messages. Yet, studying the thematic aspect of messages is crucial to having a more holistic understanding of CSR dialogue. To this end, the notion of news values can be valuable in complementing ideas in the existing CSR literature by Capriotti et al. (2021).

News value theory was originally conceived to explain the choices of newspaper editors as to what stories to cover. News values are factors that influence the newsworthiness of an event, such as its impact, relevance for audiences, its unexpectedness, and its frequency, among other things (Galtung & Ruge, 1965). On social media, for example, news stories that contain strong news values such as human interest (narrative appeals to human emotions), conflict (disagreements and clashes), or emotional arousal are shared more frequently than other types of news stories (Berger & Milkman, 2012). Indeed, editors use news values to guide their selection of newsworthy stories that they publish on social media (Lischka, 2021).

Borrowing from news values theory, a study by Ziegele et al. (2014) recruited interviewees to describe the comments on social media that they typically replied to and to explain why they would reply to certain comments over others. The researchers identified several social media discussion values found in social media comments that influenced people’s willingness to comment, including uncertainty (e.g., questions), negativity (e.g., disagreeing in a negative tone), unexpectedness (e.g., alternative perspectives, re-contextualizing an issue), and humor, among others. Two discussion values, in particular, appear to have unique synergies with CSR dialogue on social media: uncertainty and negativity.

Uncertainty (operationalized as asking questions), on its own, facilitates greater willingness to comment. According to Ziegele et al. (2014), this was because “participants who encountered questions felt committed or invited to respond to them, either because they were challenged or because they wanted to demonstrate their knowledge” (p. 1121). In contrast, negativity dampens willingness to comment, as people “perceived that responding to negative comments might not stimulate meaningful discussions” (Ziegele et al., 2014, p. 1119).

Within CSR communication on social media, the possibility that negativity makes stakeholders less willing to comment on a company’s CSR post is troubling from a normative point of view. The conundrum is that if companies are not aware of negative sentiment towards their CSR posts, the likes, shares, and positive comments they receive may lead to a positively biased echo chamber. Negative comments are also essential for conflictual stakeholder dialogue, where agonistic perspectives are valued for their roles in promoting a plurality of voices on social media (Acuti et al., 2024). Thus, it is important to test the prediction within the context of CSR posts on social media:

H3: People are more willing to comment on a social media post if the post (and its associated comments) contains the discussion value of uncertainty rather than the discussion value of negativity.

Additionally, these discussion values (i.e., uncertainty and negativity) and interactivity can produce joint effects on stakeholders’ willingness to comment on CSR posts. A key feature of uncertainty is questions, which, of course, prompt answers (Ziegele et al., 2014). A company’s answers to questions asked by stakeholders can fulfill Rafaeli’s (1988) definition of interactivity—reference to earlier messages—and they may signal a company’s sincerity in attempting to engage stakeholders in dialogue. Therefore, a company’s answers to stakeholders’ questions (i.e., uncertainty) should intensify stakeholders’ willingness to comment on social media posts as compared to when a company replies to a negative comment (i.e., negativity). In the latter scenario, stakeholders may perceive that the company has done its job in replying, and do not see a need to pile on with more negativity as it does not lead to better CSR dialogue (see Ziegele et al., 2014). Therefore:

H4: There is an interaction between company reply and discussion value, such that people are more willing to comment on a social media post if a company replies than if a company does not reply, and this effect is magnified if the post (and its associated comments) contains the discussion value of uncertainty rather than the discussion value of negativity.

Graphically, H4 is a contributory and divergent positive interaction diagrammed in Holbert and Park (2020), and with reference to that article: the x-axis represents company reply, the y-axis represents willingness to comment, the line with the steeper gradient represents uncertainty, and the line with the gentler gradient represents negativity.

A related research question is whether the news value of uncertainty (i.e., asking questions) will prompt further questions and whether negativity will cultivate further negativity in the comments. As such:

RQ2: Do people comment differently (i.e., positively, neutrally, negatively, or ask questions) on a social media post if the post (and its associated comments) contains the news value of uncertainty rather than the discussion value of negativity?

Method

A 2 (company reply: present/absent) × 2 (discussion value: uncertainty/negativity) between-subjects online experiment was conducted to test the hypothesis and research questions. Participants were communication undergraduates at the University of California, Santa Barbara. Those who volunteered to take part were each given a link that took them to the study’s webpage, which was hosted on Qualtrics. After reading the informed consent, participants were randomly assigned to view one of four possible stimuli (see detailed description of the stimuli in the next section). A timer on the webpage hid the “next” button until 1 minute had passed, ensuring that participants spent an adequate amount of time reading the stimuli. When participants clicked “next,” they were instructed to answer several questions related to the stimuli (see Measures section), after which the experiment was concluded.

Procedure

The stimuli that participants saw consisted of an Instagram post mocked up for this study, supposedly posted by Hershey’s, Mars, Ferrero Rocher, or Nestlé. These companies were selected as they were all on RepTrak’s list of 100 most reputable companies for CSR (Reputation Institute, 2019). Using stimuli attributable to four companies minimized the chances of obtaining company-specific idiosyncratic effects. Analyses were collapsed across the four companies because the company is not a relevant factor/predictor in this study. Instagram was chosen as it had the capacity for commenting and replying to comments (and thus results can reasonably be generalized to other social media apps with the same capacity), plus it was popular among the age group for the sample (see Sample subsection for details; Auxier & Anderson, 2021).

The fictitious Instagram post featured a photo of children in a classroom, and its caption contained text on how the company contributes to the environment, supports local communities, and helps children succeed. Below the post, two fictitious Instagram users (representing stakeholders) each commented on the post once, for a total of two comments. Across all four conditions, the image and caption of the post was kept constant, as were the names of the two people who commented.

For the factor of discussion value, the two comments were operationalized in line with Ziegele et al.’s (2014) conceptualization of uncertainty and negativity. The two comments in the “uncertainty” condition asked questions: “This sounds great but how do we know you’ve helped build thriving communities?” and “Could you be more specific about what you mean by being better stewards of the planet?” The two comments in the “negativity” condition criticized the company: “Nourishing the lives of children! I think you forgot something called childhood diabetes,” and “Self-praise is cheap. Admit it, the bottom line always comes first.” Both fictitious Instagram users made the same type of comment, i.e., either both users asked questions or both users lambasted the company.

For the factor of company reply, participants either saw the company’s reply to the two user comments or saw that the company did not reply. The company’s reply differed according to (and was sensitive to) the type of discussion value in the user comments. Those who saw the company reply therefore saw the original post, two user comments, and one company reply. Participants who did not see the company reply therefore saw only the original post and two user comments. See the Online Supplemental Materials for the stimuli.

Sample

Participants recruited from the University of California, Santa Barbara were awarded course credits for taking part in this study (N = 290). Their ages ranged from 18 to 29 (M = 19.52, SD = 1.47). About 71.4% of participants identified as female and 28.6% identified as male. White participants comprised 36.9% of the sample, Asian participants 30.3%, Hispanic/Latine participants 15.9%, Black participants 2.8%, while the rest indicated mixed ethnicity. Given that users of social media are overwhelmingly youths or young adults rather than older adults, as shown in a Pew Research Center report by Auxier and Anderson (2021), the student sample in the present study is appropriate.

Measures

Willingness to comment was measured using three items by Lew and C. Stohl (2023). Items were scored on a 7-point semantic differential scale: unwilling/willing, unlikely/likely, and reluctant/eager, Cronbach’s α = .80.

Perceived contingency was measured using five items by Lew and C. Stohl (2023). The question stem asked, “Thinking of the Instagram update you just saw, how much do you agree or disagree with the following statements?” The items were: “The company and the Instagram users who commented engaged in two-way communication,” “The interaction between the company and the Instagram users led to jointly produced meaning,” “Messages sent by the Instagram users and by the company seemed interconnected with each other,” “In the interaction between the company and the Instagram users, later messages recounted the relatedness of earlier messages,” and “The company’s interaction with the Instagram users who commented felt like a continuous thread or a loop.” Responses were scored from 1 (strongly disagree) to 7 (strongly agree), Cronbach’s α = .73.

To measure the valence/type of comment, participants were asked, “If you were to comment on the Instagram post you just saw, what would you write?” The two authors coded the open-ended responses for (a) positive comments towards the company, (b) neutral comments, (c) negative comments towards the company, and (d) questions (Cohen’s κ = .88). Both unvalenced responses as well as mixed-valence responses (i.e., comments containing positive and negative aspects in equal measure) were coded as neutral comments. Responses unequivocally seeking more information from the company were coded as questions, on top of responses that were literally questions and ended with a question mark. Disagreements were resolved through discussion.

Examples of positive comments include: “I appreciate the effort to make a difference” and “This is awesome. Love to see big companies giving back to the community.” Neutral comments included unvalenced statements such as: “@[a friend’s username] they look happier than we ever will be [sad face emoji].” The neutral comments category also included two-sided statements (i.e., with both positively and negatively valenced aspects), such as: “Their post made them seem like a good company but gave no reason as to what they do to make themselves good. But they do say what they value as a company.” As for negative comments, examples include: “Although this is a nice message, I do not believe the [company] is truly aligned with these values. This seems like a cheap marketing technique” and “It seems as though they are trying very hard to make themselves look good but actions speak louder than words.” Comments such as “Could you go into specifics about how your donations assist these communities?” and “Is [company] being ethical and sustainable when it comes to gathering ingredients?” were coded as questions.

Before participants saw the stimuli, they answered two questions regarding their preexisting attitudes towards the featured company. Following Lew and Stohl (2023), the first question asked, “Have you heard anything positive about [company]?” and had definitely yes (+2), probably yes (+1), probably not (0), and definitely not (0) as possible responses. The second question asked, “Have you heard anything negative about [company]?” Possible responses were definitely yes (−2), probably yes (−1), probably not (0), and definitely not (0). The values in parentheses were then added together to create a prior attitude score, which served as a control in the analyses.

Again, before participants saw the stimuli, participants answered one question on purchase frequency: “How frequently do you buy products from [company]?” Responses were scored from 1 (very infrequently) to 7 (very frequently). Purchase frequency was also used as a control variable.

Attention Checks

There were two attention check questions. One asked participants to recall the name of the candy company they saw from a list of eight candy companies (plus a ninth possible response: “I’m not sure”). The other asked participants to recall whether the company replied to the two fictitious Instagram users who commented (yes/no/not sure). Indicating “not sure” for either question resulted in a failed attention check. Participants who failed attention checks (n = 77) were excluded from analysis and were not counted in the total sample size (N = 290) reported earlier.

Manipulation Check

The manipulation check was done on a separate sample of N = 51 participants on Prolific. Participants were based in the United States. Their ages ranged from 21 to 50 (M = 32.65, SD = 8.81). 27 participants (52.9%) were female, and 24 participants (47.1%) were male. All participants declared, on Prolific, that they use Instagram. Thus, the stimuli should not be unfamiliar to them.

After signing up, participants were randomly shown one version of the fictitious Instagram post (see Online Supplemental Materials). They were told, “Observe the comments made by [usernames of the two fictitious Instagram users who commented]”. Participants were asked, “How much uncertainty is there in their comments?” Possible answers ranged from 1 (very little uncertainty) to 7 (a lot of uncertainty). Participants were also asked, “How much negativity is there in their comments?” A 7-point scale was again used, with answers ranging from 1 (very little negativity) to 7 (a lot of negativity).

An independent samples t-test showed that participants in the uncertainty discussion value condition (M = 5.54, SD = 1.07) perceived greater uncertainty in the comments than participants in the negativity discussion value condition (M = 3.68, SD = 2.10), p < .001, Cohen’s d = 1.12. Another independent samples t-test showed that participants in the negativity discussion value condition (M = 6.04, SD = 1.37) perceived more negativity in the comments than participants in the uncertainty discussion value condition (M = 3.23, SD = 1.31), p < .001, Cohen’s d = 2.10. As such, the stimuli were perceived as intended.

Results

The analysis for H1, H2, H3, and H4 was performed using Model 5 of Hayes’ (2022) PROCESS, using 10,000 bootstrapped samples. The model specified company reply as the predictor, willingness to comment as the outcome, perceived contingency as the mediator (between company reply and willingness to comment), and discussion value as the moderator (moderating the relationship between company reply and willingness to comment). Prior attitude towards the company and purchase frequency were entered into the model as controls.

H1 predicted that people were more willing to comment on a company’s social media post if the company replied than if the company did not. H2 predicted that the relationship in H1 was mediated by perceived contingency. H3 predicted that people had greater willingness to comment on a company’s social media post if uncertainty, rather than negativity, was present in the discussion. H4 predicted an interaction effect between company reply and discussion value (i.e., the combined effect of H1 and H3). Codes for the independent variables were as follows: company reply absent = −0.5, company reply present = 0.5; negativity discussion value = −0.5, uncertainty discussion value = 0.5.

Initial results showed that participants perceived greater contingency when the company replied than when the company did not reply, b = 0.94, p < .001. Both control variables—prior attitude and purchase frequency—did not significantly predict perceived contingency, p = .487 and p = .588 respectively. Participants were also more willing to comment when the discussion value of uncertainty (operationalized as asking questions) was present than when the discussion value of negativity (operationalized as criticism of the company) was present, b = .50, p = .001. All other predictors, as well as the control variables of prior attitude (p = .998) and purchase frequency (p = .054) did not significantly predict willingness to comment. There was no indirect effect between company reply and willingness to comment, via the mediator of perceived contingency, indirect effect = 0.05, SE_bootstrapped = 0.08, 95 % CI_bootstrapped = [−.110, .215].

Both control variables (i.e., prior attitude and purchase frequency) were consistently non-significant, and random assignment to condition also diminished their necessity. Therefore, applying the principle of parsimony, the control variables were dropped from the model. Model 5 of PROCESS was run again, this time with only the hypothesized variables (and no control variables). The interpretation of the results (i.e., the significance of the tested relationships) did not change.

Final results, diagrammatically represented in Figure 1, showed that the presence/absence of a company’s reply influenced perceived contingency, R² = .19, F(1, 288) = 67.57, p < .001. Participants perceived greater contingency when the company replied (M = 4.63, SD = 0.88) than when the company did not reply (M = 3.70, SD = 1.04), b₁ = 0.93, p < .001.

Figure 1.

Influence of company reply, discussion values, and perceived contingency on willingness to comment. Note. All significant paths (marked with asterisks) p ≤ .001.

Results also showed that discussion value was the only significant predictor of willingness to comment. The overall regression was significant, R² = .04, F(4, 285) = 3.34, p = .011. Neither perceived contingency (b₂ = 0.05, p = .490) nor the direct effect of company reply (b₃ = −0.17, p = .301) predicted willingness to comment. The difference between replying (M = 2.29, SD = 1.23) and not replying (M = 2.42, SD = 1.34) was non-significant. In contrast, discussion value significantly predicted willingness to comment, b₄ = .49, p = .001. Participants were more willing to comment when the discussion value of uncertainty (operationalized as asking questions) was present (M = 2.60, SD = 1.40) than when the discussion value of negativity (operationalized as criticism of the company) was present (M = 2.09, SD = 1.08). There was no significant interaction effect between company reply and discussion value on willingness to comment, b₅ = −0.19, p = .527.

Perceived contingency did not mediate the relationship between company reply and willingness to comment, indirect effect = 0.05, SE_bootstrapped = 0.08, 95 % CI_bootstrapped = [−0.099, 0.218].

As such, H3 was supported, but not H1, H2, or H4. See Figure 1.

Crosstabs were used to analyze RQ1 and RQ2, which asked about the influence of a company’s reply on the valence/type of comment and the influence of discussion values (i.e., uncertainty and negativity) on the valence/type of comment respectively. The results for RQ1 (see Table 1) were not significant, χ²(3, N = 290) = 5.69, p = .128. In contrast, the results for RQ2 (see Table 2) were significant, χ²(3, N = 290) = 29.51, p < .001.

Table 1.

Relationship Between Presence/Absence of a Company Reply and Valence/type of Comment.

	Positive comments	Neutral comments	Negative comments	Questions	Total
Company reply absent	11	14	40	71	136
Company reply present	16	11	63	64	154
Total	27	25	103	135	290

Note. χ²(3, N = 290) = 5.69, p = .128.

Table 2.

Relationship Between Uncertainty/Negativity Discussion Values and Valence/type of Comment.

	Positive comments	Neutral comments	Negative comments	Questions	Total
Disagreement	12	15	70	45	142
Uncertainty	15	10	33	90	148
Total	27	25	103	135	290

Note. χ²(3, N = 290) = 29.51, p < .001.

The pattern of results for RQ2 showed that when existing comments on a company’s CSR post conveyed negativity, participants were most likely to also comment in a negative way, more so than ask questions or comment in a positive or neutral way. Yet, when existing comments on the CSR post conveyed uncertainty, participants were most likely to ask questions of their own, rather than make negative, neutral, or positive comments. It therefore seems like existing comments set the tone for subsequent comments: negative comments may engender further negativity, and questions may lead to even more questions.

Discussion

The present research investigated how interactivity and discussion values could engender greater willingness to comment on CSR posts on social media and what sorts of comments may ensue, in terms of valence and asking questions. It found that the discussion value of uncertainty engendered greater willingness to comment on a company’s social media post about CSR than the discussion value of negativity, supporting H2. It also found that if people were to comment on a company’s social media CSR post, they will most likely build on existing comments, and comment in a similar manner (RQ2). If existing comments criticize the company, people were more willing to follow suit in their comments. If existing comments asked the company questions, people were more willing to also ask questions.

These findings should be interpreted in light of two contextual factors. First, although uncertainty fosters greater willingness to comment than negativity in a relative sense, participants’ overall willingness to comment is low in an absolute sense. The mean self-ratings of willingness to comment across the four experimental conditions ranged from M = 2.09 to M = 2.60 on a 7-point scale. This is unsurprising, considering that only about 1% of people actively create new content on social media, while the overwhelming majority of people are lurkers who read what others have written but do not make any comment of their own (Sun et al., 2014). As such, care should be taken not to interpret the results to mean that asking questions makes people highly willing to comment on CSR social media posts. Rather, a small but significant group seems motivated to comment under the right conditions, such as when there is uncertainty.

Second, the four companies chosen in this study were from the chocolate industry. The data in the present study did not permit cross-industry comparisons, but it may not make sense to generalize the results to industries that are extremely weak in CSR, such as the oil industry. Generalizations should therefore be made only within contextual parameters reasonably similar to those in the present study.

Theoretical Implications

This study showed the importance of including message content elements to existing theorizing on interactivity, so that there is a more holistic understanding of CSR dialogue. One criticism of interactivity research is its fixation on taxonomy (Bucy, 2004). For example, a taxonomy of interactivity in digital media can include different media types such as graphic resources, audiovisual resources, or hypertextual resources (Capriotti et al., 2021), or actions such as clicking, dragging, or zooming (Sundar et al., 2014). But such taxonomies of interactivity alone do not lead to greater understanding of the factors that engender CSR dialogue. Instead, to learn about CSR dialogue, interactivity should be—and has been—conceptualized as a process (Bucy, 2004) that involves elements such as reciprocity or intensity (Capriotti et al., 2021). Yet, what does the process of interactivity entail at a concrete level? This study examined two elements pertaining to interactivity as a process: perceived contingency (Sundar et al., 2016) and news/discussion values (Ziegele et al., 2014). Although only the latter element was statistically significant, the results show that the concept of uncertainty derived from news/discussion values, operationalized as asking questions at a message content level, performs a crucial role that links interactivity to dialogue on two counts. First, asking questions in a comment makes it more likely that future comments will also entail questions, maintaining the thrust-and-parry of dialogue. Second, and more importantly, when people ask questions on a company’s CSR post on social media, the increased uncertainty encourages others to be more willing to comment on the post, thus keeping the dialogue going.

Creating uncertainty by asking questions is also an important countervailing force against online echo chambers, where people and corporations, “convinced of the echo that surrounds them with their own views and preconceptions… lose the inclination to proactively discuss ideas with people or groups of a different opinion” (Spohr, 2017, p. 151). In other words, the uncertainty potentially exposes companies to different perspectives that challenge their assumptions about their own CSR activities, reversing the possible “loss of diversity of opinions and arguments” (Spohr, 2017, p. 151) if companies’ CSR communication go unquestioned.

From a broader perspective, news/discussion values are theoretically untethered to the social media context, and thus uncertainty may be able to elicit CSR dialogue even beyond social media. In Katz and Lazarsfeld’s (1955) two-step flow model of communication, opinion leaders receive information from mass communication channels and convey that information to their followers via interpersonal communication. Is a parallel possible in CSR, where opinion leaders create uncertainty—based on a company’s CSR communication—by asking questions, which foster dialogue among followers? Future research may benefit by approaching uncertainty or other discussion values through such a lens.

The discussion value of negativity presents a paradox. On one hand, negatively responding to a company’s CSR post—disagreement—is crucial for deliberative argumentation in an open dialogue and for a company to gain legitimacy (Palazzo & Scherer, 2006). Yet, as the present study shows, negativity also dampens stakeholders’ willingness to comment on those CSR posts. Normatively, this is a potential source of concern. If stakeholders are only exposed to homogeneously positive information about a company’s CSR efforts, they could end up in echo chambers, ignorant of the CSR faults of the company (Spohr, 2017). And if companies do not receive negative feedback from stakeholders, they too will be in echo chambers of their own, not having heard reasonable criticism of their CSR efforts. Indeed, the lack of comments can limit the information people are exposed to, which may lead to echo chambers (Kitchens et al., 2020).

In the present study, negativity was operationalized as, simultaneously, (a) disagreement and (b) writing in a negative tone. Future research can attempt to theoretically differentiate these two aspects of negativity. Disagreement can benefit dialogue as it compels interactants in a dialogue to provide strong reasons for their own positions (Price et al., 2002). Contrariwise, writing in a negative tone can produce a chilling effect on dialogue if people do not perceive the dialogue as meaningful (Ziegele et al., 2014). However, the greater challenge may be to differentiate disagreement and negative tone in practice, given that they often occur in tandem on social media (Hutchens et al., 2015). If a deliberative environment can ever be created—naturalistically or in controlled settings—for CSR dialogue, where companies and stakeholders focus on producing justifiable solutions based on mutual respect (Gutmann & Thompson, 2004), then the utility of the disagreement aspect of negativity for CSR dialogue can perhaps be tested in a way that is unadulterated by negative tone.

No evidence was found, for H1 and RQ1, regarding the possibility that a company’s replies can influence people’s willingness to comment or the valence/type of the comment they would like to make. This result contrasts with the findings in Lew and C. Stohl (2023), that a company replying directly led to greater perceived contingency, leading to greater willingness to comment. Speculatively, this could be due to the one-sidedness of the company’s response to comments, as the company’s reply was operationalized to only defend itself, rather than concede that the stakeholders who commented have made some valid points.

Alternatively, the non-significant finding may be due to the lack of concrete evidence in the company’s reply in defense of its CSR efforts. In the Lew and C. Stohl (2023) study, the company’s responses to criticism on social media provided an additional link to evidence—in the form of sustainability reports—that the company has walked the talk. Similar evidence or links were not present in the company’s responses in the present set of stimuli. It is possible that people want to see hard evidence for a company’s CSR activities alongside its aspirational CSR communication, without which they became reluctant to comment on the social media post.

As a whole, the non-significant results for H1 and RQ1 suggest that the relationship between interactivity and willingness to comment is conditional—perhaps upon the sidedness of a company’s response to comments or the presence of concrete evidence. This is commensurate with a study by Wise et al. (2006), which also found conditional effects for interactivity: Greater interactivity made people more willing to comment in online discussions, but only if the response rate was slow and not if it was fast. Thus, future research may consider the conditional—rather than direct—effects of interactivity on willingness to comment.

Managerial Implications

For practitioners, interactive content has become a benchmark for successful online CSR communication (Bialkova & Te Paske, 2021). In this domain, a key question concerns how CSR should be communicated to stakeholders, to show that a company “treats its [stakeholders] with sincerity and actually values what they say” (Go & Bortree, 2017, p. 732).

Given the dampening effect that negativity has on dialogue, how should companies respond to negative comments? The present study showed that companies should not be too defensive, and that replying to negative comments—in themselves—did not motivate stakeholders to become more willing to comment. Conversely, creating greater uncertainty—by asking questions—seems to be a possible strategy to engender stakeholders’ greater willingness to comment.

Speculatively, companies can also give two-sided responses to create some uncertainty. Instead of being defensive in the face of criticism, companies can admit—among other things—that some criticism was valid or even that its CSR efforts had self-serving intentions (besides altruistic intentions). In so doing, the company admits uncertainty in its original position—and this could potentially prompt greater willingness to comment. Furthermore, two-sided CSR responses not only make a company appear less self-serving, but they also enhance the credibility of the company’s message (Rim & Song, 2016). In this way, two-sided responses may also reduce echo chambers, minimize polarization, and engender future dialogue. Relatedly, one limitation of the present study was that the fictitious company reply to the two fictitious users were one-sided, in that the company only defended itself. Was this why perceived contingency did not engender greater willingness to comment in the present study (cf. Lew & C. Stohl, 2023)? Future research should therefore consider the sidedness of a company’s reply as a potentially crucial factor in furthering dialogue.

And as artificial intelligence (AI) becomes more embedded in our everyday lives, it is possible that companies begin to use AI to generate social media comments/replies based on some suggestions presented here (e.g., ask questions or give two-sided responses to foster uncertainty, disagree without being negative in tone). It will be interesting to study how AI-mediated communication (AI-MC; Hancock et al., 2020) brings new socio-psychological dynamics to CSR dialogue. Hancock et al. (2020, p. 93) ask, among other questions, whether audiences will be “aware of the fact that AI-MC is involved in communication, and if so, what are the outcomes of such awareness” and “under which design conditions will AI involvement be perceived as augmenting a sender’s agency, and when will it be viewed as usurping it?” The present study does not have the answers, but they may be important considerations for CSR practice.

Conclusion

As a whole, the results of this study suggest new pathways for studying CSR and echo chambers, which are produced when people are exposed only to information that confirms their point of view (Spohr, 2017). If social media were to function as platforms for CSR dialogue and avoid the pitfalls associated with echo chambers (see Capriotti, 2017), CSR communication needs to foster a greater willingness to comment and a stronger desire to ask questions, thereby exposing stakeholders to different perspectives that enable contingent, informed, and open dialogue. Social media responses that are univalenced—bare praising or lambasting of a company—may lead to polarization, where people in their own echo chambers adopt increasingly extreme attitudes (Sunstein, 2017). Whether social media can practically sustain a plurality of views from various stakeholders, or turn into segmented echo chambers with polarized views, may depend on this.

Future research needs to explore a greater diversity of corporate and stakeholder responses to better understand not only how social media posts promote more meaningful dialogue, but also how they engender echo chambers and polarization.

Supplemental Material

Supplemental Material - News/Discussion Values and Interactivity in Corporate Social Responsibility Communication via Social Mediasj-pdf-1-mcq-10.1177_08933189241261671

Supplemental Material for News/Discussion Values and Interactivity in Corporate Social Responsibility Communication via Social Media by Zijian Lew and Cynthia Stohl in Management Communication Quarterly

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Zijian Lew

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Zijian Lew (PhD, University of California, Santa Barbara) is an Assistant Professor at the Wee Kim Wee School of Communication and Information, Nanyang Technological University, Singapore. He studies social influence, self-effects, and meaningfulness in digitally-mediated contexts.

Cynthia Stohl (PhD, Purdue University) is a Distinguished Professor in the Department of Communication at the University of California, Santa Barbara. She is the past director of the Center for Information Technology and Society at UCSB. Her most recent work addresses a diversity of network and collective action organizations in the global context, focusing specifically on the role of new communication technologies in corporate social responsibility and visibility management.

References

Acuti

Glozer

Crane

(2024). What is (and is not) stakeholder dialogue in CSR? A review and research agenda. International Journal of Management Reviews. https://doi.org/10.1111/ijmr.12363

Auxier

Anderson

(2021, April 7). Social media use in 2021. Pew Research Center. https://www.pewresearch.org/internet/2021/04/07/social-media-use-in-2021/

Berger

Milkman

K. L.

(2012). What makes online content viral? Journal of Marketing Research, 49(2), 192–205. https://doi.org/10.1509/jmr.10.0353

Bialkova

Te Paske

(2021). Campaign participation, spreading electronic word of mouth, purchase: how to optimise corporate social responsibility, CSR, effectiveness via social media? European Journal of Management and Business Economics, 30(1), 108–126. https://doi.org/10.1108/EJMBE-08-2020-0244

Bucy

E. P.

(2004). Interactivity in society: Locating an elusive concept. The Information Society, 20(5), 373–383. https://doi.org/10.1080/01972240490508063

Buhmann

Maltseva

Fieseler

Fleck

(2021). Muzzling social media: The adverse effects of moderating stakeholder conversations online. Technology in Society, 64, 101490. https://doi.org/10.1016/j.techsoc.2020.101490

Capriotti

(2017). The world wide web and the social media as tools of CSR communication. In Diehl

Karmasin

Mueller

Terlutter

Weder

(Eds.), Handbook of integrated CSR communication (pp. 193–210). Springer. https://doi.org/10.1007/978-3-319-44700-1_11

Capriotti

Zeler

Camilleri

M. A.

(2021). Corporate communication through social networks: The identification of the key dimensions for dialogic communication. In Camilleri

M. A.

(Ed.), Strategic corporate communication in the digital age (pp. 33–51). Emerald Publishing.

Castelló

Etter

Årup Nielsen

(2016). Strategies of legitimacy through social media: The networked strategy. Journal of Management Studies, 53(3), 402–432. https://doi.org/10.1111/joms.12145

10.

Chen

Y. R. R.

Hung-Baesecke

C. J. F.

Chen

(2020). Moving forward the dialogic theory of public relations: Concepts, methods and applications of organization-public dialogue. Public Relations Review, 46(1), Article 101878. https://doi.org/10.1016/j.pubrev.2019.101878

11.

Cho

Furey

L. D.

Mohr

(2017). Communicating corporate social responsibility on social media: Strategies, stakeholders, and public engagement on corporate Facebook. Business and Professional Communication Quarterly, 80(1), 52–69. https://doi.org/10.1177/2329490616663708

12.

Dunn

Harness

(2019). Whose voice is heard? The influence of user-generated versus company-generated content on consumer scepticism towards CSR. Journal of Marketing Management, 35(9–10), 886–915. https://doi.org/10.1080/0267257X.2019.1605401

13.

Galtung

Ruge

M. H.

(1965). The structure of foreign news: The presentation of the Congo, Cuba and Cyprus crises in four Norwegian newspapers. Journal of Peace Research, 2(1), 64–90. https://doi.org/10.1177/002234336500200104

14.

Ganesh

Zoller

H. M.

(2012). Dialogue, activism, and democratic social change. Communication Theory, 22(1), 66–91. https://doi.org/10.1111/j.1468-2885.2011.01396.x

15.

Glozer

Caruana

Hibbert

S. A.

(2019). The never-ending story: Discursive legitimation in social media dialogue. Organization Studies, 40(5), 625–650. https://doi.org/10.1177/0170840617751006

16.

Bortree

D. S.

(2017). What and how to communicate CSR? The role of CSR fit, modality interactivity, and message interactivity on social networking sites. Journal of Promotion Management, 23(5), 727–747. https://doi.org/10.1080/10496491.2017.1297983

17.

Golob

Podnar

(2014). Critical points of CSR-related stakeholder dialogue in practice. Business Ethics: A European Review, 23(3), 248–257. https://doi.org/10.1111/beer.12049

18.

Grunig

J. E.

(2001). Two-way symmetrical public relations: Past, present, and future. In Heath

R. L.

(Ed.), Handbook of public relations (pp. 11–30). Sage. https://doi.org/10.4135/9781452220727.n1

19.

Gutmann

Thompson

D. F.

(2004). Why deliberative democracy? Princeton University Press.

20.

Hancock

J. T.

Naaman

Levy

(2020). AI-mediated communication: Definition, research agenda, and ethical considerations. Journal of Computer-Mediated Communication, 25(1), 89–100. https://doi.org/10.1093/jcmc/zmz022

21.

Hayes

A. F.

(2022). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (3rd ed.). Guilford Press.

22.

Holbert

R. L.

Park

(2020). Conceptualizing, organizing, and positing moderation in communication research. Communication Theory, 30(3), 227–246. https://doi.org/10.1093/ct/qtz006

23.

Hutchens

M. J.

Cicchirillo

V. J.

Hmielowski

J. D.

(2015). How could you think that?!?!: Understanding intentions to engage in political flaming. New Media & Society, 17(8), 1201–1219. https://doi.org/10.1177/1461444814522947

24.

Illia

Romenti

Rodríguez-Cánovas

Murtarelli

Carroll

C. E.

(2017). Exploring corporations’ dialogue about CSR in the digital era. Journal of Business Ethics, 146, 39–58. https://doi.org/10.1007/s10551-015-2924-6

25.

Illia

Zyglidopoulos

Romenti

(2015). Creating effective dialogue about corporate social responsibility. MIT Sloan Management Review, 57(1), 20–22.

26.

Katz

Lazarsfeld

(1955). Personal influence: The part played by people in the flow of mass communications. The Free Press.

27.

Kent

M. L.

Taylor

(2016). From Homo Economicus to Homo dialogicus: Rethinking social media use in CSR communication. Public Relations Review, 42(1), 60–67. https://doi.org/10.1016/j.pubrev.2015.11.003

28.

Kitchens

Johnson

S. L.

Gray

(2020). Understanding echo chambers and filter bubbles: The impact of social media on diversification and partisan shifts in news consumption. MIS Quarterly, 44(4), 1619–1649. https://doi.org/10.25300/MISQ/2020/16371

29.

Korschun

(2013). How virtual corporate social responsibility dialogs generate value: A framework and propositions. Journal of Business Research, 66(9), 1494–1504. https://doi.org/10.1016/j.jbusres.2012.09.011

30.

Lee

Park

(2013). Testing the impact of message interactivity on relationship management and organizational reputation. Journal of Public Relations Research, 25(2), 188–206. https://doi.org/10.1080/1062726X.2013.739103

31.

Lew

Stohl

(2023). What makes people willing to comment on social media posts? The roles of interactivity and perceived contingency in online corporate social responsibility communication. Communication Monographs, 90(1), 1–24. https://doi.org/10.1080/03637751.2022.2032230

32.

Lischka

J. A.

(2021). Logics in social media news making: How social media editors marry the Facebook logic with journalistic standards. Journalism, 22(2), 430–447. https://doi.org/10.1177/1464884918788472

33.

Morsing

(2017). CSR communication: What is it? Why is it important? In Rasche

Morsing

Moon

(Eds.), Corporate social responsibility: Strategy, communication, governance (pp. 281–306). Cambridge University Press.

34.

Morsing

Schultz

(2006). Corporate social responsibility communication: Stakeholder information, response and involvement strategies. Business Ethics: A European Review, 15(4), 323–338. https://doi.org/10.1111/j.1467-8608.2006.00460.x

35.

Palazzo

Scherer

A. G.

(2006). Corporate legitimacy as deliberation: A communicative framework. Journal of Business Ethics, 66(1), 71–88. https://doi.org/10.1007/s10551-006-9044-2

36.

Pang

Shin

Lew

Walther

J. B.

(2018). Building relationships through dialogic communication: Organizations, stakeholders, and computer-mediated communication. Journal of Marketing Communications, 24(1), 68–82. https://doi.org/10.1080/13527266.2016.1269019

37.

Price

Cappella

J. N.

Nir

(2002). Does disagreement contribute to more deliberative opinion? Political Communication, 19(1), 95–112. https://doi.org/10.1080/105846002317246506

38.

Rafaeli

(1988). Interactivity: From new media to communication. In Hawkins

R. P.

Wiemann

J. M.

Pingree

(Eds.), Advancing communication science: Merging mass and interpersonal processes (pp. 110–134). Sage.

39.

Reputation Institute (2019). 2019 Global RepTrak. https://www.reptrak.com/globalreptrak/.

40.

Rim

Song

(2016). How negative becomes less negative: Understanding the effects of comment valence and response sidedness in social media. Journal of Communication, 66(3), 475–495. https://doi.org/10.1111/jcom.12205

41.

Schoeneborn

Morsing

Crane

(2020). Formative perspectives on the relation between CSR communication and CSR practices: Pathways for walking, talking, and t(w)alking. Business & Society, 59(1), 5–33. https://doi.org/10.1177/0007650319845091

42.

Spohr

(2017). Fake news and ideological polarization: Filter bubbles and selective exposure on social media. Business Information Review, 34(3), 150–160. https://doi.org/10.1177/0266382117722446

43.

Stohl

Etter

Banghart

Woo

(2017). Social media policies: Implications for contemporary notions of corporate social responsibility. Journal of Business Ethics, 142, 413–436. https://doi.org/10.1007/s10551-015-2743-9

44.

Stohl

(2017). Human rights and corporate social responsibility. In Brysk

Stohl

(Eds.), Expanding human rights: 21st century norms and governance (pp. 115–138). Edward Elgar.

45.

Sun

Rau

P. P. L.

(2014). Understanding lurkers in online communities: A literature review. Computers in Human Behavior, 38, 110–117. https://doi.org/10.1016/j.chb.2014.05.022

46.

Sundar

S. S.

Bellur

Jia

Kim

H. S.

(2016). Theoretical importance of contingency in human-computer interaction: Effects of message interactivity on user engagement. Communication Research, 43(5), 595–625. https://doi.org/10.1177/0093650214534962

47.

Sundar

S. S.

Bellur

Jia

(2014). User experience of on-screen interaction techniques: An experimental investigation of clicking, sliding, zooming, hovering, dragging, and flipping. Human-Computer Interaction, 29(2), 109–152. https://doi.org/10.1080/07370024.2013.789347

48.

Sunstein

(2017). #Republic: Divided democracy in the age of social media. Princeton University Press. https://doi.org/10.1515/9781400890521

49.

Toubiana

Zietsma

(2017). The message is on the wall? Emotions, social media and the dynamics of institutional complexity. Academy of Management Journal, 60(3), 922–953. https://doi.org/10.5465/amj.2014.0208

50.

Weick

K. E.

(1979). The social psychology of organizing (2nd ed.). Addison-Wesley.

51.

Wise

Hamman

Thorson

(2006). Moderation, response rate, and message interactivity: Features of online communities and their effects on intent to participate. Journal of Computer-Mediated Communication, 12(1), 24–41. https://doi.org/10.1111/j.1083-6101.2006.00313.x

52.

(2023). CSR communication and social media. In O’Connor

(Ed.), The Routledge handbook of corporate social responsibility communication (pp. 152–164). Routledge.

53.

Ziegele

Breiner

Quiring

(2014). What creates interactivity in online news discussions? An exploratory analysis of discussion factors in user comments on news items. Journal of Communication, 64(6), 1111–1138. https://doi.org/10.1111/jcom.12123

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.62 MB