How can CSR help organizations to gain a license
to operate?
Master thesis
An experimental study about different communication tools and strategies for
organizations in order to gain credibility, engagement and legitimacy.
Graduate School of Communication, University of Amsterdam
Master Corporate Communication (MSc)
Author:
Fleur Vis (10544747)
Supervisor:
P.G.A. van der Rijt
2
Abstract
Being profitable as an organization is not enough these days. Stakeholders expect
organizations to be socially responsible and to care for society and the environment. Activities to care for society and the environment can be expressed in Corporate Social Responsibility (CSR) initiatives. Developing these initiatives may not be the hardest part, but effectively communicating about them is. In the end, one of the most important things that matters for organizations is survival. And legitimacy is vital for survival. Moreover, a credible
organization which people feel engaged with can result in more legitimacy. But what can organizations do in their CSR communication that is effective and results in credibility, engagement and legitimacy? This study attempted to answer this question by researching different communication tools and communication strategies in order to gain credibility, engagement and legitimacy. An online experiment among 161 participants was conducted to provide answers. The results indicate that deliberative communication tools such as Facebook lead to more legitimacy compared to instrumental communication tools such as the corporate website. When there is a possibility to leave a comment under a CSR message from an organization, it does not affect credibility, engagement or legitimacy much. However, it gets interesting when an organization responds to comments from their stakeholders. A response from an organization leads to higher credibility, engagement and legitimacy than no response. When organizations respond they can best engage into dialogue with their stakeholders. This strategy results in more credibility, and through credibility ultimately legitimacy can be achieved.
3
Introduction
Much research has been done into the benefits that organizations can achieve through their Corporate Social Responsibility (CSR) initiatives (Du et al., 2007; Fombrun et al., 2000; Lichtenstein et al., 2004; Sen and Bhattacharya, 2001, Sen et al., 2006). Through CSR, organizations can achieve positive attitudes among their stakeholders, but also the purchase intention and image of the organization can be positively influenced (Du et al., 2010). Not only CSR, but public relations in general have been studied from a functional perspective. A lot of studies focus on business benefits and how to maximize profits. But some scholars (Van Ruler & Vercic, 2005; Christensen & Cornelissen, 2011) suggest we need to look beyond this functional approach and the focus must shift from organizational goals such as maximizing profits to gaining legitimacy. An organization may only exist if it makes sense from the perspective of the public and adds some value to society, in other words: an organisation should have a license to operate (Van Ruler & Vercic, 2005).
When we apply this kind of research to CSR we acknowledge that economic benefits that can be derived from CSR initiatives are important for organizations, but so is legitimacy. Stakeholders increasingly expect organizations to fulfil their social responsibilities with regard to society (Colleoni, 2013). Organizations are aware of these responsibilities and communicate about their CSR activities to stakeholders, but there is still uncertainty about how to communicate about CSR most effectively.Because most stakeholders receive CSR messages with scepticism (Du et al., 2010). Therefore, a challenge of CSR communication is to minimize possible scepticism of stakeholders. According to Seele and Lock (2015) this can be done by credible communication. And through credible CSR communication, legitimacy can be achieved (Lock & Seele, 2015). Furthermore, legitimacy can also be achieved by stakeholders who feel engaged with the organization. Stakeholders who can co-construct and engage into dialogue with the organization, feel more engaged to the organization (Zhang &
4
Lin, 2015). A higher form of engagement will lead to more favourable stakeholder evaluations and thus to higher legitimacy (Zimmerman & Zeitz, 2002). As a result, this research will focus on credibility, engagement and legitimacy.
CSR activities are mostly presented in reports and corporate websites and not on social media (Pollach, 2015). Also, most studies have focused on CSR reports instead of using social media for CSR communication. But social media could be an effective tool in order to gain legitimacy.Seele and Lock (2015) classify reports and corporate websites as
instrumental tools. Instrumental tools have a strategic goal, namely to inform stakeholders instead of actively involving them. Social media applications can be classified as deliberative tools. Deliberative tools are open to participation and thus every stakeholder can participate in a discussion or can respond (Seele & Lock, 2015). This distinction in instrumental and
deliberative tools is understudied and it is important to know which of these tools will lead to better outcomes.
Apart from tools, organizations can also use different strategies in order to respond (or not respond) to their stakeholders. In this research four strategies will be examined.
Organizations can use 1) an information strategy (Morsing & Schultz, 2006) in which there is one-way communication, 2) a possibility to leave a comment (symbolic two-way
communication) in which stakeholders can have a say but no genuine decisional input (Lewis, 2011), 3) a response strategy (Morsing & Schultz, 2006) which is characterized by two-way asymmetrical communication, 4) dialogue (involvement strategy) which is characterized by two-way symmetrical communication (Morsing & Schultz, 2006).
This paper aims to fulfil a research gap on how to execute CSR communication most effectively. This research responds to the question from Seele & Lock (2015) to empirically test their distinction between instrumental and deliberative tools. Because of the interactivity of deliberative tools, will a CSR message distributed with this type of medium lead to more
5
credibility and engagement? And does this result in higher legitimacy? Moreover, Eberle et al. (2013) made a suggestion for future research to investigate the effects of stakeholder and corporate responses on social media. They state that when a CSR message is perceived as interactive, feelings of identification with the organization will be stronger (Eberle et al., 2013). But if this also counts for the reaction strategy of the organization remains unknown in the current literature about CSR. Which of the four strategies will lead to more credibility, engagement and ultimately legitimacy?
This research aims to answer these questions and contribute to the existing literature about CSR and legitimacy. Also, this research aims to contribute to management practice. In this moment of time stakeholders expect organizations to participate in being corporate socially responsible (Colleoni, 2013), but their license to operate is put on the spot and stories of scandals have hit the news (Unilever CEO Paul Polman's sustainability effort covers up the fact that his business practices are unsustainable, 2017). Take Unilever for example, who serves on the boards of organizations fighting climate change, poverty and social ills. However, they are accused of sexual harassment and poor housing conditions of their employees in Kenya. Stories like this can result in scepticism, because the actions of
organizations do not correspond with their words. A challenge for organizations is to reduce possible scepticism and find out how credibility can be achieved. Organizations put effort, time and money into their CSR policy, so it is important to determine which tools and strategies lead to positive outcomes. This research can help organizations in the process of choosing the right channel and the right strategy in order to communicate about their CSR policy. This research will make clear if companies should pay attention to instrumental or deliberative tools when communicating about CSR and which communication strategy is best to use in order to increase credibility, engagement and legitimacy.
6
Consequently, the following research question emerges: What kind of impact does the
communication tool and the communication strategy in a CSR message have on credibility and engagement, and does this lead to higher legitimacy?
This paper first gives an overview on the existing literature about CSR communication, after that the experimental design and sample will be explained. Later on the results of this research are presented, followed by a conclusion and discussion in which limitations and suggestions for future research are being discussed.
Theoretical Framework
Studies about CSR communication mostly focus on companies’ attempts to maximize business benefits such as a good reputation or higher purchase intention. These studies are consistent with the functional perspective in corporate communication, a strongly sender-biased view in which communication is used as a means to gain a specific goal (Christensen & Cornelissen, 2011). In this functional view, communication is something that managers do to accomplish something else (Conrad & Haynes, 2001, in Hallahan et al., 2007). But this view has gained a lot of criticism over the past years. Critics say it is too sender oriented and ignores the role of receivers in the communication process. According to those critics, we must see communication as a process in which the organization does not have control over all communication flows but where collective sense making between the organization and its stakeholders is important (Hallahan et al., 2007; Christensen & Cornelissen, 2011). Van Ruler & Vercic (2005) call this the social perspective. This perspective is about constructing
meaning and legitimacy. By means of dialogue and interaction between stakeholders and the organization, an organization gains a license to operate. According to these scholars
7
communication does no longer exist in tactical activities such as CSR to gain a specific goal, but communication is an ongoing process between the organization and its stakeholders and taps into issues of identity and legitimacy.
Suchman (1995, p. 574) defined legitimacy as “a generalized perception or assumption that the actions of an entity are desirable, or appropriate within some socially constructed system of norms, values, beliefs, and definitions”. Legitimacy can be seen as an evaluation of the organization and its activities. Stakeholders increasingly expect organizations to fulfil their responsibilities with regard to society, and no longer only justify their economic actions, but also their social and ecological actions (Colleoni, 2013). Companies perceived as
desirable or appropriate are more likely to be persistent and have a goodwill buffer during adverse times (Suchman, 1995; Colleoni, 2013). So legitimacy is vital for survival.
Research has proved organizations use CSR in order to obtain or maintain legitimacy (Arvidsson, 2010; Eberle et al., 2013). According to Seele and Lock (2015) CSR
communication is one of the most important means to convey legitimacy to the outside and inside stakeholders. However, communicating about CSR initiatives often leads to one of the biggest challenges CSR communication is facing: stakeholder scepticism (Elving, 2013). Stakeholders often see CSR communication as strategic in nature and thus not credible (Elving, 2013; Illia et al., 2013). They believe a company is acting in self-interest rather than for the benefit of society. In that case, the expectations of stakeholders about CSR are not met by the company. According to Elving’s experiment (2013), a bad reputation from an
organization automatically leads to scepticism among stakeholders compared to a good reputation. Participants in the good reputation condition identified themselves more strongly
with the company leading them to consider the company’s activities as trustworthy and true,
8
stakeholders can be minimized when an organization and its activities are positively evaluated, where stakeholders identify strongly with the organization (engagement) and corporate activities are seen as credible (Elving, 2013).
Credibility, engagement and legitimacy
Credibility can be defined as “a characteristic which is attributed to individuals, institutions or their communicative products (verbal or written texts, audiovisual displays) by somebody (recipient) in relation to something (events, facts, etc.)” (Bentele & Nothaft, 2011, p. 216). Coombs (1992) suggests that association with a credible person can build legitimacy. Credibility is a basis for legitimacy (Coombs, 1992), a credible organization will gain more legitimacy than a non credible organization. Moreover, a CSR source which is not credible will challenge a firm’s license to operate (Seele & Lock, 2015). Credibility is important for organizations in order to obtain or maintain legitimacy. Therefore this research assumes the following hypothesis:
H1: When people find the organization credible, it leads to higher legitimacy.
Engagement is defined in terms of a combination of cognitive aspects (e.g. being interested in the activities of a company), emotional aspects (having positive feelings about the activities of a company) and/or behavioural aspects (e.g. participation in the activities of a company) (Dijkmans, Kerkhof & Beukeboom, 2015). A higher form of engagement with an
organization will lead to more favourable stakeholder evaluations (Dijkmans et al., 2015), and those evaluations determine legitimacy. Zimmerman and Zeitz (2002) assume that legitimacy exists in the eye of the beholder and so legitimacy is seen as an evaluative process in which stakeholders judge an organization. Before assigning legitimacy to an organization,
9
the more favourable evaluations stakeholders will have which results in higher legitimacy. Therefore this research assumes the following hypothesis:
H2:When people have feelings of engagement with the organization, it leads to higher legitimacy.
Now the question arises how organizations, and thus their communication professionals, can use communication to gain credibility, engagement and legitimacy. Different communication tools and communication strategies are being discussed below and examined in this research to create clarity among communication professionals which tools and strategies lead to better outcomes.
Instrumental versus deliberative tools
CSR initiatives are mostly presented in CSR reports and corporate websites (Pollach, 2015). This is ineffective for legitimacy building, because only people with a prior interest in CSR will read about a companies’ CSR activities. Pollach (2015) suggests researchers to look beyond these instrumental tools. Seele and Lock (2015) made a distinction between
instrumental and deliberative tools. Instrumental tools have a strategic goal and these tools are used to ‘get the message across’ in an informative or persuasive sense (Seele & Lock, 2015). It is more about informing stakeholders rather than actively engaging them. Furthermore, instrumental tools such as CSR reports are being criticised because of their lack of credibility (Lock & Seele, 2016). Organizations choose what to report and in this way they can cherry
pick elements about their CSR policy which are favourable to report. Because of this,
stakeholders will view the organization as less or not credible.
Besides instrumental tools, Seele and Lock (2015) acknowledge deliberative tools. These tools have dialogical functions in order to facilitate two-way communication. Examples
10
are social media applications such as Facebook and blogs. “New or social media are in this view regarded as instruments to gain legitimacy for corporations, as they would further improve the dialogue and engagement toward deliberative democracy and the access to conversations” (Schultz et al., 2013, p. 684). Deliberative tools are better in achieving credibility than instrumental tools, because deliberative tools are based on deliberation and understanding between multiple societal actors in a discourse (Seele & Lock, 2015).
Interaction is central in deliberative tools and transparency is high. Not only the organization chooses what to write about their CSR policy on social media, but stakeholders can comment and give their opinions as well. When CSR messages are perceived as being interactive, credibility increases (Van Halderen, van Riel & Brown, 2011; Eberle et al., 2013).
Interactivity creates an impression of sincerity, because CSR messages are open to criticism which makes them and the organization more believable. “CSR information published through interactive online channels is more credible than purely company controlled messages” (Eberle et al., 2013, p. 733).
An organization’s credibility will not only increase, but stakeholder feelings of identification with the organization increase when CSR messages are perceived as interactive (Eberle et al., 2013). Stakeholders who can engage in a dialogue via social media feel more connected to the organization (Zhang & Lin, 2015). In contrast, one-way dissemination of information from an organization to stakeholders is often viewed as a weak form of
engagement (Burchell & Cook, 2006). Morsing and Schultz (2006, p. 335) state: “Instead of imposing corporate norms for CSR on stakeholders, the invitation to participate and co-construct the CSR message increases the likelihood that these stakeholders will identify positively with the company”. This research therefore assumes the following hypothesis:
11 H3: Exposure to deliberative tools leads to higher credibility towards the organization (H3a) and to more engagement with the organization (H3b) compared to exposure to instrumental tools.
Communication tools that have the ability to possess an open discourse, participation,
transparency and accountability will lead to legitimacy. Deliberative tools do have the ability to possess an open discourse, participation, transparency and accountability (Seele & Lock, 2015). These tools have dialogical functions in order to facilitate two-way communication, such as social media. If we take Facebook for example, everybody can participate and so there can be an open discourse between the organization and other Facebook users. Furthermore, transparency can be reached as well because the organization can not control what is written about them by other users. They can not ‘cherry pick’ elements which are favourable to report. Only accountability can not always be guaranteed. Users can have access with false profiles and can not be verified at all times. Nonetheless, deliberative tools can (almost) meet these demands compared to instrumental tools. Instrumental tools mostly lack participation and open discourse, and thus transparency and accountability. According to Seele and Lock (2015) legitimacy can be achieved when there is open discourse, participation, transparency and accountability. Therefore this research assumes the following hypothesis:
H3: Exposure to deliberative tools leads to higher legitimacy compared to exposure to instrumental tools (H3c), and this relationship is mediated by credibility and engagement (H3d).
Response strategies
The channel or tool an organization communicates with is of great importance as illustrated above, but so is the response strategy of the organization. Based on Grunig & Hunt’s (1984)
12
models of public relations, Morsing and Schultz (2006) present three types of stakeholder response strategies in relation to CSR communication: the stakeholder information strategy, the stakeholder response strategy and the stakeholder involvement strategy. The stakeholder information strategy is a form of one-way communication from the organization to its
stakeholders, with the purpose of disseminating information (Morsing & Schultz, 2006). The organization informs stakeholders about its good CSR initiatives to ensure positive
stakeholder support, but there is no possibility for stakeholders to respond. The stakeholder response strategy can be seen as two-way asymmetrical
communication, where communication flows to and from the public (Morsing & Schultz, 2006). The organization does not change itself, but it tries to change stakeholders’ attitudes and/or behaviours. The intention of the organization is convincing its stakeholders of its attractiveness. This response strategy is a form of two-way communication, but if the organization responds to their stakeholders it is still sender oriented (Morsing & Schultz, 2006). Therefore they may respond to comments from stakeholders, but they comment on them with a defensive strategy in which they will try to persuade their stakeholders of its good intentions.
The stakeholder involvement strategy can be seen as two-way symmetrical communication, which is characterized by dialogue and negotiation (Morsing & Schultz, 2006). Organizations engage into dialogue with their stakeholders in order to explore mutually beneficial actions, assuming that both parties involved are willing to change. The key factor in this type of communication is understanding between the organization and stakeholders instead of disseminating information or persuasion. Kent and Taylor (2016) introduced a new view on online CSR communication via social media: Homo Dialogicus. They state that companies need to move away from the Homo Economicus, where CSR only exists for economical benefits, and shift the attention to dialogue and relationship building.
13
This corresponds with the social perspective from Van Ruler & Vercic (2005). By means of dialogue and negotiation between stakeholders and the organization, an organization gains a license to operate (Van Ruler & Vercic, 2005).
Research on internal communication also focuses on three different kind of communication processes in organizational change. Lewis (2011) speaks about information dissemination, the symbolic approach and the resource approach to participation. Disseminating information involves the spreading of facts, details, rationale and clarifying misconstrued information. According to Lewis (2011) information dissemination is problematic because it ignores the important circulation of and creation of information among other stakeholders. We can more or less compare this communication process with the stakeholder information strategy.
In the symbolic approach to participation Lewis (2011) speaks about “managerial-driven forms of participation” in which stakeholders can have a say so that loyalty and commitment will increase and resistance will decrease, instead of seeking for genuine decisional input. Stakeholders are provided input channels in order to make them feel more involved. This communication process differs on one point from the stakeholder response strategy. The latter has the sole intention of convincing its stakeholders of the organization’s attractiveness and therefore the organization may respond to comments from stakeholders with a defensive strategy.
But just giving stakeholders a say or a seat is not enough, genuine decisional input is something an organization must strive for. The resource approach to participation is where stakeholders participate in a dialogue in which they exchange their evaluations and opinions about organizational change (Lewis, 2011). Moreover, stakeholders have a voice in how to implement changes. This communication process can be more or less compared with the involvement strategy, because both are characterized by negotiation and dialogue.
14
Possibility to comment
The symbolic approach to participation (Lewis, 2011), from now on called “the possibility to comment”, is communication in which stakeholders can have a say but no genuine decisional input. Loyalty and commitment with the organization increases when there is a possibility to leave a comment (Lewis, 2011). The possibility to comment as a stakeholder creates an impression of sincerity, because CSR messages which are open to criticism makes them and the organization more believable (Eberle et al., 2013). When there is no possibility to leave a comment, as in the information strategy, messages are not open to criticism and organizations choose what to report. In this way they can ‘cherry pick’ elements about their CSR policy which are favourable to report (Lock & Seele, 2016). When stakeholders can leave a comment under a message or post from an organization, not only will the organization be more credible but it can also stimulate identification (Thorson & Rodgers, 2006). Thorson and Rodgers (2006) empirically confirmed with their experiment among 69 students that the ability to interact fosters the creation of an intimate relationship between organization and stakeholders. Unfortunately, no direct relationship between the possibility to comment and legitimacy can be found in the literature, according to the researcher’s knowledge. But as shown in this research earlier, credibility and feelings of engagement lead to legitimacy (Zimmerman & Zeitz, 2002; Seele & Lock, 2015; Dijkmans et al., 2015). This leads to the following hypotheses:
H4: The possibility to leave a comment leads to higher credibility towards the organization (H4a) and to more engagement with the organization (H4b) compared to no possibility to leave a comment.
H4: The possibility to leave a comment will lead to higher legitimacy compared to no possibility to leave a comment (H4c), and this relationship is mediated by credibility and engagement (H4d).
15
Responsiveness
Organizations need to do more instead of just spreading messages online and giving
stakeholders the possibility to leave a comment. Organizations should respond to comments from stakeholders and it is important that stakeholders must not be left ignored (Veil, Buehner & Palenchar, 2011; Huibers & Verhoeven, 2014; Einwiller & Steilen, 2015). Interaction and responsiveness is essential in addressing misinformation and establishing the organization as a credible source, according to a literature review from Veil et al. (2011). Responding to posts demonstrates the organization cares what stakeholders think and can be trusted to address their concerns. Leaving a comment as an organization shows concern and gives the impression an organization is not shying away from discussion (Veil et al., 2011). Also Einwiller and Steilen (2015) empirically confirm that customer satisfaction requires responsiveness by the organization. Furthermore, an experiment among 437 participants studying the relationship between a webcare response versus no webcare response on engagement showed that reacting to comments and questions on social media increases engagement (Schamari & Schaefers, 2015). The success of an organization is becoming increasingly dependent on how the organization uses social media. No webcare response to stakeholders or withdrawal from a dialogue on social media is the most damaging strategy (Huibers & Verhoeven, 2014). Responding to comments can result in an organization which is perceived as a credible source, engages stakeholders and ultimately obtain or maintain legitimacy. This leads to the following hypotheses:
H5: A reply from an organization to their stakeholders leads to higher credibility towards the organization (H5a) and to more engagement with the organization (H5b) than no reply at all. H5: A reply from an organization to their stakeholders results in higher legitimacy than no reply at all (H5c), and this relationship is mediated by credibility and engagement (H5d).
16
Stakeholder response strategy versus dialogue (stakeholder involvement strategy)
The more an organization shows that they take the opinions of their stakeholders into
consideration, the more positive stakeholders will be towards its CSR commitment (Lauritsen & Perks, 2015). An organization can show they take the opinions of their stakeholders into consideration by responding to their comments. The stakeholder response strategy (Morsing & Schultz, 2006), is a form of two-way asymmetrical communication. The organization does not change itself, but it tries to change stakeholders’ attitudes and/or behaviours. The current literature suggests that CSR communication should be characterized by dialogue, or also known as the stakeholder involvement strategy (Morsing & Schultz, 2006), because this is a more ethical and moral way of communicating with stakeholders in comparison to the
stakeholder information strategy, the possibility to comment (symbolic communication) or the stakeholder response strategy (Morsing & Schultz, 2006; Johansen & Nielsen, 2011; Scherer et al., 2013). Not only the organization, but stakeholders as well have certain rights and responsibilities (Johansen & Nielsen, 2011). Dialogue ensures reaching consensus between sender and receiver, through which legitimacy can be achieved (Scherer et al., 2013).
Lauritsen and Perks (2015) found that in order to bolster credibility as an organization, the organization needs to engage their consumers via media that encourage and facilitate interaction and involvement. Through dialogue stakeholders can co-construct meaning, so that mutual understanding can be reached (Van Ruler & Vercic, 2005). Moreover, entering into dialogue makes stakeholders feel they are a valuable part of the organization and they will be more likely to function as ambassadors of CSR activities (Grunig & Hunt, 1984; Morsing & Schultz, 2006). Dialogue should be incorporated into the communication strategy from organizations in order to be credible, foster stakeholder relations and obtain legitimacy (Morsing & Schultz, 2006; Johansen & Nielsen, 2011; Scherer et al., 2013). This leads to the following hypotheses:
17 H6: Dialogue leads to higher credibility towards the organization (H6a) and to more
engagement with the organization (H6b) compared to the stakeholder response strategy. H6: Dialogue leads to higher legitimacy compared to the stakeholder response strategy (H6c), and this relationship is mediated by credibility and engagement (H6d).
Figure 1 below shows a conceptual model with all the expected relationships between the independent variables, mediators and dependent variables.
Method
Design
To investigate the effects of different CSR communication approaches of an organization on credibility, engagement and legitimacy, an online experiment with a two factor between-subjects design was conducted. One factor was the communication tool (Instrumental versus Deliberative) and the second factor the communication response strategy from the
organization (Information strategy, Possibility to comment, Response strategy and Dialogue). This second factor is a nested factor, because this factor does not occur on all the levels of the first factor. Figure 2 shows the factorial design.
18
The corporate website was used as an instrumental tool and Facebook as a deliberative tool in this research. Other social media applications could have been used as well, but Facebook is the most popular and leading social media site with more than 1.23 billion people using the social network every day, and 1.15 billion logging in from mobile devices each day (Popper & Erlick, 2017).
It is important to note this online experiment consists out of five conditions instead of eight, as shown in Figure 2 above. Because it is in Facebook’s nature to comment on content, it would be odd to create a condition in which there is no possibility to leave a comment on Facebook. Furthermore, both asymmetric and symmetric strategies are investigated in a deliberative tool only. This is due to the nature of the medium and literature, which states it is important to examine these strategies in deliberative tools because of their interactivity (Eberle et al., 2013).
Sample
This online experiment was conducted between 23rd of April and 9th of May 2018. A
convenience sample was used, which means that participants were personally contacted by the researcher (Endacott & Botti, 2005). Participants were asked to fill in the experiment by messages on WhatsApp, e-mail and Facebook. A snowball sample was used because the
19
acquaintances of the researcher were asked to share the link with family, friends and/or colleagues on Facebook, WhatsApp or by e-mail.
In total 161 participants completed this online experiment. Participants were randomly assigned to one of the five conditions. 65% of the participants were female and the age of all participants differed between 18 and 69 years old. The average age was 27 years old. 36% had a university degree and 22% had a HBO degree. Most of the participants use social media several times per day (76%) and they are somewhat interested in CSR activities (39%). 57% says it is unlikely that they will leave a comment under a message or post from an
organization.
In order to see whether the random assignment of participants was successful, a one-way ANOVA with the five different conditions as independent variables and age and education as dependent variables was conducted. Results show that the randomization was successful for age, (F(4, 156) = 0.76, p = .550, η2 = 0.02) and education, (F(4, 156) = 1.05,
p = .385, η2 = 0.03).Moreover, a Chi-square test showed that the randomization for gender was successful, X²(4, N = 161) = 2.14, p = .709.
Stimulus material
This research consisted out of five different experimental groups. In the first group (N = 32), participants were exposed to a CSR message on the corporate website in which the
organization used the information strategy, so there was no possibility to react under the message. In the second group (N = 32), participants were exposed to a CSR message on the corporate website in which there was a possibility to leave a comment under the message, but the organization did not reply to the comments. The third group (N = 31) was more or less the same as the second group, but participants were exposed to the message on Facebook. In the fourth group (N = 31), participants were exposed to a CSR message on Facebook in which the
20
organization used the stakeholder response strategy. The organization did respond to the comments under the post, but they responded with a defensive strategy. In the last group (N = 35), participants were exposed to a CSR message on Facebook in which the organization engaged into dialogue with their stakeholders (see Appendix I for the stimulus material).
Pilot test
Prior to this research a pilot test was conducted to check if the manipulations worked as intended by the researcher. 27 participants completed the pilot test and all manipulation checks were significant and thus successful, which means the participants got the manipulations as intended. The results of the pilot test can be found in Appendix II.
Procedure
First the participants were introduced to the study and after the participants read and accepted the informed consent, they were randomly assigned to one of the five conditions. Before they were exposed to the condition, they were asked to read the message and the introduction about the organization carefully. The organization used in this experiment, ICT-fix, was made up by the researcher, because prior attitudes towards the organization could bias the
credibility, engagement and legitimacy of the organization. The ICT sector was chosen because this is kind of a neutral sector were most participants would not have a very strong opinion about. After exposure the participants had to answer questions about credibility, engagement and legitimacy. After that, the participants were asked to answer questions about gender, age, education, social media use, their interests in CSR and the likelihood of leaving an online comment to messages from organizations. The online experiment ended with a manipulation check and finally the participants were debriefed about the goal and fictive nature of this experiment (see Appendix III for the online experiment).
21
Measurements
Independent variables
All five conditions were recoded into four dummy variables. Communication tool was coded into 0 = instrumental tool (Website) and 1 = deliberative tool (Facebook). Communication strategy was recoded into three different dummy variables. The first one is the possibility to leave a comment under the message or post from the organization (= 1) or not (= 0). The second one is whether the organization responds to the comments (= 1) or not (= 0). The last one is whether the organization engages into dialogue with stakeholders (= 1) or not (= 0).
Dependent variable: Credibility
Credibility can be defined as “a characteristic which is attributed to individuals, institutions or their communicative products (verbal or written texts, audiovisual displays) by somebody (recipient) in relation to something (events, facts, etc.)” (Bentele & Nothaft, 2011, p. 216). To measure the credibility of the organization, four items from the Credibility-scale of Meyer (in West, 1994) were used. Using a 5-point Likert scale (1 = strongly agree, 5 = strongly
disagree), the participants could indicate to which extent they (dis)agreed with statements as: “This organization can be trusted” and “This organization is accurate” (see Appendix IV, Table 1.1 for all items and factor loadings). A principal component analysis (PCA) showed that the four items form a single uni-dimensional scale, as only one component had an eigenvalue above 1, namely 2.77. There is a clear point of inflexion after this component in the scree plot. The total explained variance was 69.1% and this variable showed a high level of internal consistency (Cronbach’s α = .84). Finally, the four items that measure the variable credibility were combined into one scale. The values of this scale ranged from 1.75 up to 5. This variable was recoded so that the higher the score on this scale, the more credible the organization is perceived by the participants (M = 3.50; SD = .83).
22 Dependent variable: Engagement
Engagement can be defined in terms of a combination of cognitive aspects (e.g. being interested in the activities of a company), emotional aspects (having positive feelings about the activities of a company) and/or behavioural aspects (e.g. participation in the activities of a company) (Dijkmans, Kerkhof & Beukeboom, 2015). This variable was measured with five items, with two items from Dijkstra, Buijtels en Van Raaij (2005) and three items from Zaichkowsky (1985). Using a 5-point Likert bipolar adjective scale, participants could
indicate if they think the organization was “interesting/not interesting” or “relevant/irrelevant” for example (see Appendix IV, Table 1.2 for all items and factor loadings). A principal
component analysis (PCA) showed that the five items form a single uni-dimensional scale, as only one component had an eigenvalue above 1, namely3.19. There is a clear point of
inflexion after this component in the scree plot. The total explained variance was 63.8% and this variable showed a high level of internal consistency (Cronbach’s α = .86). Finally, the five items that measure the variable engagement were combined into one scale. The values of this scale ranged from 1 up to 5. This variable was recoded so that the higher the score on this scale, the more engagement with the organization is perceived by the participants (M = 3.26;
SD = .82).
Dependent variable: Legitimacy
Suchman (1995, p. 574) defined legitimacy as “a generalized perception or assumption that the actions of an entity are desirable, or appropriate within some socially constructed system of norms, values, beliefs, and definitions”. To measure this variable five items from Suchman (1995) were used. Using a 5-point Likert scale (1 = strongly agree, 5 = strongly disagree), the participants could indicate to which extent they (dis)agreed with statements as: “The activities of ICT-fixare desirable when I compare these with my own standards and values” (see
23
Appendix IV, Table 1.3 for all items and factor loadings). A principal component analysis (PCA) showed that the five items form a single uni-dimensional scale, as only one component had an eigenvalue above 1, namely3.28. There is a clear point of inflexion after this
component in the scree plot. The total explained variance was 65.5% and this variable showed a high level of internal consistency (Cronbach’s α = .86). Finally, the five items that measure the variable legitimacy were combined into one scale. The values of this scale ranged from 1.20 up to 5. This variable was recoded so that the higher the score on this scale, the more legitimacy with the organization is perceived by the participants (M = 3.78; SD = .74).
Control variables
Three demographic variables were measured in this study: gender, age and level of education. There were also other variables added to this online experiment to see if they had a possible influence on the dependent variables. Social media use was measured with the following question: “How often do you use social media (eg. Facebook, Instagram, Twitter, Snapchat)?” and exists out of the following answer options: Never, Rarely, A few times per week, Once a day, Several times a day. Participants who are familiar with social media could maybe allocate more credibility, engagement or legitimacy to deliberative tools. Interest in CSR was measured by one question, “Are you interested in activities from organizations to positively contribute to the community and environment (e.g. using solar panels, driving electric cars, donating to charities)?, ranging from 1 = very much to 5 = not at all. Participants with higher interests in CSR will be more engaged with the organization for example, because the
organization addresses their interests. The likelihood of leaving a comment was measured by one question as well: “Is it likely that you will leave a comment under an organization’s message or post on a website or Facebook page?”, ranging from 1 = extremely likely to 5 = extremely unlikely. Participants who are likely to leave comments, are maybe more aware if
24
organizations respond and how they respond. Both interest in CSR and likelihood to comment were recoded into 1 = not at all/extremely unlikely to 5 = very much/extremely likely.
Manipulation checks
To check if the manipulations in this study worked successfully different questions were asked to see if the participants got the manipulations as intended. The differences between the communication tools and communication strategies were clear. All manipulation checks were significant, which means the participants got the manipulation as intended and thus the manipulation was successful. The results of this check can be found in Appendix V.
Results
Before testing all hypotheses, it was important to explore if any control variables correlate with the dependent variables. Almost all control variables turned out to have a (very) weak correlation with credibility, engagement and legitimacy. Only the control variable Interest in
CSR turned out to have a moderate correlation with the dependent variables (see Appendix
VI, Table 2). Because of this moderate correlation, interest in CSR has been added to all analyses as a covariate. Interest in CSR will only be reported in the text in the first analyses, otherwise this results section would become too extensive. Certainly, the results of interest in CSR as a covariate in every regression analysis can be found in Appendix VI.
Hypothesis 1 examines if credibility leads to higher legitimacy and hypothesis 2 examines if engagement leads to higher legitimacy. To test both hypotheses, a multiple regression analysis was conducted with credibility and engagement as independent variables, interest in CSR as a covariate and legitimacy as dependent variable. The analysis showed the model can be used to predict the differences in legitimacy, F(3, 157) = 48.80, p = < .001. 55.1% of the variation in
25
legitimacy could be explained by credibility, engagement and interest in CSR (R² = .551). Credibility had a significant, moderately strong effect on legitimacy, b* = .39, t = 5.25, p < .001, 95% CI [.22, .48]. This means when credibility increases with one point on the scale, legitimacy increases with .35 (see Appendix VI, Table 3). Thus H1 was supported: credibility leads to higher legitimacy. It also appeared that engagement had a significant, moderately strong effect on legitimacy, b* = .31, t = 3.96, p < .001, 95% CI [.14, .41]. This means when engagement increases with one point on the scale, legitimacy increases with .28 (see
Appendix VI, Table 3). Thus H2 was also supported: engagement leads to higher legitimacy. Interest in CSR had a significant, weak effect on legitimacy, b* = .21, t = 3.60, p < .001, 95% CI [.06, .21]. This means when interest in CSR increases with one point on the scale,
legitimacy increases with .14 (see Appendix VI, Table 3).
To test all other hypotheses, a mediation analysis from Hayes (2012) was used by running a regression analysis in Process. Model 4 was used with 1.000 bootstrap samples to estimate the bias corrected bootstrap confidence intervals. All conditions are transformed into dummy variables, as explained above (see page 21).
Communication tool
H3a, the direct effect of communication tool on credibility, was analyzed. The regression model can be used to predict the differences in credibility based on the communication tool, F (2, 158) = 8.17, p < .001. The communication tool predicts for 9.4% the differences in
credibility (R2 = .094). The direct effect from communication tool (Website or Facebook) on credibility is not significant, b = .20, t = 1.53, p = .127, 95% CI [-.06; .46]. This means participants in the Facebook condition do not significantly score higher on credibility than participants in the Website condition. Thus H3a needs to be rejected: Facebook does not
26
significantly leads to more credibility. The effect from interest in CSR on credibility is significant, b = .19, t = 3.43, p < .001, 95% CI [.08; .31]. This means when interest in CSR increases with one point on the scale, credibility increases with .19.
Testing H3b, the direct effect of communication tool on engagement, showed the regression model can be used to predict the differences in engagement based on the
communication tool, F (2, 158) = 14.86, p < .001. The communication tool predicts for 15.8% the differences in engagement (R2 = .158). The direct effect from communication tool
(Website or Facebook) on engagement is not significant, b = .14, t = 1.15, p = .252, 95% CI [-.10; .39]. This means participants in the Facebook condition do not significantly score higher on engagement than participants in the Website condition. Thus H3b needs to be rejected: Facebook does not significantly leads to more engagement. The effect from interest in CSR on engagement is significant, b = .27, t = 5.06, p < .001, 95% CI [.17; .38]. This means when interest in CSR increases with one point on the scale, engagement increases with .27.
Testing H3c, the direct effect of communication tool on legitimacy, showed the regression model can be used to predict the differences in legitimacy based on the
communication tool, F (2, 158) = 22.54, p < .001. The communication tool predicts for 22.2% the differences in legitimacy (R2 = .222). The direct effect from communication tool (Website or Facebook) on legitimacy is significant, b = .26, t = 2.45, p = .015, 95% CI [.05; .48]. This means participants in the Facebook condition significantly score .26 higher on legitimacy than participants in the Website condition. Thus H3c was supported: Facebook significantly leads to more legitimacy. The effect from interest in CSR on credibility is significant, b = .27, t = 5.75, p < .001, 95% CI [.05; .48]. This means when interest in CSR increases with one point on the scale, credibility increases with .27.
Finally, H3d was tested to see if credibility or engagement mediated the relationship between communication tool and legitimacy. There was no significant indirect effect of
27
communication tool on legitimacy via credibility, b = .10, SE = .07, 95% bca CI [-.02; .24]. Credibility is no mediator of the relationship between communication tool and legitimacy. In addition, there was no significant indirect effect of communication tool on legitimacy via engagement, b = .07, SE = .06, 95% bca CI [-.04; .20]. Engagement is no mediator of the relationship between communication tool and legitimacy. Thus H3d needs to be rejected.
The above discussed results can be found in a conceptual model in Figure 3 below and in Appendix VI, Table 4 and 5.
Communication strategy I: Possibility to leave a comment
The other independent variable, communication strategy, exists out of three independent dummy variables (see page 21). The first independent variable from communication strategy
28
is possibility to leave a comment. H4a, the direct effect of possibility to leave a comment on credibility, has been analyzed. The analysis showed the regression model can be used to predict the differences in credibility based on the possibility to comment, F (2, 158) = 6.89, p < .001. The possibility to comment predicts for 8.0% the differences in credibility (R2 = .080). The direct effect from possibility to comment on credibility is not significant, b = .00, t = -.01, p = .992, 95% CI [-.32, .31]. Thus H4a needs to be rejected: the possibility to comment does not lead to more credibility.
Testing H4b, the direct effect of possibility to leave a comment on engagement, showed the regression model can be used to predict the differences in engagement based on the possibility to comment, F (2, 158) = 14.24, p < .001. The possibility to comment predicts for 15.3% the differences in engagement (R2 = .153). The direct effect from possibility to comment on engagement is not significant, b = -.08, t = -.51, p = .607, 95% CI [-.38, .22]. This means that participants in the condition with a possibility to leave a comment do not significantly score lower on engagement compared to the condition where there was no possibility to leave a comment. Thus H4b needs to be rejected: the possibility to comment does not lead to more engagement.
Testing H4c, the direct effect of possibility to comment on legitimacy, showed the regression model can be used to predict the differences in legitimacy based on the possibility to comment, F (2, 158) = 19.60, p < .001. The possibility to comment predicts for 19.9% the differences in legitimacy (R2 = .199). The direct effect from possibility to comment on legitimacy is not significant, b = .15, t = 1.13, p = .262, 95% CI [-.11; .41]. This means participants in the condition with a possibility to leave a comment do not significantly score higher on legitimacy than participants in the condition with no possibility to leave a comment. Thus H4c needs to be rejected: the possibility to leave a comment does not lead to higher legitimacy.
29
Finally, H4d was tested to see if credibility or engagement mediated the relationship between possibility to comment and legitimacy. There was no significant indirect effect of possibility to comment on legitimacy via credibility, b = -.00, SE = .07, 95% bca CI [-.16; .14].
Credibility is no mediator of the relationship between possibility to comment and legitimacy. In addition, there was no significant indirect effect of possibility to comment on legitimacy via engagement, b = -.04, SE = .08, 95% bca CI [-.20; .13]. Engagement is no mediator of the relationship between possibility to comment and legitimacy. Thus H4d needs to be rejected.
The above discussed results can be found in a conceptual model in Figure 4 below and in Appendix VI, Table 6 and 7.
Communication strategy II: Responsiveness
The second independent variable from communication strategy is responsiveness. Testing H5a, the direct effect of responsiveness on credibility, showed the regression model can be
30
used to predict the differences in credibility based on the responsiveness, F (2, 158) = 10.46, p < .001. The responsiveness predicts for 11.7% the differences in credibility (R2 = .117). The direct effect from responsiveness on credibility is significant, b = .32, t = 2.56, p = .011, 95% CI [.07; .57]. This means that participants in the condition where the organization responds significantly score .32 higher on credibility compared to participants in the condition with no response. Thus H5a was supported: responsiveness significantly leads to more credibility.
Testing H5b, the direct effect of responsiveness on engagement, showed the regression model can be used to predict the differences in engagement based on the responsiveness, F (2, 158) = 16.53, p < .001. The responsiveness predicts for 17.3% the differences in engagement (R2 = .173). The direct effect from responsiveness on engagement is significant, b = .25, t = 2.04, p = .043, 95% CI [.01; .49]. This means that participants in the condition where the organization responds significantly scored .25 higher on engagement compared to participants in the condition with no response. Thus H5b was supported: responsiveness significantly leads to more engagement.
Testing H5c, the direct effect of responsiveness on legitimacy, showed the regression model can be used to predict the differences in legitimacy based on responsiveness, F (2, 158) = 21.42, p < .001. The responsiveness predicts for 21.3% the differences in legitimacy (R2 = .213). The direct effect from responsiveness on legitimacy is significant, b = .22, t = 2.05, p = .042, 95% CI [.01; .43]. This means participants in the condition where the organization responds significantly score .22 higher on legitimacy compared to participants in the condition with no response. Thus H5c was supported: responsiveness significantly leads to more legitimacy.
Finally, H5d was tested to see if credibility or engagement mediated the relationship between responsiveness and legitimacy. There was a significant indirect effect of
31
Credibility is a mediator of the relationship between responsiveness and legitimacy.In addition, there was a significant indirect effect of responsiveness on legitimacy via engagement, b = .13, SE = .06, 95% bca CI [.01; .25]. Engagement is a mediator of the relationship between responsiveness and legitimacy. Thus H5d was supported.
The above discussed results can be found in a conceptual model in Figure 5 below and in Appendix VI, Table 8 and 9.
Communication strategy III: Dialogue
The third independent variable from communication strategy is dialogue. Testing H6a, the direct effect of dialogue on credibility, showed the regression model can be used to predict the differences in credibility based on dialogue, F (2, 158) = 13.47, p < .001. Dialogue predicts for 13.8% the differences in credibility (R2 = .138). The direct effect from dialogue on
32
credibility is significant, b = .48, t = 3.26, p = .001, 95% CI [.19; .77]. This means that participants in the condition where the organization engages into dialogue significantly score .48 higher on credibility compared to participants in the condition with no dialogue. Thus H6a was supported: dialogue significantly leads to more credibility.
Testing H6b, the direct effect of dialogue on engagement, showed the regression model can be used to predict the differences in engagement based on dialogue, F (2, 158) = 15.84, p < .001. Dialogue predicts for 16.7% the differences in engagement (R2 = .167). The direct effect from dialogue on engagement is not significant, b = .25, t = 1.73, p = .086, 95% CI [-.04; .54]. This means that participants in the condition where the organization engages into dialogue do not significantly score higher on engagement compared to participants in the condition with no dialogue. Thus H6b needs to be rejected: dialogue does not significantly leads to more engagement.
Testing H6c, the direct effect of dialogue on legitimacy, showed the regression model can be used to predict the differences in legitimacy based on dialogue, F (2, 158) = 20.12, p < .001. Dialogue predicts for 20.3% the differences in legitimacy (R2 = .203). The direct effect from dialogue on legitimacy is not significant, b = .19, t = 1.45, p = .149, 95% CI [-.07; .44]. This means that participants in the condition where the organization engages into dialogue do not significantly score higher on legitimacy compared to participants in the condition with no dialogue. Thus H6c needs to be rejected: dialogue does not significantly lead to more
legitimacy.
Finally, H5d was tested to see if credibility or engagement mediated the relationship between dialogue and legitimacy. There was a significant indirect effect of dialogue on
legitimacy via credibility, b = .26, SE = .08, 95% bca CI [.10; .43]. Credibility is a mediator of the relationship between dialogue and legitimacy.However, there was no significant indirect effect of dialogue on legitimacy via engagement, b = .13, SE = .08, 95% bca CI [-.02; .27].
33
Engagement is no mediator of the relationship between dialogue and legitimacy. Thus H6d can be partly supported: credibility is a mediator of the relationship between dialogue and legitimacy, but engagement is not.
The above discussed results can be found in a conceptual model in Figure 6 below and in Appendix VI, Table 10 and 11.
Conclusion & discussion
Central in this research is the concept Corporate Social Responsibility (CSR). The goal of this research was to examine the effects of different communication tools and strategies on
credibility and engagement, and in the end legitimacy. An online experiment with five different screenshots from a fictive organization was conducted to answer the following research question: What kind of impact does the communication tool and the communication
34 legitimacy? With this question the researcher aimed to contribute to the existing literature
about CSR and to the knowledge of organizations on how their efforts in being socially responsible can lead to the desired outcomes.
The results showed that credibility and engagement lead to legitimacy, as expected by various scholars (Coombs, 1992; Zimmerman & Zeitz, 2002; Colleoni, 2013; Scherer et al., 2013; Dijkmans et al., 2015; Seele & Lock, 2015). This finding is important, because organizations who want to strive for legitimacy need to be credible and need to engage stakeholders with their organization. The question from Seele and Lock (2015) to empirically test their distinction between instrumental and deliberative tools have been carried out in this research. Quite surprisingly, deliberative tools did not lead to more credibility or engagement. These results are contradictory to previous research because scholars indicated that
deliberative tools with their interactivity and invitation to participate will lead to more credibility and engagement in contrast to instrumental tools (Morsing & Schultz, 2006; Van Halderen, van Riel & Brown, 2011; Eberle et al., 2013; Seele & Lock, 2015; Zhang & Lin, 2015). An explanation for this result could be related to the stimulus material. Although the pilot test and all manipulation checks were successful, one of the two conditions from the instrumental tool was designed more or less as an deliberative tool (see Appendix I, Condition 2). It did not have the exact same format as Facebook, but it did have the possibility to
comment and thus could be interpreted as an deliberative tool. Moreover, Seele and Lock (2015) also assume it is best to use a mix of instrumental and deliberative tools. However, as expected by Seele and Lock (2015), this study found out that deliberative tools lead to higher legitimacy compared to instrumental tools, but this relationship is not mediated by credibility or engagement.
The possibility to leave a comment did not lead to more credibility, engagement or legitimacy as expected by Thorson and Rodgers (2006) and Eberle et al. (2013). Lauritsen and
35
Perks (2015) discovered that while participants encourage interactive CSR communication, they expect proof that their engagement or participation is acted upon and hence organizations should incorporate customer suggestions into their CSR initiatives. This could be an
explanation. In four out of five conditions in this research, there was a possibility to leave a comment. But in only one condition the organization proved stakeholders their engagement or participation was acted upon.
In line with previous research (Veil, Buehner & Palenchar, 2011; Huibers & Verhoeven, 2014; Einwiller & Steilen, 2015) the results show that a response from an
organization to its stakeholders leads to more credibility, engagement and legitimacy than no response from the organization. Moreover, the relationship between responsiveness and legitimacy is mediated by credibility and engagement. The research gap pointed out by Eberle et al. (2013) has now been fulfilled. Their suggestion to investigate CSR messages with stakeholder and corporate responses on social media has been followed up and so this study showed corporate responses are very essential.
In addition, different scholars confirmed that dialogue leads to higher credibility, engagement and legitimacy (Morsing & Schultz, 2006; Johansen & Nielsen, 2011; Scherer et al., 2013; Lauritsen & Perks, 2015). This was partly the case in this research. Dialogue leads to higher credibility and the relationship between dialogue and legitimacy is mediated by credibility. But this research did not found a significant relationship with engagement. Even though this result is not significant at the .05 level, it is important to note that this result is significant at the .10 level and thus, in other words, is marginally significant. We must acknowledge there is something going on between dialogue and engagement. Furthermore, the results show an interesting finding of the control variable which was examined in this research: interest in CSR. The more interest in CSR participants claim to have, the higher credibility, engagement and legitimacy.
36
Finally, the research question can be answered. Deliberative tools do not lead to more credibility and engagement than instrumental tools, but they do lead to more legitimacy. The possibility to leave a comment without a response from the organization (symbolic two-way communication), does not have an effect on credibility, engagement and legitimacy. However, responsiveness does lead to higher credibility, engagement and legitimacy. This finding is noteworthy and makes clear organizations should respond to comments from their
stakeholders. When an organization does not respond to comments, it seems there is no significant difference in credibility, engagement and legitimacy between conditions where stakeholders can participate by leaving a comment or when they can not. Corporate responses are important and when different response strategies are compared, engaging into dialogue (involvement strategy) seems to be more effective than the response strategy. Dialogue fosters credibility and through credibility, ultimately legitimacy will be achieved.
Limitations & future research
A few limitations have to be mentioned when interpreting the results. The sample in this study was quite small (N = 161) to generate significant generalizable results. Every condition just had about 30 participants, due to practical restrictions. Moreover, a convenience sample was used (Endacott & Botti, 2005) and thus participants were recruited by the network of the researcher. This resulted in a sample mostly consisting out of females with a HBO or university degree, causing low external validity. The sample in this study does not represent the general population. Thus, future research with less time and financial restrictions should examine a more diverse sample. Also, the language of this online experiment was in English while the mother tongue of almost all participants is Dutch. A few participants contacted the researcher afterwards to make clear there English was not that good and to verify if they had understood everything correctly. Future research should also send out the online experiment
37
in Dutch, to make sure that only Dutch speaking participants do not misinterpret the study or experience difficulties which can lead to lower internal validity.
To avoid any bias from previous experiences, a fictive organization was used. But the question arises if participants who have seen one screenshot from a fictive organization can answer all questions properly. Credibility, engagement and legitimacy can be difficult to measure after one short exposure to a fictive organization. Future research should also focus on qualitative research. For example focus groups can gain more insights in the answers participants give. Moreover, motivations whether or not participants see the organization as credible, legitimate or feel engaged to the organization can become clear. Future studies which combine mixed methods such as quantitative and qualitative research together with existing and fictive organizations will add additional value to the relationship between communication tools and strategies on one hand and credibility, engagement and legitimacy on the other. Not only mixed research methods can be interesting, but mixing different social media platforms can be valuable for research. Especially younger social media users shift away from Facebook to Instagram. Due to time restrictions only Facebook is examined, but future researchers should combine different deliberative tools and take into account that different age groups have different preferences in communication channels.
No significant results were found between the possibility to leave a comment and credibility, engagement and legitimacy. As mentioned earlier, Lauritsen and Perks (2015) discovered that participants expect proof that their engagement or participation is acted upon and hence organizations should incorporate customer suggestions into their CSR initiatives. Future research should focus on this sort of ‘proof’. It is important for organizations to know what kind of ‘proof’ stakeholders expect, what is enough ‘proof’ and how this can be
communicated. Furthermore, there seems to be something going on between dialogue and engagement. On behalf of this research strong conclusions can not be taken. The results made
38
clear the effect from dialogue on engagement is marginally significant. Future research must explore this relationship to a further extent.
This research does not only contribute to the existing literature about CSR, but also to managerial practice. CSR reports and corporate websites were seen as primary CSR
communication vehicles by CSR directors (Pollach, 2015). This research can be of great importance to management practice in order to refer CSR directors and communication professionals to other channels such as deliberative tools. It is vital communication
professionals use deliberative tools because with these tools more legitimacy can be achieved. But there is something else whith deliberative tools. The communication functions that these tools offer are important, for example to get in touch with stakeholders and start conversations with them. This study made clear corporate responses are essential. When organizations do not respond to comments from stakeholders, it does not matter anymore if stakeholders can participate online or not. Participating only adds value when the organization responds to the online comments. This is an important finding and organizations and their communication professionals must be aware that investing in webcare teams is of great importance for their credibility, engagement and legitimacy. Engaging into dialogue seems to be the most effective strategy and organizations and their communication departments should therefore incorporate dialogue in their strategy. According to Lewis (2011) it is not simple to engage into dialogue with stakeholders in practice, but this is something management practice should work out further. It would be advisable to train employees of webcare teams, CSR directors and communication professionals to engage into dialogue with their stakeholders, especially because this research made clear this strategy leads to more credibility through which legitimacy can be achieved.
All in all, this study is indicatory for how researchers as well as communication practitioners should choose and create effective communication tools and strategies in order
39
to gain credibility, engagement and legitimacy. Despite the limitations, this research has laid a foundation for future research in the field of CSR and legitimacy. Organizations who strive for legitimacy can not solely use CSR communication as a means to gain a specific goal, but it now takes two to tango: effective communication is an ongoing process between the organization and its stakeholders (Van Ruler & Vercic, 2005; Christensen & Cornelissen, 2011).
References
Arvidsson, S. (2010). Communication of corporate social responsibility: A study of the views of management teams in large companies. Journal of Business Ethics, 96(3), 339-354. doi:10.1007/s10551-010-0469-2
Bentele, G., & Nothhaft, H. (2011). Trust and credibility as the basis of corporate social responsibility. The handbook of communication and corporate social responsibility, 208-230. doi:10.1002/9781118083246.ch11
Burchell, J., & Cook, J. (2006). It's good to talk? Examining attitudes towards corporate social responsibility dialogue and engagement processes. Business Ethics: A European
Review, 15(2), 154-170. doi:10.1111/j.1467-8608.2006.00439.x
Christensen, L.T., & Cornelissen J. (2011). Bridging corporate and organizational communication: review, development and a look into the future. Management
Communication Quarterly, 25(3), 383‐414.
Colleoni, E. (2013). CSR communication strategies for organizational legitimacy in social media. Corporate Communications: An International Journal, (18)2, 228 – 248. doi:10.1108/13563281311319508
