Graduate School of Communication
CSR Communication on YouTube and the Role of Storytelling in Creating Visibility
Master’s Thesis
Master’s programme Communication Science Corporate Communication Corinne Bachmann 12334375 Supervised by Dr. Iina Hellsten Academic year 2018/2019 28-06-2019
Abstract
The growing demand of stakeholders towards corporations to implement corporate social responsibility (CSR) in a time with more information available than consumable asks for new strategies for effective communication in order to create visibility for corporation’s CSR. This study examines the eligibility of YouTube videos as a tool for CSR communication with regards to storytelling as a communication strategy.
Content analysis has been applied to analyse the global Fortune 500’s appearance on YouTube and the way they communicate about their CSR strategy or approach in videos. The findings show, that the use of YouTube as a communication platform is varying highly and European corporations are more likely to provide videos on CSR compared to their Asian counterparts. The examined CSR videos, on the other hand, share a high similarity and are mainly constructed in the form of a documentary including the storytelling elements characters, positive emotions and prospective value by addressing the corporations’ goals or how the future could be. However, only 34 out of the 159 examined videos contained a typical story structure with a beginning, middle and end, while most of them list different CSR initiatives without a narrative structure. Confirming previous findings, they are made with the purpose to inform, thus failing to engage their audience and neglecting YouTube’s quality as a tool to manage stakeholder relationships. Implications for CSR professionals and theoretical development are provided, together with suggestions for further research.
Keywords: CSR, strategic communication, storytelling, YouTube, visibility, global Fortune 500
CSR Communication on YouTube and the Role of Storytelling in Creating Visibility With formulation of the United Nations’ 17 Sustainable Development Goals in 2015 (UN, 2019) and the enactment of the Paris Agreement on climate change in 2016 (UNFCCC, 2018), issues around responsibility and sustainability reached the global agenda and therewith globally operating corporations. In a corporate context, topics addressing sustainability and responsibility are often part of an overall corporate social responsibility (CSR) strategy (Allen & Craig, 2016; Hetze & Winistörfer, 2016). Although this is still voluntary, corporations worldwide have been pushed to implement CSR as a response for accountability and
transparency claims (Nielsen & Thomsen, 2007). Stakeholders expect global organizations to fulfil growing ethical expectations and contribute to achieving “the general wealth of society” (Colleoni, 2013, p. 231). Today, 93% of the largest 250 companies worldwide report their corporate responsibility towards people, the planet and profits (KPMG, 2017), which is why governments increasingly regulate this corporate disclosure inside and outside of Europe (Lock & Seele, 2017).
CSR is defined as “the process of embracing responsibility for the company’s actions and encouraging a positive impact through its activities for the environment, consumers, employees, communities, stakeholders and all other members of the public sphere who may also be considered as stakeholders“ (Gill, 2015, p. 3). Although corporations operating globally implement CSR at a progressive rate (Wilburn & Wilburn, 2011), awareness among stakeholders of those activities is generally low (Du, Bhattacharya, & Sen, 2010). This creates a need to communicate CSR activities more effectively (Nielsen & Thomsen, 2007). CSR communication has therewith gained in importance (Araujo & Kollat, 2018), but asks for new strategies and more sophisticated techniques (Du et al., 2010; Morsing, Schultz, & Nielsen, 2008). Traditional communication channels such as press releases, CSR reports or websites have shown to be ineffective in reaching stakeholders, and media does not report on CSR
news (Pollach, 2015). Neither stakeholders nor the public is seeking CSR information about a corporation, even when the issue is regarded as important (Dawkins, 2005).
Social media specifically, is a new strategy that can be implemented by corporations in order to reach the public and the media and has gained much scholarly attention (Colleoni, 2013; Jones & Comfort, 2018). However, the research focus has thus far been on Twitter (Araujo & Kollat, 2018; Lovejoy & Saxton, 2012) or Facebook (Abitbol & Lee, 2017). The online video platform YouTube has not attracted much scholarly attention. This lack of academic research cannot be legitimized, as YouTube has become the second most visited website in the world and is continuously expanding while other social media channels are found to be ephemeral (Arthurs, Drakopoulou, & Gandini, 2018).
The narrative nature of YouTube (Pace, 2008) enables corporations to apply specific and unique communication strategies such as storytelling. Storytelling became more relevant in times of digitalization and is being used increasingly (Weissenfeld, Abramova, &
Krasnova, 2017). It can serve as an effective communication strategy, especially for CSR, since it has been shown to be a “natural medium to represent an organization’s significance, values and mission […]” (Gill, 2015, p. 670). Stories explain complex content through sensemaking and consequently enable a target audience to access information that is, due to its complexity, not that accessible otherwise (Colley, 2014; Singh & Sonnenburg, 2012).
Only few studies examined the theoretical (e.g. Gill, 2015) or empirical (e.g. Araujo & Kollat, 2018; Kim & Ferguson, 2018) link between effective CSR communication and
storytelling. The role of stories, when it comes to enhancing awareness of a message, has only been mentioned in theoretical contributions of branding literature (Woodside, 2010; Waters & Jones, 2011; Singh & Sonnenburg, 2012). No established methodology to measure
storytelling in the context of CSR or social media has been implemented so far (Singh & Sonnenburg, 2012). Considering these gaps, two goals can be formulated: First, the Analysis
of YouTube as a platform for strategic CSR communication, and second, the investigation of how storytelling as a communication strategy can be related to the visibility of CSR videos on YouTube. The main research question is: How can storytelling as a corporate communication strategy be associated with visibility of CSR videos on YouTube? To answer this question, this contribution builds on stakeholder theory within a CSR context and develops a conceptual framework to empirically examine storytelling in videos. Offering explorative insights in the use of YouTube as a social media tool for CSR communication and empirically researching the relationship between storytelling and CSR by the means of content analysis, this is adding to the growing body of CSR communication literature.
Theoretical Background
In order to answer the research question formulated above, the theoretical foundation has to be outlined. This section concludes with a conceptual model showing the hypothesized relationships of the central concepts.
CSR and the Management of Stakeholder Relationships
Corporations that operate on a global arena increasingly recognize the importance of corporate social responsibility (CSR) (Wilburn & Wilburn, 2011). The reporting rate of CSR among the largest 250 corporations worldwide has risen from 35% in 1999 to 93% in 2017 (KPMG, 2017). Research and practice seem to equally acknowledge the increasing
importance of CSR as a strategic resource with many benefits (Araujo & Kollat, 2018; Dawkins, 2005; Lee, Oh, & Kim, 2013; Nielsen & Thomsen, 2007) and the body of CSR literature is growing accordingly (Allen & Craig, 2016). The advantages of CSR activities range from the creation and strengthening of stakeholder relationships (Du et al., 2010) to contributions to internal (Gill, 2015) and external reputation (Hetze & Winistörfer, 2016; Pollach, 2015), legitimacy (Hetze & Winistörfer, 2016; Pollach, 2015) and credibility (Lock & Seele, 2017). CSR has been found to account for financial outcomes (Kim & Ferguson,
2018) by influencing investments (Du et al., 2010), consumers buying decisions (Dawkins, 2005), or the willingness to pay more for a product that is produced sustainably (de-Magistris & Gracia, 2016). Moreover, a good and unique CSR strategy can be hard to copy, thus creating competitive advantage through differentiation in a field that has been found to mimicry best practices (Johansen & Nielsen, 2012; Pollach, 2015).
Still, there is no consensus of CSR as a concept and also no established definition. This is mainly due to its contextual nature as a relationship of corporations and their environment (Nielsen & Thomsen, 2007). Most definitions focus on the contribution of corporate resources to improve societal and environmental well being and are thereby positively connoted (Du et al., 2010). In this context, Gill’s (2015) definition is the most adequate as it underlines the broad spectrum of a corporations’ internal and external relationships: “Corporate social responsibility (CSR) is the process of embracing responsibility for the company’s actions and encouraging a positive impact through its
activities for the environment, consumers, employees, communities, stakeholders and all other members of the public sphere who may also be considered as stakeholders“ (Gill, 2015, p. 3).
Stakeholders are “any group or individual who can affect or is affected by the achievement of the firm’s objectives” (Freeman, 1984, p. 25). They represent a network of investors, customers, communities, employees, suppliers, governments and political groups on which a corporation’s survival and long-term success is dependent (Donaldson & Preston, 1995; Louche & Baeten, 2006). Stakeholder theory was first connected to CSR in the 1990s (Carroll, 1999) and remained the main perspective in this field of research ever since (Frynas & Yamahaki, 2016). It implies that corporations should engage in CSR activities and
communicate them to their stakeholders in order to gain support of those stakeholders that are not only interested in financial outcomes, but also in the way a corporation addresses topics around responsibility and sustainability (McWilliams, Siegel, & Wright, 2006). CSR is
thereby often used as an umbrella term (Allen & Craig, 2016) and corporations mostly use the concepts of responsibility and sustainability in the titles of their CSR reports or CSR sections on their websites (Hetze & Winistörfer, 2016).
In sum, stakeholders provide a clear view on societal expectations towards a corporation, which legitimizes the creation, implementation and ultimately the success of CSR activities (Louche & Baeten, 2006).
CSR Communication and the Challenge of Visibility
Benefits of a CSR strategy can only be harvested if stakeholders know about it
(Coombs, 2019; Kim & Ferguson, 2018; Lee et al., 2013). Generally, stakeholder’s awareness of corporations CSR activities is found to be low (Dawkins, 2005; Du et al., 2010; Lee et al., 2013) and has recently been named to be one of the main challenges for CSR communication (Kim & Ferguson, 2018).
Consequently, more proactive efforts are needed to gain external visibility (Werther & Chandler, 2005). Visibility as a precondition for successful CSR communication (Pollach, 2015) is gained through three mechanisms: Availability (actions are converted into data and stored in a format and device), approval (legal regulations, industry-based norms and social consciousness), and accessibility (knowledge how to get the information, classification to facilitate the acquisition, skills to get the data, make sense of it, and amount of effort; Stohl, Stohl, & Leonardi, 2016). This paper suggests addressing the mechanisms of availability by providing CSR videos on YouTube and accessibility by using storytelling as a communication strategy, which is further elaborated in the following chapters. The mechanism of approval is not central for this paper and will shortly be addressed in the discussion.
Altogether, communication is the means to create and spread CSR activities of a corporation (Allen & Craig, 2016) but asks for more effective platforms and strategies in order to gain visibility.
CSR Communication on YouTube
With the rise of social media, an ideal platform for CSR communication has emerged as they enable audiences to spread and react on messages (Araujo & Kollat, 2018), thus facilitating a two-way communication. Because of this possibility to discuss CSR strategies and initiatives online, social media have been found to play a role in raising organizational visibility (Yang & Kent, 2014).
Previous studies examined CSR communication on Twitter (Araujo & Kollat, 2018; Lee et al., 2013; Lovejoy & Saxton, 2012) and Facebook (Abitbol & Lee, 2017), neglecting another key player of online social networks: YouTube. Founded in 2005, the online video platform has grown fast and steadily, now being “the second most visited website in the world” (Arthurs et al., 2018, p. 3). YouTube has the ability to provide more online traffic than a corporation’s own website (Waters & Jones, 2011). Messages can therefore potentially reach a much larger audience. Other functions, such as comments, enable to establish a dialogue between message sender and receiver (Bonsón, Bednarova, & Escobar-Rodríguez, 2014): Corporations and stakeholders.
Corporations use YouTube for educational and informational purposes, rather than the mere entertainment of their audience (Waters & Jones, 2011), which justifies the use of the platform for disseminating CSR information. As poor CSR performance has been found to be associated with low readability of CSR reports, another mode of information presentation could be more suitable (Nazari, Hrazdil, & Mahmoudian, 2017). It has been shown that communication containing verbal, vocal and visual elements can be remembered best. While pictures help to reinforce the content and words of a message, videos can create a strong impression of a corporation (Waters & Jones, 2011). Consequently, the first step to gain visibility for CSR on YouTube is to have an account and provide videos about the respective corporations CSR strategy (mechanism of availability). The first research question aims to
examine whether corporations have an official account on YouTube and if so, whether they use it to disclose information about their CSR strategy: To what extent do corporations use YouTube as a tool to communicate about their CSR strategy?
Once Upon a Time – (Re)Introducing Storytelling
In a globally connected environment, many different stakeholders need to be
addressed and communication processes have to be implemented with regards to a broad and diverse audience (Nielsen & Thomsen, 2007). This challenges current communication strategies and asks for other strategies that take into account the plurality of stakeholders and can be adjusted to different topics or contexts, effectively reaching a diverse audience (Crane & Glozer, 2016). Stories can be told in almost any communicational context (Dahlstrom, 2014) and humans have been conditioned to respond to narratives from early childhood (Kent, 2015). Strategic changes are better recognized when told as a story instead of mere
information (Fenger, Aschemann-Witzel, Hansen, & Grunert, 2015; Weissenfeld et al., 2017), which is relevant when introducing a new CSR approach. The second focus of this study lies therefore on storytelling as a communication strategy for CSR.
Storytelling is the “cultural and social activity of sharing stories” (Jones & Comfort, 2018, p. 147), and has long been recognized to be a powerful tool of communication
(Weissenfeld et al., 2017). Although the concepts of storytelling and narrative are often used synonymously (Escalas, 2004; Kent, 2015; Arthurs et al., 2018), they can be distinguished by defining the story as the production of a storyteller and the narrative as the consumption of the story by its audience (van Laer, de Ruyter, Visconti, & Wetzels, 2014). The main
characteristic of stories is their recognizable structure with a clear beginning (initial event), a crisis or turning point in the middle and a resolution or conclusion at the end. They contain at least one character and have a clear plot (Escalas, 2004). Good stories further comprise emotional content and portray a vision, purpose or prospective value (Marshall & Adamic,
2010; Gill, 2015) with a purpose of information, education or engagement (Waters & Jones, 2011). They can be told as fantasy or/ fiction, an individual experience or in the form of a documentary (Moezzi, Janda, & Rotmann, 2017).
Storytelling and narratives have been researched in depth in the fields of marketing and branding (Escalas, 2004; Woodside, 2010; Singh & Sonnenburg, 2012; Colley, 2014; Bell & Leonard, 2018; Ryu, Lehto, Gordon, & Fu, 2019), customer identification (Dessart & Pitardi, 2016; Woodside, 2010), organizational identity and image (Bell & Leonard, 2018; Gill, 2015; Pace, 2008; Waters & Jones, 2011), reputation (Dowling, 2006) and crisis
communication (Ertem Eray, 2018). Yet, in the context of CSR, they have only recently been applied in theoretical and explorative contributions (Gill, 2015; Kent, 2015; Araujo & Kollat, 2018; Jones & Comfort, 2018; Coombs, 2019; Dhanesh & Nekmat, 2019).
Storytelling and Visibility
The enabling function of stories to create visibility of CSR communication has been outlined in theoretical contributions (Dessart & Pitardi, 2016; Du et al., 2010; Singh & Sonnenburg, 2012), but not researched empirically. However, stories foster comprehension and help to make sense of a message (Colley, 2014; Singh & Sonnenburg, 2012), thus addressing elements of accessibility, the third mechanism to create visibility (Stohl et al., 2016). As narrated experiences are more likely to be repeated and recommended to others (Moore, 2012), the first hypothesis investigates whether a relationship of storytelling and visibility can be found empirically and predicts:
H1: CSR videos containing a higher level of storytelling are associated with greater
visibility on YouTube compared to CSR videos with lower levels of storytelling. The Influence of CSR Topics
Nielsen and Thomsen (2007) identify seven topics of CSR, which are more commonly divided into the three categories society, environment and economy (Nazari et al., 2017;
Rasche, Morsing, & Moon, 2017). Society includes people in general, such as communities and employees, while environment concerns issues around nature or the planet, like renewable energy, sustainable sourcing, or the reduction of CO2. Economy comprises topics around
corporate governance and accountability as well as business strategies or the reporting of CSR (e.g. invested money, amount of initiatives). YouTube videos with social content have been found to generate more engagement compared to other topics (Bonsón et al., 2014), indicating that individuals relate better to human topics. The topic of the CSR message is therefore expected to moderate the relationship of storytelling and visibility. This assumption will be tested with the second hypothesis:
H2: The relationship between the level of storytelling in a CSR message and visibility
on YouTube is moderated by the topic of the CSR message. Storytelling and Engagement
Humans’ natural affection towards narratives (Fisher, 1984) and the "power of stories to reach people” (Coombs, 2019, p. 354) make storytelling a legitimate strategy to build relationships with stakeholders. Empirical research has found a positive relationship between the use of storytelling elements and the effectiveness to create dialogues and spread CSR messages on Twitter, measured in terms of retweets and shares (Araujo & Kollat, 2018). These findings confirmed previous studies: A storytelling spot generated more engagement (in terms on likes, views and comments) than product-focused communication (Dessart & Pitardi, 2016) and the use of some storytelling elements such as characters has been
associated with more user comments on YouTube (Wen & Song, 2017). The last hypothesis builds on these findings and predicts:
H3: CSR messages containing a higher level of storytelling in videos on YouTube will
generate more a) likes, b) shares, and c) comments compared to CSR messages with lower levels of storytelling.
Figure 1. Conceptual model with the hypothesized relationships and moderation of CSR topic Research Method
This chapter documents the processes of the methodology used, sampling strategy and data collection. Further, it provides insights into the measures of the discussed constructs and other relevant variables, concluding with information about the reliability of the codebook. Sample and Data Collection
The target population encompasses all corporations on the Fortune global 500 list of 2018, which represents corporations with the highest revenues during that year (Fortune, 2019). It is a common approach for corporate communication research (e.g. Lee et al., 2013; Abitbol & Lee, 2017; Araujo & Kollat, 2018). A full and non-random sample was manually collected in May 2019, investigating all of the 500 corporations’ presence on YouTube by searching for their full name as stated by Fortune and its abbreviations. If no channel has been found, the corporation’s website was accessed to see whether a link to the corporations
YouTube channel is provided. More commonly however, corporations and especially
holdings had multiple channels divided geographically (e.g. continents, regions, countries), by products or by sub-brands (e.g. Air France-KLM has two brand channels but none for the
holding). In those cases, the channel containing the word “official”, “group” or “global” was chosen to ensure a comparable sample. Following this procedure, 432 channels (86%, N=500) were identified, of which 395 (79%) fulfilled the criteria of being an official channel. To ensure a reliable and replicable data collection, the respective channel was then accessed and a systematic search with “CSR” and the theoretically derived keywords “social
responsibility”, “responsibility”, “responsible” “sustainability” and “sustainable” was carried out. Then, a catalogue with exclusion criteria was designed and applied on all the videos the systematic search yielded (see Table 1).
Table 1
Exclusion criteria after keyword search to identify one video per corporation
Exclusion criteria Exclusion reason
1 Videos in other languages than English English can potentially reach the largest audience (Statista, 2019)
2 Commercials Not part of CSR (see Gill, 2015)
3 Identity videos (e.g. “About us”, “Our history”) Not part of CSR (see Gill, 2015) 4 Employee branding (e.g. “My work at Microsoft”) Not part of CSR (see Gill, 2015) 5 General information which is not related to the
corporation (e.g. “The world is getting warmer”)
No disclosure about the corporations CSR approach 6 Interviews with external experts (e.g. “Obama on
climate change”) not concerning the corporation No disclosure about the corporations CSR approach 7 Videos of events, summits and series with
predetermined formats of presentation
Limits or prevents strategic choice of duration and video format 8 Single products, projects or videos limited to a
geographical area (e.g. “The new sustainable Ford”)
Single projects can be a part of a CSR strategy but do not
necessarily reflect a corporation’s approach. They are thus not always the same type of video and were not included to ensure comparability of the sample 9 Choose latest video before 1st January 2019 The latest video can be assumed
to represent the current strategy Note. The list of corporations and selected videos are provided by the author upon request.
Following these criteria, 159 videos about corporations CSR strategy were identified as units of analysis, posted between 28.11.2010 and 24.12.2018 – a notably long period regarding the fact that only the newest video was included in the sample. Still, the majority of videos were uploaded in the last three years (n=126, see Appendix C, Table C2 for the distribution). Video characteristics (views, likes etc.) were collected within the time span of two days to ensure comparability, since these numbers can grow over time.
Operationalization
In order to perform the content analysis, a codebook was created (see Appendix B). It contains a new empirical framework to measure storytelling in the context of CSR videos and further questions about the videos format and content.
Storytelling. To analyse the presence of storytelling a new framework was constructed for the context of CSR videos, based on the literature review of Weissenfeld, Abramova and Krasnova (2017) and findings of research in marketing, branding,
organizational identity, image, reputation and crisis communication (see Appendix A). Six main elements were identified to be essential for the concept of storytelling and measured with regard to their presence in a video.
Each story should include at least one ‘character’ (Escalas, 2004; Dowling, 2006; Pace, 2008; Woodside, 2010; Waters & Jones, 2011; Colley, 2014; van Laer et al., 2014; Kent, 2015; Pera & Viglia, 2016; Moezzi et al., 2017; Bell & Leonard, 2018; Coombs, 2019). Moreover, a good story should have a ‘narrator’ (Marshall & Adamic, 2010; Pera & Viglia, 2016; Moezzi et al., 2017; Bell & Leonard, 2018) and more specifically a ‘hero’ (Denning, 2006), ‘villain’ and ‘victim’ (Bell & Leonard, 2018). The presence of these characters was examined with four individual questions.
The story ‘plot’ as a second element should be clear and builds the large, overall theme in the story (Denning, 2006; Dowling, 2006; Pace, 2008; Waters & Jones, 2011; van
Laer et al., 2014; Fenger et al., 2015; Kent, 2015; Pera & Viglia, 2016; Coombs, 2019; Ryu et al., 2019). There are five main plots used in public relations, namely “quest”, “adventure”, “rivalry”, “underdog” and “wretched excess” (Kent, 2015) that were operationalized as a categorical variable, asking the coder to choose the plot that fits best based on short descriptions. If no clear plot was present, the options to choose “none of the above” or “a combination of the above” was chosen.
The third element of a story is a clear ‘structure’ (Escalas, 2004; Denning, 2006; Pace, 2008; Kent, 2015), which was measured with three questions about the presence of a clear ‘beginning’ (initial event or baseline), ‘middle’ (crisis or turning point) and a resolution or conclusion in the ‘end’ (Escalas, 2004; Kent, 2015).
Moreover, a story should contain a perspective on the future, which was measured by asking whether the video contained any statements about the ‘vision or goal’ (Marshall & Adamic, 2010; Jones & Comfort, 2018; Coombs, 2019) of the corporation or general assumptions how the ‘future’ (Gill, 2015; Araujo & Kollat, 2018) could be. In addition, the presence of ‘emotional content’ (Waters & Jones, 2011; Gill, 2015; Araujo & Kollat, 2018; Bell & Leonard, 2018) was examined by asking about the mention of positive and negative emotional words (Laros & Steenkamp, 2005).
The fifth element ‘purpose’ was operationalized as the presence of any explicit reason for CSR activities (Jones & Comfort, 2018; Marshall & Adamic, 2010) and the aim of the video as a nominal variable with the categories “inform”, “educate”, “engage”, “none of the above”, or “a combination of the above” (following Waters & Jones, 2011). A last question that was asked is, whether there is any ‘call to action’, because it has been proven important for CSR messages in other contexts (see e.g. Lovejoy & Saxton, 2012; Wen & Song, 2017). Level of Storytelling. Finally, a storytelling index for further analysis was built as the sum of the abovementioned variables measuring different elements of storytelling, together
indicating the ‘level of storytelling’ applied in a video. This variable is continuous on a metric level and ranges from zero (none of the element and their determinants are present) to eight (all are present). Because of the constructs exploratory nature, principal axis factoring (PAF) with Varimax rotation was applied (Carpenter, 2018). The variables were found to be related (determinant =. 297) with correlations below .8 and the data was thus suitable for further analysis (KMO=.661, p=.000). Three factors had an Eigenvalue higher than 1 (2.44, 1.21 and 1.16) and explained together 60.1% of the variance (30.4%, 15.1% and 14.6%).
Visibility and Engagement. The variables ‘views’, ‘likes’, ‘dislikes’ and ‘comments’ were all automatically measured by YouTube and displayed below the respective video. They can reflect any positive number (discrete, metric variables). The dependent variable ‘views’ ranged from 44 (Costco) to 3,183,470 (Exxon Mobil), with an average of 26,710.53 views (SD=252,538.17). The engagement (‘likes’, ‘dislikes’ and ‘comments’) was generally lower and some corporations disabled these functions altogether (n=48 for comments and n=24 for likes and dislikes). PAF has shown that these three variables do not measure the same
construct and were therefore examined separately (KMO=.396, p=.000, one Eigenvalue above 1 of 2.08 explaining 69.3% of the variance). The amount of ‘shares’ was not included as a dependent variable, because YouTube restricts this information to the owner of the respective channel.
CSR Topic. The concept of CSR encompasses three fields of topics, namely ‘society’, ‘environment’ and ‘economy’ (Allen & Craig, 2016). They have been operationalized with regard to their presence in the video (dichotomous variables) and clarified by examples to minimize the risk of misinterpretation. To investigate whether the relationship between storytelling and visibility was stronger in the context of a specific topic, a new nominal variable ‘CSR topic’ was computed, ranging from zero (no topic present) to seven (all topics present).
Video Format. Because some of the prior studies (e.g. Waters & Jones, 2011) found other factors to enhance the success of a video, the ‘type of video’ was measured with five individual questions. Four questions asked for the presence of ‘spoken words’, ‘music’, ‘written text’ and ‘graphs or figures’ (Waters & Jones, 2011). PAF has shown, that the variables were related (KMO = .562, Determinant = .78, p=.000) with two components showing an Eigenvalue higher than 1 (1.57 and 1.01), together explaining 64.5% of the variance (39.3% and 25.2%). The fifth question asked to choose the ‘video format’ that fitted best and was measured as a nominal, categorical variable. The categories were built on the basis of Moezzi, Janda and Rotmann's (2017) concept of truth realm and were adapted to the context of this study. The categories are “fantasy or fiction”, “individual experience”,
“documentary”, and were complemented by the option that “none of the above” or “a combination of the above” was present.
Additional Variables. The regression models were completed with the three control variables ‘revenues’, ‘subscribers’ and ‘duration’. The revenues were provided by Fortune (2019) and ranged from 500,343 to 23,556 million dollars (M=67,538.9, SD=60,217.6). YouTube does show the amount of ‘subscribers’, however, 15 corporations chose to not show them. The most subscribers had Ford Motor (1’934’620), and the least with 185 the Power Corporation of Canada (M=66,098, SD=188724.39, n=144). On average, the videos lasted for 163.68 seconds (2 minutes and 43 seconds, SD=99.27) with the shortest lasting 30 seconds (Exxon Mobil and Kroger) and the longest 604 seconds (10 minutes and 4 seconds, Archer Daniels Midland). In addition, the ‘name of the coder’, a ‘video identification number’ and the ‘date of upload’ were indicated as administrative variables.
Intercoder Reliability
To ensure the reliability of the codebook and thus the results, intercoder reliability was conducted for each variable on 10% (n=16) of the final sample (N=159). The codebook was
thoroughly explained in a coder training and tested with two videos that were not part of the sample (IKEA and Novo Nordisk). After improving it with examples and more instructions for ambiguous variables or descriptions, the final codebook was discussed in another coder training, together with the decision to always watch the video twice and control specific questions (e.g. ‘Q12’, ‘Q14’ and ‘Q15’). Krippendorff’s alpha (Kalpha, α) showed values ranging from α=.628 to α=1 (see Appendix C, Table C1 for all values), which is acceptable, although low for some variables (De Swert, 2012). The main reason for low values were rare categories, which can quickly result in low values if calculating Kalpha (De Swert, 2012). Another reliability test using Fretwurst’s (2013) Lotus that takes rare categories into account confirmed this assumption. ‘Purpose’ as a categorical variable scored the lowest (α=.628) with an 87,5% agreement. Because the main interest of this variable lied in the presence of a category, this value was accepted, although it is close to the very low point of acceptance (α=.610, De Swert, 2012).
Results
The following section examines the collected data, provides an overview of its characteristics and tests the hypotheses for the defined sample.
Descriptive Statistics
To answer the first research question „to what extent do corporations use YouTube as a tool to communicate their CSR strategy?” descriptive statistics were conducted. The results showed, that 86% (n=432) of the full sample (N=500) had a channel on YouTube, while 79% (n=395) had one that fulfilled the criteria of an official channel. 70% (n=351) of the
corporations had videos in English, whereas only 32% (n=159) provided a clearly identifiable video about their CSR strategy. Interestingly, the distribution among countries varied highly and changed with the sampling criteria, resulting in a large underrepresentation of Asian corporations in comparison to European and American corporations (see Table 2).
Table 2
Distribution of countries among the sampling criteria
Country Fortune 500 Any Account Official account In English CSR video USA 25,1% (126) 28.7% (124) 29.9% (118) 33.3% (117) 34.6% (55) China 22,2% (111) 13.4% (58) 10.6% (42) 10.8% (38) 1.6 % (2) Japan 10,4% (52) 11.3% (49) 11.1% (44) 10% (35) 5% (8) Germany 6,4% (32) 6.9% (30) 7.3% (29) 7.4% (26) 8.2% (13) France 5,6% (28) 6.3% (27) 6.6% (26) 5.1% (18) 8.8% (14) Britain 4% (20) 4.6% (20) 5% (20) 5.7% (20) 5% (8) South Korea 3,2% (16) 3.5% (15) 3.3% (13) 2% (7) 0.6% (1) Netherlands 2,8% (14) 3.2% (14) 3.5% (14) 3.7% (13) 5.7% (9) Switzerland 2,8% (14) 3.2% (14) 3.5% (14) 3.7% (13) 6.9% (11) Canada 2,4% (12) 2.8% (12) 2.5% (10) 2.6% (9) 2.5% (4) Others 15% (75) 16% (69) 16.5% (65) 15.7% (55) 21.4% (34) Total 100% (500) 100% (432) 100% (395) 100% (351) 100% (159) Note. Only the ten biggest countries are presented separately.
On average the videos had 36.76 likes (SD=93.02), 6.03 dislikes (SD=35.07) and 2,07 comments (SD=4.71). The most likes (655) and dislikes (397) received the video of Ford Motor, who at the same time had the most subscribers for their channel. The most comments (27) were written on Adidas’ video. Nine videos had 0 likes, 59 had 0 dislikes and 65 received no comments. In most of the videos, a combination of formal elements was applied whereby almost a quarter (22%) did not use spoken words. On average, they contained almost seven out of ten storytelling elements. They contained a rather high level of storytelling
(Mode=6.67, n=47, 29.6%). The use of no elements or all elements however, was never present in this sample. 89.9% of the videos had at least one character present, with the
more positive emotional words (83.6%) were present than negative ones (20.1%). While most of the videos did contain a prospective value by addressing the future in general (81%) or the corporation’s vision (76.7%), only a quarter of all videos (24.5%) asked its audience
explicitly to take action (see Table 3). Table 3
Presence of video characteristics and storytelling elements
Elements Present n
Video Format Text 91.8% 146
Music 91.8% 146
Spoken words 78% 124
Graphs or figures 59.1% 94
Characters At least one character 89.9% 143
Narrator 68% 109
Hero 49.7% 79
Victim 20.1% 32
Villain 1.9% 3
Narrative style Positive emotional words 83.6% 133
Negative emotional words 20.1% 32
Vision or goal of the corporation 76.7% 122 Addressing the future in general 81% 130 CSR activities’ reason or purpose 61% 97
Call to action 24.5% 39
Note. N=159
Almost half of the videos were presented as a documentary (n=76, 47.8%), followed by fantasy or fiction (n=36, 22.6%), which consisted of hand-drawn or computer animated styles. 29 (18,2%) videos were in the form of individual experiences and 18 (11.3%) were combinations.
All of the videos addressed at least one of the topics society, environment or economy and half (54.1%, n=86) addressed each of them. 62 videos (39%) mentioned two topics, the most combination being environment and economy (n=23, 14,5%), followed by society and economy (n=21, 13.2%) and society and environment (n=18, 11,3%). Economy alone has never been discussed while 5 (3,1%) addressed society only and 6 (3.8%) the environment only, respectively.
A story structure ideally consists of a clear beginning, middle and end. Notably, only 34 videos (21.4%) had all of the three parts present while nearly half of the videos (n=72, 45.3%) only had a beginning and an end, without a change or climax in the middle. Constructively, none of the videos had a middle only, while 13 (8.2%) just had a clear beginning and 11 (6.9%) a clear end. 15 videos (9,4%) had no structure at all and each 7 (4.4%) had a beginning and middle or a middle and end.
Regarding the plots that have been used as an underlying theme, most videos represented an adventure (n=110, 69,2%), followed by quest (n=18, 11,3%) and wretched excess (n=6, 3.8%). Each one video (0,6%) used the plots of rivalry and underdog, while 23 videos (14.5%) merely named different CSR activities without using any plot. Combinations of plots have not been used. More than half of the videos were explicitly made to inform their audience (n=125, 78.6%), a combination of information with education (n=8, 5%) or with engagement (n=13, 8.2%). Only 7 (4.4%) videos had the main purpose to engage and 6 (3.8%) to educate, respectively.
Hypotheses Testing
To test the first hypothesis predicting that CSR videos containing a higher level of storytelling are associated with greater visibility on YouTube compared to CSR videos with lower levels of storytelling, a multiple linear regression was conducted, as both variables storytelling and amount of views as well as the control variables revenues, subscribers and
duration were on a metric level. Descriptive statistics showed, that while the level of storytelling was nearly normally distributed (Histogram, Skewedness = -1.1, SE=.19), the variables views (12.5, SE=.19), revenues (3.6, SE=.19), subscribers (7.6, SE=.20) and duration (1.5, SE=.19)were all skewed to the left. Although normality was thus violated, the regression could still be conducted because the central limit theorem states that this deviation does not affect the analysis if the sample consists of more than 30 data points (Field, 2014), which was clearly the case (n=159). With some precaution, the assumptions of linearity (P-Plot) could be interpreted as met and concerning homoscedasticity (Scatterplot), the data was approximately equally distributed around the 0-point, except of one outlier. The Durbin-Watson-Statistic was 2.03, meaning that the residuals did not auto-correlate and there was also no perfect multicollinearity (VIF values between 1.024 and 1.083).
The regression model 2 with amount of views as a dependent variable, the level of storytelling as an independent variable and revenues, subscribers and duration as control variables, was significant, F(4, 154) = 1.48, p<.01 (see Table 4). The model could therefore be used to predict the amount of a video’s views on YouTube, although only for two
variables: There was a weak association between the level of storytelling, revenues and amount of views, which could explain 8.5% of the total variance (R2=.085). Revenues b*=.25, t=3.23, p = .002, 95% CI [.41, 1.72] and the level of storytelling b*=-.14, t=-1.7, p=.085, 95% CI [-52516.33, 3448.83], had a significant, weak association. Per million in revenues, the amount of views increased by 1.06 and for each increase in one level of storytelling, the views decreased by -24533.75, assuming that all other independent variables were held constant. However, when not controlling for revenues, subscribers and duration, the regression model 1 with views as a dependent and the level of storytelling as an independent variable, was not significant, F(1, 157) = 2.34, p=.128 and the effect disappeared. The first hypothesis was therefore rejected.
Table 4
Regression models for variables predicting views on YouTube (N=159)
Model 1 Model 2 Variable B SE (B) β B SE (B) β Level of Storytelling -21418.74 13989.96 -1.21 -24533.75* 14164.89* -.139* Constant 155896.7 86647.32 127240.85 86469.31 Revenue 1.06*** .33*** .254*** Subscribers .014 .11 .01 Duration -155.54 202.48 -.061 R2 .015 .085 F 2.34 3.59** ΔR2 .07 ΔF 1.25 Note. *p <.1 **p <.01 ***p <.005
The second hypothesis aimed to examine whether the relationship between the level of storytelling in a CSR message and visibility on YouTube is moderated by the CSR topic. A moderated, linear regression was performed, using the Process-Macro of Hayes (2018),
because there was a small (although negative) association between the level of storytelling and views when testing Hypothesis 1 in the condition where control variables were included. However, the model was not significant, F(11, 147) = .40, p=.953 and could not be used to predict whether the CSR topic moderates the relationship between storytelling and views.
Hypotheses 3a-3c investigated whether CSR messages containing a higher level of storytelling in videos on YouTube will generate more online engagement (likes, dislikes and comments) compared to CSR messages with lower levels of storytelling. The hypotheses were tested with three separate, linear regressions as all variables had a metric level and the PAF
showed that they did not measure the same construct (see Operationalization and Table 5). Again, the preconditions to run a linear regression were examined using descriptive statistics.
Regarding the distribution of likes for H3a, the assumptions of normality (Histogram, Skewedness statistic of 4.85, SE=.21) and homoscedasticity (Scatterplot) were violated, while the assumptions of the absence of autocorrelation and multicollinearity were met (Durbin-Watson statistic 2.20, VIF=1.00). The regression model with likes as a dependent and level of storytelling as an independent variable was not significant, F(1, 157) = 2.06, p=1.53, thus making the model inoperative to predict the amount of likes a video receives and showing no support for Hypothesis 3a.
Descriptive statistics for H3b showed, that the assumption of normality was strongly violated, because the distribution for storytelling (-1.12, SE=.19) and especially for dislikes (10.56, SE=.21) was skewed. Similar to H1, the dataset had more than 30 data points (n=159) and the regression could still be performed (Field, 2014). The data was approximately linear (P-Plot) and the assumption of homoscedasticity was met, apart from one outlier (Scatterplot). There was no auto-correlation (Durbin-Watson statistic shows 2.06) and no perfect
multicollinearity (VIF=1.00). The regression model with dislikes as a dependent and
storytelling as an independent variable was significant, F(1, 157) =2.94, p < .1, which means that the model can be used to predict the amount dislikes a video gets, but the strength of the prediction was weak: 1.8% of the variation in dislikes could be predicted by the level of storytelling (R2=.018). The level of storytelling had a weak association with the amount of dislikes, b*=.14, t=1.72, p = .088, 95% CI [-.46, 6.59], meaning that for each additional point in the level of storytelling the dislikes increased by 3.06. Hypothesis 3b was thus supported.
The assumption tests in preparation to test hypothesis 3c by means of a regression with comments as a dependent and the level of storytelling as an independent variable were all violated, except of multicollinearity (VIF=1.00) and autocorrelation (Durbin-Watson 1.8).
The data was skewed (3.42, SE=2.3) and neither linear (P-Plot) nor homoscedastic
(Scatterplot). However, this will not be further elaborated as the regression model itself was not significant, F(1, 157) =.20, p=.654 and could not be used to predict the amount of
comments. This did not change when including revenues, subscribers and duration as control variables, F(14, 129) =1.14, p=.329 and Hypothesis 3c was therefore rejected.
Table 5
Regression models for variables predicting likes, dislikes and comments on YouTube (N=159)
Model 3 Model 4 Model 5
Variable B SE (B) β B SE (B) β B SE (B) β Level of Storytelling 6.82 4.75 .114 3.06* 1.79* .136* .098 .219 .036 Constant -4.32 29.42 -12.17 11.06 1.48 1.358 R2 .013 .018 .001 F 2.06 2.94* .20 Note. *p <.1 Discussion
What has been described as conceptual unclarity of CSR in theory (see e.g. Nielsen & Thomsen, 2007), is revealed in its practical manifestation online: After examining the
appearance of 500 corporations on YouTube it can be concluded that the use of the video platform as a communication tool is at high variance. While 78 corporations (14%) don’t have any channel at all, others have a dedicated channel for their CSR activities (e.g. Samsung, ConocoPhillips or Enel). The results further show that 89% among those with an official channel do provide videos in English, confirming previous findings of English being the most common language online (Statista, 2019a). One corporation (Flex) produced its CSR video in seven different languages (including Hebrew), which could indicate an acknowledgement of many and diverse stakeholder groups (Crane & Glozer, 2016).
Use of YouTube for CSR Communication
Interestingly, the sample did not shrink with language as an exclusion criteria (351 out of 395, equates to -11%), but with the criteria of having an official account (395 out of 500, equates to - 21%) and then more drastically with having a video about the corporations CSR (159 out of 351, equates to -55%). Here, country-specific differences are more prominent, with Asian corporations (China, Japan and South Korea) uploading considerably less videos about CSR (11 out of 99 with an official channel, equates -89% and totally 179 included in the Fortune 500, equates -94%) compared to their European peers (Germany, France, Britain, Netherlands and Switzerland with 55 CSR videos out of 103 official channels, equates -47%, and totally 108 corporations, equates -49%). These results show that the mechanisms to create visibility on YouTube (Stohl et al., 2016) are mainly hindered by the lack of availability of CSR information. This is surprising, as the struggle for visibility is often argued to be caused by the increasing availability of information in social media (Thompson, 2005), resulting in an information overload for stakeholders (Rodriguez, Gummadi, & Schoelkopf, 2014). Moreover, accessibility of the channels and the videos should be improved, for instance by including a link to the YouTube channel on the respective corporate website and
systematically using keywords to tag the videos, making it easier to retrieve the desired information. This is especially relevant, since YouTube has been ranked as the second most popular network worldwide (Statista, 2019b) with 1.9 million active users (after Facebook with 2.32 million active users) and could thus potentially reach many stakeholders. In addition, while other social media platforms have found to be ephemeral, YouTube is
perpetually growing (Arthurs et al., 2018) and only one corporation (Wesfarmers) embedded a link to Vimeo on their website, one of YouTube’s biggest competitors.
The first research question can thus be answered as follows: Only one third (32%) of the corporations ranked in the Fortune global 500 in 2018 use an official YouTube channel to
communicate about their CSR strategy in English, while two third (68%) either don’t have a respective video, an official channel or no channel at all. There are notable country-specific discrepancies and the results insinuate that these are not mainly due to differences of
language-use but rather of appliance or communication about CSR activities. Another reason could be related to the mechanism of approval (Stohl et al., 2016), since the use of social media platforms and legal Internet regulations differ among regions. This assumption should be researched in further studies.
Storytelling in CSR Videos
The descriptive statistics of the videos mainly confirmed previous empirical findings, for instance about the presence of characters in 90% of the videos (Waters & Jones, 2011) or the use of emotional content (Araujo & Kollat, 2018; Bell & Leonard, 2018). Other findings, like the absence of videos only addressing topics related to economy, seem intuitive – in this case seeing economy as an enabler and measurement for CSR, but not its central characteristic or goal. Four findings, however, are worth addressing more thoroughly:
First, in contradiction to Araujo and Kollat’s (2018) study, most videos did address the future in general (81%) and also the corporation’s vision or goal for the future (77%), thus clearly containing a prospective value, which they found to be present in only 9.9% of the examined tweets. A possible explanation could be, that this element is not regarded to be the most important one and thus not addressed in a tweet with limited character count but still included in a video with an average length of 2 minutes and 43 seconds. In addition, their study sample was considerably larger, which could also explain the different findings.
Second, although most videos contained many storytelling elements, the full story structure with a clear beginning, middle and end only applied to 34 videos (21%). Almost half of the videos (45%) did set the scene in the beginning and ended with a conclusion or
different initiatives and efforts without a clear central theme or reason why change or actions are needed. This might not be necessary for CSR because of its positive connotation (Du et al., 2010), but is seen to be inevitable for a good story (Denning, 2006; Escalas, 2004; Kent, 2015; Pera & Viglia, 2016; van Laer et al., 2014; Woodside, 2010). Besides, a clear structure might also assist in preventing cognitive overload cause by too much information (Rodriguez et al., 2014).
Third, there is a high similarity observable in the CSR videos: A clear dominance of presenting the information as a documentary (48%) with an adventure plot (69%) and the purpose of informing the audience (79%) is perceptible, whereby alternatives have been used distinctly and visibly less. This cognition confirms Pollach’s (2015) findings, that mimicry of best practices is common in the field of CSR communication. Moreover, the finding that corporations use social media rather to inform than to engage their audience is in line with previous studies on YouTube (Waters & Jones, 2011) and Twitter (Lovejoy & Saxton, 2012).
Fourth, the engagement in terms of likes, dislikes and comments was astonishingly low, suggesting the platform to be less eligible for creating a dialogue with stakeholders. This might also be a reason why some corporations disabled these functions entirely (48 for comments and 24 for likes and dislikes). The investment of resources for monitoring and responding seems to be of no account for an average of 37 likes and two comments found in this study. On the other hand, by not allowing the audience to react on the videos, the potential of YouTube to create a dialogue and manage relationships (Bonsón et al., 2014) is neglected and the communication remains one-way.
Relationship of Storytelling and Views
The results of the quantitative analyses of the CSR videos yielded interesting starting points for future research, but could not confirm most hypothesized relationships: The regression models with views, likes, dislikes or comments as a dependent and the level of
storytelling as an independent variable were not significant, with the exceptions of H1 and H3b. They were significant on an arguably valid level of p=.08, which could be due to the rather small sample size (N=159) and large standard errors (SE) in the distribution of the data.
However, when testing H1 including control variables, a significant but weak
association of the level of storytelling in a CSR video and the amount of views was observed, although in the opposite direction than proposed: A higher level of storytelling can be
associated with less views, indicating that increased visibility of CSR communication on YouTube is negatively related to the amount of storytelling elements applied in a video. Whether the topic of CSR has an influence on this weak relationship (H2) could not be
predicted by the model and will therefore not be discussed further. More research is needed to confirm the weak negative relationship and examine whether this association can be
understood to be causal, because the applied method and analysis cannot provide this information. In addition, further studies should investigate the significant effect of revenues found in two of the regression models and test if higher revenues can be associated with more views or if this might be a spurious relationship and more views are rather associated with a corporation having a high profile. The other exception showing an arguably significant result is H3b, which investigated the relationship between dislikes as a dependent and the level of storytelling as an independent variable and found a weak association. Again, this can only be interpreted as a mild indication of a relationship between the variables, thus requiring further research and confirmation. Still, it is incidental that such a relationship is not desired in a managerial perspective, since this would indicate that the strategic use of storytelling might be contradicting the goal of maintaining favourable relationships with stakeholders.
The answer to the main research question of this study is: The use of storytelling as a corporate communication strategy can not be associated with visibility of CSR videos on
YouTube, although there is a weak indication that the use of more storytelling elements could be related to less views of a video.
Limitations and Further Research
Notwithstanding that this study provides interesting insights, the following limitations need to be considered. First, choosing the most recently uploaded CSR video could have biased the sample, for example if the strategic use of storytelling experienced a trend within a specific time period and also because some of the videos were uploaded when CSR only started to get acknowledged. Second, although the compiled framework for storytelling adds to the existing literature by combining previous findings and applying storytelling to a CSR context, two shortcomings of the created index need to be addressed: The variables are all binary and the scale could therefore not be tested for reliability, since Cronbach’s alpha is not suitable in this regard and there’s a lack of other feasible measurements for the reliability of scales with dichotomous variables (Raykov, Dimitrov, & Asparouhov, 2010). Furthermore, most of the sub-constructs have been measured with less than the commonly suggested three variables (Carpenter, 2018). It is therefore advisable to further refine the framework by measuring each element of storytelling with multiple variables and preferably also the extent to which the individual elements are applied (e.g. with a likert scale). Third, the data is skewed, which limits the validity and value of the regression models and the statistical outcomes need to be interpreted more carefully. Yet, social media data has found to be skewed (Baym, 2013) and not following a normal distribution (Tang & Liu, 2012; Tsou, 2015), making the findings comparable with other studies.
Further research should aim for bigger samples by including all CSR videos of a corporation, to deepen the knowledge about CSR communication on YouTube and perhaps detect changes or trends among different time periods. Other fruitful approaches could be the extension of the conceptual model by investigating other variables that could have an
influence on the visibility on YouTube or to contrast the most successful videos and corporations with media coverage to see whether circumstances like organizational crisis, accidents, or natural catastrophes influence the views of a video more than its content or the way it is made. Additionally, the effects storytelling has on an audience could be researched by combining content analysis (storytelling in the videos) with surveys, interviews or experiments. These methods can provide information about causality of a relationship and also rule out the fact that the amount of views does not provide any information about how long and attentive the video has been watched.
Conclusion
This research provides new insights into the CSR communication of 500 corporations on YouTube, thus providing benchmark numbers for theory and practice. The findings are generalizable for global corporations with the highest revenues as listed by Fortune (2019), which are especially interesting because these corporations theoretically have the resources to employ CSR specialists, resulting in professional strategic communication. This is eminent in a field with high mimicry of best practices (Pollach, 2015), especially if the corporation tries to differentiate itself and gain competitive advantage through CSR (Johansen & Nielsen, 2012). Both the framework to measure storytelling and the results offer promising starting points for future research and contribute to the existing body of CSR and storytelling literature by establishing a theoretical linkage between the two concepts and empirically investing it on YouTube, which has been disregarded by previous research. Corporations are advised to consider YouTube as a platform to strategically communicate about CSR because of its unique characteristics and the second largest audience online. Yet, this study has also shown that other platforms, such as Facebook, might be more suitable to create stakeholder relationships and enter into a dialogue with the audience.
References
Abitbol, A., & Lee, S. Y. (2017). Messages on CSR-dedicated Facebook pages: What works and what doesn’t. Public Relations Review, 43(4), 796–808.
https://doi.org/10.1016/j.pubrev.2017.05.002
Allen, M. W., & Craig, C. A. (2016). Rethinking corporate social responsibility in the age of climate change: A communication perspective. International Journal of Corporate Social Responsibility, 1(1), 1–11. https://doi.org/10.1186/s40991-016-0002-8
Araujo, T., & Kollat, J. (2018). Communicating effectively about CSR on Twitter: The power of engaging strategies and storytelling elements. Internet Research, 28(2), 419–431. https://doi.org/10.1108/IntR-04-2017-0172
Arthurs, J., Drakopoulou, S., & Gandini, A. (2018). Researching YouTube. Convergence: The International Journal of Research into New Media Technologies, 24(1), 3–15.
https://doi.org/10.1177/1354856517737222
Baym, N. K. (2013). Data not seen: The uses and shortcomings of social media metrics. First Monday, 10(18), 1–16. https://doi.org/10.5210/fm.v18i10.4873
Bell, E., & Leonard, P. (2018). Digital Organizational Storytelling on YouTube: Constructing Plausibility Through Network Protocols of Amateurism, Affinity, and Authenticity. Journal of Management Inquiry, 27(3), 339–351.
https://doi.org/10.1177/1056492616660765
Bonsón, E., Bednarova, M., & Escobar-Rodríguez, T. (2014). Corporate YouTube practices of Eurozone companies. Online Information Review, 38(4), 484–501.
https://doi.org/10.1108/OIR-07-2013-0181
Carpenter, S. (2018). Ten Steps in Scale Development and Reporting: A Guide for Researchers. Communication Methods and Measures, 12(1), 25–44.
Carrol, A. B. (1999). Corporate Social Responsibility: Evolution of a Definitional Construct. Business & Society, 38(3). 268–295. https://doi.org/10.1177/000765039903800303 Colleoni, E. (2013). CSR communication strategies for organizational legitimacy in social
media. Corporate Communications: An International Journal, 18(2), 228–248. https://doi.org/10.1108/13563281311319508
Colley, D. (2014). Storytelling Brand Narratives in the Digital Age. In C. Cobanoglu & S. Ongan (Ed.), Advances in International Interdisciplinary Business and Economics. Proceedings of the Global Interdisciplinary Business-Economics Advancement Conference (pp. 714-721). Clearwater Beach, Florida, USA.
Coombs, T. (2019). Transmedia storytelling: A potentially vital resource for CSR
communication. Corporate Communications: An International Journal, 24(2), 351-367. https://doi.org/10.1108/CCIJ-11-2017-0114
Crane, A., & Glozer, S. (2016). Researching Corporate Social Responsibility Communication: Themes, Opportunities and Challenges: Researching CSR Communication. Journal of Management Studies, 53(7), 1223–1252. https://doi.org/10.1111/joms.12196
Dahlstrom, M. F. (2014). Using narratives and storytelling to communicate science with nonexpert audiences. Proceedings of the National Academy of Sciences, 111 (Supplement 4), 13614–13620. https://doi.org/10.1073/pnas.1320645111
Dawkins, J. (2005). Corporate responsibility: The communication challenge. Journal of Communication Management, 9(2), 108–119.
https://doi.org/10.1108/13632540510621362
De Swert, K. (2012). Calculating inter-coder reliability in media content analysis using Krippendorff’s Alpha. First version.
de-Magistris, T., & Gracia, A. (2016). Consumers’ willingness-to-pay for sustainable food products: The case of organically and locally grown almonds in Spain. Journal of
Cleaner Production, 118, 97–104. https://doi.org/10.1016/j.jclepro.2016.01.050 Denning, S. (2006). Effective storytelling: Strategic business narrative techniques. Strategy &
Leadership, 34(1), 42–48. https://doi.org/10.1108/10878570610637885
Dessart, L., & Pitardi, V. (2016). How Youtube storytelling can win consumers’ heart: the case of NIVEA. Advances in Consumer Research, 44, 728.
Dhanesh, G. S., & Nekmat, E. (2019). Facts Over Stories for Involved Publics: Framing Effects in CSR Messaging and the Roles of Issue Involvement, Message Elaboration, Affect, and Skepticism. Management Communication Quarterly, 33(1), 7–38.
https://doi.org/10.1177/0893318918793941
Donaldson, T., & Preston, L. E. (1995). The Stakeholder Theory of the Corporation: Concepts, Evidence, and Implications. Academy of Management Review, 20(1), 65– 91. https://doi.org/10.5465/amr.1995.9503271992
Dowling, G. R. (2006). Communicating Corporate Reputation through Stories. California Management Review, 49(1), 82–100. https://doi.org/10.2307/41166372
Du, S., Bhattacharya, C. B., & Sen, S. (2010). Maximizing Business Returns to Corporate Social Responsibility (CSR): The Role of CSR Communication. International Journal of Management Reviews, 12(1), 8–19.
https://doi.org/10.1111/j.1468-2370.2009.00276.x
Ertem Eray, T. (2018). Storytelling in Crisis Communication. Online Journal of Communication and Media Technologies, 8(2). 131–144.
https://doi.org/10.12973/ojcmt/2358
Escalas, J. E. (2004). Narrative Processing: Building Consumer Connections to Brands. Journal of Consumer Psychology, 14(1), 168–180.
https://doi.org/10.1207/s15327663jcp1401&2_19
words – Assessing the impact of short storytelling messages on consumer preferences for variations of a new processed meat product. Food Quality and Preference, 41, 237–244. https://doi.org/10.1016/j.foodqual.2014.11.016
Field, A. (2014). Discovering statistics using IVM SPSS statistics (4th ed.). London: SAGE. Fisher, W.R. (1984). Narration as a human communication paradigm: The case of public
moral argument. Communication Monographs, 51, 1–22. https://doi.org/10.1080/03637758409390180
Fortune. (2019). Global 500. Retrieved from http://fortune.com/global500/
Freeman, R. E. (1984). Strategic management: A stakeholder approach. Boston: Pitman. Fretwurst, B. (2013). Intercoderreliabilität und Expertenvalidität bei Inhaltsanalysen
Erläuterungen zur Berechnung des Reliabilitätskoeffizienten Lotus mit SPSS [PDF file]. Zürich: Institut für angewandte Kommunikationsforschung. Retrieved from www.iakom.ch/Lotus/LotusManual.pdf
Frynas, J. G., & Yamahaki, C. (2016). Corporate social responsibility: Review and roadmap of theoretical perspectives. Business Ethics: A European Review, 25(3), 258–285. https://doi.org/10.1111/beer.12115
Gill, R. (2015). Why the PR strategy of storytelling improves employee engagement and adds value to CSR: An integrated literature review. Public Relations Review, 41(5), 662– 674. https://doi.org/10.1016/j.pubrev.2014.02.012
Hayes, A.F. (2019). The PROCESS macro for SPSS and SAS. Retrieved from https://processmacro.org/index.html
Hetze, K., & Winistörfer, H. (2016). CSR communication on corporate websites compared across continents. International Journal of Bank Marketing, 34(4), 501–528. https://doi.org/10.1108/IJBM-02-2015-0022
question of differentiation and conformity. Corporate Communications: An
International Journal, 17(4), 434–448. https://doi.org/10.1108/13563281211274130 Jones, P., & Comfort, D. (2018). Storytelling and Sustainability Reporting: An Exploratory
Study of Leading US Retailers. Athens Journal of Business & Economics, 4(2), 147– 162. https://doi.org/10.30958/ajbe.4.2.2
Kent, M. L. (2015). The power of storytelling in public relations: Introducing the 20 master plots. Public Relations Review, 41(4), 480–489.
https://doi.org/10.1016/j.pubrev.2015.05.011
Kim, S., & Ferguson, M. A. T. (2018). Dimensions of effective CSR communication based on public expectations. Journal of Marketing Communications, 24(6), 549–567.
https://doi.org/10.1080/13527266.2015.1118143
KPMG. (2017). International Survey of Corporate Responsibility Reporting [PDF File]. Retrieved from https://assets.kpmg.com/content/dam/kpmg/xx/pdf/2017/10/kpmg-survey-of-corporate-responsibility-reporting-2017.pdf
Laros, F. J. M., & Steenkamp, J.-B. E. M. (2005). Emotions in consumer behavior: A hierarchical approach. Journal of Business Research, 58(10), 1437–1445. https://doi.org/10.1016/j.jbusres.2003.09.013
Lee, K., Oh, W.-Y., & Kim, N. (2013). Social Media for Socially Responsible Firms: Analysis of Fortune 500’s Twitter Profiles and their CSR/CSIR Ratings. Journal of Business Ethics, 118(4), 791–806. https://doi.org/10.1007/s10551-013-1961-2 Lock, I., & Seele, P. (2017). Measuring Credibility Perceptions in CSR Communication: A
Scale Development to Test Readers’ Perceived Credibility of CSR Reports. Management Communication Quarterly, 31(4), 584–613.
https://doi.org/10.1177/0893318917707592
Jonker & M. de Witte (Eds.), Management Models for Corporate Social Responsibility (pp. 171–179). Berlin, Heidelberg: Springer.
Lovejoy, K., & Saxton, G. D. (2012). Information, Community, and Action: How Nonprofit Organizations Use Social Media. Journal of Computer-Mediated Communication, 17(3), 337–353. https://doi.org/10.1111/j.1083-6101.2012.01576.x
Marshall, J., & Adamic, M. (2010). The story is the message: Shaping corporate culture. Journal of Business Strategy, 31(2), 18–23.
https://doi.org/10.1108/02756661011025035
McWilliams, A., Siegel, D. S., & Wright, P. M. (2006). Corporate Social Responsibility: Strategic Implications. Journal of Management Studies, 43(1), 1–18.
https://doi.org/10.1111/j.1467-6486.2006.00580.x
Moezzi, M., Janda, K. B., & Rotmann, S. (2017). Using stories, narratives, and storytelling in energy and climate change research. Energy Research & Social Science, 31, 1–10. https://doi.org/10.1016/j.erss.2017.06.034
Moore, S. G. (2012). Some Things Are Better Left Unsaid: How Word of Mouth Influences the Storyteller. Journal of Consumer Research, 38(6), 1140–1154.
https://doi.org/10.1086/661891
Morsing, M., Schultz, M., & Nielsen, K. U. (2008). The ‘Catch 22’ of communicating CSR: Findings from a Danish study. Journal of Marketing Communications, 14(2), 97–111. https://doi.org/10.1080/13527260701856608
Nazari, J. A., Hrazdil, K., & Mahmoudian, F. (2017). Assessing social and environmental performance through narrative complexity in CSR reports. Journal of Contemporary Accounting & Economics, 13(2), 166–178. https://doi.org/10.1016/j.jcae.2017.05.002 Nielsen, A. E., & Thomsen, C. (2007). Reporting CSR – what and how to say it? Corporate
