Framing of Corporate Social Responsibilities: an Analysis of Text and Photographs in Corporate Reporting

Framing of Corporate Social Responsibilities:

An Analysis of Text and Photographs in Corporate Reporting

Master´s Thesis

Graduate School of Communication

Master´s Program Communication Science: Corporate Communication

Francesca Sauer Student ID: 11571667 Supervisor: Dr. Irina Lock

1 Abstract

Corporate social responsibility reports can serve as a strategic instrument for a company to communicate what it cares about and how it cares and constitute an important resource for any stakeholder interested in gaining deeper insights into organizational procedures and impact. This paper will introduce readers through the rise of visual communication that has continued throughout the last decade. The relevance of framing for CSR communication is approached through the concept of multimodality. Fundamental dimensions of corporate responsibilities (economic, environmental and social) are discussed and framing approaches are introduced in context with the ultimate goal of credible CSR reporting. As multimodal messages are now common practice in corporate reports, this study will analyze framing of problem, activity and target in CSR communication and explore whether and how

2 Introduction

With the rise of social media platforms and, more specifically, visual platforms such as Instagram or Pinterest, there has been a continuously increasing attention for the power and possibilities that visual content has to offer. Visuals have been found to be processed faster and have a positive impact on recall and memory of the information they contain (Dan, 2017). Moreover, there has been argumentation for the persuasive power of visuals through their impact on emotions (Joffe, 2008). Multimodal messages are now more common as they are considered a better presentation of reality by appealing to multiple senses (Kress, 2012). Not only the news media, social media or advertising has drawn from this trend; the cultural development of visualization has spilled over to the realms of corporate communications, revealing new strategic benefits. Richer media content has found to have positive influence on audience trust and perceptions (Cho, Phillips, Hageman, & Patten, 2009; Burgoon et al., 2002) which shows its relevance for credible CSR communication.

The business case of corporate social responsibility has been argued for by means of its positive impact on reputation, employee satisfaction and financial performance (Du, Bhattacharya, & Sen, 2015; Minor & Morgan, 2011; Barnett & Salomon, 2012; Chernev & Blair, 2015). There has been a significant increase in corporate non-financial reporting activities over the last decades and especially European companies have acknowledged the importance of incorporating CSR communication tools (Breitbarth, Harris, & Insch, 2011). External perceptions of an organization´s CSR activities, motivations and objectives are for the most part constructed according to how these are communicated to various groups of stakeholders. Corporate reports serve as a key instrument for any company to communicate about its performance and activities across economic, environmental and social responsibility dimensions and provide the transparency so often desired by society.

Through green-washing and other strategic techniques to deceive the public when it comes to ethical and responsible business, stakeholders have become more aware and skeptical when judging about a company´s CSR motives. This very skepticism has been identified as the crucial barrier which organizations aim to overcome (Elving, 2013; Skarmeas & Leonidu, 2013). Especially disbelief in the communication of corporate social

responsibility has been classified as one of the key factors predicting skepticism (Rim & Kim, 2016). An organization´s right to operate, i.e. its legitimacy, is thus highly dependent of credible and transparent CSR communication.

We make use of framing in our everyday lives, whether it is when we tell friends about our day or by posting content on social media; the highlighting of specific events or

3 aspects is an essential part of how we tell the story of our life to others. It does not necessarily have to happen intentionally, at times we will specifically adapt the way we tell a story to the audience that is listening. However, most of the time there will likely be some kind of inner motivation in form of personal objectives which lead us to make use of framing (Dan, 2017). When it comes to corporate communication strategies, one can identify frames by looking into content created by the organization itself. While the true intentions behind those framing methods can merely be assumed, one can certainly gain a better understanding of an

organization´s identity by analyzing how it tells it´s very own story of responsibilities (Johansen & Nielsen, 2012). The method of storytelling has been found to have positive impact on the credibility of CSR (Gill, 2015).

Photographs constitute a key medium for the purpose of framing and directly link to credibility. By capturing a specific moment in time from a distinctive viewpoint and angle, a photograph can be a frame in itself; it leads the audience´s eye towards a focal point while completely holding back the aspects of a scene or setting which it does not show. This unique quality can serve as a simple, yet effective tool not only for private purposes such as social media profiles but also for strategic communication of any organization, e.g. in form of corporate reports.

The interest of this research is, therefore, to first identify which frames are most prominently used for which dimension of corporate social responsibility. Second, it aims at analyzing the use of visual material by investigating whether the majority of photographs included in sustainability reports are in fact aligning with textual content, i.e. whether

congruent frames are the common practice. This paper will discuss the reporting, framing and visualization of corporate responsibility and ultimately aim at answering the overarching question of how organizations frame corporate responsibility in text and photographs through the medium of reports.

Theoretical Framework Reporting Corporate Responsibilities

According to the triple bottom line framework first introduced by Elkington (1994), there are three fundamental spheres which organizations should pay attention to in their general practice and everyday activities: the people, the planet and profit (Henriques & Richardson, 2004). The theory still serves as a simplified, yet solid foundation for the purpose of

structuring corporate responsibility. There have been various definitions and interpretations of CSR in academic literature, which experienced many adaptations and different developments

4 over the years (Sheehy, 2014; Carroll & Shabana, 2010). However, when it comes down to the core pillars of CSR, there is higher consistency and agreement on a broader level.

The pyramid of CSR dimensions developed by Carroll (1991) often serves as a starting point for further research. It divides corporate responsibilities into economic, legal, ethical and philanthropic and “is intended to be seen from a stakeholder perspective” (Carroll, 2016, p.6). While the requirements, expectations and desires of stakeholder groups can be a large part of the motivation to conduct responsible and ethical business in general,

organizations themselves have realized that credibility, and ultimately legitimacy, are mainly achieved through genuine “public-serving” motivations. A purely “self-serving” approach developed to meet expectations and therefore achieve a better image, will lack in credibility and even increase skepticism (Forehand & Grier, 2003).

There have been different approaches to the distinctions between or overlap of CSR and the concept of sustainability or sustainable development. According to Perez-Batres, Miller and Pisani (2010), CSR operates under the scope of sustainability and both concepts are corporate activities shaped by voluntariness. A similar interplay of both concepts is argued for by van Marrewijk (2002): “CSR relates to phenomena such as transparency, stakeholder dialogue and sustainability reporting“ (p. 102), and can therefore serve as the contribution of a company towards the greater goal of sustainability, i.e. operating in a way that sustains the same conditions for the future (Castro, 2004).

From a sender-oriented perspective, i.e. the organization´s point of view, a broader categorization of CSR into economic, environmental and social responsibilities can provide appropriate guidelines, especially when it comes to one-way communication channels such as corporate reports. With the very basic foundation of economic responsibilities, aspects such as profitability and compliance, legal matters and those concerning ethical practice can be

summarized into one fundamental category directly linked to the business itself. With economic responsibility serving as the traditional “bottom line”, it is further crucial to look into additional aspects such as the society and the environment to see how they are

communicated and whether there are specific differences in organizational reporting between these dimensions.

Dahlsrud (2008) concluded from his content analysis of CSR definitions that economic, social and environmental dimensions are included in the majority of academic definitions, while the corporate environmental impact as such can be classified as a rather recent development in comparison to the other two. Furthermore, the author highlights the importance of the social context of CSR and the influence it has on the variation of

5 definitions. External factors such as industry but also cultural circumstances should be

considered when approaching the understanding / definition of CSR; a perspective also acknowledged by Carroll (2016) himself in the re-evaluation of his original framework.

Corporate reporting has been described as a strategic communication tool (Vallaster, Lindgreen, & Maon, 2012); reports are a medium through which a corporation can tell its own story and thus present its identity and culture in a certain light (Johansen & Nielsen, 2012). Congruent with the predominant CSR categories in academic literature (Dahlsrud, 2008), the Global Reporting Initiative (GRI) provides an official guideline framework for companies reporting about corporate responsibility and sustainability matters. These guidelines serve as a standardization instrument by identifying which aspects of responsibility can and should be reported on and include the three overarching dimensions of economic, environmental and social/societal responsibilities. The relevance of such official guidelines has found evidence in literature (Seele & Lock, 2016); The authors propose that credible CSR reports have the potential to “reestablish moral legitimacy” (p. 187) which is so often questioned by stakeholders and, therefore, suggest organizations to give more attention to the report´s content while structuring it according to official guidelines. Implemented in 2015, the United Nations Sustainable Development Goals are directly linked to corporate citizens, as

responsibility and contribution is required on all levels of society (Bakardjieva, 2016).This refers back to the interplay of sustainability and corporate responsibilities; globally

manifested goals as the SDGs and standardized reporting guidelines as the GRI´s can give companies the chance to emphasize aspects such as ethical compliance and its commitment to the global objective of bettering the world, e.g. by highlighting their individual contributions to specific targets (Schönherr, Findler, & Martinuzzi 2017). Seeing that official frameworks can in fact also act as a frame of corporate reporting, the following section will discuss the potential of framing of corporate responsibilities in more detail.

Framing Corporate Responsibilities

„Frames seem natural to those whose ideas, norms, or values they reference. They organize social reality, simplifying complex matters by emphasizing certain aspects and disregarding others“(Dan, 2017, p. 2). Framing itself, then, incorporates both a selective process and the attribution of salience (Entman, 1993), i.e. rendering specific information “more noticeable, meaningful and memorable to audiences” (p. 53). Hallahan (1999) argues that framing constitutes one of the most important strategic tool in corporate communication and proposes

6 that while frames come in various shapes, they are all linked by their ability of providing information with a context.

When looking at framing in the realm of CSR, especially the framing of environmental responsibility has received academic attention so far. Davis (1995) applied the three framing aspects of problem framing (gain vs. loss), activity framing (taking less vs. doing more) and target framing (current vs. future generation) to test effects on the audience´s attitudes and behavioral intentions. Results showed that a current generation loss frame would be most likely to positively impact environmental behaviors. However, Bortree, Ahern, Smith and Dou (2013) looked into the framing patterns of environmental responsibility over time by adopting Davis´s (1995) categories and found that the gain frame combined with addressing the current generation was most frequently used in CSR advertisements.

One application of framing to corporate environmental reports was implemented by Onkila, Joensuu and Koskela (2014), who looked specifically into how relationships between companies and stakeholders are depicted. Stakeholders were most frequently found to either promote sustainability or commit to sustainability measures of external parties such as NGOs. Moreover, the importance of message framing in social marketing has been discussed in the context of sustainable behavior change (Cheng, Woon, & Lynes, 2011). As evident from these findings, framing techniques have largely been applied to advertising or marketing and used to test for a message´s effect on audience perceptions, behaviors and intentions. To provide a more comprehensive picture of corporate responsibility framing (as opposed to the predominant environmental focus) the present study makes extended use of the three

identified framing categories (problem, target and activity) by analyzing how organizations present their CSR efforts in sustainability reports across the dimensions of economic, environmental and social responsibilities.

Linking back to the relevance of official frameworks for CSR reporting, a company can frame the overall orientation of their CSR activities within those frameworks (e.g. the UN SDGs), for example by emphasizing their commitment to the collective goal of making the world a better place. Furthermore, the role of the company itself in this process, e.g. how sustainability is managed by internal committees and boards, can be communicated in

different ways to stakeholders. As Elkington (2006) states: “ultimately, whatever the drivers, the TBL (triple bottom line) agenda is the responsibility of the corporate board” (p. 524). The present study therefore includes the analysis of two additional frames, the collective goal frame and the governance frame, which will give insights into how organizations frame CSR orientation and their very own role in responsibility spheres:

7 RQ1: How is the orientation of corporate responsibilities framed in reports? (collective goal frame)

RQ2: How is the company´s role regarding responsibilities framed? (governance frame)

As this study not only investigates textual content of corporate reports but also takes visual material in form of photographs into account, the next chapter will discuss the current state of findings and outline further research objectives.

Visualizing Corporate Responsibilities

The use of visual content in corporate reporting documents is certainly no recent development (Preston and Wright, 1996); however, there has been an increase in academic attention

(Ramö, 2011; Breitbarth, Harris, & Insch, 2011; Garcia and Greenwood, 2015; Chung and Lee, 2018). Ramö (2011) proposes that corporate reports do not only randomly place photographs as illustrative decorations; rather, they are visual channels of meaningful information and even give structure to content. As opposed to purely written messages, “pictures (…) are likely to provide immediate perceptual insight and understanding, and thereby increasing the coherence of the message” (p. 380). Ramö (2011) argues, therefore, that images can be especially useful for capturing complex concepts such as corporate responsibility and identifies three main categories of photographs included in corporate reports: humans, products and services, and the environment. This classification is perfectly in line with the triple bottom line framework of social, economic and environmental

responsibilities which implies that the use of photographs in reports can be an important strategic communication tool for corporations. As argued by Dan (2017), framing literature which integrates both words and visuals is still limited although there is evidence for the potential of their interplay. The question remains: do organizations in fact draw from this potential in their CSR reporting?

Through a qualitative comparison between Germany and the UK, Breitbarth et al. (2011) developed a typology of images in CSR reports and determined five prominent types of photographs used: Accountant, People, Dreamworld, Leadership and Day-to-Day Business. When examining the descriptions per type more closely, an underlying structure becomes apparent; while the Accountant, the Leadership and Day-to-Day Business categories all encompass themes related to general business practice, i.e. economic responsibilities, the

8 People and Dreamworld categories can be ascribed to the general impact on society and its habitat, i.e. social and environmental dimensions of corporate responsibilities. It can be observed that when it comes to the division of CSR and its communication, whether in textual or visual form, the triple bottom line framework still proves as a highly relevant foundation for academic research today.

CSR communication does not merely constitute a distribution of information; rather, it also has a distinct purpose of persuasion and promotion (Ramö, 2011). While reports are aimed at being perceived as a truthful and reliable representation of responsible activities, they are an approach to manifesting a beneficial organizational image and reputation among external audiences. Visual elements such as photographs constitute an important resource, as they are often considered “functional evidence” (Breitbarth et al., 2011, p. 255) by supporting claims through illustrations of reality. Seele and Lock (2016) developed a two-phase model for CSR reporting where the understandability of content constitutes a prerequisite for the remaining three claims (truth, sincerity and appropriateness) ultimately shaping credibility. It can be concluded that what reports communicate about and how they embed various topics, e.g. by textual and visual framing techniques, is highly relevant for corporate communication via reports. Furthermore, since visual material has been found to have positive influence not only on attention and recall (Dan, 2017; Joffe, 2008) but also on comprehension and reducing ambiguity (Cho, Phillips, Hageman, & Patten, 2009), the use of text and photographs in corporate reports directly links to the concept of credibility. This linkage provides additional evidence for the overall relevance of visual material in CSR communication.

Garcia and Greenwood (2015) looked into visual framing of CSR by American

multinationals as they recognized the influence of visual content on stakeholders´ perceptions of a company and its potential for framing purposes. In line with prior research (Breitbarth et al., 2011; Ramö, 2011), the authors found that the majority of photographs in corporate reports showed human subjects. Moreover, environmental sustainability was the most prominent theme used, i.e. employees were mostly depicted as caring and contributing towards preserving the environment. Garcia and Greenwood focus on this environmental theme as a frame itself, yet it is questionable to which degree this classification of content can be considered a visual framing approach. While a purely visual analysis can give crucial insights into the use of photographs in general, this study aims at identifying prominent CSR topics in both text and visuals, the frames used to communicate these topics and will

9 framing categories of problem, activity and target framing (Davis, 1995) were adopted for both text and photographs:

RQ3: How are economic, environmental and social responsibility framed through the aspects of problem, target and activity? (in text and photographs)

RQ4: Is there a fit between textual and visual frames for problem, target and activity?

Many visual framing studies either followed a rather qualitative approach (Ramö, 2011; Breitbarth et al., 2011) or solely looked into photographs (Garcia and Greenwood, 2015) without taking the context into account (i.e. topic of the chapter the photograph was included in). The present study strives to fill this gap by combining textual and visual frames and comparing across responsibility dimensions. Moreover, as cultural differences have been acknowledged to shape CSR understandings (Hartmann, Rubin, & Dhanda, 2007), the analysis of CSR framing will add to existing findings by further exploring salience and framing of responsibility issues by European companies and will thus provide a foundation for further comparison between Europe and the US (Garcia and Greenwood, 2015).

Method Unit of Analysis

The unit of analysis of the study is a chapter within an organization´s sustainability/ CSR report. A more detailed description and definition of a chapter can be found in the codebook (see Appendix A). Overall, the units were analyzed on three levels: first, the most prominent CSR dimension was identified. Subsequently, the frames used in the unit´s textual material were analyzed, and lastly, the same procedure was undertaken for each single photograph per chapter. The visual material considered for this analysis were exclusively photographs; charts, animations, graphics or drawings are not taken into account.

Sample

The sample of 201 chapters was drawn from a sample frame of 29 sustainability reports of organizations listed in the Euro Stoxx ESG Leaders 50 index as of May 7tth _{2018. The index} components represent the leading European companies according to specific environmental, social and governance criteria. For each organization, the most recent sustainability or CSR report available, ranging from 2016 to 2017, was downloaded and saved as PDF. To ensure sufficient comparability across units, only sustainability reports were considered for the

10 sample frame. This decision was made after a preliminary content investigation of a

subsample of 10 % indicated that the majority of annual or integrated reports showed a different structural composition in comparison with reports specifically tailored to corporate responsibility matters. As sustainability reports generally incorporate and differentiate

between the three main CSR dimensions in separate sections, they are the most suitable frame of reference for the interest of this research. Separate sustainability sections within annual reports were not considered and, therefore, a total of 21 organizations and three industries (real estate, financial services and food & beverage) were excluded from the sample as solely integrated or annual reports were provided. However, all countries of the original index listing are represented in the final sample.

A total of 13 industries were represented by at least one organization in the sample, the five most represented industries being: insurance, banks, personal and household goods, automobiles and parts and technology. The company size by employees ranged from 24500 to 235400 and a total of eight countries were represented: Germany, France, Italy, Spain,

Finland, Ireland, Belgium and the Netherlands. A list of all represented organizations and industries can be found in the appendix (see Appendix B).

Inter-Coder Reliability

To ensure sufficiently reliable variables, the inter-coder reliability was tested on a subsample of ca. 10 % with a second coder by calculating Krippendorff´s Alpha or the percentage of agreement. First, a short coder training session was conducted in which both coders discussed each variable throughout the coding process of a sample chapter. Then, additional

explanations were added to the codebook instructions and example photographs were

provided for the visual framing variables to further improve reliability. Judging by the general rule of thumb for KAlpha scores, a minimum of .60 is required and scores above .80 indicate good reliability (Hayes & Krippendorff, 2007). Five of the variables showed perfect reliability while only one variable, the governance frame, showed rather low agreement (58%).

However, the decision was made to keep the variable and provide an example in the codebook for further clarification. All intercoder reliability scores can be found in Appendix C.

Coding Procedure and Variables

After collecting administrative items such as the company name and ID, it´s size by

employees, country of origin and industry, each chapter was first coded based on its textual content. For text-based items, the coding procedure followed a keyword approach. The most

11 salient topic per chapter was determined by headings and subheadings. If those proved to be insufficient or ambiguous, single paragraphs were scanned for keywords and general tone of voice. The coding sheet was administered through Excel while a separate printed document offered the complete codebook including instructions and examples.

The coding of photographs followed a similar pattern as for text-based content; however, more specific instructions and examples were provided to reduce the risk of subjective interpretations. The photographs were coded in the order they appeared in the chapter and were each given a unique ID number. In the case of one photograph appearing more than once, only the first appearance was coded. If one photograph was split up into several separate parts (i.e. smaller sections) only the largest picture was coded, as it provided the most content and information. Captions and text around or in the picture itself were only taken into account in case of ambiguity about its content. For example, if the person shown could not clearly be identified as an employee or consumer, surrounding textual content served as a cue. As the three CSR dimensions of economic, environmental and social

responsibility function as a structural component of this study´s research interest, the coding procedure for the issue identification of a chapter is explained below.

Main subject of chapter. The most prominent subject or issue was determined by

means of the three pillars of corporate responsibilities: economic, environmental and social/societal responsibility. To identify the main subject of a chapter, its heading and, if provided, subheadings served as a first impression about its content. If more information was needed to clearly categorize the subject discussed, the unit was subdivided and analyzed per single page which provided a more reliable procedure in many cases. In case more than one or all three of the categories were included within a chapter, the most prominent subject was coded. Per CSR dimension, the content was further analyzed and then assigned to a category of the respective subject. The three overarching CSR dimensions as well as the sub-categories were adapted from the official GRI 4G Guidelines (see Appendix D). After a pre-coding procedure of approximately 10% of the sample, sub-aspects of economic and

environmental responsibility provided by the guidelines were partly summarized or extended and then grouped into more suitable categories for the purpose of this research. An open entry category was provided for each CSR category and the variables were coded as not applicable (= 99) for the two categories which were previously not identified as the main subject of the chapter. The frame variables were each coded according to specific instructions and keywords

12 Table 1. Text-based framing variables.

Variable Categories Description Keywords

Problem frame Gain frame Positive Potential and opportunities

Loss frame negative Risks, dangers and challenges

Target frame current Immediate impact and focus on current

generation, reporting on past

Everything until year 2017, progress

future Long term approach, commitments and

objectives for years ahead

2018 and future, long-term, objectives, commitment

Activity frame Taking less

Passive Reducing, preserving,

restraining, conservation, educate and inform

Doing More

Active Acting, contributing,

creating, innovation, improving, inducing changes

Collective Goal frame

Yes/No Official frameworks providing structure

UN SDG, Paris Agreement, UN Global Impact

Governance frame

Yes/No Strong internal management of responsibility matters

Committees, task forces, representatives, boards

Visual Variables

Each photograph per chapter was given a unique ID number for identification. The formal composition was coded by determining whether the photograph was a cover page photo (i.e. on the cover page of a chapter) and by indicating its size in relation to one whole page. The location, as in setting, of the photograph was coded as “corporate location”, such as offices, buildings, staged events, or “natural location”, e.g. landscapes, public streets, private living spaces and any other locations not related to the company itself.

13

Photo Category. The best fitting category was coded for each photograph as “people”,

“planet” or “profit”. A photograph was categorized as “people” when: a human subject was clearly the focus such as working employees, interactions between stakeholders were shown but also for smiling faces and eye contact, group pictures or employee portraits. The category “planet” included mainly photographs of landscapes or of natural resources (also those used for a product or business), but also images of solar panels or other facilities directly linked to the environment. The “profit” category was coded for photographs which highlighted a feature of power and business, such as office buildings, factory floors, or pictures which put the product or service offered itself in focus. Table 2 provides an overview how and according to which visual cues the framing variables were coded. Example pictures of each category can be found in the codebook (see Appendix A).

Table 2. Photograph-based framing variables.

Results

Data Analysis

As the main objective of this study is to gain insights into which dimensions of CSR are communicated through what types of textual and visual frames (i.e. binary categorical

dependent variables), the Chi-Square test was identified as the most adequate statistical test to Variable Categories Visual cues

Problem frame Gain frame Smiling faces, interactions, caring, people in

action, landscape in natural state, aesthetic, overall positive tone

Loss frame Situations that need solutions, subjects suffering, polluted landscapes, dead animals, negative tone

Target frame Current generation Only adults shown in picture

Future generation Children or young adults included in picture

Activity frame Taking less Calm setting, no human activity or movement visible, current state or quiet setting

Doing more Human involvement, actions, stakeholders helping each other, campaigns/projects, innovations, investments

14 find out about significant differences of distributions and Cramer´s V is reported for the strength of associations between the three issue categories of CSR and the frames used in chapters.

Textual Frames per CSR Dimension

Of the 201 analyzed chapters, the most frequent main topic of a chapter was found to be social responsibility, followed by economic responsibility and environmental responsibility (see Table 1). For the social dimension, the predominant sub-category was “labor practices and decent work”. “Product and Practice” was found to be the most frequent sub-category of the economic chapters and environmental responsibility was largely communicated through reporting about “Resources and Impact”.

Table 1. Frequency of CSR dimensions.

CSR dimension Frequency Prominent sub-dimension

Economic responsibility 59 Product and Practice (93 %) Environmental responsibility 50 Resources and Impact (90.2 %) Social responsibility 91 Labor Practices and Decent Work (47.3 %)

Collective Goal and Governance frame

Figure 1 below shows how the analyzed chapters framed the orientation of corporate responsibilities through a presence or absence of the collective goal frame and how the company´s involvement is expressed through the governance frame.

For the governance frame, a significant but weak association was found, χ2 (2) = 13.74, V= 0.26, p < .001; the frame was present in 96.6% of the economic responsibility chapters, in 70.6% of the environmental responsibility chapters and in 78 % of the social responsibility chapters. The collective goal frame showed a similar association strength, χ2 (2) = 12.29, V = 0.25, p = .002, and was used for 80.4% of chapters about environmental responsibility matters while only for 53.8% of chapters in the social category and in 50.8% of chapters in the economic category. Strong internal management was thus predominantly communicated for economic and business-related information while external frameworks and criteria were most relevant for environmental matters.

15

Activity frame

To explore how activities are framed per CSR dimension, a chi-square test was executed and indicated that there is a significant and very strong association regarding the activity frame, χ2 (2) = 141.78, V = 0.84, p < .001. Crosstabulations showed that 78.4% of chapters in the

environmental dimension of responsibility made use of the activity frame “taking less” while the economic dimension exclusively made use of the “doing more” frame and only one chapter of the social dimension showed a prominent “taking less” activity frame. As for the problem frame, there was not one chapter in the entire sample which made prominent use of the “loss” frame; i.e. no differences were found as the variable was constant across all three dimensions of CSR.

Target Frame

Regarding the target frame, a significant but moderate association was found, χ2 = 25.58, V = 0.36, p < 0.001. As demonstrated in Figure 2, the “current generation” frame was used in 74.7% of the social responsibility chapters and in 54.2% of the economic responsibility chapters. The “future generation” frame was predominantly used only in the environmental dimension with 68.6 % of chapters. When reporting about environmental responsibilities, the chapter content was mainly oriented towards the future and emphasized a long-term approach while past success and an immediate impact were most frequently communicated for social and economic responsibilities.

50.8 80.4 53.8 96.6 70.6 78 0 10 20 30 40 50 60 70 80 90 100

Economic Responsibility Environmental Responsibility Social Responsibility Collective Goal frame Governance frame

16 Figure 2. Textual target framing by CSR dimension.

Visual Variables

The visual analysis of the sample revealed a total of 609 photographs; 61.1% of which were classified as “people” photographs, 22.6 % were coded as “profit” photograph and 16.3 % as “planet” photographs. A total of 131 chapters included at least one photograph per chapter and the number of photographs ranged from 1 to 20 (M = 4.57, SD = 3.64). Comparing the use of photographs across CSR dimensions, 45.3% of the photographs were part of the social responsibility chapters, 31.7 % were found in the environmental responsibility chapters and 23 % in the economic chapters. Figure 3 below illustrates the distribution of photo category in each CSR dimension in percentages.

Figure 3. Distribution of Photo Category in Percentages.

54.2 31.4 74.7 45.8 68.6 25.3 0 10 20 30 40 50 60 70 80 90 100

Economic Responsibility Environmental Responsibility Social Responsibility Current Generation Future Generation

17

Frame Usage in Visuals

Looking at visual target framing, the “current generation frame” was the prominent visual frame in 87.9% of all economic responsibility chapters, in 88% of the environment chapters and in 80.4 % of the social responsibility-based chapters. This result indicates that the majority of photographs exclusively showed adults, while younger generations were more frequently visually depicted within social responsibility chapters in comparison to the other two dimensions.

Regarding the framing of activity through visuals, both for the economic and social responsibility chapters, there was not one chapter which made prominent use of a visual “taking less” frame, i.e. all chapters of those CSR categories included more photographs fitting with the “doing more” frame. For the environmental dimension, however, more than half (51.4%) of the chapters analyzed primarily included photographs which fit into the taking less categorization of activity framing. When looking at these numbers, one should take into account that an activity frame was coded for a total of 473 photographs, as a human subject in action had to be shown and portraits were thus excluded from this frame variable.

When it comes to problem framing, only seven out of the 609 photographs analyzed were classified as a loss frame. At a closer look, all loss frame photographs were included in separate chapters about environmental responsibility, mostly showing natural disasters such as hurricanes (see Codebook in Appendix A for examples). This observation indicates an alignment between text and photographs for problem framing in all 201 chapters analyzed.

50 34.7 85.9 11.4 40.4 2.2 38.6 24.9 11.9 0 10 20 30 40 50 60 70 80 90 100

Photos in Economic (n = 140) Photos in Environmental (n = 193) Photos in Social (n = 276) People Planet Profit

18 Frame Alignment

To explore whether the textual and visual frames within the chapters are supporting each other, e.g. using a “doing more” activity frame in both text and the majority of photographs, new variables were computed in order to obtain mean scores of all photographs per unit for each frame type (problem frame, target frame and activity frame). Subsequently, a “frame fit” variable was computed (by dividing the value of the text frame by the mean score of the visual frame) and scores were divided into two categories of “fit” and “no fit”. The visual frame was coded as fitting with the textual frame, when more than half of the photographs in a chapter showed the same frame as the textual content of that chapter. No statistical tests were computed for the problem frame fit as all textual units made use of a gain frame and only seven loss frame photographs were identified in total. Thus, a frame fit is already revealed through descriptive data. However, since all loss frame photos were included in environmental chapters, this dimension of CSR implies slightly lower congruency between text and photos for the problem frame.

Table 2. Example of establishing frame fit for chapter with four photographs.

Chapter Activity Frame Text Activity Mean Visual Recoding Mean Score Fit (Text Frame/ Visual Mean) 1 “1” (taking less) (2 + 2 + 1 + 2) / 4 = 1.75 1 to 1.49 → “1” (taking less) 1.50 or higher → “2” (doing more) 1.75 → 2 1 / 2 = 0.5 (no fit) “1” = fit

Activity frame fit

A total of 123 chapters included at least one photograph for which an activity frame was coded. The economic responsibility chapters did not include a textual taking less frame and for the one “taking less”- framed chapter of social responsibility there was no fit with the photographs it included. Hence, there was only one CSR dimension for which a frame fit was possible (i.e. the variable of the most prominent CSR dimension was a constant) and no chi-square test could be computed. A “taking less” fit between text and visual framing occurred

19 in 57.1 % of chapters in the environmental responsibility dimension.

However, a significant strong association was found between the CSR dimension type and the “doing more” frame fit between text and photographs, χ2 (2) = 33.05, V = 0.59, p < .001. The majority of economic responsibility chapters (97%) and most of the social

responsibility chapters (98.2%) showed a fit for the “doing more” activity frame while only 42.9% of the environmental chapters did.

Target frame fit

114 chapters included at least one photograph for which a target frame was coded (i.e. showed at least one human subject, n = 414). When it comes to the target frame, the “current generation” frame showed a 100% alignment between text and photographs, i.e. each chapter which was textually framed towards the current generation supported this by largely showing photographs which exclusively depicted adults and no differences were found between groups.

Yet, when comparing across CSR dimensions, there was a significant moderate association for the “future generation” frame fit, χ2 (2) = 6.76, V = 0.35, p = .034, indicating that CSR dimensions significantly differed regarding their use of “future generation” frame alignment. The majority of environmental chapters (90%) and economic responsibility chapters (86.7%) did not show a frame alignment between text and photographs for the “future generation” frame, while 42.1 % of social responsibility chapters did reveal a frame fit. All in all, the fit of activity framing in text and visuals is very frequently established in economic and social responsibility chapters while environmental chapters showed a weaker alignment. For the fit of target framing, the analysis indicates that the majority of chapters did not use congruent frames in text and photographs. The social responsibility dimension stands out here by making most use of “future generation” frame alignment.

Discussion

The purpose of this study is to contribute to framing research in the field of corporate social responsibility by exploring how different issues are framed in text and photographs of reports. Findings provide ground for various conclusions and implications; yet, the true organizational motivations and strategies behind CSR communication can merely be assumed. Patterns in content and its framing were identified and provide insights into the ways of reporting responsibility in different areas. As discussed earlier, the potential of framing and especially visual material has been argued for in CSR literature; the objective of the present study is to

20 advance the understanding of how organizations draw from this potential and to further

identify strengths, weaknesses, and opportunities by integrating results with previous findings.

Onkila, Joensuu and Koskela (2014) looked into the framing of stakeholders in environmental reports and found that the promotion (e.g. educating other stakeholders about responsibilities) and commitment (i.e. committing to externally set criteria) frames were most frequently used. They argue that companies utilize stakeholders as a way to legitimize their actions and to endorse their contributions. The present study supports these findings as it investigated the overall orientation of CSR and found that the vast majority of environmental chapters were framed as a collective goal. Many chapters were structured according to the United Nations Sustainable Development Goals by highlighting which specific objectives the company contributes to through which activities and measures. By officially committing to such globally acknowledged objectives, businesses have the opportunity to communicate to their stakeholders that and how they care; as they are not required to contribute to specific targets or to contribute at all, companies can aim at enhancing their perceived legitimacy through this voluntary act of responsible commitment.

The governance frame was specifically dominant for economic responsibilities, which implies that business-related aspects such as profitability or compliance are still considered a central corporate responsibility requiring strong management. The frame was least frequently found in the environmental chapters, linking back to the finding for the collective goal frame. While economic responsibility is more often framed as a matter of internal management, environmental responsibility is framed by the company´s acknowledgement of external criteria and its commitment as a corporate citizen to making the world a better place.

Furthermore, the governance framing of social responsibility could be explained through the predominant sub-dimension of labor practices and decent work. As many social aspects such as health, training, diversity and equal opportunities are directly connected to the stakeholder group of employees, a framing of these as a responsibility of corporate management shows how much importance a company attributes to its “own” people. As previous studies found, employees are the central subjects in corporate reports since they function as company representatives (Greenwood & Garcia, 2015; Onkila et al., 2014). They serve as a reliable source of information, can have significant influence on a company´s reputation (Gill, 2015) and thus may be framed as a valuable corporate asset in need of strategic governance.

One central interest of the present study was to explore the framing of problem,

21 lack of loss frame chapters in CSR reports aligns with the general purpose and composition of corporate reports which have been argued for as strategic communication tools in terms of reputation and credibility (Pérez, 2015). With this in mind, it is certainly not surprising that, consistent with previous research (Bortree et al., 2013), the prevalent tone of chapters across all dimensions of responsibility was found to be rather positive, with an optimistic mindset to any challenge posed or opportunity given. Furthermore, the analysis revealed how the focus of reports was largely set on the activities implemented by the company (i.e. what can we do

to contribute?) compared to the issue itself (i.e. why do we need to contribute?). This general

finding supports the universal use of the gain frame and directly links to the insights on activity framing the study provides.

Results for the textual activity framing across CSR dimensions show that a “taking less” approach is predominantly used for the environmental responsibility chapters. Related to a company´s environmental impact, i.e. through packaging materials, the use of water for production or the amounts of emissions and waste created by the entire supply chain, these sections of corporate reports are mostly pointed towards the objective of reducing. Presenting themselves in a respectful and humble way, one can assume that organizations utilize such “taking less” frames to communicate the severity of climate change. By acknowledging that we, as humans, have exploited the earth´s resources to a degree where acts of “balancing out” the negative impact with good deeds (i.e. doing more) do not suffice the damage that has already been done, companies seem to prefer showing stakeholders how they adjust to new conditions and what they do (or rather avoid) in order to honor the vulnerable environment.

In contrast, the “doing more” frame was the predominant frame in all economic and social chapters which could be explained by the sheer nature of these two responsibility dimensions and their need for a high degree of organizational agency. Fitting with the findings for the governance frame, matters like performance, innovation or ethical practice, but also human rights, safety and projects or philanthropy are for the most part consequences of active contributions on behalf of companies and require a “let´s do more” approach.

When it comes to the question of the target generation addressed in reports, this study implies that organizations prefer to present past success or current progress in the dimensions of economic and social responsibilities. Previous research identified messages directed towards the current generation and implying an immediate impact as most effective when it comes to environmental behavior change (Davis, 1995) and as the predominant target frame for environmental messages (Bortree et al., 2013). Findings of the present study, however, indicate that corporate environmental responsibilities in reports are predominantly framed

22 through an outlook on future objectives. This prospects-oriented framing of environmental responsibility should certainly be interpreted in consideration of the collective goal framing previously discussed. As environmental chapters were frequently framed within global frameworks, the target orientation of these chapters logically aligns with the time frame provided; for example, the 17 UN SDGs are set out to be reached by 2030, and thus automatically frame corporate responsibilities towards the future.

Regarding the use of photographs in corporate reports, several findings stand out to provide ground for implications. Overall, the majority of photographs was included in social responsibility chapters. Moreover, human subjects were found to be the focal point of most photos, as also found in previous research (Ramö, 2011; Garcia & Greenwood, 2015). One conclusion is that companies place importance on humanizing corporate procedures by highlighting the involvement of people and their agency within the scope of responsibilities (Breitbarth et al., 2010). The strong presence of employees in the photographs connects to the general dominance of social responsibility chapters and additionally supports the findings of textual governance framing. Employees are in fact depicted as representatives of the whole organization; general impressions of the analysis revealed that portraits of smiling employees or employees in action (e.g. interacting with other stakeholders, participating in local projects on-site, using or demonstrating the product) were especially common.

Chung and Lee (2017) found that arousing negative visuals had the most favorable effects on attitudes towards the organization and the underlying CSR motives were more likely to be perceived as serving the public (i.e. not motivated by internal factors such as reputation or profit). However, the authors tested these effects solely by messages

communicated via the social media platform Facebook, thus findings may not apply to other CSR communication channels. Since strategies behind corporate reporting most likely differ from those behind social media communication, the universally positive and optimistic framing of responsibilities in reports does imply how they are designed to enhance reputation and image. The fact that only very few photographs and no textual units were framed

negatively by showing the risks and dangers of climate change, shows that CSR reports generally focus on opportunities arising from such external developments. The loss frame (in photographs as well as sub-units of text) implied the organization´s acknowledgement of the urgency of change; it mostly served as a short introduction or side information which paved the way to corporate commitment, contribution and activity.

The CSR dimensions identified for text and photos showed overall alignment; while most chapters dealt with the social dimension of responsibility, most photographs did so by

23 showing people. Children or young adults were predominantly visually depicted for social responsibilities; often times reports focused on employee well-being, work and life balance or projects in local communities and visually framed these issues by giving attention to future generations.

The high occurrence of a frame alignment for the “doing more” frame in chapters dealing with economic and social dimensions of CSR implies how organizations aim at backing up the textual claims they make by adding visual “proof”. Again, the essential role of employees comes into play here as they are the ones who stand for the company and therefore build the core of any corporate “doing more” approach. However, it is questionable if such company-centered CSR representations will elicit the most positive audience perceptions. Public-serving motives of CSR can be crucial for the perception of credibility. Thus, a shift from employee-centered CSR reporting to a more external focus where consumers or NGO members serve as validation of a company´s activities and impact, might offer other benefits with regards to the impression a report gives. Future research could look into the effects that different stakeholder types (internal vs. external) in visual material of CSR reports have on the perceived intentions behind and the credibility of such corporate channels. Especially for environmentally sensitive sectors (e.g. Oil and Gas) which might have more difficulties in establishing legitimacy, external validation could be of assistance.

As visual material within environmental chapters largely consisted of natural settings and landscapes, the lack of a target frame fit only make sense here because no target frame could be coded for photographs without human subjects. This might also explain the dominance of the visual “taking less” frame (coded for pictures without human activity) in environmental responsibility chapters. One striking finding of this study is the complete frame alignment of the current generation target frame. Each chapter which textually communicated an immediate impact and the urgency to act now, supported this frame visually with

photographs of adults. This observation may be partly explained due to the high number of employee photos but further implies that organizations consider it important to highlight what they are doing right now to contribute instead of merely making future promises. Again, this shows how one purpose of corporate reporting is to communicate a company´s identity and provide evidence (in text and photos) for what it is and what it does. As argued by Ramö (2011), visuals can efficiently improve the understandability of information, but only if they are congruent to the textual message itself. Linking back to the principle of multimodality, the combination of text and visuals is definitely common practice nowadays; however, not all communication efforts draw from the full potential of this multimodal approach. The

24 importance of understandable reporting has been argued for in context with the ultimate goal of credibility (Seele & Lock, 2016) which demonstrates the relevance of multimodality within corporate reports.

A final implication resulting from this study´s findings is the distinct standing of environmental responsibility. This dimension of CSR revealed significantly different frame usage for both the activity itself and the target it is directed to. One explanation may be that environmental responsibilities of corporations are still a relatively recent dimension compared to economic and social aspects (Dalhsrud, 2008) which can affect the ways in which they are communicated. In a broader cultural context, however, the trend of sustainability and

especially environmentally sustainable behaviors has not only changed consumer behavior and lifestyles of many individuals; it also urged organizations to take their role as corporate citizens as serious as ever before. This rising pressure from external conditions could also be a reason for organizations to emphasize the environmental dimension in their reports.

Moreover, academic literature reflects this societal development as environmental

responsibility has seen a high degree of attention in academic CSR research. Based on the present study, future research could further look into the effects of activity and target framing in text and visuals to identify how and where congruent frames actually should be used. Concluding from the trend of “humanization” of corporate reports, especially stakeholder framing seems to be a promising field to look into.

Limitations

There are several limitations which should be taken into account when interpreting this study´s findings. To enhance the external validity of results and ensure a better representation of industries and countries, the unit of analysis of this research is determined as a chapter of sustainability reports. It has to be noted that while the majority of chapters included several different issues and frames, only the most prominent ones were coded. A smaller unit of analysis, i.e. a single page, would certainly contribute to higher internal validity, as one page more likely offers one clearly identifiable main topic. The same assumption holds for the framing variables; single paragraphs which used a loss frame, for example about the risks of climate change and the challenges it brings for businesses, were still outweighed by the gain frame used in the majority of textual content. Additionally, with the chapter as unit of analysis, the composition of the report as a whole is not taken into account; thus, results do not provide information about the salience of topics in relation to one report but rather give insights into how the three different CSR dimensions are communicated through framing.

25 Especially for the coding procedure of visual content such as photographs there are many different approaches to an interpretation of content and framing. The activity frame (“taking less” vs. “doing more”) for photographs was coded according to the instruction that a doing more frame was present when the picture showed a human involved in some kind of activity. Therefore, in case of a photograph showing a setting of corporate philanthropy, which could be interpreted as a “doing more” activity, the lack of human subjects lead to the coding of a taking less frame.

Furthermore, for the coding of photograph content and frames, the caption or surrounding text was advised in case of lacking clarity, e.g. about the stakeholder group the human subject shown belonged to or whether the setting was a corporate location or not. While this procedure certainly simplifies the coding of visuals, a completely separate analysis of text and photographs could lead to different results as coders would not be biased by textual content. However, as text was only consulted in highly ambiguous cases, this method assisted in coding as many visual items as possible. Since the unit of analysis was a chapter which often included several photographs, solely the mean score of all photos per unit could be computed for the framing variables. A frame fit between text and visuals was established if more than half of the photos within a chapter aligned with the textual frame of that chapter. This procedure certainly impedes internal validity as photographs were summarized per unit; however, it serves as a foundation for further research of visual CSR framing which is still scarce.

26 References

Bakardjieva, R. (2016). Sustainable Development and Corporate Social Responsibility: Linking Goals to Standards. Journal of Innovations and Sustainability, 2(4), 23–35. Retrieved from https://sites.google.com/site/journalinsust/home/journal-of-innovations- and-sustainability-vol-2-no-4-2016/sustainable-development-and-corporate-social-responsibility-linking-goals-to-standards

Barnett, M. L., & Salomon, R. M. (2012). Does it pay to be really good? addressing the shape of the relationship between social and financial performance. Strategic Management

Journal, 33(11), 1304–1320. https://doi.org/10.1002/smj.1980

Bergin, R. (2015). Media Richness. Department for Homeland Defense and Security.

Borkowski, S. C., Welsh, M. J., & Wentzel, K. (2010). Johnson & Johnson: a model for sustainability reporting. Strategic Finance, 09(September), 29–37.

Bortree, D. S., Ahern, L., Smith, A. N., & Dou, X. (2013). Framing environmental responsibility: 30 years of CSR messages in National Geographic Magazine. Public

Relations Review, 39(5), 491–496. https://doi.org/10.1016/j.pubrev.2013.07.003

Breitbarth, T., Harris, P., & Insch, A. (2010). Pictures at an exhibition revisited: Reflections on a typology of images used in the construction of corporate social responsibility and sustainability in non-financial corporate reporting. Journal of Public Affairs, 10(4), 238– 257. https://doi.org/10.1002/pa.344

Burgoon, J. K., Bonito, J. A., Ramirez, A. J., Dunbar, N. E., Kam, K., & Fischer, J. (2002). Testing the interactivity principle: Effects of mediation, propinquity, and verbal and nonverbal modalities in interpersonal interaction. Journal of Communication, 52(3), 657–677. https://doi.org/10.1093/joc/52.3.657

Carroll, A. B. (1991). The pyramid of corporate social responsibility: Toward the moral management of organizatrional stakeholders. Business Horizons, Elsevier, 34, 39–48. Carroll, A. B. (2016). Carroll’s pyramid of CSR: taking another look. International Journal of

27 Carroll, A. B., & Shabana, K. (2010). The Business Case for Corporate Social

Responsibility:A Review of Concepts, ijmr_275 85..106 Research and Practice Archie.

International Journal of Management Reviews, 12(1), 85–105.

https://doi.org/10.1111/j.1468-2370.2009.00275.x

Castro, C. J. (2004). Sustainable development: Mainstream and critical perspectives.

Organization and Environment, 17(2), 195–225.

https://doi.org/10.1177/1086026604264910

Cheng, T., Woon, D. K., & Lynes, J. K. (2011). The use of message framing in the promotion of environmentally sustainable behaviors. Social Marketing Quarterly.

https://doi.org/10.1080/15245004.2011.570859

Chernev, A., & Blair, S. (2015). Doing Well by Doing Good: The Benevolent Halo of Corporate Social Responsibility. Journal of Consumer Research, 41(6), 1412–1425. https://doi.org/10.1086/680089

Cho, C. H., Phillips, J. R., Hageman, A. M., & Patten, D. M. (2009). Media richness, user trust, and perceptions of corporate social responsibility. Accounting, Auditing &

Accountability Journal, 22(6), 933–952. https://doi.org/10.1108/09513570910980481

Chung, S., & Lee, S. Y. (2017). Visual CSR Messages and the Effects of Emotional Valence and Arousal on Perceived CSR Motives, Attitude, and Behavioral Intentions.

Communication Research, 009365021668916.

https://doi.org/10.1177/0093650216689161

Dahlsrud, A. (2008). How corporate social responsibility is defined: An analysis of 37 definitions. Corporate Social Responsibility and Environmental Management, 15(1), 1– 13. https://doi.org/10.1002/csr.132

Dan, V. (2017). Integrative framing analysis: Framing health through words and visuals.

Integrative Framing Analysis: Framing Health through Words and Visuals.

https://doi.org/10.4324/9781315171456

Davis, J. J. (1995). The Effects of Message Framing on Response to Environmental Communications. Journalism & Mass Communication Quarterly, 72(2), 285–299. https://doi.org/10.1177/107769909507200203

28 Dennis, A. R. (2009). Media Richness Theory. Encyclopedia of Communication Theory, 642–

643. Retrieved from

http://knowledge.sagepub.com/view/communicationtheory/SAGE.xml%5Cnhttp://dx.doi .org/10.4135/9781412959384

Du, S., Bhattacharya, C. B., & Sen, S. (2015). Corporate Social Responsibility, Multi-faceted Job-Products, and Employee Outcomes. Journal of Business Ethics, 131(2), 319–335. https://doi.org/10.1007/s10551-014-2286-5

Elkington, J. (1994). Towards the Sustainable Corporation: Win-Win-Win Business Strategies for Sustainable Development. California Management Review, 36(2), 90–100.

https://doi.org/10.2307/41165746

Elkington, J. (2006). Governance for sustainability. Corporate Governance: An International

Review, 14(6), 522–529. https://doi.org/10.1111/j.1467-8683.2006.00527.x

Elving, W. J. L. (2013). Skepticism and corporate social responsibility communications: the influence of fit and reputation. Journal of Marketing Communications, 19(4), 277–292. https://doi.org/10.1080/13527266.2011.631569

Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of

Communication, 43(4), 51–58. https://doi.org/10.1111/j.1460-2466.1993.tb01304.x

Forehand, M. R., & Grier, S. (2003). When Is Honesty the Best Policy? The Effect of Stated Company Intent on Consumer Skepticism. Journal of Consumer Psychology, 13(3), 349–356. https://doi.org/10.1207/S15327663JCP1303_15

Garcia, M. M., & Greenwood, K. (2015). Visualizing CSR: A visual framing analysis of US multinational companies. Journal of Marketing Communications, 21(3), 167–184. https://doi.org/10.1080/13527266.2012.740064

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

Global Reporting Initiative. (2014). G4 Sustainability Reporting Guidelines. Global

29 Hallahan, K. (2008). Strategic Framing. The International Encyclopedia of Communication,

4855–4860. https://doi.org/10.1002/9781405186407.wbiecs107

Hallahan, K. (1999). Seven Models of Framing: Implications for Public Relations. Journal of

Public Relations Research, 11(3), 205–242.

https://doi.org/10.1207/s1532754xjprr1103_02

Hartman, L. P., Rubin, R. S., & Dhanda, K. K. (2007). The communication of corporate social responsibility: United states and European union multinational corporations.

Journal of Business Ethics, 74(4), 373–389. https://doi.org/10.1007/s10551-007-9513-2

Hayes, A. F., & Krippendorff, K. (2007). Answering the Call for a Standard Reliability Measure for Coding Data. Communication Methods and Measures, 1(1), 77–89. https://doi.org/10.1080/19312450709336664

Henriques, A., & Richardson, J. (2004). The Triple Bottom Line: Does it all add up? Pacific

Accounting Review (Vol. 16). https://doi.org/10.1080/13549830120073275

Herzig, C., & Schaltegger, S. (2011). Corporate Sustainability Reporting. In Sustainability

Communication (pp. 151–169). https://doi.org/10.1007/978-94-007-1697-1

Joffe, H. (2008). The power of visual material: Persuasion, emotion and identification.

Diogenes. https://doi.org/10.1177/0392192107087919

Johansen, T. S., & Nielsen, A. E. (2012). CSR in corporate self-storying - legitimacy as a question of differentiation and conformity. Corporate Communications, 17(4), 434–448. https://doi.org/10.1108/13563281211274130

Jose, A., & Lee, S.-M. (2007). Environmental reporting of global corporation: A content analysis based on website disclosures. Journal of Business Ethics, 72(4), 307–321. Retrieved from http://www.jstor.org/stable/25075385

Kress, G. (2012). Key concepts in multimodality. Retrieved from

30 Lee, S. Y., & Chung, S. (2018). Effects of emotional visuals and company-cause fit on

memory of CSR information. Public Relations Review. https://doi.org/10.1016/j.pubrev.2018.02.001

Lock, I., & Seele, P. (2016). The credibility of CSR (corporate social responsibility) reports in Europe. Evidence from a quantitative content analysis in 11 countries. Journal of

Cleaner Production, 122, 186–200. https://doi.org/10.1016/j.jclepro.2016.02.060

Marrewijk, M. Van. (2002). Concepts and Definitions of CSR and Corporate Sustainability.

Journal of Business Ethics, 2–11. https://doi.org/10.1023/A

Messaris, P., & Abraham, L. (2003). The role of images in framing news stories. In Framing

Public Life: Perspectives on Media and Our Understanding of the Social World (pp.

215–226).

Milne, M. J., & Patten, D. M. (2002). Securing organizational legitimacy: An experimental decision case examining the impact of environmental disclosures. Accounting, Auditing

& Accountability Journal, 15(3), 372–405.

https://doi.org/http://dx.doi.org/10.1108/BIJ-10-2012-0068

Minor, D., & Morgan, J. (2011). CSR as Reputation Insurance: Primum Non Nocere.

California Management Review, 53(3), 40–59. https://doi.org/10.1525/cmr.2011.53.3.40

Onkila, T., Joensuu, K., & Koskela, M. (2014). Implications of managerial framing of

stakeholders in environmental reports. Social and Environmental Accountability Journal,

34(3), 134–156. https://doi.org/10.1080/0969160X.2013.870488

Perez-Batres, L. A., Miller, V. V., & Pisani, M. J. (2010). CSR, sustainability and the meaning of global reporting for Latin American corporations. Journal of Business

Ethics, 91(SUPPL 2), 193–209. https://doi.org/10.1007/s10551-010-0614-y

Pérez, A. (2015). Corporate reputation and CSR reporting to stakeholders: Gaps in the literature and future lines of research. Corporate Communications, 20(1), 11–29. https://doi.org/10.1108/CCIJ-01-2014-0003

Preston, A. M., Wright, C., & Young, J. J. (1996). Imag[in]ing annual reports. Accounting,

31 Rämö, H. (2011). Visualizing the Phronetic Organization: The Case of Photographs in CSR

Reports. Journal of Business Ethics, 104(3), 371–387. https://doi.org/10.1007/s10551-011-0916-8

Rim, H., & Kim, S. (2016). Dimensions of corporate social responsibility (CSR) skepticism and their impacts on public evaluations toward CSR. Journal of Public Relations

Research, 28(5–6), 248–267. https://doi.org/10.1080/1062726X.2016.1261702

Schönherr, N., Findler, F., & Martinuzzi, A. (2017). Exploring the Interface of CSR and the Sustainable Development Goals. TRANSNATIONAL CORPORATIONS, 24(3), 33–48.

Sheehy, B. (2014). Defining CSR: Problems and Solutions. Journal of Business Ethics,

131(3), 625–648. https://doi.org/10.1007/s10551-014-2281-x

Skarmeas, D., & Leonidou, C. N. (2013). When consumers doubt, Watch out! The role of CSR skepticism. Journal of Business Research, 66(10), 1831–1838.

https://doi.org/10.1016/j.jbusres.2013.02.004

Vallaster, C., Lindgreen, A., & Maon, F. (2012). Strategically Leveraging Corporate Social Responsibility. California Management Review, 54(3), 34–60.

https://doi.org/10.1525/cmr.2012.54.3.34

van Leeuwen, T. (2015). Multimodality. In The handbook of discourse analysis. https://doi.org/10.1353/lan.2004.0166