The Effect of Company Level Social Media Use on the Influence of CSR on Organizational Reputation

(1)

The Effect of Company Level Social Media Use on the Influence

of CSR on Organizational Reputation

Master’s Thesis

M.Sc. International Business & Management

Faculty of Economics and Business

(2)

Acknowledgements

First, I would like to thank my supervisor, dr. Rudi W. de Vries, who was always open for questions and discussion and always offered honest and straight-forward advice. He provided guidance throughout the challenge that is academic research, while at the same time encouraging me to follow my own ideas and methods. This, I am very thankful for.

I would also like to thank Marcel Werner, who assisted me with his time and expertise in programming. The field of programming nowadays is almost limitless. Having had someone to point me into the right direction and to contribute his experience has helped me to gather all the data I needed for my analysis.

I would also like to thank prof. dr. Alan R. Muller and Melih Astarlioglu, who lead the Research Seminar that prepared us all for the task of writing a Master’s thesis.

Furthermore, I would like to acknowledge co-assessor of this thesis dr. Bartjan J.W. Pennink. I am grateful for the valuable comments and a second opinion on this thesis.

(5)

Abstract

(6)

1. Introduction

Facebook’s CEO Mark Zuckerberg once said that “by giving people the power to share, we're making the world more transparent” (BrainyQuotes, 2016). In fact, his social media invention has revolutionized our everyday social lives. In the first quarter of 2016, Facebook already had more than 1.65 billion users world-wide (Statista, 2016a). In 2015 Facebook generated a revenue of 17.93 billion US$ (Statista, 2016b). Those numbers show how popular social media is world-wide, not even taking into account other social media sites like Twitter, a platform that has forever changed micro-blogging (Java et al., 2007). Digitalization offers up unique opportunities for companies to engage with existing and potential customers. According to Malcolm Adler from KPMG Australia “Organizations cannot afford not to be listening to what is being said about them, or interacting with their customers in the space where they are spending their time and, increasingly, their money too” (KPMG International, 2011:2). However, this form of communication also has some setbacks. Because exchange is mostly public and people are connected across all countries of the world, bad news travel just as fast as good news. This makes companies open for public criticism and outrage, sometime earning that much negative responses or ‘shit-storms’ that it can severely damage a firm’s reputation (Lee, Oh, & Kim, 2013). Reputation in turn is an important construct for companies, since 60% of corporate market value can be attributed to company reputation, with 49% of consumers being convinced that social networks influence consumer perceptions about companies (Weber Shandwick, 2012). This shows that consumers themselves acknowledge the fact that social media plays a crucial role in how they perceive the reputation of a firm. Nowadays, organizations therefore have to face many different threats that could seriously harm their reputation, a vulnerability that calls for adequate management (Christopher & Gaudenzi, 2009). A present example of how neglecting social issues can damage reputation is the case of Volkswagen (VW). Since news broke that the company manipulated software in order to pass emission tests, Volkswagen faces major public backlash, leading to falling stock prices (Forbes Magazine, 2017). This example clarifies that reputational losses are often accompanied by financial losses.

(7)

CSR practices and open up new perspectives about what is really important to consumers. But to which extent does CSR really influence a firm’s reputation?

Over the past few years, a remarkable growth in academic publications (both conceptual and empirical) could be observed, as academics and companies realized that a business can increase its success by doing good to society (Danilovic, Hensbergen, Hoveskog, & Zadayannaya, 2015; Falck & Heblich, 2007; Quazi, Amran, & Nejati, 2016). The popularity of social responsibility is further being increased by the growing globalization, connecting people and regions but also their respective CSR issues (Falck & Heblich, 2007). However, there is a lack of knowledge about how and through which channels exactly CSR can affect firm value (Servaes & Tamayo, 2013).

Still, research connecting innovative communication via social media with CSR and organizational reputation remains scarce and ambiguous. This gap in research leads to the following research question: How does CSR affect organizational reputation and which role does social media play for this relationship? The main objectives of this research are to investigate, if and to which extent CSR related activities can affect company reputation and to find out if social media plays a significant role in this relationship.

This thesis responds to two distinct calls for papers. The Journal of Information Processing and Management has called for research focusing at online reputation management and the possible impact on social media activities on organizational reputation (Journal of Information Processing and Management, 2015). In the same vein, the Journal for Sustainability Science encourages research “addressing the link between digital media and sustainability” (Sustainability Science, 2015:374).

(8)

2. Theoretical Background

This section provides an overview of the most relevant literature for this thesis. I first offer insights into the field of CSR, then I explain common theories on organizational reputation and social media. I conclude this section by showing how those three concepts relate to each other, leading to the formulation of two hypotheses and the development of the research model.

2.1 Corporate Social Responsibility

2.1.1 Definition and Geographical Occurrence of CSR

American economist Milton Friedman once said: „There is one and only one social responsibility of business – to use its resources and engage in activities designed to increase its profits so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud” (goodreads, 2016a). However, nowadays this principle is not applicable anymore, as CSR has become one of the most important organizational issues, academically and practically (Danilovic et al., 2015). Today’s socially conscious market environments make CSR play a crucial role on everyday business agendas (Du, Bhattacharya, & Sen, 2010). “CSR has evolved into a complex concept that is now a key component of the corporate decision-making of a number of multinationals that are considered to be the frontrunners in integrating CSR” (Torres, Garcia-french, Hordijk, Nguyen, & Olup, 2012:52). CSR encompasses all parts of an organization and is a major concern for both big and small or medium-sized companies (Danilovic et al., 2015). As such, Friedman’s attitude towards this topic is not up-to-date anymore, because it assumes that everyone will just voluntarily stick to the rules of the game. However, this thesis follows the notion that investing in CSR is indeed important and can also lead to increased profits.

(9)

relationship orientation are the main features distinguishing CSR activities from other organizational investments.

Academics often use corporate social performance (CSP) to assess how socially responsible a firm is (Barnett, 2007; Ducassy, 2013; Muller & Kolk, 2010; Van Der Laan, Van Ees, & Van Witteloostuijn, 2008). Barnett (2007:797) defines CSP as a “snapshot of a firm’s overall social performance at a particular point in time - a summary of the firm's aggregate social posture”. Van Der Laan et al. (2008) follow a more dynamic approach, stating that CSP is the outcome of socially responsive behavior, responsiveness in this context being the process through which corporations respond to social demands. As Chen & Delmas (2011) point out, CSP is mostly assesses via qualitative, soft measures, related to management practices. They declare that, while common measures include labor rights protection and transparency of social and environmental reporting, an adequate assessment is difficult. Consequently, the research field of CSP remains controversial, ambiguous and difficult to research, seeing as there is a lack of differentiation between the concepts of CSR and CSP (Wood, 2010).

Researchers also pay attention to the differences of CSR practices across countries, especially focusing on the differences between Western countries and emerging economies, though there is far less literature on the latter (Muller & Kolk, 2009). This topic is of both academic and practical relevance, seeing as multinational organization nowadays face a variety of different institutional settings, laws and standards (Falck & Heblich, 2007). Muller & Kolk (2009) concluded that companies in emerging markets engage in the same types of CSR activities as companies in more developed country environments. CSR thus can be seen as a way to deal with the absence of global governance and supranational institutional frameworks (Falck & Heblich, 2007). Carroll & Shabana (2010) support this view by pointing out that CSR is not only an important topic in countries of origin but exists in all developed nations as well as in emerging economies.

2.1.2 CSR Dimensions and Motivation

(10)

for the public good). More specifically, Muller & Kolk (2009) identified nine commonly addressed, CSR related types of activities, namely renewables, recycling, environmental training, women in management, vocational training, absenteeism, philanthropy, common relations management and internships.

Going on, the question about what drives CSR activities is another essential topic in literature, seeing as consumers do not react well to greenwashing of self-serving ethical claims (Lee, Oh, & Kim, 2013). In this context, internal motivation plays a significant role as indicator of the true level of commitment to certain CSR measures (Dare, 2016). Some academics criticize that more often than not, CSR activities are mainly driven by the fear of reputational damages to the company (Torres et al., 2012), as stakeholders tend to reward good behavior and punish steps outside of what they consider to be socially responsible (Du & Vieira, 2012). Therefore, a voluntary or discretionary motivation lies at the core of many CSR definitions and frameworks (e.g. Barnett, 2007). A possible external factor that upgraded CSR to be one of the most discussed fields is the rising level of consumer social responsibility (CmSR), meaning that consumers evaluate products not only based on prices or specifications, but also take sustainability issues into account (Quazi et al., 2016). Consumers are not only interested in just buying ethical products, instead they demand corporations to take responsibility for the way in which they bring products to market, how they address vulnerability of certain consumers and how they manage the market impacts (Caruana & Chatzidakis, 2013). The more socially responsible the consumer mindset becomes, the higher the extrinsic motivation for companies to engage in CSR activities.

2.1.3 CSR Outcomes and Benefits

(11)

(Choi, Eldomiaty, & Kim, 2007; Lamberti & Lettieri, 2009), thus having a positive effect on firm performance, market values (Danilovic et al., 2015) and competitiveness (Ducassy, 2013). Falck & Heblich (2007) see CSR as an efficient management strategy that can be a crucial factor in a company’s success. Well selected CSR practices can support not only long-term strategy but also short-term profitability (Lamberti & Lettieri, 2009). Consequently, CSR is highly beneficial in a competitive market, as it contributes to corporate and product differentiation and drives efficient use of capital (Manning, 2013).

Also, CSR leads to increased customer loyalty (turning customers into brand ambassadors willing to spread the word about their good experience), a higher willingness to pay premium prices and resistance against negative company news (Du, Bhattacharya, & Sen, 2007). Additionally, consumers are also more likely to seek employment with the company or invest in the organization (Sen, Bhattacharya, & Korschun, 2006). Furthermore, Cahan, Chen, Chen, & Nguyen (2015) found out that companies that perform well in CSR related activities, usually get more favorable media responses. Therefore, corporations striving to realize higher firm value or lower cost of capital through CSR, need positive media coverage.

Stakeholder theory, as opposed to traditional liberalist view, underlines the positive influence of social behavior on organizational financial performance, resulting from lower implicit costs (Ducassy, 2013). According to Falck & Heblich (2007) the stakeholder approach focuses on an organization’s external environment and all those actors who might be affected by its business goals and their realization. They conclude that it is critical for a company’s success to consider externalities and their impact on various stakeholders, not only shareholders. Essential stakeholders like customers, employees and investors are more inclined to reward socially responsible behavior while punishing insufficient CSR efforts (Du et al., 2010).

(12)

of intangible assets. Moreover, CSR can protect firms not only from negative actions, but also serve as safeguard for the existing assets of the company and act as a buffer against losses (Fombrun, Gardberg, & Barnett, 2000). This is in line with Godfrey, Merrill, & Hansen (2009), who are convinced that engaging in CSR related business activities builds a form of moral capital, which can serve as insurance-like protection in the case of negative events. Along the same line, Minor (2010) observed that CSR indeed helps firms to withstand the influences of negative business shocks and leads to a lesser decrease of firm value (e.g. after product recall).

2.2 Organizational Reputation

2.2.1 Definitions, Dimensions and Drivers of Organizational Reputation

Corporate, organizational or company reputation can be defined as the collective opinion of an organization, held by its stakeholders (Brammer & Millington, 2012). Others perceive it as a “long-term combination of the stakeholders’ assessment” (Cretu & Brodie, 2007:232) about “what the firm is, how well the firm meets its commitments and conforms to stakeholders’ expectations and how well the firm’s overall performance fits with its socio-political environment” (Logsdon & Wood, 2002:366). Lange, Lee & Dai (2011) compiled an extensive literature review on the various definitions of reputation in an organizational context. They classified the definitions in three general groups, depending on the respective core of the definition: being known, being known for something, and having a generalized favorability (Lange, Lee, & Dai, 2011). As such, organizational reputation stands in contrast to brand image (Cretu & Brodie, 2007) or identity (shared understandings of internal stakeholders) (Lewellyn, 2002). However, Whetten & Mackey (2002) see reputation as a special form of feedback, mirroring the company’s identity and credibility, exemplifying the inconsistency in terms of definitions.

(13)

certain standard by the stakeholders. Hence, organizational reputation is built on long-term behavioral outcomes and cannot easily be manipulated by short term actions (Cao, Myers, Myers, & Omer, 2015).

Greyser (1999) identified six key drivers of company reputation: competitive effectiveness (high caliber management, strategic R&D, financial strength), market leadership (incl. industry leadership, well-differentiated products), customer focus (offering good value, commitment and clearly defined image), familiarity/favorability, corporate culture (ethical standards, CSR and high quality employees) and communication (effective advertising, event sponsorship). However, Cao et al. (2015) sees high quality management, ethnical and talented employees or innovation capacity as the main determinants of organizational reputation. Others underline the connection to financial performance, stating that oftentimes, organizational reputation is intensely influenced by the firm’s financial performance (Brammer & Millington, 2005). But according to Barnett (2007) and Cao et al. (2012), company reputation not only depends on the financial but also social and environmental impacts attributed to the corporation over time. Kitchen & Laurence (2003) found out that corporate reputation also depends on the CEO, as chief communicator and face of the company.

Reputational risk thus can be seen as the failure to meet stakeholders’ expectations (Taewon & Amine, 2007). Consequently, reputation management is one important activity that should not be neglected. Christopher & Gaudenzi (2009) identify the protection of organizational reputation as a main objective of risk management, seeing as it is the most difficult risk of all to manage. Aligning real behavior with stakeholder expectations is especially vital, as the company will otherwise suffer in terms of reputation, business and share prices (Greyser, 1999). Heugens, Van Riel, & Van Den Bosch (2004) identify four reputation management capabilities, needed in order to provide protection from the impact of reputational threats: dialogue capabilities (build cooperative and trust-based relationship), advocacy capabilities (ability to persuade audience of the appropriateness of the company’s position in times of controversy), corporate silence capabilities (avoidance of ownership of reputational threats) and crisis communication capabilities (meaningful interaction with parties involved).

2.2.2 Outcomes of Good and Bad Reputation

(14)

(Szöcs, Schlegelmilch, Rusch, & Shamma, 2016). Furthermore, it can help reduce agency problems, as it fosters reliability and trust (Cao, Myers, & Omer, 2012). A high level of organizational reputation can enhance both customer value and loyalty (Cretu & Brodie, 2007), but also lead to a higher level in the financial marketplace (Greyser, 1999). Moreover, it can help to attract skilled employees and support international expansion (Kitchen & Laurence, 2003). According to Greyser (1999), a positive corporate reputation can also lead to preference in doing business when many firms or products are similar in quality and price and help to keep the support of the customers in times of controversy. Another benefit of good reputation is that it enhances beneficial knowledge sharing within networks (Christopher & Gaudenzi, 2009). The reputation itself and reputational changes can deeply impact the company’s relationship with its stakeholders (Lange, Lee, & Dai, 2011). Companies with good reputation are better equipped to attract stakeholders and to form stable relationships with them (Christopher & Gaudenzi, 2009).

Some academics emphasize a company’s reputation’s role for financial performance outcomes, underlining that reputation is of high importance for corporate financial performance and vice versa (Agarwal, Osiyevskyy, & Feldman, 2014; Sabate & Puente, 2003). This can be attributed to its positive influence on firm value, employee morale and productivity (Brammer & Millington, 2012). Cao, Myers, & Omer (2012) found out that reputable companies usually deliver financial reports of high quality, thus reducing information asymmetry and lowering the cost of equity (Barth, Konchitchki, & Landsman, 2013). They see reputation as a form of disclosure that can lower the cost of equity through three ways. First, reputation signals the company’s commitment and ability to act in accordance with shareholders’ interests, thereby reducing information asymmetry. Second, it leads to increased investor recognition, thus increasing stock liquidity. Third, higher earnings quality lowers cost of equity by reducing information asymmetry.

(15)

2.3 Social Media

2.3.1 Social Media Characteristics

In the last ten years, communication via social media has intensively shaped our everyday personal and business lives. As such, it now constitutes a fundamental pillar in corporate communication strategies (Castelló & Ros, 2012). Social media can be defined as a “technology-facilitated dialogue conducted through platforms” (Reilly & Hynan, 2014:749) or “the production, consumption and exchange of information across platforms for social interaction” (Dutot, 2013:55). Social media platforms are an important element of the available online word of mouth (Dijkmans, Kerkhof, & Beukeboom, 2015). As such, social media is a powerful tool that enables two-way communications, thus empowering consumers, by creating international, public platforms, where everyone can share positive and negative feedback in an open, transparent way (Boyd, McGarry, & Clarke, 2014; Dijkmans, Kerkhof, & Beukeboom, 2015). It is a form of democratized communication in so far as it has transferred marketing and PR activities to a level where individuals and communities can create, share and consume a variety of contents (e.g. blogs, tweets, movies, photos) (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011).

Social media’s unique characteristics include being dialogic, anonymous, uncontrollable (Lee et al., 2013) and the perception of “coordinated effects of uncoordinated actions” (Benkler, 2006:12). Furthermore, it provides immediate market reactions and feedback, which might be valuable information for evaluating CSR efforts (Lee et al., 2013).

(16)

availability of others, “who is online?”), relationships (relate to each other), reputation (know social standing of others) and groups (form communities). The extent to which consumers engage in those activities defines the functionality of social media.

2.3.2 Advantages and Risks

(17)

2011). In order to benefit from social media activities, companies are bound to develop special capabilities of monitoring and engaging with the community in a positive way (Benthaus et al., 2016). Active and purposeful management of a company’s social media profiles is thus essential (Miller & Tucker, 2012), while at the same time the appropriate level of engagement is hard to define (Benthaus et al., 2016)

Still, social media use does not only have perks, as communication about companies and brands happens with or without the permission of the respective firm (Kietzmann et al., 2011). A firm’s presence on social media platforms leaves room for the public to not only compliment companies but also to complain about and criticize policies (Dijkmans, Kerkhof, Buyukcan-Tetik, et al., 2015). Indeed, this open, fast, uncontrollable and public way of communication can create backlash, or “shit-storms” when companies try to create an image consumers perceive as hypocritical (Dutot et al., 2016; Lyon & Montgomery, 2013). “Shit-storm” refers to “wide-spread and vociferous outrage expressed on the internet – especially on social media platforms” (Connolly, 2013). Ollier-Malaterre & Rothbard (2015:26) go one step further and state that social media is rather “a social minefield, difference being if you step out of bounds, it’s recorded for posterity, interestingly if you don’t use it you’ll still be judged negatively”. In fact, a study revealed that 24% of communication managers fear the occurrence of shit-storms (Statista, 2016).

2.4 The Interrelatedness of CSR, Corporate Reputation and Social Media

Use

CSR related activities and company reputation are very much intertwined. Stanwick & Stanwick (1998) perceive corporate reputation as a measure of CSP, hinting at a linear relationship between both. A higher level of CSR activities thus would result in a better corporate reputation, as reputation is rooted in proactive CSR activities (Agarwal et al., 2014). Moreover according to Szöcs et al. (2016), CSR related perceptions make up 42% of a company’s overall reputation. CSR activities add value to consumer perceptions of the corporate identity and its impact on society in all countries the company operates in. Therefore, higher engagement in CSR related activities, as measured by CSP, will increase organizational reputation.

(18)

Du et al. (2010) have found that business returns to CSR largely depend on stakeholder awareness of a company’s CSR activities, with studies revealing that such awareness is usually low (Du et al., 2007). Corporate communication has an important role in sharing corporate CSR activities and engagements (Signitzer & Prexl, 2007). Social media offers room for linking companies with humanitarian causes, environmental solutions or economic issues (Kietzmann, Hermkens, McCarthy, & Silvestre, 2011). Via social media, consumers can show their support (e.g. through likes) of or criticism for organizations’ CSR activities (Boyd, McGarry, & Clarke, 2014). Stakeholders can seek consultation and confirmation of specific CSR topics within a broad online community (Colleoni, 2013). Therefore, organizations must keep the public informed about their CSR initiatives (Castelló & Ros, 2012). Social media also impacts firm reputation (Dijkmans, Kerkhof, Buyukcan-Tetik, & Beukeboom, 2015; Wang, 2016). By allowing the demonstration of commitment to certain CSR principles it thus increases favorable corporate reputation (Dutot, 2013), specifically within the group of potential customers (Dijkmans, Kerkhof, & Beukeboom, 2015). Certain ways of corporate communication can shape stakeholder interpretation of different events, with social media giving them the ability to influence reputation by mostly building relationships with companies that have a favorable reputation (Floreddu, Cabiddu, & Evaristo, 2014). Lee et al. (2013) found out that the new media innovation disproportionally favors firms with higher CSR credentials, confirming that there is in fact an important connection between CSR and social media communication. Overall, social media is the voice for a wide range of social and environmental concerns (Fieseler & Fleck, 2013), it translates a company’s values to its stakeholders (Castelló & Ros, 2012; Dutot, Galvez, & Versailles, 2016). Dijkmans, Kerkhof, Buyukcan-Tetik, & Beukeboom (2015) found out that there is a positive effect of corporate exposure to social media activities on the perception of corporate reputation. As such, a company’s level of social media use will help to “translate” its CSR efforts into organizational reputation. Considering that social media makes companies more available for public praise but also for criticism, companies that are very active on social media will be more associated with transparence and trustful intentions, thus increasing, or moderating, the effect of CSP on reputation.

Hypothesis 2: A higher level of social media activity leads to a higher effect of CSP on organizational reputation.

(19)

(20)

3. Methodology: Sample and Data Collection

This section reveals, which methods were used to test the two proposed hypotheses. First, I explain the composition of the sample. Next, I describe, which metrics I used to assess the dependent, independent and control variables.

3.1

Sample

The sample consists of companies featured in Fortune’s Global 500 list of 2016. I specifically chose the Global 500 list as opposed to the regular Fortune’s 500 list, featuring only US firms. By doing so I want to test the hypotheses in an international, heterogenous setting, thereby fully acknowledging (and controlling for) country bias. In order to be sampled, organizations also should have CSR ratings as measured by Thomson Reuters ESG-Data and a corporate reputation score, as rated by Fortune Magazine as part of the selection process for the list of “The World’s Most Admired Companies”. The latter further reinforces the focus on companies from Fortune’s Global 500 list. Furthermore, sampled companies should also have a Twitter account. Out of the 500 companies from the list, 280 also had a reputation score. 188 companies out of those also had CSR ratings. 20 out of those 188 companies did not have a Twitter account. The final sample for the main regression analysis consists out of 163 that are also active on Twitter (see Appendix A). My analysis focuses on the latest reputation and CSP scores (both from 2015) and the most recent social media activity data available.

3.2 Dependent Variable: Organizational Reputation

Company reputation is measured using Fortunes “World’s Most Admired Companies” list. It is based on a survey conducted by Fortune’s cooperation partner, the Korn Ferry Hay Group. The method is described by the Korn Ferry Institute (2016) as follows. The survey starts with 1.500 organizations (from the Fortune 1000, Fortune’s Global 500 lists). In the next step, Korn Ferry Hay Group selects the 15 largest companies for each international industry and the 10 largest for each U.S. industry, leading to a total of 652 companies from 30 countries. Then, various executives, directors and analysts are asked to rate the companies in the respective industry based on nine criteria:

- innovation

- people management (ability to attract and retain talented people) - social responsibility (to community and environment)

(21)

- financial soundness

- long-term investment value - wise use of corporate assets - quality of products and services

- global competitiveness (effectiveness in doing business globally)

I fully acknowledge that social responsibility to the community and environment is a dimension of this reputation score and thus might lead to an overlap with the CSR score. However, the assessment of the two distinct CSP measures differ greatly. The Asset4 ESG metric bases its scores on an extensive number of metrics to asses social, environmental and corporate governance activity, as will be explained below. Fortune’s reputation score however only encompasses a rank of a firm’s overall CSP in comparison with other companies from the same industry. Furthermore, the literature review revealed that CSP is considered to be the main driver of company reputation. Thus, Fortune’s reputation rankings are an appropriate measure to assess organizational reputation.

In order to be listed as one of the most admired companies, the score must rank in the top half of its industry survey. The aggregate reputation score is an average of the attribute scores, ranging from 0 to 10. Higher scores thus indicate higher reputation. The top 50 selection is then published online and in print by Fortune Magazine, the full list of contenders is not. However, to ensure a decent sample size, I personally contacted the person responsible for the listing on the website. For the sake of research, he generously provided a list of all contenders and scores from the last 4 years. By also including the contenders (thus companies with lower scores) I intend to provide enough variance among the reputation scores and thus get more valid results. The reputation scores have been published by Fortune Magazine on the 22nd of February 2016, thus are based on the perceived reputations in the year 2015.

(22)

3.3 Independent Variables

CSP Ratings

In order to measure CSR activity, I selected the Asset4 ESG scores, as expressions of CSP. The Asset4 ESG framework by Thomson Reuters (2016) rates and compares companies against 700 individual data points, combined in ca. 250 key performance indicators, which are then classified in 10 categories within 3 pillars: social, corporate governance and environmental (see Figure 2). The environmental pillar considers a company’s use of resources, emissions and innovation. The social dimension takes into account activities in terms of workforce, human rights, community and product responsibility, while the governance dimension observes the management, shareholders and overall CSR strategy. Overall scores are calculated by equally weighting all data points and comparing them against all the other ASSET4 organizations. The scores thus are relative performance percentages, normalized to better distinguish values and position the score between 0 and 100%. An example list of various metrics of those three dimensions can be found in Appendix B.

(23)

Figure 2: ESG Model and Sample Measures (Thomson Reuters, 2016)

Level of Social Media Use

(24)

average number per tweet. An example screenshot of the script interface can be found in Appendix C. I chose only official company Twitter accounts, indicated by the ‘verified’ mark (blue check mark) that can be found in the Twitter account information of each company. In case a company had several accounts, I used the global account (meaning I excluded regional Twitter accounts). If a firm had several global accounts, I used the most active one. I hereby neglected Twitter accounts that are run by company officials, mostly CEOs (e.g. Tim Cook for Apple). I collected the data between the 24th and 30th of November 2016. In line with Lee et al. (2013) I divide the gathered data in four categories: proactive adoption (number of months active on Twitter), online presence in social media (total number of followers), communication activities initiated by both the company (total number of tweets) and its followers (average number of favorites and re-tweets of the firm’s tweets by followers). With the aim of reducing data I created a sum variable from the average number of re-tweets and favorites per tweet, thus measuring the average reaction from followers per tweet. I called this variable Tweet_Reactions. I tested the reliability based on Cronbach’s alpha, which had a value of 0,85, hinting at good internal consistency. I then transformed all four dimensions logarithmically, due to the skewed distribution of the data.

3.4 Control Variables

(25)

(26)

(27)

Table 1: Industry and Country Frequency Table

Industry Frequency Percent Country Frequency Percent

1 Aerospace 7 4,3 1 Belgium 1 0,6

2 Airlines 3 1,8 2 Canada 1 0,6

3 Apparel&Cosmetics 4 2,5 3 France 14 8,6

4 Chemicals 4 2,5 4 Germany 13 8

5 Computers 6 3,7 5 Hong Kong 1 0,6

(28)

4. Analysis and Results

The main analysis of this thesis focuses on a multiple linear regression analysis, executed via SPSS. This section is structured as follows. First, I explain the descriptive statistics. Second, I describe the main findings of the correlation analysis. Third, prior to the regression analysis I also do an independent sample T-Test, with the aim of getting a general indication of whether having a Twitter account matters for the reputation score or not. Lastly, I summarize the main results from the hierarchical regression analysis, adding variables separately to the model. Multicollinearity was tested by inspecting the correlation coefficients and variance inflation factors (VIF). According to Hoerl & Snee (2012:263), issues of multicollinearity occur, if the VIF values are higher than 5. VIF values that are higher than 10 are problematic. Also, correlation coefficients higher than 0,7 could be an indicator for multicollinearity.

4.1 Descriptive Statistics

(29)

The table reveals that the mean reputation score of the sampled companies are relatively high at 6,23 out of 10. While the highest reported score was at 8,64 the lowest was 2,31. This is confirmed by the relatively low value of the standard deviation of 1,02. The same can be observed for the mean CSR score, which is 79,29 out of 100, with a relatively low standard deviation. Overall, this shows that the CSR scores of the sampled organizations are mostly located in the upper half of the scale. Interestingly, the means of the three partial scores show similarities, especially the social score (mean=81,42) and the environmental score (mean=82,93). The mean of the governance score, however, is a little lower at 73,51. Overall, those means show that the CSR activities of the sampled firms are generally favorable and in the upper half of the scale. As for the Twitter activities, here the standard deviations show that there are significant differences between the companies. The standard deviations of both the number of Tweets and Followers are higher than the respective means. This means that the values of those three variables are highly dispersed and deviate a lot from the mean. They spread out over a wide range of values. Only the fourth Twitter variable, Adoption, does not show an extreme standard deviation from the mean. On average, the companies represented in the sample have been using Twitter for 68,48 months already. To sum up, the sampled organizations mostly are similar in terms of reputation and CSR scores, but differ in their use of Twitter.

4.2 Correlations

Table 3 presents the calculated Pearson correlations for all dependent and independent variables and additionally the three dimensions of the CSR score. The significant correlations are highlighted in bold font. Organizational reputation shows a small positive correlation with revenue, meaning that higher reputation is associated with higher revenue. As for the country effects, while there is a moderate positive correlation between company reputation and GDP per capita (R2=0,432, significant at 0,01 level), there is a negative relationship with the percentage of internet users. Additionally, GDP is also significantly negatively correlated with internet use. This means that though companies from countries with a higher GDP per capita are expected to have a better reputation, they do not necessarily have more internet users than those from countries with lower GDP per capita.

(30)

is likely to score higher. As expected, the CSR score shows high significant correlation with the three individual scores (social, governance and environmental). While the correlation coefficients are relatively high, they do not vary much in their values, indicating that the average CSR score is not mainly driven by only one factor.

The environmental and social scores further indicate a positive relationship with company age, meaning that older organizations are expected to score higher on those dimensions, though the effects are small (0,238 and 0,181 respectively). Similarly, the same observation can be made for the relationship with internet use. However, while both the environmental and social scores show significant positive correlations, the governance score shows a moderately stronger, negative correlation with the percentage of internet users. This means that the more people in a country have access to internet, the higher the environmental and social scores of the company will be, but that the governance score will be lower. The other country variable, GDP per capita, shows the exact reverse effect. Here, the governance score shows a positive an adequately high positive relationship, while the other two CSR dimensions are negatively correlated. Within the three CSR dimensions, the social and environmental scores have a strong positive correlation, meaning than a rise of one is associated with a rise of the other.

Going on, the measures used to assess Twitter use also demonstrate some interesting effects. The most noticeable relationship again is with the country measures. The number of Tweets and number of Followers both show a significant negative correlation with internet use. GDP per capita on the other hand is positively correlated with the number of Tweets and months passed since the adoption of Twitter. For Twitter activities, this means that companies from countries with a higher GDP per capita tweet more often and have adopted Twitter faster than organizations from countries with a lower GDP per capita. On the other hand, the higher the percentage of people with access to the smaller the number of tweets and followers seems to get. Though again, the correlation coefficients are not specifically high, this is an interesting phenomenon to observe.

(31)

The four Twitter variables themselves are also positively correlated with each other. The number of tweets shows a high positive correlation (0,68) with the number of followers and a smaller, but also significant, positive correlation with the number of months since adoption (0,26). Furthermore, the number of followers is moderately positively correlated with the average number of reactions per tweet (0,445) and, to a smaller extent, also with the number of months since adoption. Consequently, companies that have been using Twitter for a longer period have more followers and receive more reactions (via retweets and favorites) for their Tweets. Also, more followers seem to indicate higher numbers of tweets and more reactions. Overall, the correlation coefficients are not very high. The only coefficients higher than 0,7 are between the CSR score and its dimensions. However, this makes sense, as the score is an average of the three factors. Consequently, multicollinearity is not a problem.

4.3 Independent Sample T-Test

(32)

Table 2: Descriptive Statistics (N=163)

Mean S.D. Min. Value Max. Value

(33)

Table 3: Correlation Analysis 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 1. Rep_Score 2. Industry -,126 3. Age ,046 -,178* 4. Revenue ,188* _,123 _,001 5. Internet_Use -,275** _,012 _,009 _,043 6. GDPperCapita ,432** _,017 _,023 _-,023 _-,384** 7. CSR_Score ,030 -,155* _,174* _-,118 _-,036 _,093 8. Soc_Score -,054 -,083 ,181* _-,111 _,287** _-,209** _,760** 9. Gov_Score ,114 -,044 -,024 -,079 -,497** _,460** _,581** _,055 10. Env_Score -,028 -,197* _,238** _-,056 _,276** _-,180* _,736** _,669** _-,035 11. Tweets ,146 -,112 -,003 ,075 -,246** _,188* _,171* _,003 _,150 _,172* 12. Followers ,258** _-,206** _,111 _,150 _-,179* _,146 _,252** _,035 _,193* _,260** _,680** 13. Tweet_Reaction ,120 -,070 ,109 ,164* _,013 _-,051 _,107 _,079 _,035 _,114 _-,069 _,445** 14. Adoption ,016 ,018 ,032 -,091 -,127 ,287** _,031 _-,052 _,153 _-,074 _,260** _,190* _,053

**Correlation is significant at the 0,01 level (2-tailed) * Correlation is significant at the 0,05 level (2-tailed) N=163

(34)

4.4 Regression Analysis

Table 4 shows the results from the multiple regression analysis. The first model only includes the five control variables. The results show that this model is statistically significant (R2=0,266, p<0,01), meaning that approximately 26,6% of the dependent variable can be explained with the control variables only. The adjusted R2 is only slightly smaller. All control variables except for age show significant coefficients, meaning that they influence the reputation score. However, both industry and internet use have negative coefficients, revealing a negative effect on the reputation score. GDP per capita and revenue on the other hand show highly significant but miniscule positive coefficients, meaning that the effects on the reputation score are very little. The tolerance statistics show overall favorable values. The highest VIF in Model 1 is 1,176.

In Model 2, the CSR score has been added. The analysis shows that there is no significant coefficient for this variable. More specifically, this variable does not improve the model, as the R2 remains at 0,266, meaning that there has been no change in comparison with the first model. In fact, the adjusted R2 even indicates that the model decreased in its explanatory power, when compared with the Model 1. However, this model is not statistically significant. Probably, the influence of the variable is not high enough to be observable or the relationship could be troubled by other, unobserved variables. The control variables remain comparable to the observations of the first model, the only difference being that the coefficient for age also is significant (p<0,1). The tolerance statistics do not show any extraordinary values.

Model 3 contains all four independent Twitter variables. The only significant coefficient is the number of followers (p<0,1). The number of followers has the only significant effect on the dependent variable, with a positive coefficient (B=0,223, p<0,1). The control variables do not show major differences, except for age and internet use, both not being statistically significant anymore. Overall, the variables increased the model, with R2=0,302, which indicates an increase of 0,035. Still, the addition of the four variables only marginally improves the model and is not statistically significant. The tolerance statistics show higher values than before, with the highest VIF being 3,604.

(35)

significant coefficients. Only the control variables, again, show significant coefficients, comparable to those of the prior models. The highest VIF is only slightly higher (3,647) than in the model before, thus showing that there is still no tolerance problem.

Model 5 and Model 6 show similar results. Neither the moderating effect of followers nor that of reaction to tweets have significant coefficients. The model improves only marginally, gaining 0,2% and 0,5% in explanatory power. Moreover, the adjusted R2 indicates that Model 5 indeed leads to a decrease, while Model 6 does not affect the percentage of the data explained at all. The control variables again remain comparable to the prior models in terms of significance. One noticeable change in Model 6 however, is the significant positive coefficient of the number of followers. The tolerance statistics show almost no difference in Model 5. In turn, Model 6 shows a VIF value of 6,558, which according to Hoerl & Snee (2012) can be an indicator of multicollinearity, but is not yet a problematic value.

Finally, Model 7 offers some interesting findings. The independent Twitter variable Adoption shows a negative but significant (B≈-0,157, p<0,05). At the same time, the moderation variable of CSR and Adoption shows a positive and significant coefficient (B≈0,17, p<0,1). From all control variables, only GDP per capita and revenue have significant coefficients, similar in value to the prior models. Overall, Model 7 is significant at the 10% level and explains 32,5 % of the reputation scores. Yet, the adjusted R2 is lower, indicting explanatory power of 26,1%. The highest VIF in this model amounts to 6,617, which also is not yet a problematic value. The Durbin-Watson Statistic shows a value of 2,078, which lies between 1,5 and 2,5. Consequently, there is no first order linear auto-correlation in the regression data.

4.5 Hypothesis Assessment

(36)

(37)

Table 4: Multiple Linear Regression Analysis (Dependent Variable: Organizational Reputation)

(38)

5. Discussion

5.1 Implications

Even though Hypothesis 1 was rejected and Hypothesis 2 only partially confirmed, this thesis still offers practical implications for stakeholder management.

The regression analysis could not detect a direct, positive relationship between CSP and company reputation. This could be evidence, suggesting that CSP matters for organizational reputation far less than the general consensus. In fact, the control variables explained the biggest part of the variance of the reputation scores. As a result, if a company faces criticism for subpar CSR activities, the reputation might not suffer at all or just in a very limited way. One example to strengthen this idea is Adidas, who faced major criticism in 2000 after it became known that the company was involved in child labor (The Guardian, 2000). Nonetheless, 17 years later the company is still thriving and managed to recover from the scandal.

Still, this does not mean that CSR should be paid less attention to. On the contrary, it remains an important part of each business. However, practitioners should consider that CSR does not alone constitute corporate reputation. Instead, other factors, like authenticity and company identity seem to play an important role in terms of credibility (Dutot et al., 2016). If a company is not seen as genuine in the public’s eyes, it does not matter how socially responsible it is, as long as it is suspected for greenwash (Dutot et al., 2016; Lyon & Montgomery, 2013). Consequently, practitioners should ensure that the company’s activities indeed are in line with how the company would like to be perceived. This unity creates more trust and credibility (Du, Bhattacharya, & Sen, 2011).

(39)

The positive moderating effect of Adoption indicates that the longer a company was active on Twitter the higher its organizational reputation (even if the coefficient was small). Earlier adoption of new forms of media and ways of doing things, shows that a company is innovative. According to Olson, Slater, & Hult (2005:52) innovation orientation “indicates that the firm not only is open to new ideas, but also proactively pursues these ideas in both its technical and administrative domains”. This openness might be seen favorably by the public, as the company shows interest for change but also encourage easier, two-way interaction with its stakeholders. Thus, companies should not hesitate to adopt certain new technologies, especially those that facilitate and change common old ways of communication.

5.2 Limitations

This research suffered from various limitations. The measurements used to grasp a firm’s social media activity are purely quantitative, hereby completely neglecting the content posted on Twitter. This can bias the results in multiple ways.

First, this research did not differentiate between the various purposes, the company Twitter accounts are used for. Some organizations mainly utilize social media in order to provide quick and efficient customer service (Benthaus, Risius, & Beck, 2016; Lewellyn, 2002), as was the case for the sampled company Air France KLM Group. The majority of this company’s tweets was directed in response to specific customer requests, providing information about flight delays and general travel information. While this is highly valuable content for the customers in question, those tweets (while observable for the online community) did not draw much reaction from other customers, in terms of favorites and re-tweets. Thus, by only counting the re-tweet and favorites of the last 100 tweets, the purposes of those tweets have not been paid attention to. Apple depicts a similar case. While there exists a global company Twitter account (@Apple) that has around 700.000 followers, so far, this account has not issued a single tweet. However, the company has a separate account specifically for customer service. As it is the most active account, it has been used in the data analysis. This content bias distorts the statistical analysis, especially because Apple has one of the highest reputation scores.

(40)

to distribute information about their CSR activities was not part of this research. Another restriction might be the perceived authenticity of a company. According to Beckman, Colwell, & Cunningham (2009) authenticity is the main factor of successful and accepted CSR programs, because stakeholders constantly monitor transparency and consistency in a company’s CSR profile in order to assess whether or not the motives are intrinsically or extrinsically driven. Consequently, even if a firm would tweet information about its CSR related activities to its followers, if those followers do not believe the organization to be authentic, they would refrain from re-tweeting or favoriting their messages. Therefore, a company could have many followers, tweet regularly and be active on Twitter for a long time and communicate their relevant CSR activities, but still receive only a small number of favorites and re-tweets, if the customers are not convinced about the company’s authenticity and motivation. This is especially relevant in the context of so-called shitstorms, storms of outrage via online communication tools, often combined with insulting remarks (Duden, 2016). The open and interactive nature of social media leaves little room for mistakes from the company’s side, as criticism can be quickly proclaimed and shared with other customers and stakeholders around the world. According to Gruber, Smerek, Thomas-Hunt, & James (2015), nowadays it is not uncommon for reputational crises to start online, while other, offline, problems are brought to online platforms if not solved otherwise. They conclude that Twitter, as the most important micro-blogging platform, has huge real-time power, which can turn a local problem into a global reputation crisis very fast. Consequently, if a company experienced something such as this in recent times, it will not really matter how well it advertises its own social responsibility, as its authenticity might be tarnished.

(41)

For example, it could be possible that some companies communicate their CSR efforts mainly through Facebook, as opposed to Twitter, probably attracting more active reactions from stakeholders. In this case, a firm’s Twitter activity might not grasp its overall activity on social media. Also, this research focused on online communication only, neglecting “traditional” stakeholder communication tools.

In the context of the lacking interaction between the CSR scores and company reputation, CSR might make up only a small portion of overall company reputation. The regression analysis revealed that in fact the control variables explain the biggest part of the reputation scores. As such, there might be more neglected variables that play a bigger role for company reputation than CSP. In addition, Fortune’s reputation score equally treated 9 distinct dimensions, CSP being one of them. Even though the methodologies of assessment of those two variables are very different, this overlap might lead to the surprising, yet statistically insignificant results. Moreover, Pomering & Johnson (2009) revealed that company reputation can be affected by certain communication strategies, aspiring to create a picture and positive attitude of the firm in the eyes of the public. Differences in the efficiency of those strategies could thus bias the measured impact of CSR on reputation.

5.3 Future Research

(42)

(43)

6. Conclusion

The last decade has seen an increase in research on CSR (Falck & Heblich, 2007; Quazi et al., 2016) and social media (Roshan, Warren, & Carr, 2016). This thesis contributes to the existing academic discussion by combining those two constructs, thus responding to calls for papers (Journal of Information Processing and Management, 2015; Sustainability Science, 2015). This thesis revealed that both CSR and social media have pervaded all aspects of our nowadays social and business lives. The literature review showed that there is a huge body of research centering around the concepts of CSR and company reputation. Still, social media, due to its fairly current advent, remains fairly unexplored, especially in terms of business use. There is little research about how to actually determine a “decent” level of engagement from a company’s side, thus making it difficult to quantify. The research question underlying this thesis was: How does CSR affect organizational reputation and which role does social media play for this relationship? However, there is no clear answer to this question. The analysis showed that CSR does not have a significant effect. However, one of the four social media dimensions showed a significant, yet small moderating effect on organizational reputation, meaning that social media does indeed play a role for this relationship, at least to a certain extent. The first hypothesis suggested that higher CSP scores would mean higher reputation scores. This hypothesis was rejected. Neither the correlation analysis nor the multiple regression analysis revealed statistically significant results. The second hypothesis claimed that higher activity on social media would increase this impact on reputation. This hypothesis was partially accepted.

(44)

References

Agarwal, J., Osiyevskyy, O., & Feldman, P. M. 2014. Corporate Reputation Measurement: Alternative Factor Structures, Nomological Validity, and Organizational Outcomes.

Journal of Business Ethics, 130(2): 485–506.

Barnett, M. . 2007. Shareholder influence capacity and the variability of financial returns to corporte social responsibility. Academy of Management Review., 32(3): 794–816. Barth, M. E., Konchitchki, Y., & Landsman, W. R. 2013. Cost of capital and earnings

transparency. Journal of Accounting and Economics, 55: 206–224.

Beckman, T., Colwell, A., & Cunningham, P. H. 2009. The emergence of corporate social responsibility in Chile: The importance of authenticity and social networks. Journal of

Business Ethics, 86(SUPPL.2): 191–206.

Benthaus, J., Risius, M., & Beck, R. 2016. Social media management strategies for organizational impression management and their effect on public perception. The Journal

of Strategic Information Systems, 25(2): 1–13.

Boyd, D. E., McGarry, B. M., & Clarke, T. B. 2014. Exploring the empowering and paradoxical relationship between social media and CSR activism. Journal of Business Research. http://doi.org/10.1016/j.jbusres.2015.11.009.

Brammer, S., & Millington, A. 2005. Corporate Reputation and Philanthropy : An Empirical

Analysis European, 61(1): 29–44.

Cahan, S. F., Chen, C., Chen, L., & Nguyen, N. H. 2015. Corporate Social Responsibility and Media Coverage. Journal of Banking & Finance, 59: 409–422.

Cao, Y., Myers, J. N., Myers, L. A., & Omer, T. C. 2015. Company reputation and the cost of equity capital. Review of Accounting Studies, 20(1): 42–81.

Cao, Y., Myers, L. A., & Omer, T. C. 2012. Does Company Reputation Matter for Financial Reporting Quality? Evidence from Restatements. Contemporary Accounting Research, 29(3): 956–990.

Carroll, A., & Shabana, B. 2010. The Business Case for Corporate Social Responsibility:A Review of Concepts, Research and Practice. International Journal of Management

(45)

Caruana, R., & Chatzidakis, A. 2013. Consumer Social Responsibility (CnSR): Toward a Multi-Level, Multi-Agent Conceptualization of the “Other CSR.” Journal of Business

Ethics, 121(4): 577–592.

Castelló, A., & Ros, V. 2012. CRS communication through online social media. Revista Latina

de Comunicaciòn Social, 67(47–67).

Chen, C., & Delmas, M. 2011. Measuring Corporate Social Performance: An Efficiency Perspective. Production and Operations Management, 20(6): 789–804.

Choi, C. J., Eldomiaty, T. I., & Kim, S. W. 2007. Consumer trust, social marketing and ethics of welfare exchange. Journal of Business Ethics, 74(1): 17–23.

Christopher, M., & Gaudenzi, B. 2009. Exploiting knowledge across networks through reputation management. Industrial Marketing Management, 38(2): 191–197.

Colleoni, E. 2013. CSR communication strategies for organizational legitimacy in social media.

Corporate Communications: An International Journal, 18(2): 228–248.

Danilovic, M., Hensbergen, M., Hoveskog, M., & Zadayannaya, L. 2015. Exploring diffusion and dynamics of corporate social responsibility. Corporate Social Responsibility and

Environmental Management, 22(3): 129–141.

Dare, J. 2016. Will the Truth Set Us Free? An Exploration of CSR Motive and Commitment.

Business and Society Review, 121(1): 85–122.

Dawkins, C., & Ngunjiri, F. W. 2008. Corporate Social Responsibility Reporting in South Africa. Journal of Business Communication, 45(3): 286–307.

Deephouse, D. L. 2000. Media Reputation as a Strategic Resource: An Integration of Mass Communication and Resource-Based Theories. Journal of Management, 26(6): 1091– 1112.

Dijkmans, C., Kerkhof, P., & Beukeboom, C. J. 2015. A stage to engage: Social media use and corporate reputation. Tourism Management, 47: 58–67.

Dijkmans, C., Kerkhof, P., Buyukcan-Tetik, A., & Beukeboom, C. J. 2015. Online Conversation and Corporate Reputation: A Two-Wave Longitudinal Study on the Effects of Exposure to the Social Media Activities of a Highly Interactive Company. Journal of

(46)

Dobers, P. 2009. Corporate social responsibility: Management and methods. Corporate Social

Responsibility and Environmental Management, 16(4): 185–191.

Du, S., Bhattacharya, C. B., & Sen, S. 2007. Reaping realtionship rewards from corporate social responsibility: the role of competitive positioning. International Journal of Research in

Marketing, 24: 224–241.

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.

Du, S., Bhattacharya, C. B., & Sen, S. 2011. Corporate Social Responsibility and Competitive Advantage: Overcoming the Trust Barrier. Management Science, 57(9): 1528 – 1545. Du, S., & Vieira, E. T. 2012. Striving for Legitimacy Through Corporate Social Responsibility:

Insights from Oil Companies. Journal of Business Ethics, 110(4): 413–427.

Ducassy, I. 2013. Does Corporate Social Responsibility Pay Off in Times of Crisis? An Alternate Perspective on the Relationship between Financial and Corporate Social Performance. Corporate Social Responsibility and Environmental Management, 20(3): 157–167.

Duden. 2016. Shitstorm, der. http://www.duden.de/rechtschreibung/Shitstorm.

Dutot, V. 2013. A New Strategy for Customer Engagement: How Do French Firms Use Social CRM? International Business Research, 6(9): 54–67.

Dutot, V., Galvez, E. L., & Versailles, D. W. 2016. CSR communications strategies through social media and influence on e-reputation. Management Decision, 54(2): 363–389. Falck, O., & Heblich, S. 2007. Corporate social responsibility: Doing well by doing good.

Business Horizons, 50(3): 247–254.

Fieseler, C., & Fleck, M. 2013. The Pursuit of Empowerment through Social Media: Structural Social Capital Dynamics in CSR-Blogging. Journal of Business Ethics, 118(4): 759–775. Floreddu, P. B., Cabiddu, F., & Evaristo, R. 2014. Inside your social media ring: How to

optimize online corporate reputation. Business Horizons, 57(6): 737–745.

(47)

106.

Forbes Magazine. 2017. Volkswagen’s Net Dieselgate Spending In The U.S. Rises Above $20

Billion.

http://www.forbes.com/sites/greatspeculations/2017/01/19/volkswagens-net-dieselgate-spending-in-the-u-s-rises-above-20-billion/#4b20cc516d12.

Godfrey, P. C., Merrill, C. B., & Hansen, J. M. 2009. The relationship between corporate social responsibility and shareholder value: An empirical test of the risk management hypothesis.

Strategic Management Journal, 30(4): 425–445.

goodreads. 2016a. Milton Friedman Quotes. http://www.goodreads.com/quotes/240845-there-is-one-and-only-one-social-responsibility-of-business-to.

goodreads. 2016b. Warren Buffet Quotes. http://www.goodreads.com/quotes/148174-it-takes-20-years-to-build-a-reputation-and-five.

Greyser, S. a. 1999. Advancing and enhancing corporate reputation. Corporate

Communications: An International Journal, 4(4): 177–181.

Gruber, D. A., Smerek, R. E., Thomas-Hunt, M. C., & James, E. H. 2015. The real-time power of Twitter: Crisis management and leadership in an age of social media. Business

Horizons, 58(2): 163–172.

Hatcher, T. 2000. The Social Responsibility Performance Outcomes Model: Building socially responsible comapnies through performance improvement outcomes. Performance

Improvement, 39(7): 18–22.

Heugens, P. P. M. a R., Van Riel, C. B. M., & Van Den Bosch, F. a J. 2004. Reputation Management Capabilities as Decision Rules. Journal of Management Studies, 41(8): 1349–1377.

Hoerl, R., & Snee, R. D. 2012. Statistical thinking: improving business performance. John Wiley & Sons.

Java, A., Java, A., Song, X., Song, X., Finin, T., et al. 2007. Why we twitter: understanding microblogging usage and communities. International Conference on Knowledge

Discovery and Data Mining, 9.

The Effect of Company Level Social Media Use on the Influence of CSR on Organizational Reputation