1
The effect of green highlighting on the
credibility of companies in the apparel industry
An experimental study into the effectiveness of the self-serving and other-serving
CSR motives.
Student name: Annick Perotti Student number: 10572066
Study programme: MSc. Business Administration - Marketing track Institution: University of Amsterdam
Supervisor: L. Moratis
Date of submission: 26-01-2018 Version: Final thesis
2 Statement of originality
This document is written by student, Annick Perotti, who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no other sources than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Abstract
Green consumerism is trending and more and more companies invest in corporate social responsibility (CSR) these days. The use of CSR however could lead to greenwashing, which means that companies communicate to engage in environmental issues without adressing them in real actions (Walker & Wan 2011). Greenwashing could have negative financial results for the company and increases consumer’s skepticism. Recent literature however introduced the more nuanced form of green highlighting which is proposed as an effective way of communicating CSR to consumers. Nevertheless, the effectivenes of green highlighting is context specific. The aim of this study therefore is to expand the literature on the relationship between green highlighting and corporate credibility, by specifically focusing on the apparel industry. To discover how green highlighting can most effectively increase the credibility of companies, an explanatory research design is conducted. Consumer-company identification is included in this research as potential mediator and the factors consumer skepticism and sustainability orientation are the two moderators. Green highlighting is manipulated in this study and consists of two different variants, a self-serving CSR motive and an other-serving CSR motive. Data is collected by means of an online survey and the respondents are randomnly assigned to one of the two CSR motives. In total 218 valid respondents could be used for the analysis. The results show that the type of CSR motive does not result in any significant difference of corporate credibility or consumer-company identification. Some direct effects however are found. More skeptical consumers for example tend to evaluate the company as less credible and identify less strongly with the brand. In line with the literature moreover, consumers with a high level of sustainability orientation will show higher values of consumer-company identification. These findings are interesting for managers, to gain strategic benefits from CSR it is useful to understand how consumers respond to specific CSR actions.
4 Table of contents Statement of originality 2 Abstract 3 1. Introduction 6 1.1 General context 7 1.2 Research goal 7
1.3 Academic and managerial relevance 9
1.4 Structure 11
2. Literature review 12
2.1 Corporate credibility in the apparel industry 12
2.2 Corporate social responsibility 14
2.3 Self or other-serving CSR motives 15
2.4 Greenwashing 16 2.5 Green highlighting 18 2.6 Consumer-company identification 19 2.7 Consumer skepticism 21 2.8 Sustainability orientation 22 2.9 Conceptual model 24 3. Method 25 3.1 Research design 25 3.2 Manipulation 25 3.3 Data collection 26 3.4 Pre-test 27 3.5 Sample 29 3.6 Measurement of variables 30 3.7 Statistical procedure 32 4. Results 33 4.1 Preliminary steps 33
4.2 Validity and reliability 33
4.3 Correlation analysis 35
5
4.5 Hypotheses testing 37
4.5.1 Moderation analysis 39
4.5.2. Moderated mediation analysis 40
5. Discussion 42
5.1 Discussion of results 42
5.1.1 Effect of CSR motive on corporate credibility 43
5.1.2 Effects of CSR motive on consumer-company identification 44
5.1.3 Effects of the moderators 45
5.2 Implications 47
5.3 Limitations and future research 48
6. Conclusion 51
References 53
Appendix 60
Appendix A: Pre-test survey 60
Appendix B: Sample pre-test 2 67
Appendix C: Sample pre-test 2 68
Appendix D: Final survey 69
6 1. Introduction
Confused by all the ‘green’ claims of products on store shelves and in marketing advertisements? There is a good reason to be. Especially in the apparel industry it seems that every brand incorporates some kind of social or environmental awareness into their strategy these days. Consumers are overwhelmed by the amount of corporate social responsibility messages and are not sure what to believe (Arrigo, 2013).
Retailer H&M for example launched World Recycle Week on 18 April in 2016 (Siegle, 2016). The plan was to capture 1.000 tons of unwanted garments during the week. H&M created new textile fibre out of the clothes and the consumers got vouchers to use at H&M. The retailer marketed this event as the company’s biggest sustainability campaign yet. It sounds like a win-win situation, but in reality this plan is quite contradictory. H&M would need 12 years to recycle 1.000 tonnes of old garments, while it produces the same amount of clothes in only 48 hours (Siegle, 2016). On top of that, the vouchers often fuel more purchasing of clothes.
This practical example shows the increasing problem of greenwashing these days. Greenwashing refers to the promotion of green-based environmental initiatives of companies without the implementation of practices that actually minimize the societal or environmental impact (Delmas & Burbano, 2011). This study will focus on the more nuanced form of green highlighting, which does not include misrepresentation but selective representation of information (Walker & Wan, 2011).
In this chapter the general context will be illustrated, followed by an explanation of the specific goal of this research. The third subparagraph will explain the managerial and academic relevance and the introduction chapter ends with an illustration of the structure of the rest of the research.
7 1.1 General context
Today’s marketplace is characterized by many products of similar service, quality and price (Brønn & Vrioni, 2001). To meet consumer expectations and to differentiate themselves and their products, companies invest more in corporate social responsibility (CSR) nowadays (Kotler, 2011). Companies investing in environmental management however could sometimes lead to greenwashing. This happens when a company communicates to be environmentally friendly, but in fact the practices are not as green and sustainable as pretended (Lyon & Montgomery, 2015; Walker & Wan, 2011). Companies are misleading consumers by means of greenwashing, which could have an impact on their confidence in the firm (Chen & Chang, 2012).
Greenwashing however is an umbrella term for behaviors that induce consumers to hold overly positive views of a company’s environmental performance (Lyon & Montgomery, 2015). Green highlighting for example is a relatively new and nuanced form of greenwashing introduced by Walker and Wan (2011) that includes a combination of both symbolic and substantive actions. The firm discusses environmental responsibility in terms of what they are doing currently or have done (substantive action), and what they plan to do in the future (symbolic action). Green highlighting is not necessarily a bad thing, because compared to greenwashing it is a true and fair representation of actual company performance (Walker & Wan, 2011). Since the study of green highlighting is still young and the effectiveness of a CSR communication strategy is context specific, more research remains to be done.
1.2 Research goal
To build upon the article of Walker and Wan (2011) the aim of this research is to further investigate the relationship between green highlighting and the credibility of companies. Two different CSR motives are included within this research that companies can use to communicate
8
their sustainable message. Becker-Olsen and Hill (2005) provided two primary motives in how consumers examine a firm’s CSR, either serving or other-serving. By means of a self-serving CSR motive companies tend to focus on their own goals and the benefits for the consumer within advertising. While with the use of other-serving CSR companies focus more on benefits with regard to other parties, such as the environment.
Furthermore, three different variables are investigated within this research that might influence the relationship between green highlighting (self- or other-serving) and corporate credibility. First of all, the level of sustainability orientation of the consumer is taken into account. The degree to which individuals feel responsible for the environment and society might affect the effectiveness of green highlighting (Delmas & Burbano, 2011). Secondly, consumers expect more CSR from companies but at the same time are more skeptical about their use of CSR in advertisements (Brønn & Vrioni, 2001). The level of skepticism of the consumer therefore might also be relevant to include. Finally, consumer-company identification could mediate the relationship between green highlighting and the credibility of the firm. Highly identified consumers are likely to feel committed to the firm and tolerate corporate failures and misbehavior up to a certain point (Einwiller, Fedorikhin, Johnson & Kamins, 2006).
Since the effectiveness of green highlighting is context-specific, this research will expand the concept by focusing on the apparel industry. The apparel industry within this research includes companies that design, manufacture and market brands for men’s and women’s clothing (Kim & Hall, 2015). This industry has faced public criticism due to lack of commitment to CSR and would therefore be interesting to further investigate. Fashion moreover involves waste and short life cycles of products, but at the same time brands try to convince consumers of their sustainability (Morgan and Birtwistle, 2009).
This study focuses on the CSR communication strategy green highlighting. An experimental study is conducted with the use of a questionnaire, in which the independent
9
variable ‘green highlighting’ is manipulated. Two different versions of the questionnaire are distributed, one with a self-serving CSR motive and another version concerning an other-serving CSR motive. The main research question that is central in this research is as follows: ‘How can organizations influence the perceived credibility of the firm by means of green highlighting?. To help answering the research question as described, the following two sub questions are formulated:
1. What is the impact of the type of CSR motive (self- or other-serving) on the relationship between green highlighting and corporate credibility? And does it influence the mediating effect of consumer-company identification?
2. How do the level of sustainability orientation and skepticism of the potential consumer influence the relationship between green highlighting and corporate credibility? And do these moderators affect the relationship between green highlighting and consumer-copmany identification?
1.3 Academic and managerial relevance
Corporate social responsibility (CSR) has become more and more important for companies in the last years. Consumers are aware of the climate change and companies have to take their responsibility. Business leaders believe that CSR is an economic imperative in today’s global marketplace (Murray and Vogel, 1997). Because of the increasing emphasis on CSR in the marketplace, more research focused on the effects of CSR actions on consumers in the last years (Sen & Bhattacharya, 2001). Earlier research suggests that there is a positive relationship between a firm’s CSR actions and consumers’ attitudes toward that company (Brown and Dacin, 1997; Ellen, Mohr and Webb, 2000). It is important for marketers to understand how
10
and why various customer segments are likely to respond to specific CSR actions to reap strategic benefits from CSR (McGee, 1998).
Together with the rise of the importance of corporate social responsibility more research paid attention to potential side effects, such as greenwashing. More firms have been combining poor environmental performance with positive communication about environmental performance (Dahl, 2010). Delmas and Burbano (2011) for example examined the external, internal and organizational drivers of greenwashing and offered recommendations for managers and policymakers. Because of the growth of green business practices it might also be interesting to investigate the effect of more nuanced forms of CSR communication on consumer’s attitudes, such as green highlighting.
Walker and Wan (2011) argue that green highlighting is more beneficial compared to other methods of CSR communication. But since the concept of green highlighting is relatively new, current literature find inconsistent results on how consumers respond to it. Companies and academics have been looking for the holy grail of CSR communication ever since the rise of green marketing, but the question of how to effectively communicate it to stakeholders remains unanswered (Du, Bhattacharya & Sen., 2010). More research within the concept of green highlighting is needed to see how managers can gain strategic benefits from it. The effect of green highlighting moreover is expected to be context specific (Walker & Wan, 2011). By focusing on the apparel industry this research expands the literature regarding the CSR communication methods. The apparel industry is interesting because of its conflicting nature, offering low life cycle products while claiming to be sustainable. On top of that, distinguishing green highlighting into two CSR motives to test the credibility of the firm provides a new angle in academic literature.
11 1.4 Structure
In order to answer the research question thoroughly, the remainder of this study is structured as follows. First, a literature review will explore the field of CSR, greenwashing, green highlighting, consumer-company identification, sustainability orientation and consumer skepticism. Hypothesis will be formulated and the theoretical framework results in a conceptual model that represents the interrelations among the key constructs. Subsequently, in chapter three the data collection and research method is outlined. Followed by a representation of the results in chapter four. The results are discussed in the next section, together with the implications and limitations of this study and suggestions for future research. At last, in chapter six the final conclusions will be drawn.
12 2. Literature review
The literature review starts broadly with information about corporate credibility, which is applied to the apparel industry for this study. Subsequently, information about the importance of corporate social responsibility (CSR) is illustrated. Thereafter the two different CSR motives are introduced that companies can use within advertisements, the self- and other serving CSR motives. The fourth subparagraph contains information about why CSR could sometimes lead to greenwashing, followed by an explanation of the nuanced concept ‘green highlighting’ that is central in this research.
In the next paragraph, the three variables that influence the relationship between green highlighting and the credibility of the firm are introduced. Starting with an explanation of the mediating effect of consumer-company identification. Thereafter the first moderator consumer skepticism is illustrated, which is closely related to green highlighting. Followed by a clarification of the second moderator sustainability orientation. The literature review ends with the presentation of the conceptual.framework.
2.1 Corporate credibility in the apparel industry
Corporate credibility contributes to the overall reputation of the firm and holds that a company can be relied to do what it promises (Goldsmith, Lafferty & Newell, 2000). The credibility of the company influences the purchase intention of the consumers and their attitudes towards the advertisement and the brand. Credibility could also be explained as the extent to which consumers believe in the company’s expertise and trustworthiness (Erdem, Swait & Louviere, 2002).
Overall, corporate social responsibility (CSR) initiatives enhance consumers’ perceptions of credibility of organizations (Maignan & Ferrell, 2004). Skepticism about marketing actions, however can increase distrust and decrease the credibility of the firm (Brønn
13
& Vrioni, 2001). Since credibility is based on an organization’s intention, actions will either confirm or disconfirm beliefs in the firm (Herbig & Milewicz, 1993). Mixed signals tend to decrease credibility and inconsistent mixed signals can even erode a firm’s reputation. Especially within the apparel industry, credibility has become more important for brands. The apparel industry is the second largest polluter in the world after the oil industry. An increasing number of apparel brands however is placing environmental sustainability at the center of their strategy (Kim & Hall, 2015). Zara for example launched eco-efficient stores recently, Puma came up with a biodegradable InCycle collection and H&M introduced their Concious collection (Arrigo, 2013).
The largest group of consumers in this sector, the millenials, seem to care more about the environment than previous generations (Fernandez, 2016). The green credibility of fashion however is often paradoxical. Susan Scafidi, founder of the Fashion Law Institute at Fordham, states: "We replace clothing for style reasons rather than reasons of necessity and so fashion does have a credibility gap when it comes to any kind of environmental responsibility claims."
While fashion involves waste and short life cycles of products, green promotes recycling of products and sustainability (Morgan and Birtwistle, 2009). Decisions related to fashion products moreover are associated with social acceptance, individual preferences as well as evaluation of quality, value and sustainability (Kim & Hall, 2015).
The apparel industry has also faced some scandals in history, which have had an impact on their reputation. Some companies were negatively presented in the media because of outsourcing production to the world’s lowest wage economies and exploitation of human capital (Arrigo, 2013). The athletic footwear and apparel brand Nike for example was accused of reported abuse and unfair wages in international factories.
14 2.2 Corporate social responsibility
Corporate Social Responsibility (CSR) has become more important for business. One-third of Americans say that after price and quality, a company’s responsible business practices are the most important factor in deciding whether or not to buy a brand (Brønn & Vrioni, 2001). Where CSR started as a business tool used by a few organizations to distinguish themselves, nowadays it is a core part of the business strategy of many organizations. Organizations relate CSR goals to their own values and competences (Van de Ven, 2008). CSR has become part of modern companies’ activities because governments, consumers and other stakeholders are increasingly interested in the contributions companies should and can make to society (Skarmaes & Leonidou, 2013).
CSR has not only become more important for companies itself, but also emerged in recent years as an important academic construct (Sen & Bhattacharya, 2001). There are many definitions of CSR in literature.According toWerre (2003) CSR can be defined as the strategic choice of the organization to take responsibility for the impact of its business on environmental, societal and economic dimensions. Recent research suggests that there is a positive relationship between a company’s CSR actions and consumers’ attitudes towards the firm (Sen & Bhattacharya, 2001). Due to these positive outcomes of CSR, it is often used by companies to develop or recover the organizational reputation.
To gain the benefits of CSR the company should communicate its CSR initiatives effectively. Benefits can be realized when companies communicate the CSR motives, the organizational CSR commitment and how the company can have a positive influence on the society and environment (Du, Bhattacharya & Sen, 2010). The level of skepticism of consumers plays an important role in this case. The impact of CSR campaigns depends on how consumers perceive the motives for a company’s involvement in cause-related programmes. Illia et al. (2013) propose that it is important to communicate and implement CSR throughout the entire
15
organization, irrespective of the department. Multiple marketing communications channels can be used to communicate CSR initiatives, such as TV commercials, product packaging, social media, annual reports, a CSR section on the corporate website etc. (Du, Bhattacharya & Sen, 2010).
2.3 Self or other-serving CSR motives
One key challenge in CSR communication is to reduce stakeholder skepticism (Brønn & Vrioni, 2001). While most CSR messages focus on a firm’s involvement in social causes, there are several other facts that a company can emphasize. For instance, a firm could point out why it engages in particular social initiatives, also referred to as the CSR motives (Du, Bhattacharya &.Sen,.2010).
Becker-Olsen and Hill (2005)defined two ways in which consumers examine a firm’s CSR communication; self-serving and other-serving CSR. With the use of self-serving CSR motives firms are egoistic and focused on increasing the organization’s profit (De Roeck & Delobbe, 2012). This motive typically belongs to the classical CSR perspective, in which firms especially pay attention to economic and legal responsibilities (Pinkston & Carroll, 1996). Nowadays ethical responsibilities become more important and according to the modern CSR perspective organizations focus more on other-serving CSR motives. The main goal is no longer to increase profit, but doing business without harming the society and environment. Emphasizing the welfare of the stakeholder has become essential for organizations (Maon, Lindgreen & Swaen, 2009).Maignan, Ferrell and Ferrell (2005) describe other-serving CSR as a stakeholder-oriented subject. In other words, the situation in which firms have to adopt values and norms along with the organizational processes to minimize their negative impact and maximize their positive impact on important issues that stakeholders face.
16
own interest (Du et al. 2010). Elving and van Vuuren (2011) state that stakeholders perceive the firm as profit-driven when it emphasizes self-serving CSR motives, which affects the credibility of the firm with regard to their CSR activities. Consumers’ perceptions of the underlying motivations for a company’s action drive their evaluations and impact attitudes, beliefs and intentions (Becker-Olsen, Cudmore & Hill, 2006).
Literature confirms that the use of other-serving CSR motives leads to more positive outcomes than the use self-serving CSR motives (Du et al, 2010, Elving & van Vuuren, 2011, White & Peloza, 2009). Other-serving CSR motives are often value-driven and designed to help others, while self-serving CSR motives primarily focus on maximising profit for the firm itself. According to White and Peloza (2009) other-serving appeals are more effective in encouraging donation intentions than self-benefit appeals, when public self image concerns are heightened. Forehand and Grier (2003) however argue that consumers sometimes are willing to accept self-serving motives in CSR communication as long as they perceive the company as sincere in serving public interests. Finally, the two motives can also can be combined. A mixed strategy of intrinsic and extrinsic CSR may be most beneficial to a company according to Ellen, Mohr and Web (2006).
2.4 Greenwashing
Corporate claims about environmental performance have increased rapidly in recent years (Christensen, Morsing & Thyssen, 2013). Companies now take an active role in
environmental management, but this sometimes leads to greenwashing. Greenwashing means that companies adopt to engage in symbolic communications of environmental issues without substantially addressing them in actions (Walker & Wan 2011).For some organizations it is much easier to claim to be environmentally and socially responsible than actually
17
communication that misleads people into forming overly positive beliefs about organization’s environmental practices or products (Lyon & Montgomery, 2015).In summary, the
organization’s CSR policy is not supported by the daily organizational actions.
Firms are able to greenwash because there is an information asymmetry between the organization and the stakeholders. Sustainability is regarded as a credence good, because consumers are not able to directly evaluate the environmental or societal performance of a company(Lyon & Montgomery, 2015). Stakeholders have to rely on the firm’s messages and therefore cannot make fully informed decisions. Another reason for greenwashing is ambiguity around the concept of ‘sustainability’ (Bowen & Aragon-Correa, 2014). There are many different ways in which firms incorporate sustainability in their businesses which confuses the consumers. Lastly, the pressure from the industry and governments to improve the firm’s sustainable image also increased the use of greenwashing by firms (Delmas & Burbano, 2011). The questions is whether inconsistency between words and action is harmful to the company. Most writings treat lack of consistency between organizational CSR talk and action as a serious problem, but other researchers argue that this hypocrisy is not necessarily a bad thing (Christensen, Morsing & Thyssen, 2013). As corporate green claims have increased in the last years, consumers have become more skeptical about their authenticity. The inconsistency between words and action could challenge the moral standards of organizations and could negatively influence their brand image (Lyon & Montgomery, 2015). Greenwashing is often related to lying and misleading and could negatively affect consumers’ identification with the organization. Moreover, greenwashing can diminish employees’ organizational trust (Cheng.&.Chang,.2012).
Christensen et al. (2013) however argue that discrepancies have the potential to stimulate CSR improvements. Misalignment between words and actions could improve the organization’s CSR goals, standards and practices. CSR as aspirational talk could motivate
18
employees to implement the CSR strategies and fulfill the organization’s promises. In this way, the discrepancy between green talk and walk could be seen as an important driver to create organizational change. Brunsson (2003) moreover argues that conflicting values and ideas allow organizations to address different audiences in different ways.
2.5 Green highlighting
Whereas greenwashing mostly has a negative appearance, there are also milder forms of CSR communication strategies that might not necessarily be badly perceived by consumers (Lyon & Montgomery, 2015).The relatively new concept ‘green highlighting’ is introduced by Walker and Wan (2011). It represents the combination of symbolic and substantive actions, where the firm discusses environmental responsibility in terms of what they are doing currently or have done (substantive action), and what they plan to do in the future (symbolic action). Green highlighting is different from greenwashing, because this strategy is not about disinformation but about information that is not backed by substantive actions. Green highlighting is stressing particular CSR achievements while deliberately playing down others. It does not include misrepresentation but selective representation of information (Walker & Wan, 2011).
Walker and Wan (2011) break down environmental performance into symbolic and substantive actions, and find that only symbolic actions affect firms financially in a negative way. Symbolic actions are especially harmful in visibly polluting industries, such as the apparel industry investigated within this research. Firms in these industries are subject to increased monitoring and greater stakeholder pressures. These symbolic actions could be harmful if green talk is not backed up with any green walk, but could be effective when performance is difficult to measure (Christmann & Taylor, 2006).
Literature furthermore indicates that greenwashing and green highlighting have different possible consequences. According to the research of Walker and Wan (2011),
19
greenwashing (discrepancy between green talk and green walk) has a negative effect on financial performance and green highlighting (concentrated efforts of the talk and walk) has no effect on financial performance. This is because green highlighting is not considered as deliberately fooling the public, while greenwashing is mostly associated with feelings of distrust and betrayal. Schons and Steinmeier (2015) also found that greenwashing impacts the firm’s performance negatively, but they concluded that green highlighting can even positively affect firm performance. These authors took the proximity level of the stakeholders into account and suggested that misleading low-proximity stakeholders can actually pay off for firms. Since these stakeholders have limited ability to distinguish between firm’s symbolic and substantive CSR actions. According to the findings of previous research, this study hypothesizes the following:
H1: Green highlighting has a direct positive relationship with corporate credibility, this
positive relationship is stronger for the use of an other-serving CSR motive compared to the use of a self-serving CSR motive.
2.6 Consumer-company identification
Green highlighting not only affects the credibility of companies, but can also influences the level of identification between the consumer and the company. Sen and Bhattacharya (2001) argue that some of the strongest consumer-company relationships are based on consumer's identification with the company that help them satisfy their self-definitional needs. Consumer-company identification can be defined as the degree to which consumers feel connected to the firm, which occurs when both parties share the same values (Einwiller, Fedorikhin, Johnson & Kamins, 2006). When consumers strongly identify with a company, they are likely to feel committed to the firm and become more loyal.
20
According to the social identity theory, people tend to classify themselves and others into various social categories to be able to define or locate themselves in the social environment. Social identification is the perception of belongingness to a group of persons which leads to activities that are congruent with the identity (Asforth & Mael, 1989). People also seek out firms for identification purposes, even when they are not formal organizational members (Scott & Lane, 2000). Consumers are likely to find a company’s identity more attractive when it matches their own values, identification moreover causes people to become psychologically attached to the company (Sen & Bhattacharya, 2001).
Negative publicity however has the potential to create negative corporate associations, because negative information is given greater weight by consumers compared with the same amount of positive information (Klein & Dawar, 2004). But the level of consumer identification with a company might moderate the impact of this effect. When the firm is associated with negative publicity, strongly identified consumers try to maintain a positive view of the company and protect their beliefs. Therefore these consumers dismiss the negative publicity and process the information in a biased manner. But this happens up to a certain point, when the information is extremely negative even highly identified consumers will not tolerate it any more (Einwiller, Fedorikhin, Johnson & Kamins, 2006).
According to the motivated reasoning theory, people can be motivated to either arrive at an accurate conclusion or at a desired conclusion when processing information (Kunda, 1990). When consumers are motivated to reach a particular conclusion, they tend to evaluate the information in a biased manner. Strong consumer-company identification therefore stimulates the motivation to protect positive associations with the company. When consumers are strongly identified with the company it is expected that they will evaluate the company as more credible. Based on this information, the following is hypothesized:
21 H2: Green highlighting has a positive indirect relationship with corporate credibility
mediated by consumer-company identification.
• H2a: Green highlighting has a positive relationship with consumer-company
identification, so that this relationship is stronger for the use of an other-serving CSR motive compared to the use of a self-serving CSR motive.
• H2b: Consumer-company identification has a positive relationship with
corporate credibility.
2.7 Consumer skepticism
As stated earlier, companies engage in CSR initiatives more often nowadays to make a positive contribution to society and to support their strategic goals. At the same time, the increased use of CSR initiatives leads to a rise in consumer skepticism. Consumers doubt the extent to which companies live up to their standards and fear that CSR is just a tool that firms use to manipulate (Skarmeas & Leonidou, 2013).
In general skepticism refers to an individual’s tendency to disbelieve, doubt and question. In this study the term is used to refer to distrust or disbelief of green marketer actions (Forehand & Grier, 2003). Skepticism is conceptualized as a stable belief that increases distrust of marketing communications (Obermiller & Spangenburg, 1998).
Skepticism occurs when consumers suspect that the company is acting in its own interest and not regarding sincere principles (Du et al., 2010). Skepticism towards CSR activities mainly derives from distrust and cynicism towards advertising. Therefore, it is fundamental for companies to know how consumers perceive the brand and what they expect in return for their support (Brønn & Vrioni, 2001). It is expected that consumers with a high level of skepticism will be less likely to respond positively to CSR campaigns. This leads to the following hypotheses:
22 H3: The positive relationship between green highlighting and corporate credibility is
moderated by consumer skepticism, so that this relationship is stronger for
participants with a low level of skepticism compared to a high level of skepticism.
H4: The positive relationship between green highlighting and consumer-company
identification is moderated by consumer skepticism, so that this relationship is stronger for participants with a low level of skepticism compared to a high level of | skepticism..
2.8 Sustainability orientation
Consumer’s interest in sustainability has emerged over the last decades. They feel responsible for the environment and society and express these feelings in terms of their purchase behavior (Schuhwerk & Lefkoff-Hagius, 1995). Consumers who were more highly environmentally involved tended to be more critical to green advertisements (D’Souza & Taghian, 2005). Newell (2008) also found that people that are highly concerned with the environment, have the highest level of awareness about the content of corporate brand claims. The level of commitment to the environment thus influences the way in which CSR communication is perceived. According to the elaboration likelihood model, involvement of a person influences the processing of CSR communication (Petty, Cacioppo & Schumann, 1983). When consumers are both motivated and able to process the information presented about the product they will follow the central route. Behavior and attitudes developed through the central route are relatively persistent over time and more predictable. But, when consumers follow simple decision rules that allow them to evaluate the advocacy quickly these people process information according to the peripheral route. This route requires less mental effort and is strongly influenced by
23
context and peripheral cues. Vermeir and Verbeke (2006) also claim that the more involved the consumer is, the more positively they will evaluate a green service or product. Consumers who are less involved, have fewer concerns about the environment and therefore will be less critical about CSR initiatives.
Processing information and the resulting evaluations depend on the personal values of consumers (Bigné-Alcañiz, Currás-Pérez & Sánchez-García, 2009). Sen and Bhattacharya (2001) moreover argue that consumer’s personal investment in a CSR message could moderate their responses to CSR. Being involved with sustainability-oriented topics could lead to further elaboration of the message, which could influence the relationship between consumer-company identification and the perceived credibility of the firm. The following hypotheses are formulated accordingly:
H5: The positive relationship between green highlighting and corporate credibility is
moderated by sustainability orientation, so that this relationship is stronger for participants with a high level of sustainability orientation compared to a low level of sustainability orientation..
H6: The positive relationship between green highlighting and consumer-company
identification is moderated by sustainability orientation, so that this relationship is stronger for participants with a high level of sustainability orientation compared to a low level of sustainability orientation..
24 2.9 Conceptual model
To expand the literature regarding the effectiveness of green highlighting, the conceptual model as shown in figure 1 is researched. All the six hypotheses introduced in the previous paragraphs are illustrated in this model. This research will make a distinction between two different types of CSR motives in which green highlighting can be presented; self- and other serving CSR.
The level of perceived consumer skepticism and sustainability orientation of the consumer are taken into account as potential moderators. And the degree of consumer-company identification is included as possible mediator in the relationship between green highlighting and corporate credibility.
Figure 1: Conceptual model
Sustainability orientation Consumer-Company identification Green highlighting 1. Self-serving CSR motive 2. Other-serving CSR motive Corporate credibility Consumer skepticism H1(+) H2b(+) H3(-) H4(-) H5(+) H6(+) H2a(+)
25 3. Method
This chapter represents the start of the empirical part of this study. First an explanation of the research design used in this study will be given, followed by a description of the manipulation and data collection. Thereafter, an explanation of the pre-test that is conducted to check for inconsistencies in the materials. Next, the characteristics of the collected sample will be presented. Followed by the measurements of the included variables in the questionnaire and the corresponding reliabilities. In the last methodology paragraph a brief description is given of the statistical approach that is taken in order to test for the expected relationships as discussed in the previous chapter.
3.1 Research design
To discover how green highlighting can most effectively increase the credibility of companies, an explanatory research design is conducted. The aim of this type of research is to discover “the extent to which two or more variables covary, that is, where changes in one variable are reflected in changes in the other” (Creswell, 2008, p. 358). The independent variable green highlighting is manipulated in this study and consists of two different variants, a self-serving CSR motive and an other-serving CSR motive. This research design therefore fits best with the purpose of the study, to discover which type of CSR motive is most effective. A quantitative experimental study is conducted to answer the proposed research question by means of an online survey.
3.2 Manipulation
Two different versions of the webpage of the fictitious brand ‘Chestnut’ are created, one with a self-serving CSR motive and the other version concerns an other-serving CSR motive. The website contains information about the values and sustainability goals of the brand ‘Chestnut’
26
and respondents are randomly assigned to one of the two webpages. Based on the research of Becker-Olsen and Hill (2005) the webpage including the self-serving CSR motive focuses on the goals and benefits for the company and consumer itself. While the other-serving CSR story presents more information about the benefits with regard to other parties, such as the environment.
Participants were informed about the apparel brand ‘Chestnut’ through a short introduction text, the company pays attention to corporate social responsibility and a sustainable production. It is also mentioned that ‘Chestnut’ is a fictitious apparel brand, the webpage is based on examples from reality but the firm itself does not exist. A non existing brand is included to prevent that participants have perceptions upfront that could influence the results. The webpage created for this research is based on different webpages of existing apparel brands, such as Kuyichi, G-Star and Levi’s. The lay-out of the webpage should look professional but not distract from the content. The manipulation was checked by means of two pre-tests, which will be explained in the next paragraph. Participants were asked to what extend the advertisement focused on advantages in favor of the company itself or for the society as a whole.
3.3 Data collection
Data is collected by means of a survey and administered online from November 22th till December 4th 2017. The questionnaire is distributed via social media, i.e. Facebook, LinkedIn and e-mail. The survey starts with a short introduction text to inform the respondents on what they can expect from this questionnaire. The broad purpose of this study is mentioned as ‘corporate social responsibility’ and how companies can most effectively respond to this trend. The term ‘green highlighting’, which is the real subject of this research, is not mentioned to avoid distrust. After this introduction respondents are asked to indicate to what extent they agree with statements about sustainability and the environment. The question thereafter examines
27
how skeptical the respondents are with regard to sustainability statements in advertisements. All the questions within this survey could be answered on a 7-point Likert scale varying from 1 (strongly disagree) to 7 (strongly agree).
In the middle of the survey one of the two versions of the website of the fictitious apparel brand ‘Chestnut’ is shown to the respondents. The remaining questions are related to the information provided on the webpage. First, to minimize careless responses a control question is added to the survey to see if the participants have really looked at the information on the webpage. Thereafter the respondents have to indicate to what extent they are able to identify with the brand ‘Chestnut’ and how credible they perceive this company to be. The questions related to the webpage do not differ among the two versions, only the green highlighting content of the webpage is manipulated. The survey ends with some general questions regarding basic demographics of the respondents, namely age, gender and education level.
3.4 Pre-test
Before distribution of the survey a pre-test is performed to check for inconsistencies in the materials. All the items that need to be tested according to the conceptual framework of this research are included and at the end of the survey the respondents could indicate how they would rate this questionnaire (see appendix A for the first version of the survey).
The first pre-test consisted of 17 participants from the personal network of the researcher (Appendix B). The participants were randomly assigned to either the self-serving or other-serving CSR webpage via Qualtrics. Questions concerning the clarity and structure of the statements and webpage were included. Respondents also had to indicate how their perceived the information presented on the webpage, in the interest of the company itself (self-serving) or within a broader perspective to make it a better world (other-serving). Furthermore,
28
participants were asked via which device they filled in the survey, via laptop/computer, mobile phone or tablet. At the end of the survey there was room for specific feedback.
Based on the participants recommendations small changes were made regarding the webpage and statements used in the survey. First of all, the difference between self- and other-serving information had to more explicit. Fourteen out of the total of seventeen respondents in the test audience (82%) assigned the correct CSR motive to the webpage. This percentage needs to be improved for the final version of the survey. Furthermore, some participants were confused by the lay-out especially when they filled in the survey on a mobile device. It should be more clear that the information in the middle of the page is important to read. A lot of participants also mentioned that the structure of the survey and the aim of the fictitious webpage was not clear to them. To increase the amount of information a short paragraph is included on the webpage, called ‘about us’. On top of that, within the introduction text above the webpage it is emphasized that the company ‘Chestnut’ does not really exist. Participants have to pretend like ‘Chestnut’ is an existing company and indicate to what extent they agree with the remaining statements. During the pre-test it was not clear enough for participants that ‘Chestnut’ is just made up for this research. Furthermore some statements are slightly changed into a more nuanced form. For example: ‘I have a sense of connection with Chestnut’ is changed into ‘I could have a sense of connection with Chestnut’, which makes it apparently easier for participants to answer the question.
After the first pre-test a second pre-test was conducted with 8 participants (Appendix C). An additional test was used to see whether the adjustments made the survey more clear, before sending out the final version. The rates derived from the second pre-test were more positive, all the respondents were able to assign the right CSR motive to the webpage. The use of more nuanced items worked better according to the test panel. The survey is now ready for further distribution (see appendix D for the final version of the survey).
29
3.5 Sample
The population of interest in this research consists of Dutch consumers. Since this is a large population and the sampling frame is unknown, this research is based on a non-probability convenience sample. A snowball sampling technique was used to recruit respondents within the researchers network. The participants in turn shared the survey within their own network and in this way as many people as possible were reached. In total 297 respondents filled in the survey online via Qualtrics. For 63 respondents however more than 10% of the information was missing and therefore could not be included in this research. In the end 234 respondents completely finished the survey.
Furthermore, a control question was included within the survey to see whether the respondents have really looked at the information on the webpage. Another 16 respondents answered the question in the wrong way. Either they have not understood the question or they were not able to answer the question because they have not carefully read the information on the webpage. These respondents had to be eliminated from the dataset. In the end 218 respondents (N=218) are used for further analysis, 109 within the firm-centered condition (self-serving CSR motive) and 109 within the other-centered condition (other-(self-serving CSR motive).
In table 1 on the next page the demograpic characteristics of the respondents are shown. Most of the respondents are between 21 and 30 years old (34.6%), but the distribution of the different age categories is small (Table 1). The average age is between 41 and 50 years old (M
age= 4.11, SD age=1.455) and 57.7% of the respondents is female. The majority of 35.9% of the
sample completed HBO as the highest education level. Followed by respondents with a Bachelor’s degree (14.5%) or Master’s degree (17.5%).
30 Table 1: Demographic characteristics of respondents
Item Characteristics Frequency Percentage
Gender Woman Men 135 99 57.7% 42.3%
Age Below 20 years old
21-30 years old 31-40 years old 41-50 years old 51-60 years old 61-70 years old 71-80 years old 81 years and older
26 81 29 48 42 6 1 1 11.1% 34.6% 12.4% 20.5% 17.9% 2.6% 0.4% 0.4%
Highest education Primary school
Secondary school MBO HBO University (Bachelor) University (Master) 9 32 34 84 34 41 3.8% 13.7% 14.5% 35.9% 14.5% 17.5% CSR Motive Firm-centered Other-centered Missing 109 109 16 46.5% 46.5% 6.8% 3.6 Measurement of variables
The questionnaire is developed in English since all the items derive from earlier research validated in this language. The final version of the survey however is translated into Dutch, to avoid misunderstandings by respondents due to unfamiliarity with the English language. The translations of the statements are checked by a native English speaker using the translation-back-translation procedure (Brislin, 1980). For all the items validated Likert scales will be used on a 7-point scale at interval level (1 = completely disagree, 7 = completely agree). See table 2 on the next page, for a complete overview of the items used in the survey.
Ìn the first section of the survey the two moderators of this research will be tested. First the validated four item scale by Schuwerk and Lekoff-Hagius (1995) will be used to measure the sustainability orientation of the participants. This measure produced a Cronbach’s alpha of .90, which implies adequate internal consistency and reliability. Secondly, the level of
31
perceived consumer skepticism is measured by the four item scale of Mohr, Eroǧlu & Ellen (1998)with a reported Cronbach’s alpha of .79.
After the presentation of the webpage of Chestnut, the level of consumer-company identification of the participants will be measured by using the validated seven item 7-point Likert scale by Einwiller, Fedorikhin, Johnson and Kamins (2006)that has an Cronbach’s alpha of .92. The seven statements derived from earlier research are slightly changed to fit with the fictitious apparel brand ‘Chestnut’. According to the pre-test it was better to use more nuanced expressions in this research.
Eventually, to measure the dependent variable ‘corporate credibility’ the four item scale by Becker-Olsen, Cudmore and Hill (2006) is used, with a Cronbach’s alpha of .85. The questionnaire ends with questions concerning general information of the participants, such as gender (nominal variable), age (ratio variable) and educational background (ordinal variable).
Table 2: Survey items Sustainability Orientation (SO) 4 items SO1 SO2 SO3 SO4
I am concerned about the environment.
The condition of the environment affects the quality of my life. I am willing to make sacrifices to protect the environment. My actions impact on the environment.
Skepticism (Skep) 4 items Skep1 rSkep2 rSkep3 rSkep4
Most environmental claims made in advertising are true.
Because environmental claims are exaggerated, consumer would be better off if such claims in advertising were eliminated.
Most environmental claims in advertising are intended to mislead rather than inform consumers.
I do not believe most environmental claims in advertising.
Consumer-Company Identification (CCi) 7 items CCi1 CCi2 CCi3 CCi4 CCi5 CCi6 CCi7
I am somewhat associated with Chestnut I have a sense of connection with Chestnut
I consider myself as belonging to the group of people who are in favor of Chestnut
Customers of Chestnut are probably similar to me Employees of Chestnut are probably similar to me Chestnut shares my values
Being a customer of Chestnut is part of my sense of who I am
Credibility (Cred) 4 items Cred1 Cred2 Cred3 Cred4
Chestnut is a firm I can trust
Chestnut is a firm that cares about its customers Chestnut has a strong value system
32 3.7 Statistical procedure
The survey was developed in Qualtrics, as indicated earlier the final version is portrayed in appendix D. First of all the data is transferred from Qualtrics to the Statistical software Package for Social Sciences (SPSS) to perform the statistical analyses. Counter-indicative items were recoded and scale reliabilities, kurtosis, skewness and normality tests were computed. Both the Q-Q plot and bell shape indicate that all the variables are normally distributed.
Afterwards, the hypotheses are tested. First, an independent samples t-test was conducted to see if there is a statistically significant difference between the means of the two groups of the independent variable (self-serving CSR motive vs. other-serving CSR motive). Thereafter, PROCESS model ten was applied since it corresponds best with the conseptual model of this study (Hayes, 2013). With the use of PROCESS moderation and mediation analyses can be computed as well as their combination as conditional process model (Hayver, 2012). . PROCESS model ten (appendix E) is runned to test whether the type of CSR motive of green highlighting influences the moderating and mediating effects of consumer-company identification, consumer skepticism and sustainability orientation.
33 4. Results
In this section the results of testing the hypotheses are described. In the first paragraph the preliminary steps are explained that had to be taken before starting the actual analysis. Thereafter, validity and reliability checks are done for all the items used in this study. Followed by a correlation analysis and independent samples t-test. The results chapter ends with testing the hypotheses. PROCESS model ten (Hayes, 2012) is used and divided into the moderation analysis and moderated mediation analysis to structure the results found.
4.1 Preliminary steps
First, the data is cleared of missing data and careless responses as indicated earlier. The total of 218 respondents fully completed the survey and did not contain any missing data. Before starting with any analyses, the reverse-scaled items of ‘skepticism’ needed to be recoded. Three statements were counter-indicative and therefore needed to be recoded (rScep 2,3,4). On top of that the independent variable ‘green highlighting’ was translated into a dummy variable, to see to which of the two CSR motives the respondent was exposed. The firm-centered CSR motive was labeled with ‘0’ and the other-centered CSR motive was named ‘1’. Finally, the scale of the control variable ‘gender’ was changed, male (0) and female (1).
Finally, frequencies were checked to look for any errors or outliers in the data. For all variables descriptive statistics, skewness, kurtosis and normality tests were examined. The variables were all normally distributed, the levels of both skewness and kurtosis were between -1 and 1 which justifies the normality assumption (Brown, 1997).
4.2 Validity and reliability
The factor analysis is done using a Principal Component Analysis (PCA) and Promax rotation method to assess the reliability of the scale. After the first analysis SPSS found 4 factors, which
34
corresponds to the conceptual model of this study. The loading of Cred4 however is problematic, since the difference between components is less than .3 (0.604 -0.637). This means that Cred4 loads on both the factor ‘credibility’ and the factor ‘consumer-company identification’. Therefore, this item is deleted and a new matrix is made which resulted in four valid factors. All the factor loadings in this study are above .4 and therefore the matrix shows evidence for convergent validity (Field, 2009). The 4-factor model explains 71.18% of the variance. In addition, all the communalities are above .3 with the lowest level for Scep1 of .543 (table 3). The Kaiser–Meyer–Olkin (KMO) measure verified the sampling adequacy for the analysis. The KMO in this study is acceptable above the .6 threshold (.897). Bartlett’s test of sphericity is significant χ² (153) = 2586.549, p < .001, which indicates that correlations between items were sufficiently large (Hair et al., 2010).
Table 3: Rotated Component Matrix*
Construct 1 2 3 4 Communalities Consumer-Company Identification CCi1 CCi2 CCi3 CCi4 CCi5 CCi6 CCi7 0.776 0.794 0.777 0.841 0.838 0.734 0.759 0.711 0.718 0.729 0.758 0.710 0.707 0.699 Skepticism Scep1 Scep2 Scep3 Scep4 0.681 0.808 0.873 0.893 0.543 0.678 0.767 0.813 Sustainability Orientation SO1 SO2 SO3 SO4 0.798 0.766 0.756 0.756 0.735 0.631 0.680 0.615 Credibility Cred1 Cred2 Cred3 0.841 0.757 0.691 0.782 0.827 0.709
35
Reliability analyses are done for all the items of the variables (table 4). The Cronbach’s alpha is used to assess the reliability of the scales, where a higher value than .7 is considered as acceptable (Field, 2009). The Cronbach’s alpha’s of sustainability orientation, skepticism, consumer-company identification and credibility are all above the threshold of 0.7 (table 4). Credibility had an acceptable reliability (α = .795), after deleting Cred4 the Cronbach’s Alpha substantially improved to α = .852. The corrected item-total correlations indicates that all the items have a good correlation with the total score of the scale (all above .30).
4.3 Correlation analysis
New variables are created by calculating the mean of all the items that were used to describe the variable. The means, standard deviations and correlations are shown in table 5 on the next page. It can be assumed that multicollinearity is not a problem in this study, since no Pearson’s correlation coefficients are above .9 (Field, 2009).
The highest correlation of r = .651 shows that consumer-company identification (CCI) is the highest predictor of corporate credibility with a significance value less than 0.01. This strengthens the mediation effect of consumer-company identification. Unfortunately, no significant correlations are found between the type of CSR motive (self-serving or other-serving) and other constructs.
The two moderators sustainability orientation and consumer skepticism however significantly correlate both with consumer-company identification and credibility. Sustainability orientation is the strongest predictor for consumer -company identification (r =
Table 4: Reliability
Variable N of items Cronbach’s Alpha*
Consumer-Company Identification (CCI) 7 0.930
Skepticism (Skep) 4 0.849
Sustainability Orientation (SO) 4 0.804
Credibility (Cred) 4 0.852
36
0.427, p < .01) and also positively correlates with credibility (r = 295, p < .01). Secondly, skepticism negatively correlates with both consumer-company identification (r = -.202, p < .01) and credibility (r = -.369, p < .01). On top of that, skepticism also negatively correlates with sustainability orientation (r = -.180, p < .01). The more skeptical the respondents are, the less sustainable orientated they became.
Furthermore, table 5 shows that skepticism and consumer-company identification are the only two constructs that significantly correlate with gender. Skepticism negatively correlates with gender (r = -.156, p<0.05), which means that women are less skeptical compared to men. In addition, women can more easily identify with the fictitious company compared to men (r = .209, p < .01).
Age significantly correlates with credibility, sustainability orientation and skepticism. First of all, age negatively correlates with corporate credibility. The older the respondents get, the less credible they perceive the fictitious company to be (r = -.137, p < .05). Age moreover positively correlates with sustainability orientation (r = .141, p < .05) and skepticism (r = .203, p < .01), the older the respondents get the more sustainable orientated and skeptical they are. Finally, education does not correlate with any of the constructs relevant for the study, except from age (r = 0.147, p < .01). The older the respondents get, the higher level of education they have succeeded.
Table 5: Means, Standard Deviations, Correlations
M SD 1 2 3 4 5 6 7 8 1. Credibilty 4.65 1.14 1 2. CCI 3.88 1.25 .651** 1 3. CSR Motive 0.50 0.50 .052 .086 1 4. Sustainability Orientation 4.75 1.18 .295** .427** .069 1 5. Skepticism 3.66 1.20 -.369** -.202** .020 -.180** 1 6. Gender 0.58 0.50 .127 .209** 0.000 .065 -.156* 1 7. Age 4.12 1.42 -.137* .028 -.113 .141* .203** -.063 1 8. Education 4.00 1.39 -.045 -.013 .017 .088 .078 .011 .147* 1
** Correlation is significant at the 0.01 level (2-tailed). * Correlation is significant at the 0.05 level (2-tailed).
37 4.4 Independent t-tests
According to the correlation analysis, no significant correlations are found between the two types of CSR motives. An independent samples t-test was conducted to see if there is a statistically significant difference between the means of the two groups of the independent variable (self-serving CSR motive vs. other-serving CSR motive). The independent t-test was conducted for green highlighting and its relationship with corporate credibility, consumer-company identification, sustainability orientation and skepticism. Unfortunately, as shown in table 6 there is no significant difference in the reported levels for those within the self-serving or other-serving condition on any of these constructs (p > .05). These tests indicate that there might be an error within the two CSR motives of green highlighting. This potential issue will be discussed in chapter 5.
4.5 Hypotheses testing
In order to test the moderating and mediating effects of the overall conceptual model, the PROCESS macro written by Andrew F. Hayes for SPSS (2012) is used. PROCESS model ten (appendix E) is runned to test whether the type of CSR motive of green highlighting influences the moderating and mediating effects. X represents the independent variable (green highlighting), Y the dependent variable (corporate credibility), M the moderator
(consumer-Table 6: Independent t-tests
Levene’s test for Equality of Variances
T-test for Equality of Means
F Sig. t Sig.
(2-tailed)
95% Confidence Interval of the Difference
Lower Upper
SO Equal variances assumed 0.172 0.678 1.023 0.308 -0.151 0.477
Equal variances not assumed 1.023 0.308 -0.151 0.477
Skepticism Equal variances assumed 0.684 0.409 0.296 0.767 0.27210 0.368
Equal variances not assumed 0.296 0.767 0.27210 0.368
CCI Equal variances assumed 5.400 0.021 1.275 0.204 -0.118 0.550
Equal variances not assumed 1.275 0.204 -0.118 0.550
Credibility Equal variances assumed 0.622 0.431 0.772 0.441 -0.185 0.424
38
company identification) and W (sustainability orientation) and Z (consumer skepticism) are the moderators. XW and XZ furthermore are the products of combined variables that both have an influence.
In the next two paragraphs the results will be presented, to structure the findings PROCESS model ten is divided into the moderating analysis and the moderated mediation analysis. In figure 2 and table 7 the SPSS output of PROCESS model ten is illustrated in numbers. These results will be clarified and related to the hypotheses in the text.
** Correlation is significant at the 0.01 level (2-tailed). Figure 2: Process model 10
* Correlation is significant at the 0.05 level (2-tailed). Insignificant paths are indicated by a dotted line.
CCI CSR Cred SO Skep CSR*SO CSR*Skep -.024 .420** -.112* -.063 .038 .559** .160 .119 .044 -.209** .017
39 4.5.1 Moderation analysis
The first hypothesis assumes that the positive relationship between green highlighting and corporate credibility is stronger for the use of an other-serving CSR motive compared to the use of a self-serving CSR motive. R square shows that 49.9% of the total variance of credibility is explained by the overall model (p < .01). As shown in table 7 respondents that saw the other-centered CSR story scored 0.024 lower on credibility compared to the firm-other-centered group. Unfortunately, this direct effect is not statistically different from zero (p = .833), the 95% BC bootstrap confidence interval crosses the zero point (-.247 to .197). Based on this information, it can be concluded that there is no significant difference between the self-serving and other-serving CSR motive on credibility. Consequently, the first hypothesis is rejected.
The relationship between green highlighting and credibility however is also moderated by skepticism and sustainability orientation. The factor skepticism itself significantly affects corporate credibility (c3’= -.209, p < .01, CI= -.308 to -.110), more skeptical respondents tend to evaluate the corporate credibility lower. Interaction effect four combines the CSR motive and skepticism in relation to credibility. For a higher level of skepticism, the difference in credibility between the firm and other condition increases with 0.04 (figure 2), but this effect is not significant (c5’ = .044, p = .644, CI = -.145 to .234). The third hypothesis stated that the
Table 7: Process model 10
Consequent
CCI (M) Credibility (Y)
Antecedent Coeff. SE p Coeff. SE p
CSR Motive (X) a1 .160 0.15 0.298 c1’ -.024 0.11 0.833 CCI (M) - - - b .559 0.05 0.000 SO (W) a2 .420 0.07 0.000 c2’ .017 0.05 0.753 Skepticism (Z) a3 -.112 0.07 0.037 c3’ -.209 0.05 0.000 CSR Motive*SO (XW) a4 -.063 0.13 0.638 c4’ .119 0.10 0.226 CSR Motive*Skep (XZ) a5 .038 0.13 0.768 c5’ .044 0.10 0.644 Constant i1 3.71 0.33 0.000 i2 2.95 0.31 0.000 R² =.233 R² =.499 F =7.92, p = .000 F = 23.03, p = .000
40
relationship between green highlighting and corporate credibility is moderated by skepticism, and based on these results cannot be supported.
The second moderator sustainability orientation however does not directly affect corporate credibility (c2’ = .017, p = .75, CI = -.090 to .124). Respondents higher in sustainability orientation do not score significantly higher on corporate credibility. Interaction term three furthermore combines the CSR motive with the sustainability orientation of the respondents and is again insignificant (c4’ = .119, p = .226). The 95% bootstrap confidence interval further reveals that this effect is not significant because it crosses the zero point (-.074 to .313). Based on this information it cannot be concluded that sustainability orientation moderates the relationship between green highlighting and corporate credibility. Hypothesis five can therefore be rejected. It can be concluded that there is no difference in credibility between firm- or other-centered CSR motives with higher sustainability orientation or skepticism levels.
4.5.2. Moderated mediation analysis
The last hypothesis assumes that the relationship between green highlighting and corporate credibility is mediated by consumer-company identification. R square shows that 23.26% of the total variance of consumer-company identification is explained by the overall model (p<0.01). Table 7 shows that respondents who saw the other-centered webpage instead of the firm-centered webpage scored 0.16 higher on identification, but this effect is not significant (p = .298, CI = -.142 to .463). Therefore hypothesis 2a cannot be supported.
The mediator consumer-company identification however has a strong significant relationship with corporate credibility (b = .559, p < .01). This direct effect is statistically different from zero with a 95% BC bootstrap confidence interval with a lower bound of .458 and an upper bound of .659. Respondents who can identify with the company tend to evaluate
