How green to be?
The effects of CSR disclosure statements in green advertising on consumers’ brand
perceptions and purchase intentions
Name: Jildau Stoelwinder Student ID: 10760512 Master’s Thesis
Graduate School of Communication
Master’s Programme Communication Science Supervisor: Dr. E. de Waal
Abstract
With the disclosure of corporate social responsibility (CSR) policies becoming more common, this study attempts to research the effects of companies’ increased public accounting for green
initiatives in green advertising. This research focuses specifically on the incorporation of CSR disclosure statement (CSR-DS) in advertisements. A CSR-DS highlights the green commitment of the company in an extensive form, explicating companies’ green aims and efforts. This pervasive and detailed focus on green initiatives is expected to support a green ad by rendering it more credible, therewith enhancing consumers’ brand attitudes, brand image, and purchase intentions. The 2x2 ANOVA results suggest that brand image is not affected by the inclusion of a CSR-DS in a green advertisement. However, brand attitude and purchase intentions are enlarged when a CSR-DS supports a green ad. Multiple regression analyses further demonstrate that the level of credibility has not been found to explain these effects for the CSR-DS, but brand credibility has been found to contribute significantly towards brand attitudes, image, and purchase intentions when it comes to green advertising. Additionally, the impact of product involvement and green involvement on the evaluation of the CSR-DS were assessed. Both product involvement and green involvement were found not to moderate the effect of the CSR-DS on brand credibility, and therewith attitudes, brand image, and purchase intentions. As such, the projected supportive function of the CSR-DS has not been established across all outcomes. Only when considering carefully whether a green product should be bought does the additional information of companies’ green initiatives affect consumers. Further research ventures into this finding and limitations of the study are discussed.
Keywords: green advertising, CSR, CSR disclosure, brand credibility, consumer trust, brand image,
Introduction
After several spike years of environmental concern, Heather Rogers’ bestseller Green Gone Wrong:
How Our Economy Is Undermining the Environmental Revolution (2010) heralded the demise of
the green marketing strategy that had managed to get consumers to consider the environment in making purchase decisions. With the exposure of numerous greenwashing practices, i.e. when green advertisements are found to be intentionally misleading or confusing, consumers’ trust of green claims diminished tremendously (Do Paco & Reis, 2012; Terra Choice Group, 2010). Green advertising, which aimed to “create awareness and positive attitudes toward environmentally friendly companies,” had disillusioned consumers across the globe (Do Paco & Reis, 2012, p.148). As a result, consumers have started to question companies’ intentions continuously, creating a consumer culture that is inherently skeptic toward green advertising (Alves, 2009). Consumer research reflects this changed attitude, pointing out that more than half of the consumer population is hesitant about the truthfulness of green advertising and questions the motives behind green advertisements (Nielsen, 2011).
Nevertheless, green advertising does performs well, with many firms still investing in greening their company image and products alike (Atkinson & Rosenthal, 2014; Do Paco & Reis, 2012). Research continues to contribute to the idea that embracing green is beneficial on a company level, resulting in a better brand image (Brown & Dacin, 1997, as cited in Raska & Shaw, 2014; Nagar, 2016), stronger brand attitudes and greater purchase intentions (Becker-Olsen, Cudmore, & Hill, 2006; Khanelwal & Bajpai, 2011; Montoro-Rios, Luque-Martinez & Rodiguez-Molina, 2008). As such, it is no surprise that many marketers still see promoting green as a valid strategy, and an increase of companies investing in the research and development of green products can still be witnessed (Kotler, 2011; Ottman, 2011).
Thus, green marketing finds itself at a crossroad where two paradoxical forces are at play. As skepticism towards green advertising rises considerably, so do the investments in green products and their advertising. This development has created a complex arena for green advertisers in all
kinds of consumer markets, with at its heart the obtainment of consumers’ trust (Do Paco & Reis, 2012). With skepticism at its seeming peak when it comes to green advertising, researchers have attested to the necessity to bolster credibility in consumers’ eyes (Atkinson & Rosenthal, 2014; Raska & Shaw, 2012). Credibility of corporations’ green initiatives has proven to be one of the most effective drivers of green consumption behavior (Kalafatis & Pollard, 1999). Many consumers may wish to purchase more sustainable and environmentally friendly products but fail to do so because they believe the company or brand is incredulous and untrustworthy (Delmas & Burbano, 2011).
In order to increase credibility and restore trust in green advertising, several communicative strategies have been developed. These strategies mainly rely on credulity certifications, such as eco-labels, independent seals of approval, licenses, environmental organization cause-relation programs, and self-certification programs (Ottman, 2011). These certifications effectively strengthen
companies’ credibility (Atkinson & Rosenthal, 2014). Unfortunately, certifications are hard to come by, - or, in case of self-certification, are hard and laborious to develop - and not every marketer benefits from certifications’ persuasive power (Ottman, 2011).
Companies’ green initiatives can also be communicated differently. A current example is that of open communication of corporate social responsibility strategies, i.e. “a company’s integration of social and environmental concerns in [its] business operations,” undertaken by
companies who vouch to sell green, eco-friendly, and sustainable products and services (Cox, 2008; Lichtenstein, Drumwright, & Braig, 2004; Sheth, Sethia, & Srinivas, 2011). As Vanhamme,
Lindgreen, Reast, & van Popering (2012, p.259) have suggested, “corporate social responsibility has moved from ideology to reality.” The impressive growth from 41% in 2004 to 73% in 2015 in openly communicating corporate social responsibility (CSR) initiatives toward stakeholders may well be interpreted as a reflection of the positive corporate results, such as increased credibility, improved reputation, and increased consumer loyalty, that have been achieved though the application of this strategy (Cox, 2008; KPMG, 2015; EY, 2013).
Nonetheless, despite the positive effects of extending public access to CSR reports, many consumers - who are amongst the stakeholders - never read CSR reports (Cone Communications, 2015). The reports are oftentimes specific and technical, and do not interest consumers as a primary audience. Companies such as TOMS (TOMS, 2010) have made attempts to augment the
accessibility of their environmental commitment by distinctly packaging them in their mission statements. Others, such as Starbucks (Starbucks, 2016), are visibly highlighting their efforts in CSR disclosure statements online. This latter CSR disclosure statement (CSR-DS) is similar to an environmental claim, which, in the words of Scammon and Mayer (1995, p.33) is “a statement by a seller regarding the impact of one to more of it brand attributes on the natural environment.” but a CSR-DS also moves beyond this as it highlights what the company does to reduce the its
environmental impact and how this benefits not only the company but also the wider society. CSR-DS are often a few sentences long and attempt to explain the company’s commitment to green initiatives.
Particularly online, the CSR-DS seems to effectively communicate companies’ CSR
perspective to the consumer (Rolland & O’Keefe Bazzoni, 2009). One of the caveats of the reliance on online CSR information provision is that consumers should actively desire to engage with the product or brand advertised (Stigler, 1961). As Ottman (2011) has identified, only 19% of the consumers would consider themselves as real green purchasers and would also be willing to delve up elaborated CSR information about a green product. Nonetheless, the greater target group of semi-conscious green consumers (approximately 40%) would need to be addressed more obviously. The apparent search-factor declines the chances of CSR-DS online to be read by the majority of potential green consumers as it renders exposure largely reliant on the level of involvement, with the environment, and the brand and the product advertised alike, displayed by the consumer (Ottman, 2011).
Therefore, the looming question is why CSR-DS have not yet been incorporated in advertisements. When used in advertising rather than online, the CSR-DS offers an intriguing
communication potential that could greatly enhance the perceived credibility of companies. By incorporating the green initiative forthright in an advertisement, a company’s environmental commitment can immediately be presented to the customer. TOMS’ trademark “One for One” slogan, aided by the elaboration that “with every pair you purchase TOMS will give a pair of new shoes to a child in need,” is the quintessential success-story in this respect. The slogan and
elaboration’s straightforward CSR communication has centralized environmental commitment unambiguously, stimulating greater credibility assessments and therewith beneficial consumer responses, exemplified by TOMS’ continuously growing resale numbers (Marquis & Park, 2014).
The current study sets out to uncover whether benefits can be obtained from introducing CSR-DS in print advertisements. Little research has yet been performed on the impact of CSR disclosures in advertising. Gupta, Pirsch, and Grau (2007) have corroborated a positive impact of promotional CSR, i.e. the introduction of a cause as a reflection of CSR policy in print advertising, on purchase intentions and company attitudes. However, consumers have indicated that despite feeling more positively to the brand and product advertised, promotional CSR is regarded as a marketing ploy. In comparison with promotional CSR, messages containing institutionalized CSR felt more honest, and generated more positive outcomes on both company attitudes and purchase intentions, but also consumer loyalty (Gupta, Pirsch, & Grau, 2007). Nevertheless, the potential of institutionalized CSR messaging as a CSR-DS in promotional format, i.e. print advertising, has not yet been researched. As more firms take on corporate social responsibility and make investments in environmentally sustainable practices throughout their entire organization, there is a need to
discover whether these investments pay off (Ottman, 2011). With the success of TOMS’
communication regarding their environmental initiatives, it should be tested whether CSR policies aimed directly at the public can generate positive brand evaluations and purchase likelihood. These tests can broaden the understanding of green advertising and its efficacy, and thus presents a purposeful expedition for the business and academic field alike. As such, the following questions will be answered in the remainder of this research:
To what extent does a green advertisement with a CSR disclosure statement affect greater credibility among consumers and consequently brand attitudes, brand image perceptions, and purchase intentions?
Theoretical background
Green claims: credence attribution
Many consumers feel misled by environmental claims, characterizing that much of the terminology used, such as ecologically friendly, biodegradable, and organic, lack clear and accepted definitions and meanings (Do Paco & Reis, 2012; Furlow, 2010; Leonidou, Leonidou, Palihawadana, & Hultman, 2011). Banerjee, Gulas, & Iyer’s (1995) long-standing definition of green advertising evidences the great leeway which marketers have in constructing green advertisements. Following the definition, green advertising must either (1) “explicitly or implicitly address the relationship between a product/service and the biophysical environment,” (2) “promote a green lifestyle with or without highlighting a product/service”, or (3) “present a corporate image of environmental
responsibility.” (Banerjee, Gulas, & Iyer, 1995, p.22). These criteria, nonetheless, present no guidelines - nor many restrictions - to developing green claims in green advertising.
A fundamental problem with green advertising is that green claims cannot be tested by consumers and should therefore be taken at face value. Green claims should be understood as credence claims, which cannot be substantiated through personal experience (Atkinson &
Rosenthal, 2014). For instance, a consumer would not be able to ascertain that Nespresso cups are fully recyclable just by using the cups to brew coffee (Nespresso, 2016). Instead, as is the case with experience goods, consumers have to rely on a company’s communication attesting to the
truthfulness of the credence claim made or they have to embark on independent third-party information searches in order to evaluate the product accordingly (Bei, Chen, & Widdows, 2004).
Claims containing credence attributes force the consumer to trust the company that makes the claim, but need to establish that trust with asymmetric access to information regarding the product. (Consumer Affairs Victoria, 2010). Following the concept of the economics of information,
consumers never possess the same information as manufacturers (Ford, Smith & Swasy, 1990; Spence, 1973). Because of asymmetric access, consumers have to make a credibility assessment before they purchase, in order to reduce potential functional, financial, and/or egotistical risks (Lee, Kim, & Chan-Olmsted, 2011). As Raska & Shaw (2012, p.340) have attested, “perceived sincerity [is] an essential element in the development of promotional messages that generate positive
consumer responses to green marketing strategies.”
To assess the credibility and accuracy of green claims made by firms, consumers are reliant on information. Attribution theory suggests that through information people attempt to find causal explanations for the behavior of others, including firms (Nyilasy, Gangagharbatla, & Paladino, 2013; Phau & Ong, 2007). In their evaluation of green advertisements, consumers display attributional processes as they assign, amongst others, levels of credibility, integrity, and
consistency to a firm based on its communicative efforts (Ellen, Webb, & Mohr, 2006). The level of credence attribution on the basis of communication steers the extent to which a consumer trusts a brand and could thus potentially be manipulated by adapting the communicative content.
Signalling: CSR disclosure statements
In line with previous research on the effectiveness of eco-labels and certification seals as signals of the credibility of a brand or company (Rosenthal & Atkinson, 2014), CSR disclosure statements can function as content-based catalysts of brand credibility. A CSR-DS in advertising takes the form of a signal, which can function as “a marketer-controlled, easy-to-access informational cue, extrinsic to the product itself, that consumers use to form inferences about the quality or value of that product” (Bloom & Reve, 1990, as cited in Atkinson & Rosenthal, 2014, p.34) and the brand or company behind the product.
However, a signal only functions effectively when considered clear and credible (Spry, Pappu & Cornwell, 2011). A signal’s clarity refers to a lack of ambiguity with regard to the brand’s communicative content (Erdem & Swait, 1998), whereas the credibility of the signal represents how truthful and dependable the information conveyed through the message is received (Spry, Pappu, & Cornwell, 2011). The CSR-DS, which presents a match between the advertisement’s green credence claim and a company’s CSR policy by highlighting CSR initiatives and commitment, lauds clarity and dependability on a content level. With the CSR-DS exemplifying the environmental
investments of the company advertising, it reflects a content-based attempt to support the green product claim made in an advertisement. By providing ample and detailed information on its green commitment, a company’s CSR-DS can corroborate the credence attributes inherent to the green claim by simply providing additional information that makes up for consumers’ asymmetric knowledge with regard to credence attributes (Brammer & Pavelin, 2006). Previous research has established that cues with an elaborate argumentative character are preferred in supporting green claims (Manrai, Manrai, Lascu & Ryans, 1997). Increased tangibility and specificity of a green message has been found to feed positive evaluations of the product and company alike (Davis, 1993). Consumers apparently evaluate firms more favorably when they are open about their
environment initiatives, even despite the fact that CSR initiatives may at time be highly self-serving (Cho, Guidry, Hageman, & Patten, 2012; Forehand & Grier, 2003; Raska & Shaw, 2012).
Additionally, in the cause related marketing field, a supporting message element that
matched credence claims has evidenced more positive brand attitudes, a better corporate image, and influenced brand choice (Vanhamme, Lingreen, Reast, & van Popering, 2012). Research has also established that company programs which propagate a long term, comprehensive CSR initiative instead of a short cause-relation CSR initiative, generate better brand loyalty, purchase intentions, company attitude, and decrease skepticism (Gupta, Pirsch, & Grau, 2007). Moreover, providing information about green initiatives is essential next to the eco-labeling in order to promote green consumption effectively (Daugbjerg, Smed, Andersen & Schvartsman, 2014). Stand-alone logos
have been characterized as simple marketing ploys (Teisl et al., 2002). Considering the level of fit between the CSR-DS and the green advertisement, it is then also expected that a green
advertisement with a CSR-DS will affect brand evaluations in the form of brand attitudes, image, and purchase intentions more considerably than a green advertisement without a CSR-DS.
H1. A green advertisement with a CSR disclosure statement will result in more positive brand image evaluations, a stronger brand attitude, and greater purchase intentions than a green advertisement without a CSR disclosure statement.
Next to that, the CSR-DS should heighten the perceived credibility of a brand. The support that the CSR-DS lends to the green claim by providing information attests to the “willingness and ability of firms to deliver what they promise” (Erdem, Swait, & Valenzuela, 2006, p. 34), and therewith feeds levels of perceived credibility of a firm. The CSR-DS particularly strengthens one of the interacting variables of which credibility is composed: trustworthiness (Hovland, Janis, & Kelly, 1953;
Metzger & Flanagin, 2013). Referring to the degree in which the maker of the claim is capable and genuine, trustworthiness is heralded by the CSR-DS by portraying the commitment of the company and its efforts in aiding the society (Ayeh, Au, & Law, 2013; Dou, Walden, Lee, & Lee, 2012; Pornpitakpan, 2004). In green advertising, reinforcing trustworthiness is essential because green advertising, due to its definitive ambiguity, demands exact, concise, truthful environmental claims in order to establish credibility (Khandelwal & Bajpai, 2011).
Previous research has confirmed that CSR messages do not only play a role in brand and product evaluations (Klein & Dawar, 2011), but also increase brand credibility across
communication (Hur, Kim, & Woo, 2014). Phau & Ong (2007) have additionally established that message elements reflecting the company’s environmental credentials inspire more credibility than messages focused on a product’s environmental attributes. With the CSR-DS focus on the
company’s CSR intention, a heightened credibility can be expected in advertisements that feature the CSR-DS than those without. Atkinson & Rosenthal (2014) also validated that substantial
claims, as the CSR-DS provides, lead to higher levels of consumer trust. Research has, nevertheless, also pointed out that credibility will only be established when there is a match between perceived corporate performance and the green claim in an advertisement as it reduces consumer skepticism towards green advertising (Nyilasy, Gangadharbatla, & Paladino, 2013). The CSR-DS represents this optimization by confirming the consistency between the green credence claim of the brand and the company’s performance. Therefore, the following hypothesis about the effect of CSR-DS in green advertisements is proposed:
H1. A green advertisement with a CSR disclosure statement will be perceived as more credible than a green advertisement without a CSR disclosure statement.
The amount of credibility awarded to a brand is also a common factor underlying both brand
attitudes and brand image. A brand’s credibility reflects upon the company’s ability to deliver on its environmental promises, indicating its value as a green brand (Ng, Butt, Khong, & Ong, 2014). Both brand attitudes and brand image strengthen as a result of this perceived capability of a company to commit to green. A brand’s perceived credibility can project onto a brand’s image, decreasing the potential for risk incurrence and influencing the purchase willingness in the green product market (Ng, Butt, Khong, & Ong, 2014). In relation to brand attitudes, research has demonstrated that an increase in credibility assessments impacts brand attitudes positively (Goldsmith, Lafferty & Newell, 2000). This finding has been replicated over a range of studies related to green marketing (Atkinson & Rosenthal, 2014; Nyilasy, Gangadharbatla, & Paladino, 2013; Phau & Ong, 2007). With high credibility assessments consumers consistently evaluate brands more positively than with low credibility assessments.
On the other hand, research on the effects of credibility on purchase intent is relatively less consistent. Baek, Kim and Yu (2010) recently indicated that brand credibility functioned as a driver of purchase intentions. However, no positive effects on purchase intentions have yet been found in the green advertising context (Atkinson & Rosenthal, 2014; Kalafatis & Pollard, 1999; Peattie,
2010). An explanation for this lack of evidence can be found in the apparent attitude-behavior gap for green product. Many consumers project pro-environment attitudes, but fail to purchase green products (Carrington, Neville & Whitwell, 2010). However, advertising research has indicated that feelings of trust increase purchase likelihood among consumers (Chaudhuri & Holbrook, 2001). Therefore, it remains worthy of exploration.
H2. A credible green advertisement will affect brand image evaluations, brand attitudes and purchase intentions stronger than an incredible green
advertisement.
If all three hypotheses stated above should be validated, a possible mediation effect of brand credibility could manifest. Vlachos, Tsamakos, Vrechopoulos, and Avramidis (2009) have previously demonstrated that the impact of CSR evaluations is mediated by brand trust. They discovered that credibility is an essential underlying process of evaluation attribution. Therefore the following hypothesis will also be considered:
H4. A green advertisement with a CSR disclosure will affect more positive brand attitudes, brand image evaluations, and greater purchase intentions than a green advertisement with our a CSR disclosure statement because it is perceived as more credible.
Product involvement and green involvement
Product involvement is commonly defined in terms of purchase importance, suggesting that the level of involvement is determined by the importance of the product class to the consumer (Muncy & Hunt, 1984). This implies, in the words of Assael (1981), that “high involvement purchases are purchases that are important to the consumer… low involvement purchases represent purchases that are not important to the consumer.” This level of involvement with a product can be underscored by the sense of risk, whether functional, financial or egotistical, that is perceived in buying the product
(Lee, Kim, & Chan-Olmsted, 2011). The level of involvement a consumer displays towards a product has consequences for the extent to which advertising messages are interpreted (Petty and Cacioppio, 1981;Nagar, 2016).
As elaborated in Petty and Cacioppio’s (1986a, 1986b) Elaboration Likelihood Model (ELM), consumers can take two routes of processing an advertisement. The model proposes that several factors influence which route to processing a consumer takes, among which product
involvement (Atkinson & Rosenthal, 2014). The level of product involvement, ranging from high to low, is said to determine how much time and energy consumers spend on the purchase decision process (Nagar, 2016). Following the model, research has substantiated that consumers who are highly involved with a product spend more time on processing an advertisement compared to low involved consumers (Montoro-Rios, Luque-Martinez, & Rodiguez-Molina, 2008). Furthermore, for highly involved consumers the tendency to search for information and engage with the argument of the message is considerably enlarged vis-à-vis low involved consumers (Liu & Shrum, 2009; Zaichkowsky, 1985)
Seeing as the CSR-DS has a relatively elaborate character, it is expected that highly involved consumers will engage more actively with CSR-DS than low involved consumers. The engagement with CSR-DS will engender labored scrutiny of the message, questioning the match between the green advertisement’s claim and the company’s CSR initiatives. As the CSR-DS provides an optimized match between CSR performance and green credence claim, it is predicted that high-involved consumers reduce skepticism and will assign greater credibility to the message (Nyilasy, Gangadharbatla, & Paladino, 2013). On the contrary, low involved consumers, who rely strongly on peripheral cues, are expected to attribute less scrutiny to the CSR-DS (Petty, Cacioppo, & Schumann, 1983). The match between CSR initiative and credence claim will not be optimally processed and credibility is not expected to strengthen.
H5. High product involvement will result in stronger credibility assessments of the green advertisement with a CSR disclosure statement than a green advertisement
without a CSR disclosure statement. This effect will not occur in the low involvement condition.
Green involvement can be explained as the degree to which a consumer considers the environment in their consumption behavior (Matthes, Wonneberger, & Smuck, 2014). Consumers have been found to move along a continuous spectrum of green considerations, becoming both more or less involved with the environment and more or less susceptible to green advertising across their lifespan (Ottman, 2011). Green involvement as a concept is argued to include “three relevant conceptualisations, (1) environmental concern, (2) attitudes toward green products, and (3) green purchase behavior” (Matthes, Wonneberger, & Smuck, 2014). Of these three, particularly
environmental concern has been argued to conceptualize green involvement (D’Souza & Taghian, 2005; Shuhwerk & Lefkoff-Hagius, 1995) and found to be explanatory for consumer effects, such as purchase intentions, after exposure to green advertisements (Grimmer & Bingham, 2013;
Schwartz & Miller, 1991). Therefore, environmental concern can be considered to be an appropriate predictor of green involvement.
Green involvement levels have been found to impact the effectiveness of green
advertisements in a similar way as product involvement. Following the Persuasion Knowledge Model (Friestad & Wright, 1994), the higher level of knowledge and motivation to process environmental claims that high green involved consumers possess will moderate the persuasive abilities of the CSR-DS (Bickart & Ruth, 2012). Previous research has established that high green involved consumers react positively to environmental claims that are supported with, voluntarily provided, arguments (Matthes, Wonneberger & Smuck, 2014; Petty & Cacioppo, 1986a, 1986b). Tucker, Rifon, Lee and Reece (2012) have found that high green involved consumers particularly react positively to supporting arguments by attributing more credibility to green advertisements. On the other hand, low green involved consumers have been found to be largely indifferent to
additional information presented (Bickart & Ruth, 2012). Therefore, the following hypothesis has been formulated:
H6a. High green involvement will result in stronger credibility assessments of the green advertisement with a CSR disclosure statement than a green advertisement without a CSR disclosure statement. This effect will not occur in the low
involvement condition.
Figure 1.
Conceptual model
Methods
Research design
To test the relationships between CSR disclosure statements, credibility, product involvement, and green involvement and their effects on brand image, brand attitude, and purchase intentions, an experimental study was developed. The study featured a 2 (green advertisement: CSR-DS v. brand historical statement) x 2 (product involvement: high v. low) between subjects experimental design. All participants were randomly assigned to any of the four conditions.
Green Advertisement a. with CSR-DS b. without CSR-DS Brand Credibility Brand Image Brand Attitude Purchase Intention Product Involvement Green Involvement
Sample
The sample of the study consisted of a fairly heterogeneous group of consumers who were 18 years or older. Participants were recruited through a Facebook post on the researcher’s wall and in a variety of peer groups, such as the graduate communication class of the University of Amsterdam. 134 participants filled out the survey. 19 respondents were excluded from the sample as 12 did not complete the survey, one participant recorded the same response across measures, and 6
respondents indicated to be extensively familiar with one of the two brands, coffee brand Green Mountain Coffee (hereafter GMC) and car brand Acura, used in the experiment.
The final sample consisted of 115 participants, of which 77.4% were female and 22.6% male. The average age of the participants was 26, within a range of 47 (18-64) years. Most of the participants were Dutch (69,6%). Interestingly, nearly half (42,84%) of the internationals did reside in the Netherlands at the time the survey was conducted. The sample thus mainly concentrates on Dutch consumer culture. The sample also presented itself as fairly highly educated with three quarters of the participants (73,9%) indicating to have attained education on a tertiary or post-tertiary level. The remainder of the sample completed post-secondary education (11,3%), upper secondary education (11,3%) or lower secondary education (3,5%). With regards to employment, most of the respondents were students (53,9%). The other respondents were either employed (32,2%), self-employed (6,1%), unemployed (5,2%), or unable to work (2,6%). None of the participants stated to be retired.
Stimuli
The main between-subject factor, CSR-DS, was manipulated on two levels through a subtext that presented either a a). CSR disclosure statement or b). brand historical statement. Product
involvement, the secondary between-subjects factor, was also manipulated on two levels with coffee brand GMC as a low-involvement product category, and the car brand Acura as a high-involvement product category. This manipulation of product high-involvement was informed by the
perceived varying financial risk in purchasing either the fast-moving coffee brand or the technological and sophisticated car brand (Atkinson & Rosenthal, 2014; Zaichowsky, 1985).
Pre-test
For the CSR-DS carriers, the green advertisements, a choice was made to slightly modify existing advertisements for the two actual brands, GMC and Acura. The choice for existing brands was motivated by an effort to strengthen of the external validity of the experiment. The chosen brands were further selected on a basis of their potential familiarity with expected target group, Europeans. The brands operate in American and Asian markets, but are not activated in Europe. Seeing as brand familiarity has been found to affect the effectiveness of advertisements, a pre-test was run to validate the choice for these brands. This pre-test simultaneously assessed the greenness of the advertisements (i.e., the advertisement chosen were used in support of green campaigns but this does not guarantee that they will be identified as such by respondents from the target audience), and the effectiveness of the two brand-tailored experimental subtexts (i.e., CSR-DS v. brand historical statement) which were composed on the basis of the brand’s actual history and corporate
responsibility initiatives.
Ten respondents, who were recruited via Facebook, completed the pretest. Two participants indicated to be familiar with GMC. None of the participants knew the car brand Acura. The
familiarity with GMC extended merely to the brand’s name. No awareness of the brand’s products or environmental mission was reported. The choice for both GMC and Acura can thus be
considered justified on the basis of familiarity for the purpose of this research. The familiarity measure has been repeated in the actual experiment to eliminate participants who presented bias.
The pre-test further demonstrated that the subtexts, the CSR-DS and the brand historical statement, did not vary as to their intended function in the GMC conditions. Contrarily, the function of the CSR-DS and brand historical statement was identified correctly in the Acura conditions. The incorrect identification of the subtext function in the GMC conditions could be a side effect of the
name. The brand’s name refers to “green” and may have affected the brand’s assessment as being green as well. Function identification will be reassessed in the actual experiment.
Consistent with this predisposition towards assigning a greener evaluation to GMC than Acura on the basis of the paired subtexts, is the difference found between the assessment the greenness of the GMC and Acura advertisements. Acura (M = 2.3) scored significantly lower on greenness than GMC (M = 4.3). To account for GMC’s greener evaluation on both accounts, the Acura ad was adjusted. The green background of the advertisement was digitally enhanced, making the green more visceral. Additionally a more profound and explicit referral to the car’s green character was made by including the tagline, “high performance, low emissions, zero guilt.”
Selection
After the pre-test, stimuli for the experiment were designed by a professional designer to ensure that advertisements would maintain their original format, slogans, and outlook and would be considered highly realistic. The designs presented possible print advertisements that consumers could
encounter in magazines and/or papers.
The GMC advertisement presented the slogan “Great coffee, good vibes, pass it on” and demonstrated a cup of coffee against a drawn green mountain backdrop. The chosen Acura
advertisement highlighted an Acura car front of a luscious, but faded green background. Seeing as car advertisements are not commonly presented in front of green background, it was expected that the green background would reflect onto the car, suggestively indicating that the car could be considered eco-friendly. In both the GMC and Acura brand historical statement conditions, a tailored history statement about was devised. In the CSR-DS condition, both brand’s commitment to the environment and greening its community was explained. In all four brand conditions,
consumers were also invited to find out more about the history and/or corporate social responsibility of the company online. This “Want to find out more? Please visit us on …” tagline was added to accentuate that CSR information is usually presented online. As such, the subtexts were the only
differing variable. All other characteristics of the two advertisements, including the ‘find out more’ tagline, were held equal so as to minimize the impact of extraneous factors (see Appendix A).
Procedure
The study was conducted in the form of an online survey, distributed through Qualtrics research software. A link to the online survey was included in the Facebook messages sent out. Upon opening the survey, all respondents were informed about the subject of study (brand
communication). No references to green products and advertising were made so as to reduce response bias to a minimum. All participants were requested to give informed consent as to their voluntary participation in the experiment. The informed consent granted legitimized use of all data within the terms of the University of Amsterdam’s Communication Science Master’s Thesis program. The informed consent statement highlighted that participants could at any point revoke their data as well as contact the university, thesis supervisor, or thesis conductor with complaints, questions, or concerns.
After consent was granted, the participants were randomly assigned to either one of four conditions: (a) CSR-DS x low involvement, (b) CSR-DS x high involvement, (c) brand historical statement x low involvement, (d) brand historical statement x high involvement. Respondents were asked to consider the advertisements carefully. Subsequently, all participants filled out the
dependent measures: brand attitude, brand image, and purchase intentions. Thereafter, the respondents assessed statements regarding the credibility of the brand. Next, the manipulated moderating factor product involvement was assessed as to control and confirm successful
manipulation of involvement across the two product categories. Additionally, green involvement was examined. In order to establish that the participants did not display brand attitudinal prejudice, respondents’ brand familiarity was assessed as a control variable. Thereupon, a manipulation check was administered in order to ensure that respondents had read (if only partially) the manipulated subtexts. Lastly, participants filled out several demographic questions. In the end, all participants
were thanked for their time and effort, notified of the purpose of the study, given the opportunity to leave behind questions and/or comments, and requested to fill out their mail address to partake in a €10 gift certificate raffle set out as an incentive to participate in the study.
Measures
Dependent variables
Brand attitude - Brand attitude was measured using the validated scale of Spears & Singh
(2004). The scale consists of five items, measured on a 7-point semantic differential scale
(unappealing/appealing, bad/good, unpleasant/pleasant, unfavorable/favorable, unlikeable/likeable). All items loaded onto one factor (EV = 3.93, R2 = .78), justifying the averaging of scores on the five separate items to calculate an overall brand attitude score
(α= .93, M = 4.76, SD = 1.12).
Brand Image - Brand image was assessed using Nagar’s (2016) recently developed brand
image scale. The scale presents a comprised measure of brand image in comparison to the often used Malhotra (1981) scale and Aaker’s (1997) brand image derivative brand personality scale. The scale consists of 7 items, which where tailored to the two brands: the brand GMC / Acura (1) is of high quality, (2) has a superior function, (3) is trustworthy, (4) is a leading brand, (5) has a good reputation, (6) symbolizes a high social status, (7) gives me a comfortable feeling. Participants rated these 7 items on a 7-point Likert scale (1= extremely disagree, 7= extremely agree). The scale presented itself unilateral, EV = 3.37, R2 = .48, and reliable, α= .82, M = 4.33, SD = .87.
Purchase Intention - Purchase intention was measured using three altered items from
Grewal, Monroe, & Krishnan’s (1998) willingness to buy construct. The three items, (1) I would consider buying GMC / Acura, (2) I would probably buy GMC / Acura, and (3) I would definitely purchase GMC / Acura were assessed on a 7-point Likert scale (1= extremely agree, 7= extremely disagree). The three individual items loaded onto one factor (EV = 2.44, R2 = .81) and presented to
Mediating variables
Brand credibility - Brand credibility was examined through adapted items from Erdem and
Swait (1998). Participants rated their level of agreement with 7 items on a 7-point Likert scale (1= extremely disagree, 7 = extremely agree). The items presented were as follows: I feel that GMC / Acura (1) has a name you can trust, (2) seems competent to me, (3) has the ability to deliver what it promises, (4) makes product claims that are believable, (5) does not pretend to be something it is not, (6) delivers what it promises, (7) leads me to expect it keeps its promises. The scale proved internally consistent (EV = 4.94, R2 = .70) and reliable (α = .93, M = 4.50, SD = 1.00).
Moderating variables
Product involvement – The measure for product involvement was adopted from
Zaichowsky’s (1985) product involvement scale. To ensure applicability of items to the brands under research, 10 out of 20 items from the original scale were used. The items were tested on a 7-point semantic differential scale applicable to either GMC or Acura (of no concern to me/of concern to me, means a lot to me/means nothing to me*, trivial/fundamental, valuable/worthless*,
vital/superfluous, appealing/unappealing*, mundane/fascinating, essential/non-essential*,
undesirable/desirable, beneficial/not beneficial*).1 After reverse coding, the items were discovered to load onto two separate factors (factor 1: EV = 4.88, R2 = .48, factor 2: EV = 1.43, R2 = .14). However, combining both factors did offer a reliable scale at α= .76. Considering the possible enhancement of this reliable scale, the component matrix demonstrated that one item
(vital/superfluous) loaded negatively on the first factor. When extracting this item from the analysis, two factors remained but all items did load positively onto the two factors (factor 1: EV = 4.37, R2 = .48, factor 2: EV = 1.25, R2 = .14). For reasons of inclusivity and explanatory power, all items
loading onto the first factor were combined to construct a product involvement scale, EV = 2.97, R2
= .59, α= .82, M = 3.72, SD = 1.11. Using a median split, participants were divided across high and low product involvement to mirror the original manipulation.
Green involvement - Green involvement was assessed through the use of Shuhwerk &
Lefkoff-Hagius (1995) validated green concern scale, which corresponds with the established New Ecological Paradigm measure of green involvement (Dunlap, van Liere, Mertig, & Jones,
2000).The scale consists of four items: (1) I am concerned about the environment, (2) the condition of the environment affects the quality of my life, (3) I am willing to make sacrifices to protect the environment, and (4) my actions impact on the environment. All items were examined on a 7-point Likert scale (1 = extremely disagree, 7 = extremely agree). PCA testing confirmed that all items loaded onto one factor (EV = 2.76, R2 = .68) and that the scale was reliable (α= .84, M = 5.6, SD =
.97). A median split was used to divide participants across high and low involved groups.
Manipulation check
Product involvement - The product involvement measure examined above was also
administered to check whether the manipulation of product type across conditions (coffee brand v. car brand) altered product involvement levels among participants. Testing demonstrated that the manipulation of product involvement through the use of a coffee and car brand failed because no significant differences in product involvement were found across these conditions, t (113) = .47, p = .64. Consequently, the brand condition will not be considered in the following analyses. Instead, the product involvement measure will be used in assessing the possible moderation of relationships between the green advertisement with CSR-DS, brand credibility, brand attitudes, brand image, and purchase intentions.
Exposure to statement - The effectiveness of a CSR-DS can only be established through
exposure. Therefore, exposure to the subtext, both CSR-DS as well as the brand historical
it was perceived as intended. Respondents were requested to indicate what the function of the subtext was with a choice from the following options; (a) the subtext highlights the history of GMC / Acura, and (b) the subtext indicates the green/environmental mission of GMC / Acura.
Control variables
Brand familiarity - Brand familiarity was assessed in two steps. Firstly, respondents were
asked whether they ever heard of the brands before exposure to the advertisement in the survey (answer options: yes, no). Merely 15 participants expressed that they ever heard of either of the two brands. These respondents indicated to which extent they were familiar with the brands by selected (a) I know GMC / Acura just by name, (a) I know more about GMC / Acura than just its name (e.g. I am aware of its products, mission statement, etc.), or (c) I use GMC / Acura in my daily life). Respondents who knew the brand just by name (N = 10) were retained in the sample, whereas those who knew more about the brand (N = 5) or used the brand on a daily basis (N = 1) were deleted from the sample to avoid extraneous attitudinal brand bias.
Demographics
Lastly, age, gender, nationality, country of residence, education (answer options: primary or less, lower secondary, upper secondary, post-secondary, tertiary and post-tertiary, and other, please specify) and employment status (answer options: student, employed, self-employed, unemployed, retired, and unable to work) were assessed.
Results
Randomization check
To confirm that the green advertising condition (CSR-DS v. brand historical statement) assessment occurred randomly, several randomization checks were run in order to establish that sample
characteristics did not impact the condition distribution negatively and should therefore be controlled for in further analyses. From analyses it can be concluded that there are no significant differences between the conditions on age (t(113) = -.89, p = .38), gender (X2 (1, N = 115) = .25, p
= .62), education (X2 (1, N=115) = 3.09, p = .27), and employment status (X2 (1, N=115) = 3.03, p =
.55). On the other hand, nationality (Dutch v. international) (X2 (1, N=115) = 5.25, p = .02) and country of residence (Netherlands v. international) (X2 (1, N=115) = 8.97, p > .005) should be
controlled for as groups were found to differ significantly across the two green advertising conditions.
Manipulation check
Respondents did not always correctly identify the function of the subtext (see Table 1).
Respondents did, nonetheless, identify the corresponding function of the CSR-DS quite accurately, with 63,0% correctly matching CSR-DS and indication of a brand’s green mission. However, the brand historical statement was often mismatched, with only 25,0% accurately pairing it with a history highlight. Reverting to the results of the pre-test in which indications of greenness and brand history were made correctly for all pairs, except for GMC with a brand historical statement, a further check was done per condition to assess the impact of the subtext-condition pairing. Results indicate that 15% mismatched GMC with a brand historical statement as presenting a green mission statement. Comparatively, 16,7% mismatched Acura with a brand historical statement as presenting a green mission statement. The effect of the GMC name thus seems nihil in engendering green associations where none are presented by the subtext. On the basis of these results and in combination with the pretest, the manipulation can thus also be considered to hold.
Table 1.
Function assessment of subtexts (condition CSR-DS v. brand historical statement) in green advertisement.
Control variables
Brand familiarity was tested before proceeding with the multiple regression analyses in order to confirm whether it could potentially affect outcomes. A chi-square test demonstrated that there were no significant differences between the groups assigned to the four experimental conditions on the basis of familiarity with either the brand GMC or Acura, X2 (N = 115) = .001, p = .98.
Therefore, brand familiarity is disregarded as a source of impact in further testing.
Hypothesis testing
Main effects
In order to establish the main effect of condition (CSR disclosure statement v. brand historical statement) on the three dependent variables (brand attitude, brand image, and purchase intention), a one-way ANOVA was run (see Table 2). Results indicate that for both brand attitude (F (1,113) = 3.91, p = .05, η2 = .03) and purchase intentions (F (1,113) = 7.24, p = .008, η2 = .06) there are significant, but weak effects of the CSR-DS. Brand attitudes of consumers are thus stronger when exposed to CSR-DS (M = 4.96, SD = 1.15) than to the brand historical statement (M = 4.55, SD = 1.05). Similarly, the CSR-DS affects greater purchase intentions for green products (M = 3.82, SD = 1.43) than the brand history statement (M = 3.18, SD = 1.08). For brand image, no significant effect
Function A: history highlight Function B: indication of green mission N Condition 1: CSR-DS 0,0% 63,0% 46 Condition 2: brand historical statement 25,0% 15,9% 44
of the CSR-DS was found (F (1,113) = 1.86, p = .18)2. H1, which states that a green advertisement with a CSR disclosure statement will result in more positive brand image evaluations, a stronger brand attitude, and greater purchase intentions than a green advertisement without a CSR disclosure statement is therefore only partially supported.
Table 2.
Results of a one-way ANOVA for condition on brand attitude, brand image, and purchase intent (N = 115).
Brand attitude Brand image Purchase intent
Sum of squares 4.75 1.39 11.56 df 1 1 1 Mean square 4.75 1.39 11.56 F 3.91 1.86 7.24 p η2 .05 .03 .18 .02 .008 .06
Effects on brand credibility
Assessing H2, results of a regression indicate that a green advertisement with a CSR disclosure statement is not perceived as being more credible than a green advertisement without a CSR-disclosure statement, F (1,113) = .50, p = .68.3 This finding is supported by a t-test performed on condition and credibility. Results show that means between the CSR-DS (M = 4.60, SD = 1.03) condition and the brand historical statement condition (M = 4.40, SD = .94) do not differ
significantly, t (113)= 1.08, p = .29. Therewith, any effect of the CSR-DS on the assessment of a brand’s credibility can be discarded.
Mediation effects on brand attitude and purchase intentions
Despite the insignificant effect of the CSR-DS condition on brand credibility assessment, a regression analysis was run on brand attitude, brand image, and purchase intentions to check whether brand credibility could possibly explain variation in these variables in addition to the
2 For brand image, Levene’s test of homogeneity of variances was significant, p = .004. Equal variances could thus not be assumed.
DS. Testing for mediation of brand credibility, and controlling for nationality and country of residence, the multiple regression models indicate that the CSR-DS only predicts purchase intentions significantly, b* = .21, t = 2.44, p = .02, 95% CI [0.10, 0.97]. As such, the reading of a CSR-DS in a green advertising increases the scoring on purchase intention with .53, keeping other variables constant (see Table 3).
Table 3.
Regression models predicting brand attitudes, brand image, and purchase intent (N = 115) (unstandardized regression coefficients)
Brand attitudes Brand image Purchase intentions
Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
B B B B B B Constant 1.75*** 1.01* 1.81*** 1.41*** .47 -.10 CSR-DS .31 .23 .16 .14 .54* .43* Nationality .21 .14 -.06 -.10 .07 .02 Country of residence -.35 -.25 -.19 -.15 -.17 -.07 Brand credibility .63*** .51*** .55*** .53*** .61*** .45** Product involvement .42** .18 .41* Product involvement x credibility -.09 -.06 -.04 Overall model (F) 15.35*** 12.69*** 19.77*** 13.99*** 10.66*** 8.66*** R2 .36 .41 .42 .44 .28 .33 ΔF 5.09** 1.83 3.63* ΔR2 .06 .02 .05 Notes. For CSR-DS, 0 = no CSR-DS, 1 = CSR-DS. * p <.05. ** p <.01. *** p <.001.
Whereas the ANOVA indicated that the CSR-DS impacted attitude scoring of consumers, this effect has been usurped by the predictive strength of credibility (see Table 4). Brand credibility significantly accounts for scores across brand attitudes, brand image, and purchase intentions. It explains respectively 42, 36, and 28 percent of the variance in brand image, brand attitudes, and purchase intentions, presenting moderately strong associations for all variables, b*brand image = .63, t
brand credibility affects brand attitudes, brand images, and purchase intentions positively. Seeing as the condition, CSR-DS v. brand historical statement, did affect the dependent variables purchase intent, but did not impact brand credibility scoring significantly, no meditational model of brand credibility can be established. Therefore, H4 should be rejected.
Effects of product involvement and green involvement
Following H5 and H6, the impact of high product involvement and high green involvement on brand credibility between the CSR-DS and brand historical was tested. An independent sample t-test performed on both conditions, the CSR-DS and brand historical statement, revealed that high product involvement and high green involvement insignificantly affect brand credibility
assessments (see Table 4). Rejecting H5, respondents who are highly involved with a product do not attribute higher credibility scores to an advertisement with a CSR-DS statement than an advertisement with a brand historical statement, t (54) = .74, p = .46. However, H5 did accurately project that no differences in brand credibility scoring would occur for low involved consumers,
t(57) = .08, p = .94. Similarly, partially rejecting H6, high green involvement does not present a
better score for the CSR-DS condition than the brand historical condition, t (51) = .42, p = .65.4 As projected, no effects of brand credibility manifested in the low involvement condition, t(60) = .88, p = .39.
Table 4.
Brand credibility scores for high product involvement and high green involvement in CSR-DS and brand historical conditions.
Mean Standard
Deviation
N High product involvement
- CSR-DS 4.93 .94 32
- brand history 4.75 .84 24
High green involvement:
- CSR-DS 4.63 .99 30
- brand history 4.53 .49 23
4
Levene’s test for equality for variances was found to be significant at p = .008. Therefore equal variances were not assumed in the reporting of this statistic.
Extending beyond the hypotheses, an effect of product involvement on brand credibility across both conditions did occur, t(113) = - 4.01, p < .001. Therefore, product involvement was added as an additional variable to the multiple regression models used to explain the previous effects of brand credibility found on brand attitude, brand image and purchase intentions. Including product involvement as a co-varying predictor did not adversely affect the brand attitude, brand image and purchase intentions models. Whereas high product involvement was found to be a significant predictor of both brand attitudes and purchase intentions, no interaction effects between product involvement and credibility where established for brand attitudes and purchase intentions (see Table 3). The significant predictions of high involvement were moderately strong for both brand attitude, b* = .37, t = 3.16, p = .002, and purchase intentions, b*= .31, t = 2.43, p = .02. Both models presented no change in variance by adding the interaction variable separately from the product involvement variable, indicating that no interaction effects between product involvement and credibility could be attributed, ΔR2brand attitude = .02 , p = .09, ΔR2 purchase intent = .003, p = .50. Noteworthy is that the condition, CSR-DS v. brand historical statement, does significantly predict purchase intention again in this model, b* = .19, t = 2.25, p = .03. The function of the CSR-DS thus withstands when it comes to predicting purchase intentions.
Discussion and conclusion
This study aimed to test the extent to which the incorporation of a corporate social responsibility disclosure statement (CSR-DS) in a green advertisement would result in more positive brand attitudes, brand image, and purchase intentions among consumers. The results of this research have demonstrated that there is no effect of CSR-DS on brand image perceptions. However, when it comes to forming brand attitudes and purchase intentions, the inclusion of a CSR-DS in a green advertisement does seem to contribute to consumers’ consideration sets. An explanation for the lacking effect on brand image could be a direct derivative of testing on brands that are unknown to the consumer. Brand image is argued to be built over continued exposure to brands (Aaker, 1997).
When faced with unfamiliar brands an initial image is constructed on the basis of the advertisement exposed to. Apparently, the CSR-DS and the brand historical statement seem to attribute to brand image perceptions somewhat identically as no differences between mean brand image scores have been found. Testing on familiar brands may reflect a differential impact of the CSR-DS and brand historical statement as they are added to the pre-existing brand image of the consumer. In this stage, nonetheless, both conditions have fed into the baseline of brand image perceptions.
Furthermore, the effects of CSR-DS in predicting purchase intentions should be considered carefully. Whereas the effect of CSR-DS on brand attitudes gets usurped when considering credibility, the effect of CSR-DS on purchase intentions remains. The withstanding of the effect may indicate that consumers do consider a fair presentation of a product, elaborating its impact on the environment, as an important factor in making a final purchasing decision. The impact of argumentative elaboration is seemingly less strong for brand attitudes. The effacing of a CSR-DS effect in merely the purchase intention condition suggests that additional CSR information is key in choosing for a green product. The additional elaboration offered by the CSR-DS may just highlight the environmental commitment of the company enough to give it the selling edge.
In spite of the expectation that the level of credibility would inform the impact of the CSR-DS, no effects of the CSR-DS on brand credibility were found. Although this hypothesis was grounded in clear theoretical reasoning and results in other fields did showcase this relation (Hur, Kim, & Woo, 2014; Phau & Ong, 2007), the attribution-match between the CSR-DS and the green advertisement did not incite a stronger credibility perception. An explanation of this effect is rooted in greenwashing literature. In the face of greenwashing practices, research on signaling theory has predominantly established that signaling works if the sign is short and supported by an external partner, i.e. a government-body or independent third-party. This independent validation transfers additional trust to the brand (Rosenthal & Atkinson, 2014). Whereas it was expected that the extensive and explanatory content-based nature of the CSR-DS could also deliver this additional trust (Raska & Shaw, 2008), no such effects were attributed. An additional explication for this
effect is the use of a general brand credibility measure and the brand historical statement as control condition in comparison to the CSR-DS. A confounding effect of the brand historical statement could have manifested in relation to the general brand credibility measure, with the history statement displaying as much genuinity and capability of the company in complying with their green claims as the CSR-DS (Ayeh, Au, & Law, 2013; Erdem, Swait & Valenzuela, 2006; Dou, Walden, Lee, & Lee, 2012; Pornpitakpan, 2004). If brand credibility had been measured in relation to green or environmental performance, the CSR-DS would arguably contribute more than the history statement due to its focus on environmental performance.
Additionally, the lack of heightened credibility in the CSR-DS condition could also be explained because of the methodical choice to use unfamiliar brands. The choice to use unfamiliar brands (GMC and Acura) for participants was informed by the consideration to minimize bias in the sample. Nevertheless, as research on credibility has demonstrated before, forming a credible image of a brand relies largely on trustworthiness. Trustworthiness can only be established when the company behind the advertisement is considered capable, honest, and correct (Ayeh, Au, & Law, 2013; Dou, Walden, Lee, & Lee, 2012; Pornpitakpan, 2004). In order to establish a trustworthy brand image, the development of brand attitudes provides a necessary first step. However, forming capability and honesty attitudes without familiarizing with the brand is relatively uncommon. This is not to argue that no attitudes are constructed upon seeing the brand, but attitudes are constructs that develop over time and therewith also only inform credibility assessments over time. Simply portraying the behavior of the firm within the CSR-DS may not have been enough to offset a constructive credibility image (Nyilasy, Gangagharbatla, & Paladino, 2013).
Importantly, the positive effects of brand credibility on brand attitudes, brand image, and purchase intentions that this research has established are a further validation of previous research (Atkinson & Rosenthal, 2014; Baek Kim and Yu, 2010; Erdem, Swait, & Valenzuela, 2006;
Nyilasy, Gangagharbatla, & Paladino, 2013; Phau & Ong, 2007). Adding to the extensive literature on brand credibility, this research once again demonstrates that the level of trust and credulity a
consumer develops towards a brand impacts the brand image, the attitudinal perspective and the decision-making process of consumers when faced with green advertising. Green advertising keeps presenting itself as a field which demands the attribution of credence before positive consumer relations can be established. The green claim thrives on credibility assignment and companies’ should thus keep investing in establishing credible and trustworthy images and relations with consumers if they want to become successful players in the green, environmental market.
The inconsequential effects of both product involvement and green involvement on brand credibility are in part connected to the abovementioned explanations. However, the lack of
moderation of green involvement could also be derived from the high scores on this measure, M = 5.6, SD = .98 (1-7 scale). The high scores across conditions suggest that most consumers feel greatly for the environment and may thus be susceptible to green advertising. Seeing as respondents were presented with a green advertising in both conditions (CSR-DS v. brand historical statement) there was only a nominal incentive to award more credibility to the CSR-DS advertisement over the other: the match of CSR-DS. The high overall attitude toward the environment should lead to greater scrutiny of the message among all respondents. Expectantly, this would result in differences of CSR-DS. Nonetheless, because the centralization of people highly on the scale, too little variance may have existed to accurately expose differences. If more variation across the measure would have occurred, moderation would have been more likely.
Conversely, the lack of moderation of product involvement sparks a re-evaluation of this variables’ inclusion in the tested model. Whereas Atkinson & Rosenthal (2014) did attribute a moderating effect to product involvement on credibility, the increased level of scrutiny that is hypothesized to efface in high involvement situations (Petty and Cacioppio, 1981) does not necessarily reflect on brand credibility assessments. Instead, high product involvement, as compared to low, does seem to impact brand attitudes and purchases intentions. As Nagar (2016) has demonstrated, product involvement is predictive in explaining the level of information search and absorption of the message presented in the advertisement. However, this message does not
necessarily have to be considered credible. Apparently, the conceptual fit between CSR-DS and the green advertisement does not instigate credulity appointment, but it does affect more direct
consumer responses for those highly involved. Future research should then also reconsider the impact of product involvement, re-evaluating its impact on the relationships presented in this model.
To strengthen any future research ventures into this research subject, it is recommended to carefully consider the statements produced for testing. Although the ambiguous nature of the CSR-DS should remain intact so as to harness its ecological strengths, more precision may steer more respondents to clearly delineate its function as representing the green or environmental mission of the company (see Table 1). The clarity of the signal is an important factor in establishing credibility and any further effects derivative of this (Erdem & Swait, 1998). Furthermore, future research should consider applying the research methods set out above to brands that participants are familiar with to uncover whether this may result in effects on brand attitudes and brand image. Despite possible bias, familiarity with the brands may inform more natural credibility, attitudinal, and image scores. Moreover, in spite of the relative overlap of the sample with susceptible green consumers who have money to spend and seek consumption value beyond the product itself (Hyllegaard et al., 2010; Koch & Elmore, 2006; International Institute for Sustainable Development, 2012), future samples should extend beyond this research’s student-based, highly educated sample. More variation should be aimed for, especially since this may cause greater variation in green involvement among consumers (Ottman, 2011; Hyllegaard et al., 2010).
In order to strengthen this research methodologically, it is recommended to adjust for the several accounts when performing a replication of the study. On account of the function of the subtext, not only the text itself, but also the wording of the manipulation check should be corrected. The mixed results in the assessment of the function of the function of the CSR-DS and brand historical statement may well result from wrongful interpretation of the question asked. It should be made more explicit what the subtext entails by, for instance, including a visual. Furthermore,
although the impact of the name GMC, which includes a direct reference to “green”, has been controlled for, the use of a brand without a naming reference would be prescribed. The current results suggest that adjusting the Acura ad to become more green has minimized the bias of “green” in the coffee brand’s name, but this may not always occur. As mentioned above, the control group condition with a brand historical statement should be reevaluated as it may interferingly has contributed to credibility scoring. Additionally, the brand credibility measure could be more specifically tailored to the green context.
Notwithstanding the lack of promising effects, this research has exposed that much of brand communications strength is dependent of the level of credibility assigned to the brand. As such, forming a credible and trusting image is key in the field of credence-claim-filled green advertising. Furthermore, this research highlights that communicative elaboration on the environmental
commitment of a company affects brand attitudes, but most importantly also the decision-making process. As such, investing in corporate social responsibility disclosure statements may be wise, but its use should be considered carefully. For elaborate communication about green and environmental commitment to fare well, it is best to present it to the consumer during the decision-making process. The provision of CSR-DS in an environment that is accessed for the purposes of information search that is inspired by a pro-con evaluation in the purchasing process thus seems appropriate.
Transferring the CSR-DS to print green advertising and therewith presenting it directly to the consumer may thus be wise, but should be tailored to the consumers’ advancement in the decision process. The elaborate explanation supports the green claim the company may wish to get across, suggesting that displaying commitment to green remains key.
References
Aaker, J.L. (1997). Dimensions of brand personality. Journal of Marketing Research, 34(3), 347-356.
