Going Green: Can the Effects Be Foreseen?
An Experimental Study on the Role of Product Characteristics in Green Advertising
Fleur D. Sachse 11109939
Master’s Thesis
University of Amsterdam
Graduate School of Communication
Master’s Programme of Communication Science Supervisor: Dr. S.C.M. Welten
Abstract
Despite promoting products through environmental advertising is widely practiced nowadays, studies have shown mixed empirical findings regarding the effectiveness of such advertising strategies. The present study argued that the mixed results are possibly the result of
examining a diversity of products, which differ on many characteristics. Based on the processes underlying ELM and the associative network model, it was expected that both individual and combined effects of product involvement and a product’s environmental impact can explain inconsistencies in green advertising research. Using a 2 (product involvement: low vs. high) x 2 (environmental impact: low vs. high) between-subjects factorial design, green advertising effects are measured on attitude towards the ad, attitude towards the brand, purchase intention and level of scepticism of 227 participants (61.7% female, Mage = 29.64, 90.3% Dutch). Results demonstrate that, contrary to the expectations,
the combination of high environmental impact and low product involvement, as well as the combination of low environmental impact and high product involvement leads to the most positive persuasion outcomes and to the lowest levels of scepticism. Moreover, exposure to a green ad presenting a high environmental impact product leads to significantly higher
purchase intentions than exposure to a green ad presenting a low environmental impact product. Additionally, unlike previous studies in green advertising, no significant effects are present of product involvement on attitude towards the ad, attitude towards the brand, purchase intention and scepticism. Lastly, additional analyses reveal that perceived hedonic products lead to significantly higher levels of scepticism than perceived utilitarian products. Directions for future research as well as theoretical and practical implications are discussed.
Keywords: green advertising, product characteristics, product involvement,
Going Green: Can the Effects Be Foreseen?
An Experimental Study on the Role of Product Characteristics in Green Advertising Global warming should be limited well below 2°C in the next years to avoid drastic climate change (“Paris Agreement,” n.d.). Facing global climate challenges, public concern of the environment has risen to a higher level than ever (Xue & Muralidharan, 2015). The public’s environmental concern affects consumer behavior, such that consumers are more likely to evaluate products that positively contribute to the environment as more favorable than regular products (Kangun, Carlson & Grove, 1991; Kong & Zhang, 2013; Nyilasy, Gangadharbatla & Paladino, 2014). To meet consumers’ environmental consciousness, organizations started showing their corporate social responsibility (CSR) through, for example, the development of environmentally friendly products (Bolton & Mattila, 2014; Xue & Muralidharan, 2015). To promote green products among consumers, brands have initiated the use of “green advertising”, which involves ads that implicitly or explicitly emphasize pro-environmental aspects of a product (Alniacik & Yilmaz, 2012; Banerjee, Gulas & Iyer, 1995).
Despite 80% of the practitioners plan to increase their budget on green advertising in the future (Kong & Zhang, 2014), a univocal solution of whether green advertising works is lacking. Some studies found positive effects of green ads on persuasion outcomes (e.g., Kong & Zhang, 2013; Kong & Zhang, 2014; Nyilasy et al., 2014), whereas others claim that green ads do not yield favorable persuasion outcomes (e.g., Borin & Krishnan, 2011; Grimmer & Woolley, 2014). Due to mixed empirical findings, researchers started looking into
circumstances under which green ad effects hold. A crucial possible explanation for the inconsistent results in green advertising was the diversity of products that had been examined, as products can differ on many characteristics (Kong & Zhang, 2013). As a consequence, studies started to take into account the role of product type. Product type has
often been translated into product involvement, which concerns the degree to which consumers perceive products to be relevant and important (Drossos, Kokkinaki, Giaglis & Fouskas, 2014). However, findings on product involvement in green advertising are, again, contradicting. While some researchers (e.g., Nagar, 2015) find stronger positive effects of green ads in the context of high involvement products, others (e.g., Kong & Zhang, 2013) argue that green ads are more effective when presenting low involvement products. Considering these contradicting findings, it is conceivable that other product characteristics also play a role (Kong & Zhang, 2013), however, research on product characteristics other than product involvement remains mostly ignored. Yet, one study revealed a factor that seems promising in green advertising research: a product’s
environmental impact (Kong & Zhang, 2014). It was found that consumers perceive products that do not, or only slightly, damage the environment, to be incongruent with green ads. In this situation, they wonder how such products can actually be green, which may subsequently lead to scepticism (Alniacik & Yilmaz, 2012; Kong & Zhang, 2014). While congruity of products with ads has been mainly neglected in green advertising research, this specific product characteristic was proven to affect persuasion outcomes in other contexts of advertising research (e.g., Simpson, Horton & Brown, 1996). Besides both environmental impact and product involvement seem important variables in green advertising, the processes underlying congruity and involvement theories imply that especially the combination of these factors is promising in explaining contradicting findings in the field.
Concluding, prior research has failed to give a univocal solution of whether green advertising is effective and under what conditions. Specifically, research has demonstrated contradicting findings on product involvement and, while findings remain inconsistent, the role of other product characteristics in combination with product involvement has been ignored. Nonetheless, the combination of product involvement and environmental impact
seems a promising solution to the contradicting findings of previous studies on green advertising.
The present study aims to shed light on the debate of green advertising effects and in particular the role of product characteristics. The study contributes to the literature on green advertising by examining the combination of two product characteristics, product
involvement and environmental impact, rather than a single one, as was done by previous studies. Furthermore, results of the study could guide practitioners in making decisions about the adoption of green advertising strategies based on product characteristics, such that they can generate favorable persuasion outcomes and avoid possible backfire effects (Borin & Krishnan, 2011; Grimmer & Woolley, 2014). The following research question is formulated: “In green advertising, what is the effect of product involvement on attitude towards the ad, attitude towards the brand, purchase intention and scepticism, and to what extent does this effect depend on a product’s environmental impact?”
Theoretical Framework Green Advertising
In today’s highly competitive business environment, brands attempt to show
competitive advantage through practicing CSR initiatives, as such initiatives are considered effective ways to retain and attract consumers (Bolton & Mattila, 2014; Du, Bhattacharya & Sen, 2010; Menon & Menon, 1997). CSR programs can take many forms, ranging from charity donations to the use of sustainable packaging (Bolton & Mattila, 2014; Charter, 1998). CSR programs as the latter are often promoted through “green advertising,” which is defined as “any ad that explicitly or implicitly addresses the relationship between a
product/service and the biophysical environment, promotes a green lifestyle with or without highlighting a product or service, and presents a corporate image of environmental
responsibility” (Banerjee et al., 1995, p. 22).
Although green initiatives are practiced widespread across sectors (Olsen, Slotegraaf & Chandukula, 2014), the effectiveness of green advertising remains uncertain. Findings in green advertising research are rather contradicting: While some studies found positive effects on persuasion outcomes (e.g., Kong & Zhang, 2014; Nagar, 2015), others claim that green advertising is not more beneficial for consumer outcomes than non-green advertising (e.g., Gray-Lee, Scammon & Mayer, 1994; Grimmer & Woolleey, 2014). Additionally, some authors argue that green advertising can even result in scepticism towards such advertising (Alniacik & Yilmaz, 2012; Elving, 2013), as perceptions of “greenwashing” may arise, which concerns the process of deceiving consumers with false claims about the brands’ pro-environmental activities (Nyilasy, 2014).
In response to the contradicting findings on green advertising, researchers started looking into circumstances under which green advertising is effective, as they may explain when green advertising works (Kong & Zhang, 2013). Most studies focused on message and source characteristics (e.g., Hartmann, Apaolaza Ibáñez & Forcada Sainz, 2005; Olsen et al., 2014) and consumers’ environmental concern (e.g., Kong & Zhang, 2013; Rahman, Park & Chi, 2015), but even when considering such moderators, the effectiveness of green
advertising remains uncertain.
Product Characteristics in Green Advertising
Inconsistent results may be a consequence of previous studies’ limitations. One of the most remarkable limitations is that most studies on green advertising tested only one specific type of product (Kong & Zhang, 2013), while products differ on many characteristics and therefore require different advertising approaches (Murphy & Enis, 1986). To illustrate,
Grimmer and Woolleey (2014) investigated the effects of regular and green ads presenting a water bottle and found that green ads were not more beneficial in increasing purchase intention. Reversely, Olsen et al. (2014) studied fast-moving consumer goods including household products, and discovered that green ads positively affect brand attitude. Although both studies measured different outcome variables, it is striking that the products they studied differ on a variety of characteristics, which may have affected green advertising effects. While classifying products based on product characteristics has been mostly ignored in green advertising research, other fields of advertising research used product characteristics to classify products, for example purchase behavior (convenience, shopping and specialty goods, e.g., Jain & Hackleman, 1978), product function (utilitarian versus hedonic, e.g., Geuens, Pelsmacker & Faseur, 2011) and product involvement (high versus low, e.g., Dens & De Pelsmacker, 2010). Of these, product involvement seems most promising as this specific typology was predominantly used as product classification in previous research (e.g.,
Belanche, Flavián & Pérez-Rueda, 2017; Dens & De Pelsmacker, 2010; Flores, Chen & Ross, 2014; Te’eni-Harari, Lehman-Wilzig & Lampert, 2009; Xue & Phelps, 2013).
Product involvement. Product involvement concerns the extent to which consumers perceive products to be relevant and important (Drossos et al., 2014; Zaichkowsky, 1985). High involvement products but not low involvement products assume perceived economic or psychological risks, including loss of money and time and feelings of worry or regret after the purchase (Percy & Rossiter, 1992). Significant positive effects of product involvement on message persuasion were demonstrated by previous studies in other advertising contexts (emotional advertising contexts; e.g., Dens & De Pelsmacker, 2010; celebrity endorsement contexts, e.g., Karasiewicz & Kowalczuk, 2014).
The elaboration likelihood model (ELM, Petty et al., 1983) could explain the positive role of product involvement in prior research. The model assumes that attitude-formation
depends on the level of message elaboration, which is in turn determined by consumers’ motivation and ability to do so. When receivers have both the ability and motivation to elaborate, they process via the central route, which involves high elaboration: critical thinking, thoughtful consideration and evaluation of information, leading to an enduring attitude shift. Such critical thinking is more likely to occur when consumers perceive products to be relevant and important (Drossos et al., 2014) and when economic or social risks come along with the purchase (Percy & Rossiter, 1992). In contrast, when either ability or motivation is absent, processing goes via the peripheral route, which involves low
elaboration, relying on accessible peripheral cues. Message processing via the peripheral route leads to no or only temporary attitude shifts and is more likely to occur when consumers perceive products to be less relevant and important (Drossos et al., 2014) and when having low probability of economic or social risks coming along with the purchase (Montoro-Rios, Luque-Martínez & Rodríguez-Molina, 2008; Percy & Rossiter, 1992; Petty et al., 1983).
Based on the rationale of ELM, it is argued that different levels of involvement related to products presented in green ads, cause variations in message processing. These variations ultimately affect persuasion outcomes (Kong & Zhang, 2013). Despite this reasoning, research on product involvement in the context of green advertising is scarce. However, a couple of studies have made important contributions, of which most support the ELM. For example, Ling-Yee (1997) studied product involvement in the context of health food and found that consumers highly involved with the product engaged in more extensive green information search and purchased green products more often. Recently, Nagar (2015)
examined the influence of product involvement on the relationship between green advertising and attitude towards the ad, attitude towards the brand and brand image and concluded that in case of high involvement products, green ads positively affect persuasion outcomes. On the
other hand, when being presented green ads showing low involvement products, persuasion outcomes are less affected. Although these findings on product involvement in green advertising contexts may seem consistent, a recent study by Kong and Zhang (2013)
demonstrated contradicting results: Green appeals lead to more positive attitude towards the ad and attitude towards the brand when presenting a low involvement product than when presenting a high involvement product.
To conclude, the findings of (the few) studies that investigated product involvement in green advertising are inconsistent. Nonetheless, findings on product involvement in green advertising as well as in other advertising research indicate that product involvement is an important variable to understand how green advertising effects are affected. Following ELM and most prior research, it is expected that green ads presenting high involvement products result in more positive ad and brand evaluations and in higher purchase intentions than green ads presenting low involvement products. As mentioned before, it is argued that green advertising can lead to scepticism towards such advertising (Alniacik & Yilmaz, 2012; Elving, 2013), which is negatively related to ad and brand attitude, as well as to purchase intention (Chen & Leu, 2011). Thus, additionally, green ads with high involvement products lead to lower levels of scepticism than green ads with low involvement products. The
following hypothesis is posited:
H1: Green ads with high involvement products lead to (i) more positive attitudes towards the ad, (ii) more positive attitudes towards the brand, (iii) higher purchase intentions and (iv) lower levels of scepticism than green ads with low involvement products.
Environmental impact. To increase our knowledge about the circumstances under which green advertising is effective, product characteristics other than product involvement
should be considered. A product characteristic that seems important understanding in green advertising effects is environmental impact, as its underlying mechanisms were proven to affect different types of advertising effects, as illustrated below.
A product’s environmental impact involves the degree to which the product is
associated with environmental problems (Kong & Zhang, 2014). Alniacik and Yilmaz (2012) conceptualize a product’s environmental impact as the extent to which the presented product is relevant to the green ad. According to the researchers, highly environmentally relevant products are easily associated with environmental problems (e.g., cars), whereas low environmentally relevant products are less easily associated with environmental problems (e.g., pens). They argue that ads presenting highly environmentally relevant products increase ad effectiveness because they meet consumers’ expectations of the product’s environmental aspect, making the green ad relevant to the product. This perspective on product relevance is supported by the principle of congruity in human thinking (Osgood & Tannenbaum, 1955), which assumes that “changes in humans’ evaluation are always in the direction of increased congruity with the existing frame of reference” (p. 43). This means, the higher the perceived congruity between observed objects, the more positive someone’s evaluations and attitudes will be, whereas the lower the perceived congruity between objects, the more negative someone’s evaluations and attitudes will be. Applying this principle to green advertising, a product should be congruent (e.g., a car) rather than incongruent (e.g., a pen) with a green ad to generate positive advertising effects (Lee, Kim & Yu, 2015).
While research on product congruity in green advertising remains mostly ignored (except for Alniacik & Yilmaz, 2012; Kong & Zhang, 2014), evidence was found for the effectiveness of product congruity in non-green advertising contexts. For example, Simpson et al. (1996) examined product congruity with nudity in ads and concluded that persuasion outcomes were more positive when nudity was presented with a congruent product (body oil)
than when presented with an incongruent product (wrench set). Likewise, Lee et al. (2015) examined congruity of products with visual art advertising. They demonstrated that
respondents had higher purchase intentions when visual art ads presented a congruent, hedonic product, than when they presented an incongruent, utilitarian product. Hedonic products are considered to be congruent with visual art ads because the emotional experience related to hedonic products fits the emotions implied in visual art. Utilitarian products, in contrast, profit more from cognitive than emotional advertising (Lee et al., 2015).
The associative network model explains underlying mechanisms of the positive effect of product congruity in advertising research (Henley, Philhours, Ranganathan & Bush, 2009). The model assumes our memory consists of nodes, representing memories, and links,
representing the connection between nodes (Lang, 2000). Memories can be activated through associations. When two nodes are strongly connected, spreading activation from one node to the other becomes automatic, making it easier for the memory network to engage in cognitive processing (Grunert, 1996). Thus, strong associative links, in other words highly congruent associative links, are considered more persuasive than weak, incongruent associative links. Based on this reasoning, when consumers encounter a green ad presenting a product that damages the environment, like a car, a strong associative link between the green ad and the polluting aspect of the car will be built automatically. As a result of the strong associative link, consumers consider the ad to make sense to the product. In other words, high congruity exists between the green ad and the product, which leads to the message being more
persuasive. In contrast, when consumers encounter a low environmental impact product, like a pen, associative links between the green ad and the environmental aspect of the pen are weaker, as the environmental aspect of a pen is not explicit. Consumers may therefore become sceptical, wondering how a pen can actually be green. Concluding, low congruity exists between the product and the green ad, leading to the green ad being less persuasive
(Grunert, 1996; Kong & Zhang, 2014; Lang, 2000; Xue & Muralidharan, 2015).
Kong and Zhang (2014) based their study on the role of environmental impact on the associative network model rationale and findings were in favor of the model. According to the researchers, in case of high environmental impact products, respondents had more a more positive brand attitude and higher purchase intentions than in case of low environmental impact products because the product and green ad were considered congruent. Alniacik and Yilmaz (2012) found similar effects and both studies suggest that green ads presenting low environmental impact products may have elicited scepticism.
Concluding, following the associative network model, it is expected that green ads presenting products with high environmental impact have high congruity with the green ad and are therefore more persuasive. Reversely, green ads presenting products with low environmental impact have low congruity and are therefore less persuasive. The following hypothesis is formulated:
H2: Green ads with high environmental impact products lead to (i) more positive attitudes towards the ad, (ii) more positive attitudes towards the brand, (iii) higher purchase intentions and (iv) lower levels of scepticism than green ads with low environmental impact products.
The fact that Kong and Zhang (2013) found low involvement products to be more effective on brand attitude and purchase intention whereas other researchers (e.g., Ling-Yee, 1997) found high involvement products to be more persuasive, may be the result of
examining products with different levels of environmental impact. To illustrate, Kong and Zhang (2013) tested a green ad presenting a camera as a high involvement product, versus a green ad presenting detergent as a low involvement product. However, considering the rationale of the associative network model, it could be that the detergent elicited stronger
associations with environmental problems than the camera, and as previously discussed, such strong associations, thus high congruity, lead to messages being more persuasive. On the other hand, a camera may have elicited weaker associations with environmental problems, resulting in low congruity and ultimately in messages being less persuasive. Respondents might even have become sceptical as they critically evaluated how a camera can be green. In contrast to Kong and Zhang, other studies on product involvement in green advertising tested products with higher environmental impact in the high compared to the low product
involvement conditions, for example a car (high involvement and high environmental impact, Xue & Muralidharan, 2015) versus napkins (low involvement and low environmental impact, Xue & Muralidharan, 2015). Thus, the possible strong associative links between the high involvement products and their environmental aspects could explain the positive persuasion outcomes for these high involvement products. This phenomenon may indicate that the combination of product involvement and environmental impact is important in green advertising.
It seems essential to explore the combined effects of the two factors as they may explain inconsistencies in findings on product involvement in green advertising. As previously discussed, Kong and Zhang (2014) demonstrated the moderating influence of environmental impact on the relationship between green appeals and brand attitude and purchase intention, such that green ads with low environmental impact products led to less favorable outcomes, whereas high environmental impact products led to more favorable outcomes. In their study, they tested two low involvement products. Alniacik and Yilmaz (2012) found similar effects for the moderating role of environmental impact, but then for two high involvement products. While both studies tested products with a similar level of product involvement, neither of them included product involvement as a factor. Nonetheless, as ELM assumes, ads with high involvement products are critically evaluated and should
therefore be more effective on persuasion outcomes than ads with low involvement products. However, following the associative network model, this is only the case when the product’s environmental impact is high, as the product is congruent with the green ad and the ad thus makes sense. On the other hand, when products have low environmental impact, the critical thinking may rather result in scepticism, as consumers may critically wonder how products with low environmental impact can actually be green.
Despite this reasoning, no study has looked into the combined effects of product involvement and environmental impact in green advertising before. However, based on the processes underlying product involvement and product congruity, it is expected that
environmental impact moderates the effect of product involvement on persuasion outcomes. The following hypotheses are tested:
H3: In green advertising, the effect of product involvement on attitude towards the ad, attitude towards the brand, purchase intention and scepticism is moderated by a product’s environmental impact. Specifically,
H3a: When a product’s environmental impact is high, it is expected that green ads with high involvement products lead to (i) more positive attitudes towards the ad, (ii) more positive attitudes towards the brand, (iii) higher purchase intentions and (iv) lower levels of scepticism than green ads with low involvement products.
H3b: When a product’s environmental impact is low, it is expected that green ads with high involvement products lead to (i) less positive attitudes towards the ad, (ii) less positive attitudes towards the brand, (iii) lower purchase intentions and (iv) higher levels of scepticism than green ads with low involvement products.
Figure 1. Conceptual model.
Methods
The proposed study featured a 2 (product involvement: low vs. high) x 2 (environmental impact: low vs. high) between-subjects factorial design. To choose the appropriate stimulus material for the main experiment, a pre-test was conducted. The goal of the pre-test was to select four products that differed in terms of product involvement and environmental impact, such that four experimental conditions could be created. It was crucial to select four different products in order to elicit the previously discussed weak and strong associations with the environmental aspects of the products, while reassuring similarities on other product characteristics.
Pre-test
Design and participants. The pre-test consisted of a within-subjects design. Within-subjects designs require fewer participants (Field, 2013), which was advantageous regarding the study’s time frame. Twenty-six participants (61% female, Mage= 24.04 years, SD = 2.13,
age-range: 19-29) were recruited using a convenience sample on social media and through personal communication.
Materials, procedure and dependent measures. For each of the four conditions of the main experiment, two products were selected, such that the pre-test material consisted of 8 products: a dishwasher, a dryer, a mobile phone, a bike, a light bulb, a battery, a pack of cereal and a pen (see Figure A1). The products were selected based on previous studies (e.g., Alniacik & Yilmaz, 2012; Kong & Zhang, 2014) and on own insights, considering the previously discussed definitions of the two factors. The products appeared in the same order for all participants. Moreover, to avoid sequence effects, the products were mixed up in terms of their involvement and environmental impact.
After participants had given their informed consent, an explanation of the test
followed, stating they were about to evaluate 8 products. Subsequently, participants rated the dependent measures for each product. First, they rated products’ level of involvement on three 7-point Likert items, adopted from a scale from Drossos et al. (2014). Items included “choosing this product involves an important decision”, “decision requires a lot of thought” and “a lot to lose if you choose the wrong brand” (1 = strongly disagree to 7 = strongly agree). The results of factor and reliability analyses of product involvement for each product are shown in Table 1. Although some Cronbach’s alpha’s were below .60 and could be improved when deleting an item for some products, no items were deleted to keep the number of items measuring product involvement constant.
Table 1.
Results of principal component analyses and reliability analyses of product involvement for each product.
Products Eigenvalue Explained variance Cronbach’s α
Dishwasher 1.53 51.13% .46 Cereal 1.52 50.67% .59 Light bulb 1.72 57.57% .62 Bike 2.06 68.53% .76 Pen 2.22 74.02% .82 Battery 1.64 54.81% .54 Dryer 1.75 58.48% .61 Phone 1.52a 50.54% .68
Note. Items without superscript loaded on one component. a Item loaded on two components, with second component EV = 1.07, explained variance = 86.31%.
Second, participants rated products’ environmental impact on a single item assessing the extent to which the product harms the environment (1 = very little to 7 = very much; Kong & Zhang, 2014). Third, participants rated products’ utilitarian versus hedonic character by a single 7-point semantic differential scale (1 = functional product to 7 = entertainment product; Kempf, 1999). Products’ utilitarian versus hedonic character was measured to determine whether it was a potential confounding factor, as previous studies on product advertising found confounding effects for this variable (e.g., Çal & Adams, 2014). The means and standard deviations of the dependent measures for all of the products are presented by Table A1.
effects of the 8 products on product involvement and environmental impact. There was a significant, large effect of product on product involvement, Wilks’ Lambda = .06, F (7, 19) = 219.87, p < .001, ηp2= .94, as well as a significant, large effect of product type on
environmental impact, Wilks’ Lambda = .12, F (7, 19) = 40.79, p < .001, ηp2 = .99. Pairwise
comparisons were used to make post hoc comparisons between the 8 products, of which results are shown by Table A1.
The first step of the product selection was evaluating highest and lowest means on product involvement and environmental impact in each experimental condition (see Table A1). On product involvement, the dryer and phone scored highest (Mdryer = 5.55, SD = .86; Mmobilephone = 6.27, SD= .49), while the battery and the pen scored lowest (Mbattery= 2.42, SD
= 1.05; Mpen = 1.26, SD = .40). On environmental impact, the dryer and the battery scored
highest (Mdryer= 5.42, SD= 1.10; Mbattery= 5.42, SD= 1.27), whereas the bike and the pen
scored lowest (Mbike= 2.19, SD= 1.23; Mpen = 2.46, SD= 1.27). The second step was
determining significant differences between the other level of both product involvement and environmental impact. As can be seen in Table A1, significant differences between those levels for the dryer, the battery and the pen were indeed present, whereupon these products were selected for the main experiment.
Regarding products in the high involvement, low environmental impact condition, however, results were less straightforward. As mentioned, the mobile phone scored highest on product involvement. Nonetheless, the mobile phone also scored quite high on
environmental impact (M = 4.85, SD = .78), being not significantly different from the high environmental impact conditions (see Table A1). The bike, on the other hand, scored high on product involvement (M = 4.59, SD = 1.22), being significantly different from the two low involvement conditions, the battery and the pen, but also low on environmental impact (M = 2.19, SD = 1.23), being significantly different from the two high environmental impact
conditions, the dryer and the battery. Therefore, the bike was chosen as fourth product. To sum up, based on significantly different scores between the products on product involvement and environmental impact, the dryer, the bike, the battery and the pen were selected as products for the stimulus material of the main experiment. All four products were environmentally friendly advertised before, see Figures B1, B2, B3 and B4.
Main Experiment
Design and participants. A 2 (product involvement: low vs. high) x 2
(environmental impact: low vs. high) between-subjects factorial design was used to test the proposed model. 236 participants were recruited through social media and personal
communication, resembling the recruitment method of the pre-test. 9 participants were
excluded from the dataset because they did not complete the dependent variables (n = 2), they were familiar with the brand name (n = 5), they exactly guessed the goal of the study (n = 1) or they chose having seen the wrong product (n = 1).1 The new total of participants was 227 (61.7% female, Mage= 29.64, SD = 10.65, age-range: 19-64, 90.3% Dutch, 48.5% working).
Materials, procedure and dependent measures. The stimulus material consisted of a green ad presenting one of the four products that were determined by the pre-test and was identical through all conditions (see Figure C1). The ad’s background consisted of grass and a blue sky. The product was presented in the middle of the grass and in the right bottom corner of the ad, a green logo was placed saying ‘eco-friendly’. Also, in the upper middle of the ad, a slogan was presented stating ‘Go green. Go Clover’. ‘Clover’ as the brand name was made up to avoid confounding influences of prior brand associations.
The complete survey as presented to the participants can be found in Appendix D. People who followed the link to the survey were given an introduction that stated the study’s goal and provided contact information. After participants had given their informed consent, a short explanation of the experiment’s procedure was shown. Participants were then randomly
assigned to one out of four conditions, in which they had to look at the green ad for a couple of seconds. Afterwards, participants were guided through the dependent measures, which are discussed below. Finally, the survey ended with a debriefing, stating that the ad and the brand, as shown in the ad, were fictional.
Attitude towards the ad. Attitude towards the ad was measured using a 7-point
semantic differential scale from Xue and Muralidharan (2015). Four adjective items were presented: bad/good, unpleasant/pleasant, unfavorable/favorable and negative/positive, (1 = most negative to 7 = most positive, loading on 1 component with EV = 2.99; explained variance = 74.74%, Cronbach’s α = .88, M = 4.64, SD = 1.22).
Attitude towards the brand. Attitude towards the brand was measured using a 7-point
semantic differential scale (Alniacik & Yilmaz, 2012; Kong & Zhang, 2013), including bad/good, unfavorable/favorable and negative/positive (1 = most negative to 7 = most positive, loading on 1 component with EV = 2.63; explained variance = 87.76%, Cronbach’s α = .93, M = 4.91, SD = 1.24).
Purchase intention. Purchase intention was measured using a scale of Alniacik and
Yilmaz (2012), including unlikely/likely, improbable/probable, impossible/possible (1 = most negative to 7 = most positive, loading on 1 component with EV = 2.62; explained variance = 87.46%, Cronbach’s α = .93, M = 4.37, SD = 1.41).
Scepticism. In this study’s context, scepticism is operationalized as scepticism
towards green advertising. A 7-point Likert scale of Finisterra do Paço and Reis (2012) was used to measure scepticism with four items including “most environmental claims that are presented in advertising are true” (1 = strongly disagree to 7 = strongly agree, loading on 1 component with EV = 2.54; explained variance = 63.40%, Cronbach’s α = .81, M = 4.24, SD = 1.24).
character was measured as a control variable. Besides, demographic information was
collected: Participants were asked for their gender, age, nationality and employment status, as well as the recruitment method, as recruiting through personal communication may have affected how they filled in the survey.
Manipulation check. To explore whether significant differences existed between perceptions of the products’ level of involvement and environmental impact, manipulation checks were done. Participants were asked to evaluate the level of involvement and
environmental impact related to the product they were exposed to, by evaluating the same statements as proposed in the pre-test: one item measuring environmental impact (Kong & Zhang, 2014) and three items measuring product involvement (Drossos et al., 2014; loading on 1 component with EV = 2.60; explained variance = 86.66%, Cronbach’s α = .92, M = 3.26, SD = 1.68).
Results Preliminary Analyses
Randomization check. To check whether the randomization process through conditions was successful, a MANOVA test was performed with product involvement and environmental impact as factors and gender, age, nationality, employment status, recruitment method and products’ utilitarian versus hedonic character as dependent variables. Results showed no significant effect of product involvement on the dependent variables, Wilks’ Lambda = .98, F (6, 218) = .69, p = .655. Moreover, no significant results were found for environmental impact on the dependent variables, Wilks’ Lambda = .99, F (6, 218) = .31, p = .932. These non-significant results indicate that the randomization process through the
conditions was successful.
Covariates check. To discover whether the variables gender, age, nationality, employment status, recruitment method and products’ utilitarian versus hedonic character
correlate with attitude towards the ad, attitude towards the brand, purchase intention and scepticism, a correlation analysis was conducted. Results of the analysis are shown in Table 2 below and revealed only one significant correlation between products’ utilitarian versus hedonic character and attitude towards the brand, whereupon this variable was included as covariate in the main analysis of attitude towards the brand.2
Table 2.
Results of bivariate correlation analysis of control variables with dependent variables.
Control variables r Attitude towards the ad r Attitude towards the brand r Purchase intention r Scepticism Gender .03 .10 .09 .02 Age .06 -.06 .08 .11 Nationality -.02 .02 -.00 .02 Employment status -.05 -.19 .00 -.01 Recruitment method -.04 .06 -.01 -.12 Products’ utilitarian/hedonic character -.00 -.14* -.02 .11
Note. * Correlation is significant at the p < .05 level.
Manipulation check. Manipulation checks were carried out using independent samples t-tests. For product involvement, the homogeneity of variances was violated by Levene’s F test, F (225) = 7.23, p = .008. Nonetheless, results showed a successful manipulation for this variable, t (209.72) = -13.10, p < .001. Thus, participants in the low product involvement condition rated the level of product involvement significantly lower (M
= 2.16, SD = 1.10) than participants in the high product involvement condition (M = 4.37, SD = 1.43). Besides, the manipulation check for environmental impact was successful, t (225) = -12.18, p < .001, demonstrating that participants in the low environmental impact condition evaluated the level of environmental impact significantly lower (M = 2.56, SD = 1.31) than participants in the high environmental impact condition (M = 4.89, SD = 1.56). For
environmental impact, the assumption of homogeneity of variances was satisfied, F (225) = .14, p = .705.
Main Analyses
Table 3 summarizes the previously stated hypotheses that were tested using two-way ANOVA’s for attitude towards the ad, purchase intention and scepticism and a two-way ANCOVA for attitude towards the brand.
Table 3.
The proposed hypotheses.
H1 Green ads with high involvement products lead to (i) more positive attitudes towards the ad, (ii) more positive attitudes towards the brand, (iii) higher purchase intentions and (iv) lower levels of scepticism than green ads with low involvement products. H2 Green ads with high environmental impact products lead to (i) more positive attitudes
towards the ad, (ii) more positive attitudes towards the brand, (iii) higher purchase intentions and (iv) lower levels of scepticism than green ads with low environmental impact products.
H3a When a product’s environmental impact is high, it is expected that green ads with high involvement products lead to (i) more positive attitudes towards the ad, (ii) more positive attitudes towards the brand, (iii) higher purchase intentions and (iv) lower levels of scepticism than green ads with low involvement products.
H3b When a product’s environmental impact is low, it is expected that green ads with high involvement products lead to (i) less positive attitudes towards the ad, (ii) less positive attitudes towards the brand, (iii) lower purchase intentions and (iv) higher levels of scepticism than green ads with low involvement products.
Effects on attitude towards the ad. A two-way ANOVA with product involvement and environmental impact as factors and attitude towards the ad as dependent variable (M = 4.65, SD = .08) revealed no significant main effect of product involvement on attitude towards the ad, F (1, 223) = .02, p = .900, ηp2 = .00. Thus, participants in the high product
involvement conditions did not have more positive attitudes towards the ad (M = 4.64, SD = .11) than participants in the low product involvement conditions (M = 4.66, SD = .11), thereby rejecting hypothesis 1(i). Second, there was no significant main effect of
environmental impact on attitude towards the ad, F (1, 223) = 1.42, p = .234, ηp2 = .01. This
means, participants exposed to a high environmental impact product did not have
significantly more positive attitudes towards the ad (M = 4.74, SD = .12) than participants exposed to a low environmental impact product (M = 4.55, SD = .11) and thus hypothesis 2(i) was not accepted. Lastly, there was no significant interaction effect between the two factors on attitude towards the ad, F (1, 223) = 3.59, p = .060, ηp2 = .02, thereby rejecting hypotheses
3a(i) and 3b(i), as summarized in Table 3. Despite the interaction effect being not significant, it is interesting to look at the pattern in Figure 2 below as it demonstrates opposite direction of 3a(i) and 3b(i): Participants who saw a high environmental impact product did not have more positive but instead less positive attitudes towards the ad when being exposed to a high involvement product (M = 4.58, SD = .16) than participants being exposed to a low
involvement product (M = 4.91, SD = .16). Similarly, participants who saw a low
towards the ad when being exposed to a high involvement product (M = 4.69, SD = .16) than when being exposed to a low involvement product (M = 4.41, SD = .16).
Figure 2. Non-significant interaction effect of product involvement and environmental
impact on attitude towards the ad.
Effects on attitude towards the brand. A two-way ANCOVA with product
involvement and environmental impact as factors and attitude towards the brand as dependent variable (M = 4.92, SD = .08), controlling for products’ utilitarian versus hedonic character, revealed no significant main effect of product involvement on attitude towards the brand, F (1, 222) = .49, p = .486, ηp2 = .00. Thus, participants in the high product involvement
condition did not have significantly more positive attitudes towards the brand (M = 4.97, SD = .11) than participants in the low involvement condition (M = 4.86, SD = .11) and therefore, hypothesis 1(ii) was rejected. Moreover, there was no significant main effect of
environmental impact on attitude towards the brand, F (1, 222) = .44, p = .508, ηp2 = .00. This
means, participants exposed to a high environmental impact product did not have
significantly more positive attitudes towards the brand (M = 4.97, SD = .12) than participants 1 2 3 4 5 6 7
Low environmental impact High environmental impact
A tt it ude tow ar ds the ad
Low product involvement High product involvement
exposed to a low environmental impact product (M = 4.86, SD = .11), thereby rejecting hypothesis 2(ii). Also, the ANCOVA showed a significant effect of the covariate, products’ utilitarian versus hedonic character, on attitude towards the brand, F (1, 222) = 6.23, p = .013, ηp2 = .03. Finally, the ANCOVA showed a significant, small interaction effect of the
two factors on attitude towards the brand, F (1, 222) = 8.31, p = .004, ηp2 = .04. However,
Figure 3 below shows that the effect was opposite to what hypotheses 3a(ii) and 3b(ii) stated (as summarized in Table 3): Participants who saw a high environmental impact product did not have more positive but instead less positive attitudes towards the ad when being exposed to a high involvement product (M = 4.80, SD = .16) than participants being exposed to a low involvement product (M = 5.20, SD = .16). Similarly, participants who saw a low
environmental impact product did not have less positive but instead more positive attitudes towards the ad when being exposed to a high involvement product (M = 5.12, SD = .16) than participants being exposed to a low involvement product (M = 4.57, SD = .16). Based on the reversed direction of the results, hypotheses 3a(ii) and 3b(ii) were rejected.
Figure 3. Significant interaction effect of product involvement and environmental impact on
attitude towards the brand. 1 2 3 4 5 6 7
Low environmental impact High environmental impact
A tt it ude tow ar ds the br and
Low product involvement High product involvement
Effects on purchase intention. A two-way ANOVA with product involvement and environmental impact as factors and purchase intention as dependent variable (M = 4.38, SD = .09) showed no significant main effect of product involvement on purchase intention, F (1, 223) = .07, p = .786, ηp2 = .00. Thus, participants in the high product involvement condition
did not score higher on purchase intention (M = 4.35, SD = .13) than participants in the low involvement condition (M = 4.40, SD = .11). Based on this, hypothesis 1(iii) was rejected. Nonetheless, there was a significant, small main effect of environmental impact on purchase intention, F (1, 223) = 6.18, p = .014, ηp2 = .03. Thus, participants scored significantly higher
on purchase intention when seeing a high environmental impact product (M = 4.61, SD = .13) than when seeing a low environmental impact product (M = 4.15, SD = .13), supporting hypothesis 2(iii). Lastly, there was a significant, small interaction effect of the two factors on purchase intention, F (1, 223) = 4.56, p = .034, ηp2 = .02. However, the effect was reversed of
what hypothesis 3a(iii) and 3b(iii) stated (see Table 3). As presented in Figure 4 below, participants in the high environmental impact conditions did not score higher but instead lower on purchase intention when seeing a high involvement product (M = 4.39, SD = .18) than when seeing a low involvement product (M = 4.83, SD = .18). Moreover, participants in the low environmental impact conditions did not score higher but instead lower on purchase intention when seeing a low involvement product (M = 3.98, SD = .18) than when seeing a high involvement product (M = 4.32, SD = .19). Thus, hypothesis 3a(iii) and 3b(iii) were rejected.
Figure 4. Significant interaction effect of product involvement and environmental impact on
purchase intention.
Effects on scepticism. A two-way ANOVA with product involvement and
environmental impact as factors and scepticism as dependent variable (M = 4.24, SD = .08) showed no significant main effect of product involvement on scepticism, F (1, 223) = .07, p = .311, ηp2 = .01. Participants in the high product involvement conditions did not have
significantly lower levels of scepticism (M = 4.15, SD = .12) than participants in the low involvement condition (M = 4.32, SD = .12), thereby rejecting hypothesis 1(iv). There was also no significant main effect of environmental impact on scepticism, F (1, 223) = 1.58, p = .211, ηp2 = .01, which means that participants exposed to a high environmental impact
product did not score significantly lower on scepticism (M = 4.14, SD = .12) than participants exposed to a low environmental impact product (M = 4.34, SD = .12). Hypothesis 2(iv) was therefore also rejected. Lastly, the ANOVA revealed no significant interaction effect of the two factors on scepticism, F (1, 223) = 3.37, p = .068, ηp2 = .02, thereby rejecting hypotheses
3a(iv) and 3b(iv), as summarized in Table 3. Despite the interaction effect not being 1 2 3 4 5 6 7
Low environmental impact High environmental impact
Purc has e int ent ion
Low product involvement High product involvement
significant, it is interesting to look at the reversed pattern that appears from the results. As can be seen in Figure 5, participants in the high environmental impact conditions did not have lower but instead higher levels of scepticism when seeing a high involvement product (M = 4.20, SD = .16) than when seeing a low involvement product (M = 4.07, SD = .17). Also, participants that saw a low environmental impact product did not have higher but instead lower levels of scepticism when seeing a high involvement product (M = 4.12, SD = .16) than when seeing a low involvement product (M = 4.57, SD = .16).
Figure 5. Non-significant interaction effect of product involvement and environmental
impact on scepticism.
Discussion and Conclusion
The present study examined the effect of product involvement on attitude towards the ad, attitude towards the brand, purchase intention and scepticism and the extent to which this effect depends on a product’s environmental impact. Results demonstrate no significant effects of product involvement on attitude towards the ad, attitude towards the brand, purchase intention and scepticism. However, a significant positive effect of environmental
1 2 3 4 5 6 7
Low environmental impact High environmental impact
Scept
ici
sm
Low product involvement High product involvement
Pen Battery
Dryer Bike
impact on purchase intention is present, while the effect of environmental impact on attitude towards the ad, attitude towards the brand and scepticism stays out. Moreover, results show a significant reversed interaction effect of product involvement and environmental impact on attitude towards the brand and purchase intention. While it was expected that the combination of high environmental impact and high product involvement would lead to more positive persuasion outcomes than the combination of high environmental impact and low product
involvement, results show that instead, for high environmental impact products, low
involvement products lead to more positive attitudes towards the brand and higher purchase intentions than high involvement products. Moreover, while it was hypothesized that the combination of low environmental impact and high product involvement would lead to less positive persuasion outcomes than the combination of low environmental impact and low
product involvement, results demonstrate that instead, for low environmental impact
products, high involvement products lead to more positive attitudes towards the brand and higher purchase intentions than low involvement products. Finally, additional analyses reveal a significant positive effect of products’ utilitarian versus hedonic character on scepticism.2 Unexpected Findings and Directions for Future Research
Product involvement. To start with, results demonstrate no significant effects of product involvement on any of the dependent variables. This is not in line with ELM and most previous studies on product involvement in green advertising (e.g., Ling-Yee, 1997; Nagar, 2015), which found more positive persuasion outcomes for high involvement products than for low involvement products. A possible explanation for the absence of significant effects is the type of product involvement related to the utilitarian versus hedonic character of the experimental products. Mittal (1989) argues that different types of product involvement can be distinguished for utilitarian and hedonic products. Hedonic products are linked to enduring product involvement, which concerns products’ perceived importance and
relevance. On the other hand, utilitarian products are linked to purchase-decision
involvement, which involves the level of involvement at the time of the purchase. In this study, it was expected that, as a result of differences in levels of enduring product
involvement (i.e., perceptions of products’ relevance and importance), variations in message processing would occur, eventually leading to more or less positive persuasion outcomes (ELM, Drossos et al., 2014; Petty et al., 1983). However, because of the experimental products’ perceived utilitarian character,3
the products were conceivably linked to purchase-decision involvement and not to enduring product involvement. As a consequence,
insufficient variations in message processing may have existed, leading to non-significant results. Unfortunately, this speculation cannot be supported by the data as the manipulation check of product involvement measured the construct of purchase-decision involvement and not enduring product involvement,4 thus, conclusions about the intended enduring product involvement cannot be drawn. Future studies can solve the above-mentioned problem by distinguishing between Mittal’s (1989) different types of product involvement while considering experimental products’ utilitarian versus hedonic character.
Environmental impact. Second, a significant positive effect is present for environmental impact on purchase intention. This is in line with findings from previous studies on green advertising (Alniacik & Yilmaz, 2012; Kong & Zhang, 2014), and findings on product congruity in other contexts of advertising research (Lee et al., 2015; Simpson et al., 1996). Following the reasoning of the associative network model, strong associations, thus high congruity, may have existed between the green ad and the high environmental impact products. Subsequently, high congruity leads to more positive persuasion outcomes (Lang, 2000; Osgood & Tannenbaum, 1955). Nevertheless, we should be careful drawing conclusions about the role of congruity in this study as the processes underlying congruity were not measured. In order to demonstrate a causal relationship between congruity and
persuasion outcomes, more research is needed. Future studies could use the implicit association test to measure the associative strengths between green ads and products’ environmental aspects, which is considered a reliable method to assess the strength of associations (Teige-Mocigemba & Klauer, 2015). Because the principle of congruity in human thinking as well as the associative network model indicates that high congruity between associations is crucial to generate positive persuasion outcomes (Osgood & Tannenbaum, 1955; Lang, 2000), measuring associative strengths in future studies would deepen our understanding of the role of product congruity in (green) advertising research.
It is striking that while results demonstrate a significant effect of environmental impact on purchase intention, significant effects on attitude towards the ad, attitude towards the brand and scepticism stay out. It could be that participants’ decision-making throughout the experiment was impulsive rather than deliberate. This idea may be supported by the previously discussed speculation about product involvement: Because participants might not have been enduringly involved in the products, they processed via the peripheral route, which is related to impulsive decision-making. Central message processing, on the other hand, is related to deliberate decision-making (Petty & Brinol, 2008). It is worth noting that, although not being significant, the effects of environmental impact on attitude towards the ad, attitude towards the brand and scepticism point in the same direction as the significant effect on purchase intention. To explore whether the effects become significant, future studies could replicate this study considering the aforementioned recommendations about product involvement. The successful use of enduring product involvement as a variable possibly enables the occurrence of both impulsive and deliberate decision-making processes (Petty & Brinol, 2008).
Moderation effect. Results furthermore demonstrate significant, reversed interaction effects of product involvement and environmental impact on attitude towards the brand and
purchase intention. First, contrary to our expectations, results show that for high
environmental impact products, the low involvement product (battery) leads to more positive attitudes towards the brand and higher purchase intentions than the high involvement product (dryer). This reversed effect may be explained by the lack of textual claims in the stimulus material. To clarify, some researchers argue that the use of textual, environmental claims can lead to positive advertising responses (e.g., Hartmann et al., 2005; Xue & Muralidharan, 2015). According to Alniacik and Yilmaz (2012), especially utilitarian products profit from textual claims because they are chosen based on functional product attributes provided by textual information. Lee et al. (2015) support this argument by suggesting that utilitarian products, in contrast to hedonic products, profit from cognitive more than emotional advertising. These researchers’ arguments, in combination with the experimental products’ utilitarian character,3 make it plausible that textual claims would have been a beneficial addition to the stimulus material. While textual claims thus seem important for both the dryer and battery, there are reasons to believe that the dryer would have profited more from textual claims than the battery, potentially explaining why the battery ad was more persuasive. Houston and Rothschild (1977, in Mittal, 1989) claim that the higher the level of purchase-decision involvement, the more extensive the search for information. Following Houston and Rothschild’s claim and the manipulation check showing differences in the products’ level of decision involvement, it is likely that the dryer, associated with higher purchase-decision involvement than the battery,4 would have profited more from textual claims than the battery. In general, this view implies that not only product characteristics but also
message characteristics may be important in green advertising effects. More specifically, the aforementioned arguments of prior researchers (e.g., Alniacik & Yilmaz, 2012; Lee et al., 2015) about textual claims in green ads imply that message characteristics (e.g.,
explaining mixed findings in the field. Further research on the role of message characteristics in combination with product characteristics is therefore needed.
Second, contradicting the expectations, results demonstrate that for low environmental impact products, the high involvement product (bike) lead to more positive attitudes towards the brand and higher purchase intentions than the low involvement product (pen). A possible explanation for this reversed interaction effect is the level of product congruity. While it was intended to evoke weak associative links and low congruity between the green ad and the environmental aspect of the products, it is conceivable that instead, participants perceived the bike to be having high congruity with the ad. To clarify, the environmental aspect of a bike may be common knowledge: It is straightforward how the bike contributes to a better environment and it is well-known that the bike is an environmentally friendly vehicle compared to alternative vehicles (Horton, 2006). Consequently, strong rather than weak associations, thus high congruity, may have existed between the green ad and the
environmental aspect of the bike. Following the associative network model, this conceivable high congruity resulted in the bike ad being persuasive. Regarding the pen, in contrast, it may be less straightforward how this product contributes a better environment. Unlike the bike, this product is not well-known as an environmentally friendly product. Subsequently, as intended, weak associations between the green ad and the product may have existed, resulting in less positive persuasion outcomes. The problem regarding the product congruity of the bike supports the previously mentioned need for measuring product congruity in future research. Aside from drawing conclusions about the causal role of product congruity, measuring congruity in pre-tests of future studies may help researchers to select appropriate products for their experiments, such that expected results are more likely to be found. Finally, products’ utilitarian versus hedonic character seems important in green advertising: This variable predicts scepticism in the way that perceived hedonic products lead
to higher levels of scepticism than perceived utilitarian products.2 Moreover, as previously explained, interaction effects may exist between products’ utilitarian versus hedonic character and message characteristics. Therefore, further research is deemed necessary.
Theoretical and Practical Implications
The present study adds to the literature on product congruity by providing evidence for the effectiveness of product congruity in green advertising, thereby extending theory on product congruity in other advertising contexts. Also, the study expands green advertising literature by proving that products’ utilitarian versus hedonic character affects scepticism. Furthermore some important practical implications derive from this study. Primarily, this study provides support for the view that promoting products by using green advertising is not an unconditional solution to enhance consumer attitudes and purchase intentions. Rather, green advertising appears to be successful under certain circumstances that practitioners should consider to avoid scepticism and potential perceptions of greenwashing. While many of these circumstances remain unclear, the study does provide evidence for the importance of considering products’ environmental impact before adopting advertising strategies. Green ads with high environmental impact products lead to higher purchase intentions, while
practitioners may prefer to choose a different strategy than green advertising when the product to be promoted has low environmental impact. Findings furthermore suggest that practitioners selling hedonic products should be more careful with the use of green advertising than practitioners selling utilitarian products, as hedonic products lead to the highest levels of scepticism.
In conclusion, the study stresses the necessity of further research on green advertising and the role of product characteristics, potentially in combination with message
characteristics. Insights from numerous studies together will have to point out if and how the effects of going green can be foreseen.
References
AAA Batterij [Online image]. (n.d.). Retrieved November 26, 2016 from https://www.allekabels.nl/aaa-batterij/4571/1289491/aaa-batterij.html
Alniacik, U., & Yilmaz, C. (2012). The effectiveness of green advertising: Influences of claim specifity, product’s environmental relevance and consumers’ pro-environmental character. Amfiteatru Economic, 14(31), 207.
Baggrys [Online image]. (n.d.). Retrieved November 26, 2016 from
http://www.medicalhealthtips.com/top/best-corn-flakes-brands-in-india/
Banerjee, S., Gulas, C. S., & Iyer, E. (1995). Shades of green: A multidimensional analysis of environmental advertising. Journal of Advertising, 24(2), 21-31.
Belanche, D., Flavián, C., & Pérez-Rueda, A. (2017). Understanding interactive online advertising: Congruence and product involvement in highly and lowly arousing, skippable video ads. Journal of Interactive Marketing, 37, 75-88.
BIC ECOlutions Clic Stick Retractable Ballpoint Pen [Online image]. (n.d.). Retrieved January 18, 2017 from http://greenapplesupply.org/bic-ecolutions-retractable-ballpoint-pen-box-of-12
Bolton, L. E., & Mattila, A. S. (2015). How does corporate social responsibility affect consumer response to service failure in buyer–seller relationships?. Journal of
Retailing, 91(1), 140-153.
Borin, N., Cerf, D. C., & Krishnan, R. (2011). Consumer effects of environmental impact in product labeling. Journal of Consumer Marketing, 28(1), 76-86.
Çal, B., & Adams, R. (2014). The Effect of hedonistic and utilitarian consumer behavior on brand equity: Turkey–UK comparison on Coca Cola. Procedia-Social and Behavioral
Sciences, 150, 475-484.
Products (pp. 57-68). Springer US.
Chen, F. P., & Leu, J. D. (2011). Product involvement in the link between scepticism toward advertising and its effects. Social Behavior and Personality: An International
Journal, 39(2), 153-159.
Dens, N., & De Pelsmacker, P. (2010). Consumer response to different advertising appeals for new products: The moderating influence of branding strategy and product category involvement. Journal of Brand Management, 18(1), 50-65.
Drossos, D. A., Kokkinaki, F., Giaglis, G. M., & Fouskas, K. G. (2014). The effects of product involvement and impulse buying on purchase intentions in mobile text advertising. Electronic Commerce Research and Applications, 13(6), 423-430. Du, S., Bhattacharya, C. B., & Sen, S. (2010). Maximizing business returns to corporate
social responsibility (CSR): The role of CSR communication. International Journal of
Management Reviews, 12(1), 8-19.
Eco Friendly [Online image]. (n.d.). Retrieved January 18, 2017 from http://greenstarbikes.com/eco-friendly/
Elving, W. J. (2013). Scepticism and corporate social responsibility communications: The influence of fit and reputation. Journal of Marketing Communications, 19(4), 277-292.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage.
Flores, W., Chen, J. C. V., & Ross, W. H. (2014). The effect of variations in banner ad, type of product, website context, and language of advertising on Internet users’
attitudes. Computers in Human Behavior, 31, 37-47.
Geuens, M., De Pelsmacker, P., & Faseur, T. (2011). Emotional advertising: Revisiting the role of product category. Journal of Business Research, 64(4), 418-426.
environmental marketing claims. Journal of Public Policy & Marketing, 13(1), 155-159.
Grimmer, M., & Woolley, M. (2014). Green marketing messages and consumers' purchase intentions: Promoting personal versus environmental benefits. Journal of Marketing
Communications, 20(4), 231-250.
Grunert, K. G. (1996). Automatic and strategic processes in advertising effects. The Journal
of Marketing, 88-101.
Hartmann, P., Apaolaza Ibáñez, V., & Forcada Sainz, F. J. (2005). Green branding effects on attitude: Functional versus emotional positioning strategies. Marketing Intelligence &
Planning, 23(1), 9-29.
Henley, W. H., Philhours, M., Ranganathan, S. K., & Bush, A. J. (2009). The effects of symbol product relevance and religiosity on consumer perceptions of Christian symbols in advertising. Journal of Current Issues & Research in Advertising, 31(1), 89-103.
Hollandia Royal Dutch 26 inch omafiets zwart [Online image]. (n.d.). Retrieved November 26, 2016 from http://www.fietsbezorgd.nl/hollandia-royal-dutch-26-inch-omafiets -zwart.html
Horton, D. (2006). Environmentalism and the bicycle. Environmental Politics, 15(1), 41-58. Jain, S. C., & Hackleman, E. C. (1978). How effective is comparison advertising for
stimulating brand recall?. Journal of Advertising, 7(3), 20-25.
Kangun, N., Carlson, L., & Grove, S. J. (1991). Environmental advertising claims: A preliminary investigation. Journal of Public Policy & Marketing, 10(2), 47-58. Karasiewicz, G., & Kowalczuk, M. (2014). Effect of celebrity endorsement in advertising
activities by product type. International Journal of Management and
Kempf, D. S. (1999). Attitude formation from product trial: Distinct roles of cognition and affect for hedonic and functional products. Psychology & Marketing, 16(1), 35-50. Kong, Y., & Zhang, A. (2013). Consumer response to green advertising: The influence of
product involvement. Asian Journal of Communication, 23(4), 428-447.
Kong, Y., & Zhang, L. (2014). When does green advertising work? The moderating role of product type. Journal of Marketing Communications, 20(3), 197-213.
Lang, A. (2000). The limited capacity model of mediated message processing. Journal of
Communication, 50(1), 46-70.
Lee, J., Kim, J., & Yu, J. (2015). Effects of fit of product, visual image, and
consumer self-image on art infusion advertising. Social Behavior and Personality: An
International Journal, 43(10), 1725-1740.
LG Wasdroger CD9BPWN [Online image]. (n.d.). Retrieved November 26, 2016 from http://www.visound.nl/124/lg/83/lg%2Bwasdroger%2Bcd9bpwn.html
LG Stainless Steel Dishwasher LD-14AT2 [Online image]. (n.d.). Retrieved November 26, 2016 from http://www.comparison.com.au/dishwashers/lg/LD-14AT2
Ling-Yee, L. (1997). Effect of collectivist character and ecological attitude on actual environmental commitment: The moderating role of consumer demographics and product involvement. Journal of International Consumer Marketing, 9(4), 31-53. Menon A., & Menon, A. (1997). Enviropreneurial marketing strategy: the emergence of
corporate environmentalism as market strategy. The Journal of Marketing, 61(1), 51-67.
Miele T8164WP Warmtepompdroger. [Online image]. (n.d.). Retrieved January 18, 2017 from http://www.kwik-rep.nl/bespaar-nu-extra-energie-en-geld-door-de-aanschaf-van-onze-energiezuinige-producten..html
