Graduate School of Communication
Faculty of Social and Behavioural Sciences Master’s programme Communication Science
Master’s Thesis
Green today, green tomorrow? An analysis of the persistence of green branding appeal effects on consumers’ brand attitude and purchase intention and the mediating role of
cognitive and affective responses.
Isabel Riffel
Date of completion: 2nd February, 2018 Supervisor: Dr. Barbara Schouten
Word count: 8389 UVA-ID: 11368357
Abstract
The present study was the first to examine the persistence of the effects of distinct green branding appeals (functional vs. emotional vs. combined) on consumers’ brand attitude and purchase intention over time. Furthermore, it analysed the mediating role of two cognitive (i.e. “perceived utilitarian environmental brand benefits”) and affective (i.e. “virtual nature experience”) consumer responses. An online experiment with a 3(between-subject factor “green branding appeal”: functional vs. emotional vs. combined) × 2(within-subject factor “time”: immediate measure vs. delayed measure) factorial design with N = 101 (N = 581
) participants was conducted. In contrast to previous research, results showed no difference in brand attitude and purchase intention across conditions, measured immediately after
exposure. Mediation analysis revealed that the effects of functional appeals were serially mediated through cognitive responses and the liking of the ad. Furthermore, effects of combined appeals were serially mediated through affective responses, cognitive responses and the liking of the ad. However, in contrast with theory and previous research, a mediation of the effects of emotional appeals through affective responses was not found. Additionally, the one-week delay follow-up study could not proof any differences in the persistence of brand attitudes and purchase intentions across conditions over time.
Keywords: green branding, green marketing, green washing, repeated measure, cognitive responses, affective responses
1 The first value describes the number of participants of the first study, measuring immediate effects. The second number describes the subsample of participants that answered the second study, measuring delayed effects. In the following, measures for the follow up study (N = 58) will always be reported in parentheses after measures for the first study (N = 101).
Green today, green tomorrow? An analysis of the persistence of green branding appeal effects on consumers’ brand attitude and purchase intention and the mediating
role of cognitive and affective responses.
International concerns about climate change and sustainability have become a key political interest over the past decades. Resulting, consumers’ requests for products that are environmentally sound have grown dramatically, challenging companies to integrate
sustainability into their business strategies and making it a driver of innovation (Dangelico, & Vocalelli, 2017; Dunlap, & Scarce, 1991; Nidumolu, Prahalad, & Rangaswami, 2009).
Accordingly, marketers have realized the need to implement environmental concerns into the marketing mix. They have become aware of the necessity to develop more ecological
products and distribution channels (Rex, & Baumann, 2007), as well as of the potential that the promotion of such goods and positioning as a green brand has (Easterling, Kenworthy, & Nemzoff, 1996; Schmidt, & Donsbach, 2012). Brand positioning is defined as the part of the brand identity and value proposition that is actively communicated to the consumer.
Therefore, it is responsible for shaping consumers’ brand perceptions in a competitive market (Aaker, 1996). Thus, green branding aims to establish a brand position that makes a brand distinct from competition by proposing that the brand is more environmentally sound than competing brands. Consequently, positioning a brand as a “green brand” entails an active communication of environmentally sound attributes to the consumer (Hartmann, Apaolaza Ibáñez, & Forcada Sainz, 2005).
Based on the work by Aaker (1996) on classification schemes of positioning
strategies, Hartmann et al. (2005) were the first to classify three distinct types of green brand positioning appeals. The authors distinguished a) functional appeals, highlighting physical and sustainable product attributes, b) emotional appeals, making use of visual stimuli that represent pleasant natural scenery and c) combined appeals – a combination of the two
aforementioned. Hartmann et al. (2005) furthermore analysed the effects of these three appeals on brand attitude. Brand attitude is a person’s internal evaluation of an object such as a brand, based on the person’s salient beliefs about that object at a given point in time
(Fishbein, & Ajzen, 1975). Hartmann et al. (2005) found emotional appeals to have a stronger positive effect on brand attitude than functional appeals, while the combination of both yielded the most positive effect. Their results found support in several studies (e.g. Hartmann, & Apaolaza-Ibáñez, 2009; Matthes, Wonneberger, & Schmuck, 2014; Schmuck, Matthes, Naderer, & Beaufort, 2017). Aiming to explain these findings and the underlying mechanisms of brand attitude formation in the context of green branding appeals, further research has analysed consumers' affective and cognitive response to green branding. While a cognitive response to green branding entails judgement of a brand’s environmentally sound attributes, affective responses are based on feelings which are experienced in response to the ad (Schmuck et al., 2017). It was found that both, cognitive responses (i.e. “perceived utilitarian environmental brand benefits”) as well as affective responses, in the form of positive feelings and connectedness to nature imagery in green ads (i.e. “virtual nature experience”), can positively influence the formation of brand attitudes (Hartmann, &
Apaolaza-Ibáñez, 2008, 2009, 2012; Hartmann, Apaolaza- Ibáñez, & Alija, 2013). However, it was not yet tested how these two responses mediate the effects of green branding appeals on brand attitude in a study simultaneously comparing functional, emotional and combined green branding conditions. Hence, the first aim of this study is to simultaneously compare mediation through affective and cognitive responses across functional, emotional and combined green branding appeals in order to establish a better understanding of how persuasion for the distinct appeals is established.
Furthermore, no study has yet analysed long-term effects of green branding appeals. Following the logic of dual-process models, it can be assumed that emotional and functional
appeals are processed differently with regards to the use of reflective and impulsive systems (Strack, & Deutsch, 2004), which in turn should have an influence on the persistence of consumer responses over time. However, as purchase decisions are often not made
immediately and advertising is mostly not present at the point of sale (Belch, Belch, Kerr, & Powell, 2014), marketers should be interested in creating long-term effects. Hence, the second aim of this study is to analyse the persistence of green branding effects in a repeated measure design.
Moreover, extant research has mostly focussed on measuring brand attitudes as a response to green branding. Yet, based on the theory of planned behaviour (TPB) attitudes impact on behaviour through intentions (Ajzen, 1991; Conner, & Sparks, 2005). “Intentions capture the motivational factors that influence behaviour, how hard people are willing to try, how much effort they would exert to perform the behaviour […]” (Ajzen, 1991, p. 181) – in this case, the intention to make a purchase. Therefore, in order to better predict the likeliness that a consumer will make a purchase as a response to green branding, purchase intention should be measured as an outcome variable. Still, so far purchase intention in response to green branding has either been assessed as an item, measuring brand attitudes on a multiple-item scale (Hartmann et al., 2005; Hartmann, & Apaolaza-Ibáñez, 2009) or in a study not simultaneously comparing all three appeals (Schmuck et al., 2017). Hence, the third aim of this study is to assess purchase intention as an autonomous variable in response to emotional, functional and combined appeals in order to establish a better understanding of their effects on consumer reactions.
In sum, the current study aims to answer the following research questions:
RQ1: “How is the relationship between distinct green branding appeals (functional versus emotional versus combined) and consumers’ attitude towards the brand and purchase intention mediated through cognitive and affective responses?
RQ2: “How persistent are brand attitude and purchase intention in response to distinct green branding appeals over time?”
Hence, this study should close a relevant gap in research by being the first to analyse the persistence of green branding effects over time. Furthermore, this study should contribute to the level of knowledge about underlying mechanisms of these green branding appeal effects and therefore, offer valuable insights for researches as well as for marketers.
Theoretical Framework
Green branding appeal effects on brand and mediation through cognitive and affective responses
Functional appeals. A green positioning appeal that is based on functional attributes
(i.e., functional green branding appeal), communicates information about advantages in the production, usage or elimination of a product, which make it more environmentally sound than competing and conventional goods (Hartmann et al., 2005). Information can be communicated through both, persuasive argumentation and through labels, referring to a product’s performance or characteristics (Bickart, & Ruth, 2012; Schmuck et al., 2017). Since consumers are more environmentally concerned nowadays and have stronger requests for sustainable products, a functional appeal should offer additional perceived product benefits (Hartmann et al., 2005; Hartmann, & Apaolaza-Ibáñez, 2009; Hartmann et al., 2013; Roberts, 1996; Schmuck et al., 2017). According to the TPB, attitudes are based on beliefs and the evaluation of these beliefs (Ajzen, 1991). Therefore, the experienced benefits should lead the consumer to form positive beliefs about the outcomes of buying the brand, which in turn, should positively influence the process of forming brand attitudes (Ajzen, 2015). Hence, based on previous findings and the TPB (Ajzen, 1991) it is expected that the effect of
functional green branding appeals on brand attitude is mediated by a cognitive response: the perception of “utilitarian environmental brand benefit”.
H1: The effect of functional green brand positioning on brand attitude is fully mediated by the perception of “utilitarian environmental brand benefits” with functional appeals leading to higher levels of “utilitarian environmental brand benefits”, which consequently positively influences brand attitude.
Emotional appeals. On the other hand, emotional green branding appeals aim at
inducing affective responses to an ad (Edell, & Burke, 1987; Matthes et al., 2014). In the context of green branding advertisements, sustainable product attributes are often represented through pleasant natural background imagery (Hartmann, & Apaolaza-Ibáñez, 2008, 2009). The importance of such imagery in green branding can be explained by a general human attraction towards and affinity for nature (Hartmann, & Apaolaza-Ibáñez, 2009). Contact with nature has always played an essential role in human life (Hartmann, & Apaolaza-Ibáñez, 2009; Maller, Townsend, Pryor, Brown, & St Leger, 2006) and should, as a consequence, generally lead to positive emotional reactions. However, not only contact with nature, but also contact with natural imagery – stored in memory as a visual cognition and highly associated with positive feelings – can cause such emotional replies (Hartmann, & Apaolaza-Ibáñez, 2009). In line with classical analysis of aesthetic perception and affective language, Lang, Greenwald, Bradley and Hamm (1993) found that affective evaluations of images are related to physiological responses such as facial expression and skin conductance.
Confirmatory, Chowdhury, Olsen, and Pracejus (2008) showed that a single positive image in an advertising context can elicit positive or negative affect. Moreover, in the context of green branding, Hartmann et al. (2013) found that emotional ratings from participants who were exposed to ads containing pleasant nature imagery did not significantly differ from the
emotional responses of participants exposed to real nature. Therefore, elicited by the
exposure to natural images, consumers should experience positive feelings of well-being and connectedness to nature and thereby, experience an additional hedonic benefit through emotional green branding (Hartmann, & Apaolaza-Ibáñez, 2008, 2009). Based on the TPB, this experienced benefit should lead the consumer to form positive beliefs about the outcomes of buying the brand and consequently, positively influence the process of forming brand attitudes (Ajzen, 2015). Thus, it is assumed that the effect of emotional green branding appeals on brand attitude is mediated through an affective response: a “virtual nature experience”.
H2: The effect of emotional green brand positioning on brand attitude is fully
mediated by a “virtual nature experience”, with emotional appeals increasing the level of “virtual nature experience”, which consequently positively influences brand
attitude.
Combined appeals. According to Hartmann et al. (2005) combined green branding
appeals feature both functional information about environmental advantages of a product and pleasant nature imagery. Attention restoration theory proposes that spending time or looking at nature can improve concentration and overcome attention fatigue (Kaplan, 1995). Based on this theory, Hartmann et al. (2013) suggested that ads featuring pleasant nature imagery would not only lead to similar feelings as being in real nature, but beyond, increase attention to the advertising message and improve message processing (Hartmann et al., 2013). As these expectations were supported, it can be anticipated that the exposure to nature imagery does not only elicit affective responses but also strengthens the attention to functional attributes, featured in combined appeals. Nature imagery in a combined green branding appeal should consequently not only lead to a “virtual nature experience” but additionally, to more in-depth
processing of the advertising message. In this case, information about environmentally sound product attributes and hence, the perception of “utilitarian environmental brand benefits” should be increased. Therefore, multiple assumptions can be made: As combined appeal feature nature images as well as factual information, they should be mediated through both, “virtual nature experience” (i.e., affective responses) and “perceived utilitarian environmental brand benefits” (i.e., cognitive responses). Furthermore, following the logic of attention restoration theory, this mediation should be serial with higher levels of “virtual nature experience” leading to higher levels of “perceived utilitarian environmental brand benefits”. Moreover, as attention to the advertising message is expected to be strengthened for
combined appals, consumers should put more focus on the promoted environmental product attributes. Consequently, combined appeals should lead to greater “perceived utilitarian environmental brand benefits” than solely functional green branding appeals. In sum, the following hypothesis is proposed:
H3: (a) The relationship between combined green branding appeals and brand attitude is mediated by “utilitarian environmental brand benefits” and a “virtual nature
experience” in a serial mediation, (b) with “virtual nature experience” leading to higher levels of “utilitarian environmental brand benefits” compared to functional appeals, which, in turn, increases positive brand attitude.
Based on the discussed above, it can be expected that combined green branding appeals lead to the formation of more positive brand attitudes than functional green branding appeals or emotional green branding appeals alone. In total, through the experience of “utilitarian environmental brand benefits” (i.e., cognitive responses) and a “virtual nature experience” (i.e., affective responses) they offer the most benefits to the consumer and hence, should create more positive beliefs about the outcomes of buying the brand (Ajzen, 2015).
Support for this assumption stems from previous research (Hartmann et al., 2005; Hartmann, & Apaolaza-Ibáñez, 2009; Matthes et al., 2014; Schmuck et al., 2017) which has found combined green branding appeals to lead to more positive brand attitudes than functional and emotional green branding appeals alone.
H4: Combined green branding appeals lead to stronger positive effects on brand attitude than both, functional appeals and emotional appeals alone.
Purchase intention
Based on the theory of planned behaviour (TPB) attitudes do not have a direct effect on a person’s behaviour but instead, they have an impact on behavioural intentions which in turn, influence actual behaviour (Ajzen, 1991; Conner, & Sparks, 2005). Grounded on the TPB it is expected that positive brand attitudes, based on emotional, functional and combined green branding appeals, correspondingly lead to more positive intentions to purchase the advertised brand. Support stems from a study by Hartmann and Apaolaza-Ibáñez (2012), finding the influence of psychological benefits and environmental concern on purchase intention to be mediated through brand attitudes.
H5: The more positive the brand attitude, the higher the intention to purchase the advertised brand.
Persistence of brand attitudes and purchase intentions
Based on the reflective-impulsive-model (RIM), two systems of information processing operate in parallel. When information is processed in the reflective system decisions are based on knowledge and facts, whereas in the impulsive system decisions are formed based on associative links and simple inference making (Strack, & Deutsch, 2004). However, while the impulsive system is always active in processing incoming information,
activation of the reflective system depends on the intensity of an incoming stimulus and the attention it receives (Strack, & Deutsch, 2004). Since more attention has been drawn to environmental issues and people have generally become concerned with the topic in the past years (Dangelico, & Vocalelli, 2017, Dunlap, & Scarce, 1991; Nidumolu et al., 2009), it can be expected that incoming information transported through functional and combined green branding appeals will activate processing in the reflective system (Strack, & Deutsch, 2004). However, in the absence of information – as in emotional green branding appeals – it is more likely that the incoming stimulus is processed only through the impulsive system. Hence, the formation of brand attitude and purchase intention as a response to functional or combined green branding appeals should be formed on the basis of knowledge and facts. Propositional representations in a consumer’s mind should therefore be formed by connecting elements through the establishment of relational schemata and attaching a truth value (Strack, & Deutsch, 2004). In contrast, when a consumer sees pleasant nature images but information and facts are lacking, attitude formation and purchase intention are proposed to be based on sole processing in the impulsive system. Consequently, associations between these images and positive feelings should be activated automatically without the person reflecting actively on the appeal.
Based on the elaboration-likelihood-model (ELM) the formation of brand attitude through central (i.e., reflective) or peripheral (i.e., impulsive) processing has an effect on its persistence (Belch et al., 2014; Petty, & Cacioppo, 1986). While „attitude change that occurs through central processing is relatively enduring [...] the ELM views attitudes resulting from peripheral processing as temporary. So favourable attitudes must be maintained by continual exposure to the peripheral cues, such as through repetitive advertising” (Belch et al., 2014, p.174). Correspondingly, the RIM proposes that associative links between elements in the impulsive system change only gradually through a process of learning and under the premise
of repetition (Strack, & Deutsch, 2004). Hence, the distinct processing of a) emotional and of b) functional and combined green branding appeals through the impulsive and reflective system should have an influence on the persistence of brand attitudes and consequently, on the persistence of purchase intention in response to green branding appeals (Ajzen, 1991; Conner, & Sparks, 2005). Accordingly, studies by Dahlén, Friberg and Nilsson (2009) and by Sengupta, Goodstein and Boninger (1997) showed that the effects of unrelated cues on attitudes vanish one week after exposure. Therefore, as functional or combined appeals should be processed through the reflective system and brand attitude and purchase intention are based on propositional representations formed in the consumers’ mind, these should be consistent – even in the absence of the original stimulus. In contrast, emotional appeals are expected to be processed through the impulsive system only and hence, through the
activation of associative links in response to a cue. As associative links in the impulsive system are changed and formed only gradually through repetition, the formation of brand attitude should mainly rely on the spreading of activation in the presence of a stimulus (e.g., pleasant nature imagery). Thus, brand attitude and purchase intention in response to
emotional green branding appeals are expected to be temporary and vanish one week after exposure in the absence of the stimulus.
H6: Brand attitude and purchase intention in response to combined and functional green branding appeals are consistent over time and in the absence of the stimulus.
H7: Brand attitude and purchase intention in response to emotional green branding appeals are not consistent over time and vanish one week after exposure in the absence of the stimulus.
Figure 1. Conceptual model of concepts and hypotheses
Method Participants
A snowball sampling strategy was chosen due to limited financial resources and time. The sample was recruited over the social networks of the researcher. This sampling strategy allowed to generate a relatively large sample in a short amount of time and at little cost. To identify the number of participants needed for the study in order to secure sufficient statistical power, a priori power analysis was conducted, using the G*Power analysis tool (Faul,
Erdfelder, Lang, & Buchner, 2007). Expecting a medium effect size (η2 = .06), power analysis proposed that a total sample of N = 42 participants would allow for an 80% chance of correctly rejecting the null hypothesis (α = 5%). An actual sample of N = 101 participants was recruited of which N = 58 completed both surveys, achieving power of 92% in a repeated measure, within-between interaction design. For analysis of the first set of hypothesis, not including a repeated measure, the total set of N =101 participants was used in order to increase statistical power. To enhance representativeness of the sample, there were no
restrictions to who was allowed to participate. The mean age in the sample was 31 years [M = 30.96, SD = 10.40 (M = 30.78, SD = 10.55)], 60.4% (58.6%) were female, 80% (82.8%) had completed an undergraduate degree or higher and 56.6% (55.2%) of the sample were
German.
Design
Table 1
Experimental design
The study employed a 3 × 2 factorial design (Table 1) with “green branding appeal” (3 levels: functional appeal vs. emotional appeal vs. combined appeal) as experimental, between-subject variable and the factor “time” as within-subject variable (2 levels: immediate measure vs. delayed measure).
Procedure
After entering the study through the link shared by the researcher, participants were introduced, read a brief cover story about the research and gave consent. The cover story suggested that the study would help the new sneaker brand “Nubii” to understand how consumers received and liked the brand. Furthermore, participants were informed that the study consisted of two surveys and that it was very important for the success of the research that both surveys were completed. Following, participants were randomly assigned to the manipulation material – three print ads either containing a functional, emotional or a
combined green branding appeal. Participants were instructed to imagine that they had come across this ad in a magazine that they frequently read and to carefully look at the ad
Within subject factor: time
Between subject factor: green branding appeal
Immediate measure Delayed measure
Functional Immediate measure/
functional appeal
Delayed measure/ functional appeal
Emotional Immediate measure/
emotional appeal
Delayed measure/ emotional appeal
Combined Immediate measure/
combined appeal
Delayed measure/ combined appeal
for eight2 seconds. After eight seconds participants were automatically transferred and requested to indicate their brand attitude and purchase intentions for the advertised product and brand. Furthermore, participants had to indicate their responses to the ad and answer several manipulations check and control variable questions. After answering questions about demographics – age, gender, nationality and education – participants were thanked for their participation and asked to submit their e-mail address for a follow up questionnaire. An incentive was placed, offering the opportunity to enter the raffle for one of three 20€ Amazon gift cards, if the second questionnaire was completed. One week later participants were approached via e-mail and asked to again answer questions regarding their brand attitude, purchase attention and responses to the ad. Afterwards participants were thanked and debriefed in order to reinsure that they were aware that the scenario was fictitious and their data would not be misused.
Stimulus Materials
Green branding appeal. Based on the initial study by Hartmann et al. (2005) and a
series of pre-tests (see Appendix A), the present study contained three versions of a print ad for a fashionable sneaker3 by the fictional brand “Nubii”. A fictional brand name was chosen to avoid previous associations of the brand with sustainability and environmental concerns, as these might have influenced results. In the functional green branding appeal condition (see Figure B1 in Appendix B), participants saw the sneaker, the brand name and information about the sustainable product attribute “Made from 100% Recycled Materials”4
in textual form in front of a neutral background. In the emotional green branding appeal condition (see Figure B2 in Appendix B) participants saw the same sneaker and product name in front of a
2 The time that participants took to look at an ad (M = 7.28, SD = 3.02) was assessed in the pre-test for the final manipulation materials. N = 6 participants were instructed to look at the three print ads carefully before they carried on with the questionnaire.
3
A fashionable sneaker was chosen based on pre-test results (see Appendix A).
mountain lake5 but without information about sustainable attributes. In the combined green branding appeal condition (see Figure B3 in Appendix B) the ad contained the same product, product name and both, the nature imagery and the textual information about the sustainable product attribute.
Time. The within-subject, experimental factor “time” was measured on two levels:
“immediate measure” and “delayed measure”. For the “immediate measure” condition, the dependent variables brand attitude and purchase intention as well as the mediation variables “perceived utilitarian environmental brand benefits” and “virtual nature experience” were assessed immediately after exposure to the manipulation materials. For the “delayed measure” condition, a follow up survey was sent out to participants one week after
manipulation to reassess the outcomes. This time frame was chosen as previous research by Dahlén et al. (2009) and by Sengupta et al. (1997) showed that the effects of unrelated cues on attitudes disappeared when measured at a seven-day delay.
Measures
A list of all variables, including mean values and standard deviations, can be found in the Appendix (see Table C1 in Appendix C).
Utilitarian environmental brand benefits. Based on the studies by Hartmann and
Apaolaza-Ibáñez (2012) and Schmuck et al. (2017), the perception of “utilitarian
environmental brand benefits” was measured through two items on a seven-point Likert scale (1 = strongly disagree; 7 = strongly agree): “Nubii” respects the environment; A sneaker by “Nubii” is better for the environment than other sneaker brands [r = .72; M = 4.92; SD = 1.18 (r = .78; M = 4.91; SD = 1.15)]. The scale proved high reliability in previous research
(Schmuck et al., 2017; r = .92).
Virtual nature experience. Based on the studies by Hartmann and Apaolaza-Ibáñez
5
The image of a mountain lake was chosen based on pre-test results and previous research (e.g., Hartmann et al.,2005) (see Appendix A)
(2012) and Schmuck et al. (2017), the mediator “virtual nature experience” was measured through two items on a seven-point Likert scale (1 = strongly disagree; 7 = strongly agree): “Nubii” makes me feel close to nature; “Nubii” makes me think of nature, fields, forests and mountains [r = .79; M = 4.48; SD = 1.49 (r = .79; M = 3.96; SD = 1.42)]. The scale proved sufficient reliability in previous research (Schmuck et al., 2017; r = .84).
Brand attitude. Brand attitude was measured with five items on a seven-point
semantic differential scale: unattractive–attractive; not likable–likable; negative–positive; boring–interesting; not recommendable–recommendable [Cronbach’s α = .90; M = 4.54; SD = 1.21 (Cronbach’s α = .92; M = 4.27; SD = 1.06)]. The scale proved high reliability in a previous study by Schmuck et al. (2017) (r =.94).
Purchase intention. In line with Grimmer and Woolley (2014) and Schmuck et al.
(2017), purchase intention was measured as a single seven-point Likert scale item (1 = extremely unlikely; 7 = extremely likely): How likely is it that you will purchase a sneaker by “Nubii“ in the future? [M = 3.60; SD = 1.82 (M = 3.31; SD = 1.70)].
Manipulation check
Manipulation check. Additionally, two manipulation check questions were posed,
employing seven-point Likert scales (1 = strongly disagree; 7 = strongly agree): In the ad I have seen information about sustainable product attributes; In the ad I have seen pleasant images of nature.
Control variables
Attitude toward green products. As Matthes et al. (2014) found attitude toward
green products to moderate effects of green branding appeal on brand attitude, environmental concern was added as a control variable. Based on Chang (2011) attitude towards green products was assessed with four items on a seven-point Likert scale (1 = strongly disagree; 7 = strongly agree): I like green products; I feel positive toward green products; Green
products are good for the environment; I feel proud when I buy/use green products
(Cronbach’s α = .83; M = 5.67; SD = .87). The scale proved sufficient reliability in previous research (Matthes et al., 2014; Cronbach's α = .82).
Ad liking. As Mitchell and Olson (2000) found that a person’s attitude toward an
advertised brand might be affected by the person’s evaluation of the ad as a whole, this variable was added as a control variable. In line with the authors (Mitchell, & Olson, 2000) attitude towards the ad was measured through four items on a seven-point semantic scale: good–bad; like–dislike; irritating–not irritating; interesting–uninteresting (Cronbach’s α = .80; M = 4.42; SD = 1.16). This scale proved to be sufficiently reliable in the study by Mitchell and Olson (2000) (Cronbach's α = .87)
Demographics. Additionally, participants were asked to indicate their age (M =
30.96, SD = 10.41), gender (60.4% female) highest completed level of education (42.0% Bachelor degree) and nationality (56.6% German) on multiple choice scales (see Appendix D).
Results Randomization Check
To test whether randomization of participants between the groups was successful, chi square tests, for gender (χ2
(2) = 3.46, p = .178), education (χ2 (6) = 4.21, p = .648) and nationality (χ2
(36) = 39.04, p = .335) were conducted, showing no significant differences across conditions. Furthermore, univariate analysis of variance (ANOVA) for age, ad linking and green attitude was run. Results revealed a significant difference in age between the three groups (F (2, 97) = 3.11, p = .049, 𝜂p2 = .06). Correlation analysis between age and the
dependent variables brand attitude, purchase intention, “virtual nature experience” and “utilitarian environmental brand benefits” was not significant (all p > .05). Therefore, age was not added as a control variable in further analysis. Moreover, ad liking significantly
differed across conditions, (Welch’s F(2, 62.32) = 6.09, p = .004, 𝜂p2 = .13). Correlation
analysis between ad liking and the dependent variables brand attitude (r = .72, p < .001), purchase intention (r = .63, p < .001), “virtual nature experience” (r = .55, p < .001) and “utilitarian environmental brand benefits” (r = .40, p < .001) was significant. Therefore, ad liking was added as a covariate in further analysis. For green attitude no significant difference across conditions was found (F (2, 98) = .72, p = .490).
Manipulation Check
Two univariate ANOVAs were conducted to analyse if the manipulation had worked sufficiently, using the green branding appeal conditions (functional vs. emotional vs.
combined) as independent variables and the two manipulation check items as dependent variables. For the first item, descriptive statistics indicated participants from the functional green branding appeal condition (M = 4.90, SD = 1.42) and the combined green branding appeal condition (M = 4.43, SD = 1.68) to believe to have seen more factual information about environmental benefits of the product than participants from the emotional condition (M = 2.31, SD = 1.32). Leven’s test was not significant (F (2, 98) = 1.54, p = .220).
Homogeneity of variance could be assumed. Univariate analysis showed a significant main difference between the conditions, (F (2, 98) = 28.48, p < .001, 𝜂p2 = .37). A Bonferroni
post-hoc test revealed no significant differences between the functional and the combined green branding appeal condition (Mdifference = .51, p = .503) but between the functional and the emotional condition (Mdifference = 2.59, p < .001) and between the combined and emotional condition (Mdifference = 2.07, p < .001).
For the second item, descriptive statistics indicated participants from the emotional green branding appeal condition (M = 6.20, SD = 1.30) and the combined green branding appeal condition (M = 5.65, SD = 1.36) to more strongly believe to have seen pleasant images of nature than participants from the functional condition (M = 2.80, SD = 1.71). Leven’s test
was significant, (F (2, 98) = 3.37, p = .038). The assumption of homogeneity was therefore violated. A Welch’s test and Games-Howell post-hoc tests were conducted. The ANOVA was significant, (Welch’s F(2, 62.25) = 41.27, p < .001, 𝜂p2 = .50). As expected,
Games-Howell post-hoc tests revealed no significant differences between the emotional and the combined green branding appeal condition (Mdifference = .53, p = .220) but between the
emotional and the functional condition (Mdifference = 3.40, p < .001) and between the combined and functional condition (Mdifference = 2.87, p < .001). Therefore, manipulation had worked sufficiently.
Hypotheses Testing
To test H1, H2 and H3(a), mediation analyses were conducted, using the PROCESS macro for SPSS (Baron, & Kenny, 1986; Hadi, Abdullah, & Sentosa, 2016; Hayes, 2012). H1 and H2 were tested using model 4, while H3(a) was tested using model 6 for serial mediation. Ad liking was added as a covariate for all analyses.
H1. Testing the mediation effect of utilitarian environmental brand benefits on the
relationship between functional green branding appeal and brand attitude, the total effect of the independent variable functional green branding appeal on the dependent variable brand attitude was not significant, b = .09, t(98) = .48, p =.629. There was a significant positive effect of functional green branding appeal on the mediator, “perceived utilitarian
environmental brand benefits”, b = .73, t(98) = 3.17, p =.002. However, the regression of the mediator on the dependent variable brand attitude was not significant, b = .08, t(97) = 1.04, p = .303. Therefore, “perceived utilitarian green brand benefits” did not mediate the
relationship between functional green branding appeal and brand attitude (see Table 2, p. 21). H1 was rejected.
Results of mediation analysis for H1, using PROCESS model 4
Table 3
Results of mediation analysis for H2, using PROCESS model 4
Table 4 R² F df b SE 95% CI Model 1 X on Y .52 54.32*** (2, 98) .09 .18 [-.27, .45] Model 2 X on M .24 17.60*** (2, 98) .73** .23 [.28, 1.19] C on M .34*** .09 [.16, .53] Model 3 M on Y .52 37.35*** (3, 97) .08 .08 [-.08, .25] X on Y .03 .20 [-.36, .42] C on Y .71*** .08 [.56, .86]
Note. X = Functional green branding appeal (dummy coded), Y = brand attitude, M = perceived utilitarian environmental brand benefits and C = ad liking as covariate, *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001. R² F df b SE 95% CI Model 1 X on Y .52 55.34*** (2, 98) .05 .09 [-.13, .22] Model 2 X on M .32 31.62*** (2, 98) .24* .12 [.00, .47] C on M .77*** .10 [.58, .98] Model 3 M on Y .53 37.25*** (3, 97) .12 .08 [-.04, .27] X on Y .02 .09 [-.15, .19] C on Y .67*** .10 [.48, .86]
Note. X = Emotional green branding appeal (dummy coded), Y = brand attitude, M = virtual nature
Results of mediation analysis for H3, using PROCESS model 6
H2. Analysing the mediation effect of “virtual nature experience” on the relationship
between emotional green branding appeal and brand attitude, the total effect of the independent variable emotional green branding appeal on the dependent variable brand attitude was not significant, b = .05, t(98) = .54, p = .593. A significant positive effect of emotional green branding appeal on the mediator, “virtual nature experience”, was found, b = .24, t(98) = 2.02, p = .046. However, the regression of the mediator on the dependent variable brand attitude was not significant, b = .12, t(97) = 1.49, p = .139. Thus, “virtual nature
experience” did not mediate the relationship between emotional green branding appeal and brand attitude (see Table 3, p. 21). H2 was rejected.
H3(a). Analysing the serial mediation effect of “virtual nature experience” and
R² F df b SE 95% CI Model 1 X on Y .52 54.60 (2, 98) -.05 .06 [-.18, .07] Model 2 X on M1 .35 27.23*** (2, 98) .22* .08 [.07, .38] C on M1 .67*** .11 [.45, .89] Model 3 M1 on M2 .19 8.60*** (3, 97) .06 .09 [-.12, .25] X on M2 .12 .07 [-.02, .26] C on M2 .34** .10 [.14, .54] Model 4 M1 on Y .55 28.60*** (4, 96) .14 .08 [-.03, .30] M2 on Y .09 .08 [-.06, .25] X on Y -.10 .07 [-.24, .04] C on Y .63*** .10 [.44, .82]
Note. X = Combined green branding appeal (dummy coded), Y = brand attitude, M1 = virtual nature
experience, M2 = perceived utilitarian environmental brand benefits and ad liking as covariate; *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001.
“perceived utilitarian environmental brand benefits” on the relationship between combined green branding appeal and brand attitude, the total effect of the independent variable
combined green branding appeal on the dependent variable brand attitude was not significant, b = -.05, t(98) = -.85, p =.396. The regression of combined green branding appeal on the first mediator, “virtual nature experience”, showed a significant positive effect, b = .22, t(98) = 2.84, p = .006. However, the regression of the first mediator on the second mediator
“perceived utilitarian environmental brand benefits”, was not significant b = .06, t(97) = .67, p = .507. Hence, a serial mediation of the relationship between combined green branding appeal and brand attitude through “virtual nature experience” and “perceived utilitarian environmental brand benefits” was not found (see Table 4, p. 22). H3(a) was rejected.
Additional Mediation Analysis. In the mediation analysis reported above, no
mediation of the relationship between the distinct green branding appeals and brand attitude through either “virtual nature experience” or “perceived utilitarian environmental brand benefits” could be found. However, it is noteworthy that the covariate ad liking showed significant effects on all mediators and dependent variables in all three analysed models (see Table 2-4, p.22-23). Therefore, additional analysis was performed, exploring the role of ad liking. Ad liking was included as a mediator, testing serial mediation with PROCESS model 6 (Baron, & Kenny, 1986; Hadi et al., 2016; Hayes, 2012).
Functional green branding appeals. Testing the serial mediation effect of “perceived
utilitarian environmental brand benefits” and ad liking on the relationship between functional green branding appeal and brand attitude, the total effect of functional green branding appeal on brand attitude was marginally significant, b = .51, t(99) =3.92, p =.051. The regression of functional green branding appeal on the first mediator, “perceived utilitarian environmental brand benefits”, showed a significant positive effect, b = .93, t(99) = 4.04, p = .001, as did the regression of the first mediator on the second mediator, ad liking, b = .36, t(98) = 3.87, p <
.001 and the regression of the second mediator on the dependent variable, brand attitude, b = .71, t(97) = 9.35, p < .001. All other paths were non-significant (see Table E1 & Figure E1 in Appendix E). Therefore, full serial mediation of the relationship between functional green branding appeal and brand attitude through “perceived utilitarian environmental brand benefits” and ad liking was found (indirect effect = 0.24, boot SE = .09, 95% CI [.10; .44]).
Emotional green branding appeals. Analysing the serial mediation effect of “virtual
nature experience” and ad liking on the relationship between emotional green branding appeal and brand attitude, the total effect of emotional green branding appeal on brand attitude showed a significant, negative effect, b = -.28, t(99) = -2.24, p =.024. While the regression of emotional green branding appeal on the first mediator, “virtual nature
experience”, was not significant, b = -.10, t(99) = -.63, p = .528., a significant negative effect was found for the regression of emotional green branding appeal on the second mediator, ad liking, b = -.39, t(98) = -3.83, p = .002. Ad liking had a significant, positive effect on brand attitude, b = .67, t(97) = 6.98, p < .001. All other paths were non-significant (see Table E2 & Figure E2 in Appendix E). Therefore, the model showed full mediation of the relationship between emotional green branding appeal and brand attitude through ad liking (indirect effect = -.26, boot SE = .08, 95% BCBCI [-.44; -.13]). However, there was no mediation found through “virtual nature experience”.
Combined green branding appeals. Analysing the serial mediation effect of “virtual
nature experience”, “perceived utilitarian environmental brand benefits” and ad liking on the relationship between combined green branding appeal and brand attitude, the total effect of combined green branding appeal on brand attitude was not significant, b = .03, t(99) = .39, p =.695. The regression of combined green branding appeal on the first mediator, “virtual nature experience”, showed a significant positive effect, b = .30, t(99) = 3.18, p = .002, as did the regression of the first mediator on the second mediator, “perceived utilitarian
environmental brand benefits”, b = .21, t(98) = 2.33, p = .022, the regression of the second mediator on the third mediator, ad liking , b = .26, t(97) = 2.88, p < .001, and the regression of the third mediator on the dependent variable, brand attitude, b = .63, t(96) = 6.43, p < .001. Furthermore, the first mediator, “virtual nature experience”, had a significant, positive effect on the third mediator, ad liking, b = .38, t(97) = 5.00, p < .001. All remaining path were not significant (see Table E3 & Figure E3 in Appendix E). Therefore, two significant indirect effects of combined green branding appeal on brand attitude were found: 1) The relationship was partially, serially mediated through “virtual nature experience”, “perceived utilitarian environmental brand benefits” and ad liking (indirect effect = .01, boot SE = .01, 95% BCBCI [.00; .03]) and 2) partially serially mediated through “virtual nature experience” and ad liking (indirect effect = .07, boot SE = .09, 95% BCBCI [.10; .44]).
H3(b). To compare effects of the three green branding appeals on perceived utilitarian
brand benefits, a univariate ANOVA was conducted, controlling for ad liking. Descriptive statistics (see Table C1 in Appendix C) indicated “perceived utilitarian environmental brand benefits” to be higher for participants in the functional green branding appeal condition (M = 5.57, SD = .99) than for participants in the combined green branding appeal condition (M = 5.26, SD = .92). For the emotional condition, “perceived utilitarian environmental brand benefits” were the lowest (M = 4.00, SD = 1.00). Leven’s test was not significant (F (2, 98) = .48, p = .619). Therefore, homogeneity of variance could be assumed. The ANOVA showed a significant main effect of the conditions on “perceived utilitarian environmental brand
benefits” (F (2, 97) = 16.82, p < .001, 𝜂p2 = .26). A Bonferroni post-hoc test revealed
significant mean differences between the functional and the emotional condition (Mdifference = 1.57, p < .001) and between the combined and emotional condition (Mdifference = 1.26, p < .001), however, not between the functional and the combined condition (Mdifference = .30, p = .628). Therefore, combined green branding appeals did not lead to greater “perceived
utilitarian environmental brand benefits” than functional green branding appeals. Thus, H3(b) was rejected.
H4. To compare effects of the three green branding appeals on brand attitude a
univariate ANOVA was performed, controlling for ad liking. Descriptive statistics indicated brand attitude to be higher for participants in the functional green branding appeal condition (M = 4.89, SD = 1.14) than for participants in the combined green branding appeal condition (M = 4.60, SD = 1.17) and the emotional green branding condition (M = 4.17, SD = 1.22). Leven’s test was not significant (F (2, 98) = .43, p = .653). Therefore, homogeneity of
variance could be assumed. The ANOVA showed no significant main effect of the conditions on brand attitude (F (2, 97) = .407, p = .666). Therefore, combined green branding appeals did not lead to more positive effects on brand attitude than both, functional appeals and emotional appeals. H4 was rejected.
H5. To test H5, a linear hierarchical regression analysis was conducted with purchase
intention as dependent variable and with the control variable ad liking and brand attitude as independent variables.
Table 5
Standardized regression coefficients predicting purchase intention (N = 101)
Model 1 Model 2 b* b* Ad liking .63*** .26** Brand attitude .51*** R2 .40 .53 F 66.22*** 55.10*** ΔR2 .13 ΔF 26.76*** Note. * p <.05. ** p <.01. *** p <.001.
The regression model as a whole was significant, F(2, 98) = 55.10, p < .001. Brand attitude significantly predicted purchase intention b = .78, t(98) = 5.17, p < .001. For each additional unit on the scale of brand attitude the level of purchase intention increases by 0.51.
Together, the two variables, brand attitude and ad liking predicted 52% of the variance in purchase intention (R2 = .53, F(2,98) = 55.10, p < .001). Hence, the more positive the brand attitude, the higher the intention to purchase the advertised brand. H5 was supported.
H6 and H7. Two repeated measure ANOVAs were conducted, including a
Greenhouse-Geisser test, to determine the effects of the independent within subject-variable time and its interaction effects with the independent between-subject variable condition on brand attitude and purchase intention. Ad liking was added as a control variable. For the first analysis, Greenhouse-Geisser revealed no statistically significant differences in brand attitude between the two time points for the three conditions (F(2, 54) = .38, p = .683). For the second analysis, Greenhouse-Geisser as well revealed no statistically significant differences in purchase intention between the two time points for the three conditions (F(2, 54) = 1.49, p = .234). Hence, while H6 was supported as brand attitude and purchase intention in response to functional and combined green branding appeals stayed consistent over time and in the absence of the stimulus, H7 was rejected. Brand attitude and purchase intention were consistent for all three conditions over time and in the absence of the stimulus and did not vanish for the emotional green branding appeal (see Table 6).
Table 6
Statistics for brand attitude & purchase intention at immediate & delayed measure
Additional analysis was performed to explore the mediating role of “perceived
Immediate measure Delayed measure
M SD M SD
Brand attitude Functional appeal 4.76 1.04 4.71 1.05
Emotional appeal 4.01 1.14 3.92 1.06
Combined appeal 4.30 1.17 4.09 .95
Purchase intention Functional appeal 4.09 1.85 4.14 1.75
Emotional appeal 3.00 1.72 2.83 1.65
Combined appeal 3.50 1.83 2.78 1.31
utilitarian environmental brand benefits” and “virtual nature experience” at the delayed measure (see Appendix F). A summary of all findings of this study can be found in Figure 2 and Table 7 below.
Table 7
List of hypotheses and findings
Hypothesis Support Alternative findings
H1: The effect of functional green brand positioning on brand attitude is fully mediated by the perception of “utilitarian environmental brand benefits” with functional appeals leading to higher levels of “utilitarian environmental brand benefits”, which consequently positively influences brand attitude.
Not supported
Even if H1 could not be supported, additional analysis showed a full serial mediation of the relationship between functional green branding appeal and brand attitude through “utilitarian environmental brand benefits” and ad liking.
H2: The effect of emotional green brand positioning on brand attitude is fully mediated by a “virtual nature experience”, with emotional appeals increasing the level of “virtual nature experience”, which consequently positively influences brand attitude.
Not supported
No mediation of the relationship between emotional green branding appeal and brand attitude through “virtual nature experience” could be found. However, additional mediation analysis instead showed ad liking to be a significant mediator.
H3: (a) The relationship between combined green branding appeals and brand attitude is mediated by “utilitarian environmental brand benefits” and a “virtual nature experience” in a serial mediation, (b) with “virtual nature experience” leading to higher levels of “utilitarian environmental brand benefits” compared to functional appeals, which, in turn, increases positive brand attitude.
Not supported
Even if H3(a) could not be supported, additional analysis showed a full serial mediation of the relationship between combined green branding appeal and brand attitude through “utilitarian environmental brand benefits”, “virtual nature experience” and ad liking.
No significant mean difference in “utilitarian environmental brand benefits” between functional and combined appeals was found H3(b).
H4: Combined green branding appeals lead to stronger positive effects on brand attitude than both, functional appeals and emotional appeals alone.
Not supported
No significant main effect of branding appeal on brand attitude was found. However, descriptive statistics show highest mean for functional condition and lowest mean for emotional condition.
H5: The more positive the brand attitude, the higher the intention to purchase the advertised brand.
Supported
H6: Brand attitude and purchase intention in response to combined and functional green branding appeals are consistent over time and in the absence of the stimulus.
Supported
H7: Brand attitude and purchase intention in response to emotional green branding appeals are not consistent over time and vanish one week after exposure in the absence of the stimulus.
Not supported
No significant interaction effect of condition and time. Brand attitude and purchase intention are consistent over time for all three appeals.
Figure 2. S tatist ica l Mod el * p < .05 ** p < .005 *** p < .001
Discussion Discussion and Theoretical Implications
The results of this study showed only limited support for the hypotheses established in the theoretical framework (see Table 7). No significant main effect of green branding appeal on brand attitude was found. In contrast with findings of previous research (Hartmann, & Apaolaza-Ibáñez, 2009; Matthes et al., 2014; Schmuck et al., 2017), combined green branding appeals did not lead to a significantly more positive brand attitude than emotional and functional appeals. Descriptive statistics even indicated brand attitude to be most positive for functional appeals, while the mean value for the emotional condition was the lowest, although differences did not reach significance. Unexpectedly, it was found that ad liking significantly differed between conditions. Participants in the emotional green branding group liked the ad significantly less than participants in both other groups, while participants in the functional group liked the ad the most. A possible explanation for these findings might be that with the increased concern about environmental topics in recent years (Dangelico, & Vocalelli, 2017), the public’s need and appreciation for factual information about green product attributes has increased substantially. As the mean value in brand attitude and ad liking for combined green branding appeals was lower than for functional appeals (even if both include factual information about environmental benefits of a product) it might
furthermore indicate that with the increase in green washing practices among marketers using misleading nature images (Parguel, Benoit-Moreau, & Russell, 2015), consumers have become sceptical towards nature imagery in advertising. This assumption would moreover explain the lowest mean values for the emotional appeal condition.
When introducing ad liking in the mediation model, a positive indirect effect of functional green branding appeal on brand attitude was found, mediated through “perceived utilitarian environmental brand benefits” and ad liking in a serial mediation. Thus, results
offer some support for previous research, proposing “perceived utilitarian environmental brand benefits” to mediate effects of functional green branding appeals on brand attitude (Hartmann, & Apaolaza-Ibáñez, 2009; Schmuck et al., 2017). Moreover, findings support the TPB as they indicate that brand attitude was based on the beliefs about the brand and their evaluation (Ajzen, 1991; Conner, & Sparks, 2005).
In contrast, the effect of the emotional green branding appeal on brand attitude was only found to be mediated through ad liking, but not through a “virtual nature experience”. Furthermore, the effect of emotional green branding appeal on ad liking as well as the total effect on brand attitude was negative. Findings are in conflict with previous research (e.g., Hartmann, & Apaolaza-Ibáñez, 2008, 2009), suggesting full mediation through affective responses and a positive total effect of emotional green branding appeals on brand attitude. Additionally, findings challenge theory about affective responses to images (Chowdhury et. al, 2008) and classical analysis of aesthetic perception and affective language (Lang et. al ,1993) as both would suggest positive affective responses to pleasant nature imagery in emotional appeals. However, results do offer support for dual process models. Following the logic of the RIM and ELM (Belch et al., 2014; Petty, & Cacioppo, 1986; Strack, & Deutsch, 2004), emotional appeals should be processed impulsively. Hence, it might be that due to people not actively reflecting on the ad that they have seen, they could not recall having experienced a “virtual nature experience” and thus, only a direct effect on ad liking could be found. Nonetheless, as the effect of the emotional green branding appeal on brand attitude was negative, this supports the assumption that consumers have developed an increased need and appreciation for information about sustainable product attributes in green branding and scepticism towards persuasion through nature imagery only.
Additionally, it was found that the effect of the combined green branding appeal on brand attitude was mediated through a “virtual nature experience”, “perceived utilitarian
environmental brand benefits” and ad liking in a serial mediation, offering some support for attention restoration theory (Kaplan, 1995). That is, even if the combined appeal did not lead to higher “perceived utilitarian environmental brand benefits” than the functional appeal, the effect of the combined appeal on “perceived utilitarian environmental brand benefits” was partially mediated through a “virtual nature experience”. The observation that only combined green branding appeals were mediated through “virtual nature experience” while emotional appeals were not, again offer support for the RIM and ELM (Belch et al., 2014; Petty, & Cacioppo, 1986; Strack, & Deutsch, 2004). As combined appeals should be processed through the reflective system, people actively thought about the appeal and hence, were able to recall a “virtual nature experience”. Again, an alternative explanation might be that consumers have become sceptical to green branding that only makes use of visual stimuli. While an image of nature alone was therefore not able to create the sensation of feeling in or connected to nature, the combination of that same image with factual information about environmental product benefits was. Hence, it could be assumed that only when paired with factual information, nature imagery can create a “virtual nature experience” or overcome resistance to such.
Moreover, it has been found that brand attitude significantly increased purchase intention in the domain of green branding, supporting the TPB (Ajzen, 1991; Conner, & Sparks, 2005). However, no significant interaction effects of time and green branding condition on brand attitudes and purchase intentions were found. All three green branding appeals produced consistent brand attitudes and purchase intentions at a one-week delay. These findings are not in line with implications made from dual process models (Belch et al., 2014; Petty, & Cacioppo, 1986; Strack, & Deutsch, 2004) proposing effects of emotional appeals to be temporary. However, this study had several limitations with regards to the
repeated measure design that might have influenced results and will be discusses in the limitations section of this chapter.
Practical Implications
For marketing practitioners this study offers new insights into the effects of distinct green branding appeals and their underlying mechanisms. Particularly and in contrast with previous research, results suggest that emotional green branding appeals, making us of visual stimuli, are not able to elicit affective responses and furthermore, lead to a significantly less favourable judgement of the ad than functional or combined green branding appeals.
Therefore, marketers should aim to include factual information into their green branding appeals as ads featuring such information lead to higher levels of ad liking and – even if not statistically significant – a more positive brand attitude. From a consumer protection
perspective, the results offer hope that consumers are less easily misled by the use of nature imagery in green branding, aiming to promote an ecological brand image through simple association. Thus, there should be fewer opportunities for green washing (Parguel et al., 2015). Furthermore, being the first study to analyse long-term effects of green branding, results offer valuable insights for marketers regarding the persistence of brand attitudes and purchase intentions. Findings suggest that all three green branding appeals induce persistent consumer responses over time and are therefore valuable tools for marketers aiming to not only change consumers’ spontaneous but rather long-term brand attitudes and purchase intentions.
Limitations and Suggestions for Future Research
As mentioned previously, this research had several noteworthy limitations. The repeated measure design only included one repetition at one point in time, with a one-week delay. Even if previous research by Dahlén et al. (2009) and by Sengupta et al. (1997) proposed the effects of unrelated cues to vanish after one week, more repetitions and
especially more time might have been needed to find effects. This is because the study made use of an online experiment and was not completed in a natural setting, and thus, message processing might have been affected (Meyers-Levy, & Malaviya, 1999). Participants might have focussed more on the stimulus materials than they would have in a natural situation, which in turn might have increased their memory for the ad and the persistence of their brand attitude and purchase intention over time. Furthermore, the non-natural setting is a substantial limitation for the study’s external validity (Campbell, 1957). Additionally, as the follow-up survey was sent out via e-mail, it was not possible to control when and with what delay participants would complete the second questionnaire. Therefore, the amount of time that passed between studies, most likely differed somewhat across participants and possibly, across conditions. As has been argued above, effects might differ over a longer period of time. Future research should therefore implement a more natural setting and analyse effects over a longer period of time, holding the time that has passed between exposures steady across conditions. Furthermore, the research employed a snowball sampling strategy, making it difficult to control and monitor the type of chains, number of participants or data quality and making it less externally valid and generalizable (Biernacki, & Waldorf, 1981, Campbell, 1957). In particular, with almost 80% of participants having finished an undergraduate/ Bachelor’s degree or higher, the sample was not representative for the general public.
Furthermore, participants’ high level of education could significantly have influenced results, as a highly educated group should also have a higher need for cognition and might be more sceptical towards green branding (Parguel et al., 2015). Future research should reassess findings for a less educated group of participants. In addition, no control condition was included in this research. It is therefore only possible to compare the effects of distinct green branding appeals to one another but not to a neutral condition or other, non-green branding
practices. Research should therefore include a control condition when reassessing the effects of green branding appeals.
References
Aaker, D. A. (1996). Building strong brands. New York, NY: Free Press.
Aaker, D. A., & Joachimsthaler, E. (2000). The brand relationship spectrum: The key to the brand architecture challenge. California Management Review, 42(4), 8-23.
https://doi.org/10.2307/41166051
Ajzen, I. (2015). Consumer attitudes and behavior: The theory of planned behavior applied to food consumption decisions. Rivista di Economia Agraria, 70(2), 121-138.
http://doi.org/10.13128/REA-18003
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211. https://doi.org/10.1016/0749-5978(91)90020-T Belch, G. E., Belch, M. A., Kerr, G. F., & Powell, I. (2014). Advertising: An Integrated
Marketing Communication Perspective (10th global ed.). North Ryde, N.S.W.: McGraw-Hill Education.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182.
http://dx.doi.org/10.1037/0022-3514.51.6.1173
Bickart, B. A., & Ruth, J. A. (2012). Green eco-seals and advertising persuasion. Journal of Advertising, 41(4), 51-67. https://doi.org/10.1080/00913367.2012.10672457 Biernacki, P., & Waldorf, D. (1981). Snowball sampling: Problems and techniques of chain
referral sampling. Sociological Methods & Research, 10(2), 141-163. https://doi.org/10.1177/004912418101000205
Campbell, D. T. (1957). Factors relevant to the validity of experiments in social
Chang, C. (2011). Feeling ambivalent about going green. Journal of Advertising, 40(4), 19-32. https://doi.org/10.2753/JOA0091-3367400402
Chowdhury, R. M., Olsen, G. D., & Pracejus, J. W. (2008). Affective responses to images in print advertising: Affect integration in a simultaneous presentation context. Journal of Advertising, 37(3), 7-18. https://doi.org/10.2753/JOA0091-3367370301
Conner, M., & Sparks, P. (2005). Theory of planned behaviour and health behaviour. Predicting Health Behaviour, 2, 170-222.
Dahlén, M., Friberg, L., & Nilsson, E. (2009). Long live creative media choice. Journal of Advertising, 38(2), 121-129. https://doi.org/10.2753/JOA0091-3367380208 Dangelico, R. M., & Vocalelli, D. (2017). “Green Marketing”: An analysis of definitions,
strategy steps, and tools through a systematic review of the literature. Journal of Cleaner Production, 165, 1263-1279. https://doi.org/10.1016/j.jclepro.2017.07.184 Dunlap, R. E., & Scarce, R. (1991). Poll trends: Environmental problems and protection. The
Public Opinion Quarterly, 55(4), 651-672. https://doi.org/10.1086/269288
Easterling, D., Kenworthy, A., & Nemzoff, R. (1996). The greening of advertising: A twenty-five year look at environmental advertising. Journal of Marketing Theory and Practice, 4(1), 20-34. https://doi.org/10.1080/10696679.1996.11501714 Edell, J. A., & Burke, M. C. (1987). The power of feelings in understanding advertising
effects. Journal of Consumer Research, 14(3), 421-433. https://doi.org/10.1086/209124
Faul, F., Erdfelder, E., Lang, A. G., & Buchner, A. (2007). G* Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical
sciences. Behavior Research Methods, 39(2), 175-191. https://doi.org/10.3758/BF03193146
Funder, D. C., Levine, J. M., Mackie, D. M., Morf, C. C., Sansone, C., Vazire, S., & West, S. G. (2014). Improving the dependability of research in personality and social
psychology: Recommendations for research and educational practice. Personality and Social Psychology Review, 18(1), 3-12.
https://doi.org/10.1177/1088868313507536
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. https://doi.org//10.1080/13527266.2012.684065 Hadi, N. U., Abdullah, N., & Sentosa, I. (2016). Making sense of mediating analysis: A
marketing perspective. Review of Integrative Business & Economics Research, 5(2), 62-76.
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. https://doi.org/10.1108/02634500510577447
Hartmann, P., & Apaolaza-Ibáñez, V. (2008). Virtual nature experiences as emotional benefits in green product consumption: The moderating role of environmental attitudes. Environment and Behavior, 40(6), 818-842.
https://doi.org/10.1177/0013916507309870
Hartmann, P., & Apaolaza-Ibáñez, V. (2009). Green advertising revisited: Conditioning virtual nature experiences. International Journal of Advertising, 28(4), 715-739. https://doi.org/10.2501/S0265048709200837
Hartmann, P., & Apaolaza-Ibáñez, V. (2012). Consumer attitude and purchase intention toward green energy brands: The roles of psychological benefits and environmental concern. Journal of Business Research, 65(9), 1254-1263.
