INNOVATION DIFFUSION AS A GREEN OMEN?
THE ROLE OF ANTICIPATED SOCIAL EMOTIONS IN
SUSTAINABLE CONSUMPTION OVER THE DIFFUSION PROCESS
Christian Koch
INNOVATION DIFFUSION AS A GREEN OMEN?
THE ROLE OF ANTICIPATED SOCIAL EMOTIONS IN
SUSTAINABLE CONSUMPTION OVER THE DIFFUSION PROCESS
Christian Koch
University of Groningen Faculty of Economics and Business
MSc Marketing Master Thesis
June 2016
Jupiterstraat 29, 9742 ES Groningen, Netherlands +49 162 6144749
c.koch.1@student.rug.nl Student number 2716704
TABLE OF CONTENTS
MANAGEMENT SUMMARY ...II LIST OF FIGURES ... III
1 INTRODUCTION ... 1
2 THEORETICAL FOUNDATIONS... 3
2.1 Innovation Diffusion ... 3
2.2 Social Emotion Anticipation ... 5
2.3 Message Positioning ... 7
2.4 Conceptual Model ... 8
3 RESEARCH METHODOLOGY ... 9
3.1 Participants & Data Collection ... 9
3.2 Research Design & Procedure ... 10
3.3 Manipulations & Variable Measurements ... 10
3.3.1 Manipulation of Innovation Diffusion Stage & Message Positioning ... 10
3.3.2 Measurement of Likelihood to Adopt Green Products ... 12
3.3.3 Measurement of Anticipated Social Embarrassment... 12
4 RESULTS ... 13
4.1 Preliminary Analyses ... 13
4.2 Results for Anticipated Social Embarrassment ... 14
4.3 Results for Likelihood to Adopt Green Products ... 14
5 DISCUSSION ... 16
6 LIMITATIONS AND FUTURE RESEARCH DIRECTIONS ... 18
7 MANAGERIAL IMPLICATIONS ... 21
REFERENCES ... 23
MANAGEMENT SUMMARY
LIST OF FIGURES
1 INTRODUCTION
Step by step, the green wave is conquering our world. From upcycling 3D-printers through leased jeans up to shoes made of ocean plastic waste; sustainable products shape our daily way of life (Vlahov 2015). Originally started off as a sporadic pattern in the 1980s after a thousand years of exploitation and unconcern, sustainable consumption became a trend in the 21st century and can nowadays even be considered a fashionable lifestyle (Chappells & Trentmann 2015). This way of life is, without doubt, highly reliant on the continuous innovation of green products and their (successful) spread among the society, a process named innovation diffusion. That phenomenon describes the communication of new products through certain channels among a specific group over time (Rogers 2003). Recent research has found that this communication implies much more than only a firm’s marketing actions, in fact, it also contains social influences between the consumers (Peres, Muller & Mahajan 2010).
My thesis will contribute to this branch of study by examining whether we, in the role of sustainable consumers, predicate our publicly visible adoption decision on social emotions, which we expect along the diffusion route. In other words; does the diffusion process have an impact on our likelihood to publicly buy / consume green products through the mediator of anticipated negative social emotions?
In fact, I suppose that people will experience less social embarrassment when the product diffuses among society because popularity signals a high degree of social acceptance. This in turn is expected to lead to a higher likelihood to adopt the green product. However, the present paper accounts for another important factor influencing social embarrassment and thereby the adoption likelihood: message positioning. I argue that the level of anticipated social embarrassment not only depends on the social acceptance / popularity of a product but also on the way the product is advertised since this publicly signals your intrinsic buying motivation. More specifically, theory suggests that consumers act strategically when they assume that their social membership is at stake and do so by communicating either selfish or unselfish goals as their buying motives to others (Johnstone & Tan 2015; Vermeir & Verbeke 2006; Mead et al. 2011).
There is already a number of studies that investigated the sources of social influence on a person’s adoption behavior. Some of them found that specific groups like family and friends can contribute to one’s decision for sustainable products (Salazar, Oerlemans & van Stroe-Biezen 2013) while others showed that pioneers in general may have an impact on sustainable consumption (Bolderdijk, Brouwer & Cornelissen 2016). These analyses already head towards the direction of innovation diffusion but do not consider the whole diffusion process and furthermore do not include the fact that people may anticipate negative social emotions as a mean for their decision-making. Smaldino et al. (2015) recently addressed these deficits by developing a model that considers the entire innovation diffusion and incorporates identity signaling as a social-psychological mediator affecting the likelihood to adopt green products. However, their study only shows that people are more likely to adopt a green product if more fellow people do so as well. They do not explain why exactly people use identity signaling for their decision making. My paper will investigate this gap by arguing that it is the level of social embarrassment (together with the message positioning) that guides consumers in their adoption decisions.
Hence, this study will be - to the best of my knowledge - the first one to consider the whole innovation diffusion process as a source of consumers’ likelihood to adopt green products through the anticipation of a negative social emotion (embarrassment), moderated by advertising message positioning.
effects of the different stages on the likelihood to adopt green products. Secondly, it explains how the adopters socially influence the emotional state of a consumer and how this in turn affects the people’s inclination to publicly buy and consume green products. That enables firms to generate a particular empathy that is of use when it comes to the set-up of a marketing campaign for green products. Thirdly, the investigated anticipated embarrassment is an emotion that occurs in a social context, i.e. when people are being watched by others during their buying or consumption process. Hence, my findings indicate whether it might be more useful for a company to offer their consumers the option to adopt the green good in anonymity / privacy. Finally, the knowledge of how to position their green products is a key for firms to increase the adoption likelihood and therewith the speed of diffusion.
The paper proceeds with the literature review which builds the foundation of the present piece of research and gives a more profound insight on recent findings resulting in concrete hypotheses. After visualizing the conceptual model, the methodology section portrays the experimental setting. This part is followed by the results and discussion section depicting and interpreting the outcomes. The thesis closes with managerial implications, limitations and directions for future research.
2 THEORETICAL FOUNDATIONS
2.1 Innovation Diffusion
The theory of innovation diffusion was primarily characterized by Everett Rogers in his famous book ‘innovation diffusion’ (2003). He argues that four main factors influence the diffusion of a new product: the innovation itself, the time, communication channels and a social system. In the present context, the innovation itself are new sustainable / green products, which are defined here as goods that optimize the social (health, workers’ welfare) and environmental (resource use, ecological burden) effects of their acquisition, usage and disposal (OECD, 2008). However, the focus of this research lies on the social system as a success factor for the likelihood to adopt green products. It refers to all influences a person is exposed to such as the media, friends or opinion leaders. It can be summarized as the collectivity of all external factors that constitute and convey the social norm.
(Cialdini & Goldstein 2004). This entails that people would prefer to follow the mainstream by adopting the consumption behavior of the majority. Recent studies have found proof for this inference. They showed that the adoption of new green products is subject to social contagion / influence (Iyengar, Van den Bulte & Valente 2011; Kulviwat, Bruner & Al-Shuridah 2009). Namely, the authors demonstrated that the size of a social network and the usage volume of product users is positively related to the adoption likelihood of non-users.
This social contagion can be viewed as a temporal course that originates from different types of individuals within the social system (Rogers 2003). The innovation diffusion starts with (1) innovators, i.e. people who are the first to adopt the new product, followed by (2) early adopters who are considered as the opinion leaders. These two segments, grouped here as influentials, represent the introduction stage. This phase is characterized by low product awareness and scarce market penetration (Mahajan, Muller & Srivastava, 1990). After around 10-25% of the market adopted the product a certain critical mass is reached implying that the product can self-sustain from now on. This constitutes the point where the (3) early and (4) late majority engage in the products’ consumption. These two segments, named as imitators, represent the maturity
stage in which the product popularity, awareness and penetration reaches its peak. The social
contagion increases as the diffusion proceeds.
H1: Consumers are more likely to adopt the green product during the maturity stage than during the introduction stage.
2.2 Social Emotion Anticipation
As the previous section elaborated, the main effect of innovation diffusion on consumers’ likelihood to adopt green products is supposed to be caused by external social influences. However, this only insufficiently explains the underlying psychological mechanisms. In order to understand what triggers the consumer’s desire to conform to the majority, especially in sustainable consumption processes, one has to analyze the emotional level.
As briefly introduced in the first chapter, green products can elicit both positive and negative feelings. Research has found that so-called ‘moral emotions’ like pride and elevation or shame and embarrassment can be the cause for behaving in a certain moral way (Tangney, Stuewig & Mashek 2007). Antonetti & Maklan (2014) expanded these findings by applying them to the matter of sustainable consumption and showing that they play a crucial role in consumers’ adoption processes. Especially negative emotions gained attention by researchers who bridged these feelings to the case of green products and one’s social environment. Johnstone & Tan (2015) revealed that consumers refused to buy green products because they do not want to be grouped into an unfavorable, stereotypical outside-class by their ‘normal’ friends, which would lower their self-esteem. The authors of this qualitative study demonstrated that people somehow forecast a social exclusion from consuming publicly visible green goods in times when those products are unpopular.
MacDonald & Leary (2005) links those findings to the final adoption behavior by showing that people feel painful when being socially excluded and that this pain affects their eventual thoughts and behavior (MacDonald & Leary 2005).
To sum it up, rarely diffused green products, which would be consumed publicly visible, can cause the fear of being socially not accepted and therefore prevent people from adopting them. I summarize these negative feelings under the term social embarrassment which is in accordance with the outcomes of the large study on emotions in consumer behavior by Laros & Steenkamp (2004). Social embarrassment is here defined as an awkward and uncomfortable emotional state resulting from events that trigger the threat of undesired assessment (e.g. negative stereotypes) by a social audience (Dahl, Manchanda & Argo, 2011).
Yet, as one may have recognized, the critical point here is that these emotions are not solely self-sufficient but rather interpersonal and related to one’s social environment. Emotions help us in establishing social relationships with others and occupying a certain social position in our societal network (Fischer & Manstead 2008). In turn, the relation to significant others, i.e. people who are deeply influential in our life, defines our thoughts, feelings, behaviors and finally our sense of self (Andersen & Chen 2002). This implies that our (sustainable) consumption decisions are more or less deliberately contingent on the emotions that others express to ourselves. A study by Phillips & Baumgartner (2002) has proven this conclusion by showing that people are not only able to but actually do anticipate these consumption emotions in order to come up with a satisfying buying decision.
To get to the heart of all findings: anticipated emotions (and in particular the negative feeling of social embarrassment) explain why social standards have such an important impact on the likelihood to adopt green products. People will not engage in sustainable consumption until green becomes the social norm and their negative emotions turn into positive ones. This conclusion leads us to the subsequent hypotheses which are based on the presumption that anticipated social embarrassment mediates the effect of an innovation diffusion stage on consumers’ likelihood to adopt a green product.
H2: Consumers are more likely to anticipate social embarrassment when a green product has not been diffused (introduction stage) compared to when it has been diffused (maturity stage).
2.3 Message Positioning
The reasons people engage in sustainable consumption can be basically grouped into two categories; selfish and unselfish motives (Thøgersen 2011; Griskevicius, Tybur & Van den Bergh 2010). While selfish motives stress the economic trade-off or personal outcomes such as health and social status, unselfish motives refer to altruistic values like environmental concerns and moral standards. These two adoption motivations have the potential to differentially affect people’s anticipated social embarrassment. As the previously introduced study by Johnstone & Tan (2015) has shown, people anticipate a social exclusion from adopting publicly visible green goods which are not yet widely diffused. This is not an accidental human reaction but rather very closely connected to the way a product message is positioned.
As we have seen, the do-gooder derogation theory first and foremost stresses the moral dimension. The societal behavior of devaluing someone else not only occurs when people encounter any outlier but in particular when they meet a person whom they consider as a moral outlier. The reason is that a deviation from moral (compared to e.g. artistic grounds) basically conveys that it is wrong for anybody else to act in a different way (Monin, Sawyer & Marquez 2008). Thus, moral prescriptions are considered by humans as universal and are therefore prone to social conflicts.
Hence, emphasizing the moral dimension of a green product by appealing to unselfish motives in the introduction stage may unintentionally enhance the anticipated social embarrassment, which is associated with the adoption of novel products, even further. This assumption is in line with an intriguing finding by Parks & Stone (2010) who showed that overly unselfish people are unwelcome in a rather selfish social group.
In contrast, positioning sustainable products as “healthy” and thereby evading the undesirable stereotype that is associated with morality may lower the anticipated social embarrassment (Vermeir & Verbeke 2006). The second main result of Johnstone & Tan’s study (2015) underscores this: as long as public moral behavior is not constituting the social norm, people place individual goals over collective goals. In other words, justifying the adoption of an unpopular green product by highlighting one’s selfish motivations may minimize the threat of being judged as a moral outlier by peers.
Mead et al. (2011) found out that people consume strategically when their social affiliation is at stake. More concretely, consumers are likely to adopt a product that symbols group membership when being socially excluded. The social exclusion in the maturity stage applies to all those consumers who have not yet adopted the popular green product. As a result, these people will try to fill the social gap by buying and consuming it. However, in order to make sure that they are fully conform to the social norm, they will not just adopt it but they will specifically choose the morally positioned option. One reason for that behavior is that already the mere exposure to green products, which is an inherent part of the maturity stage, makes people subsequently act more altruistically than the exposure to conventional products (Mazar & Zhong 2010). Another reason is that people think that only the unselfish / moral option signals their unconditional affiliation desire to the majority and that this intrinsic motive will make them feel much happier than pursuing egoistic values in consumption decisions (Krishna 2011). Based on these theoretical examinations, I hypothesize that message positioning plays a crucial role in the model under investigation. On the one hand, as a moderator of the relation between innovation diffusion and anticipated social embarrassment and, on the other hand, as an interaction term with the innovation diffusion affecting anticipated social embarrassment.
H4: During the introduction stage, positioning the green product as moral (“ecologically friendly”) elicits more social embarrassment compared to positioning it as selfish (“healthy”).
In contrast, during the maturity stage, positioning the green product as moral (“ecologically friendly”) elicits less social embarrassment compared to positioning it as selfish (“healthy”).
2.4 Conceptual Model
FIGURE 1
Conceptual Framework and Hypotheses
3 RESEARCH METHODOLOGY
3.1 Participants & Data Collection
The data was collected via an online questionnaire and thus distributed through online channels (see appendix 1 for the whole survey). A convenience sample was drawn from spread via e-mail and social media. However, to also reach individuals that are more distant from the researchers’ social sphere, the survey was distributed among employees of a multi-national company and in online forums. On aggregate, 206 of 229 participating people completed the survey (drop-out rate = 10%). After considering attention and manipulation checks, 142 responses turned out to be valid (69%) with 92 male (65%) and 50 (35%) female respondents. The age distribution showed that 80.3% were between 18-34 years. The survey was online accessible for ~1 week from 22 April - 28 April in English and German. The survey data was extracted and analyzed by means of IBM SPSS.
Innovation
Diffusion Stage
Likelihood to
Consumers’
3.2 Research Design & Procedure
The present study was performed as an experimental research in order to determine causal relation. This implied the random assignment of participants to the manipulated innovation diffusion stages and message positionings. Each person was only exposed to one of the four treatment level combinations (2x2 between-participants factorial design). I chose organic potato chips as the green product that people were supposed to adopt. The first reason for this choice was the assumption that all potential participants would be familiar with this product category. Secondly, potato chips are a type of product one could easily think of consuming it in a social gathering. The latter constituted a major selection criterion with regard to the measured anticipated social embarrassment.
As the underlying pattern was expected to be too complicated for participants to disclose, no cover story was introduced. Instead, the respondents were told that they host an evening with friends (in order to stress the social aspect for anticipated social embarrassment) and that they need to buy potato chips for this gathering. Financial budget was said to be of no importance. Before the final spread of the survey, an intensive but successful pre-test was conducted among 80 respondents to make sure people were not biased or aware of the real purpose and to adapt the survey where necessary.
3.3 Manipulations & Variable Measurements
3.3.1 Manipulation of Innovation Diffusion Stage & Message Positioning
FIGURE 2
Manipulation of Innovation Diffusion Stage
Next to the data about the market diffusion, respondents were exposed to a picture of an organic bag of chips, adapted from the Albert Heijn “Biologisch Groentechips” (Albert Heijn 2016). This bag of chips was either labeled as “ecologically friendly” in the moral condition or as “healthy” in the selfish condition (see Figure 3). Again, the efficiency of this manipulation was assessed by a manipulation check at the end of the survey asking for a recall of the label.
FIGURE 3
3.3.2 Measurement of Likelihood to Adopt Green Products
After being exposed to one of the four treatment levels, the participants were asked to indicate their likelihood to choose the green product on a 7-level Likert item (extremely unlikely – extremely likely). By means of the before mentioned manipulation checks, I made sure that only participants who comprehended the situation entered the final data set.
3.3.3 Measurement of Anticipated Social Embarrassment
The chosen setting, introduced at the beginning of the survey, was one that was supposed to be easily imagined by the participants. By requesting them to imagine that they will host an evening with friends, a situation was created in which people could be concerned about the image they portrayed of themselves to others. Hence, the introduction included the potential to elicit awkward social feelings, i.e. embarrassment. The friends were representing the significant others which were ought to have high social influence on the respondents (see chapter 2.2). The particular mention that the participants plan to buy chips for themselves and their friends was of high importance as well because this signals that their adopted product can be publicly seen, consumed and judged by their peers. Finally, the additional information about the unlimited budget was also crucial. Therewith, I made sure that the price perception, which is often higher for green products, does account as little as possible for the variance in consumers’ adoption likelihood (Chang 2011). Hence, the gathering seemed to be an ideal setting to measure anticipated social embarrassment.
In order to figure out if and to what extent the respondents would anticipate social embarrass-ment after indicating their likelihood to adopt green products, I asked them how they feel about serving this environmentally-friendly produced / healthy product to their friends. On a 3-item 7-point Likert-scale (not at all – very), adapted from Dahl, Manchanda & Argo (2001), participants had to indicate the degree to which they felt embarrassed, awkward and uncom-fortable. As a 4th item on that scale, I included an attention check requesting the respondents to tick a certain bubble / value. A reliability analysis was conducted in order to determine whether the three Likert items (embarrassment, awkwardness, uncomfortableness) measure the same underlying construct, i.e. anticipated social embarrassment (see appendix 2). The scale had a high level of internal consistency, as determined by a Cronbach's alpha of 0.857 (Manticipated social embarrassment = 2.23, SDanticipated social embarrassment= 1.19). Hence, the three items constituted a reliable
can already see, the reported mean is rather low implying that people did not experience strong anticipated social embarrassment overall. This finding will be discussed further in the results section.
4 RESULTS
4.1 Preliminary Analyses
First of all, the data was reviewed for attention checks revealing that 8.25% did not pass it and thus had to be excluded. Next, the success of the innovation diffusion manipulation, i.e. the fictitious market report, was examined. At the end of the survey, respondents had to indicate on a 6-level Likert item (1= extremely unpopular, 6= extremely popular) how popular the product in their opinion was. Participants facing the introduction stage condition indeed considered the product as very unpopular (Mintroduction stage = 1.94, SDintroduction stage = 1.03). The manipulation for
the maturity stage was not that successful but still effective (Mmaturity stage = 3.8, SDmaturity stage =
1.9). A cut-off value was set at 3.5 meaning that in the introduction stage scenario only those respondents who indicated a 1-3 were taken over to the final data set (93.8%) whereas in the maturity stage scenario only those who indicated a 4-6 were accepted (64.13%).
Third, the success of the moderator manipulation was reviewed, i.e. the message positioning. In the moral condition 91.4% correctly recalled the product as being labeled “environmentally-friendly produced”. In the selfish condition 87.5% accurately recalled that the chips were advertised as “healthy”. This shows that the moderator manipulation worked well. The few respondents not correctly recalling the message were excluded.
All in all, the final set of 142 valid responses consisted of 84 respondents that were exposed to the introductions stage (42 in moral condition, 42 in selfish condition) and 58 respondents that faced the maturity stage (29 in moral condition, 29 in selfish condition).
4.2 Results for Anticipated Social Embarrassment
The mediator variable was analyzed by means of the PROCESS macro for SPSS (Hayes 2013; see appendix 4). The regression depicts that there is no significant effect of the innovation diffusion stage on anticipated social embarrassment (β = 0.03, t(142) = 0.16, p = .87). The coefficient, which is approaching a value of 0, suggests that, holding message positioning constant, neither of the two stages is eliciting more anticipated social embarrassment than the other. Likewise, there is a non-significant effect of message positioning on anticipated social embarrassment (β = -0.23, t(142) = -1.13, p = .26). The slightly negative coefficient indicates that, holding the diffusion stage constant, the anticipated social embarrassment is .23 points lower (on a scale of 1-7) for people being exposed to the morally positioned product. The hypothesized interaction effect of both, the independent variable and moderator, on the mediator is also not significant (β = .23, t(142) = .55, p = .59).
These findings already indicate that there is no statistically significant evidence that anticipated social embarrassment functions as a mediator between innovation diffusion stage and the likelihood to adopt a green product. Indeed, this is confirmed by the result that the bootstrapped confidence interval includes the value of 0 with 95% CI [-0.17, 0.29], implying that there is a non-significant indirect effect (see appendix 4).
These results on mediation and moderation do not provide sufficient evidence for hypothesis 2 and 4, which is why these assumptions have to be rejected.
4.3 Results for Likelihood to Adopt Green Products
a scale form 1-7) than respondents in the introduction stage. On average people in the introduction stage indicated a lower likelihood (Mintroduction stage = 3.13, SDintroduction stage = 1.6) than
people in the maturity stage (Mmaturity stage = 4.17, SDmaturity stage = 1.57) to adopt the green product
(compare appendix 5).
In order to detect further group differences, the data was again checked for normal distribution, this time separately for the two stages. The assumption of normality for the 'likelihood to adopt green products' scores was not satisfied for either of the two groups as assessed by Shapiro-Wilk's test (p < .05; see appendix 6). Thus, A Mann-Whitney U test was run to determine if there were differences in the ‘likelihood to adopt green products’ scores between the introduction and maturity stage (see appendix 7). Distributions of the likelihood scores for introduction and maturity stage were not similar, as assessed by visual inspection. The likelihood scores in the maturity stage (mean rank = 86.52) were statistically significantly higher than in the introduction stage (mean rank = 61.13), U = 3307, z = 3.669, p < .001. This is another evidence for my theoretical assumptions regarding the main effect.
All these outcomes together confirm hypothesis 1.
More than that, the PROCESS regressions showed a significant direct and negative effect of anticipated social embarrassment on the likelihood to adopt green products (β = -0.35, t(142) = -2.57, p < .05; 95% CI [-0.63, -0.08]; see appendix 4). This effect means that the more (less) social embarrassment people anticipate, the less (more) likely they are to adopt the green product. This interestingly points out that even though anticipated social embarrassment is not a significant mediator of innovation diffusion stage on the likelihood to adopt green products, it still significantly affects the dependent variable. This finding is in line with hypothesis 3 which is why that assumption can be accepted.
which is drawing near 0, implies that neither of the two message positions leads to a stronger likelihood to adopt green products than the other.
These and all other results will be discussed and interpreted in the next chapter.
5 DISCUSSION
The present thesis investigated the research question whether the diffusion process has an impact on our likelihood to adopt green products through the mediator of anticipated negative social emotions, moderated by message positioning. The results show that there is indeed a significant main effect of the diffusion stage on consumers’ adoption likelihood. More precisely, the outcome depicts that people are more likely to adopt a green product when it has reached the maturity stage compared to the introduction stage. This is congruent with H1 and the theory which suggests that people wait until a green good becomes the social norm before they adopt it (see chapter 2.1). The experimental research of my study used a market report and the product’s adoption rate as a signal for the social norm. Hence, the participants’ adoption tendency in favor of the maturity stage can be seen as a result from social contagion (Iyengar, Van den Bulte & Valente 2011). People actually used the percentage rate as a decision anchor for their adoption likelihood. These findings are supported by Ozaki (2010) who figured out that people are particularly likely to refuse green innovations when they experience a lack of strong social norms.
However, this paper goes beyond determining the main effect. It aims at investigating the underlying reasons why people wait until green becomes the social norm. More specifically, I predicted that anticipated social embarrassment is the factor which is largely accounting for that effect. The results of the experiment provided me with some interesting insights. In general, they have shown that there is no statistically significant evidence that anticipated social embarrassment explains the effect of the product diffusion stage on people’s adoption likelihood. However, anticipated social embarrassment indeed plays a role as a determinant of people’s adoption likelihood. In line with H3, I found that there is a direct and negative effect of anticipated social embarrassment on adoption likelihood. This implies that people who are forecasting a feeling of social embarrassment, which would arise from adopting the green product, are less likely to buy and consume it.
In order to determine what raised the feeling of anticipated social embarrassment, the impact of message positioning was examined closer. It came out that this part of my model does not play a significant role at all, which leads to a lack of statistical evidence to accept H4. The way a product is positioned in its advertising message has neither an impact on the adoption likelihood nor on the anticipated social embarrassment that one may experience. Apparently, the participants did not feel that the labels on the product will strongly exert social embarrassment on themselves. One reason for that unexpected result might be that the messages on the products were not representing the underlying dimensions, i.e. the moral and selfish motivations, clearly enough. This explanation implies that the people did not establish the cognitive association between the message “environmentally friendly produced” (“healthy”) and a moral (selfish) consumption motive. They simply screened the message without any further cognitive or affective evaluation of it. This subsequently led to the fact that the respondents did not unintentionally feel and act according to the moral do-gooder theory. This means that if they did not view the labels as a signal for moral or selfish consumption behavior, they also did not feel as a potential moral outlier and therefore did not anticipate social embarrassment.
shown a significant impact on anticipated social embarrassment if the people would have considered both independent variables in one go. However, in the present study people probably did not establish that intellectual linkage because the visual and textual manipulations of the two variables were displayed too separately. Both possible explanations will be considered further in the limitations chapter.
In any case, it entails that neither the presence vs. absence of the social norm nor the moral vs. selfish product message elicited the anticipated negative social emotions. That raises the question; what caused the significant relationship between anticipated social emotions and the adoption likelihood?
The only plausible explanation is that the participants entered the experiment with an already existing certain association of green products in general or organic potato chips in particular. This implies two scenarios. On the one hand, the respondents may already have experienced the negative emotion of anticipated social embarrassment when encountering this type of product (category) in their previous life. On the other hand, the respondents may have had rather positive experiences with this product (category) and hence were influenced by these kinds of experiences. In both cases the underlying phenomenon is the same; the people developed beforehand a basic attitude towards green products in general or the potato chips in particular and transferred this attitude into the new situation while neglecting the aspects of message positioning and innovation diffusion.
Whatever the reasoning may be, the fact remains that anticipated social embarrassment influence people’s adoption likelihood of green goods that are bought and consumed in public. The more social embarrassment one anticipates, the less likely he or she is to adopt the green product. This conforms to the theoretical evidence of Phillips & Baumgartner (2002) who have shown that people use their ability to anticipate social emotions in order to come up with a satisfying adoption decision that is in line with one’s sense of self.
6 LIMITATIONS AND FUTURE RESEARCH DIRECTIONS
the four conditions does also not allow for inferences about gender-specific behavior. One might examine potential demographical differences within this field of research further. Besides that, the study is also limited with regard to its theoretical contributions. First, the usually four different diffusion stages were compressed into two stages, namely the introduction and maturity stage. Hence, the different types of adopters were also grouped together. Especially for marketers it may be of value to further differentiate their customers in order to target them more precisely.
Next, the good that was chosen here as a prime example of green / sustainable products was a bag of organic potato chips. Although this type of product constituted a proper sample to convey a familiar situation which may elicit social embarrassment, it is still narrowed. The results for adoption likelihood and anticipated social embarrassment may be completely different for other product (categories). Especially the fact that there is no significant effect of the diffusion stage on the anticipated social embarrassment while the latter has an effect on the adoption likelihood shows that people probably already had a certain association with this kind of product (category). Hence, it would not only be interesting to examine the conceptual model with a different product for the sake of completeness but also to analyze the reasons why one half of the proposed mediation did not emerge.
The discussion on the product type also calls for consideration of the product’s way of consumption. In my paper I opted for a product whose adoption, i.e. buying and consumption, is publicly visible. This was necessary because otherwise I could not have created a setting in which anticipated social embarrassment plays a role. This means that the findings of this paper are restricted to green products that are bought and / or consumed in public. However, as the adoption of many products takes places in a publicly visible manner, these restrictions are rather marginal.
With the wisdom of hindsight, there are finally two methodological limitations which simultaneously constitute a potential for improvement of the study. First, I would have slightly adapted the diffusion stage manipulation. In the survey the diffusion stages were represented by a market report stating facts about the products’ popularity. Even though this is a good indicator for a product’s diffusion, it is not the only one. Future research might also include data about e.g. the good’s past performance, prospective sales figures or consumer awards in order to substantiate the current diffusion position. Besides that, one might also think of making the report’s visual appearance more credible. This would be particularly useful with regard to the maturity stage since the manipulation of this condition showed room for optimization. Secondly, the aspect that the message positioning had no significant effect on anticipated social embarrassment even though the manipulation worked well and theory suggested that, gives researchers another hint. The labels “environmentally friendly produced” and “healthy” are either not an adequate representation of moral vs. selfish consumption motives, were not considered closely enough in conjunction with the diffusion stage manipulation, or their underlying concepts just do not contribute to people’s anticipated social emotions. Hence, I would have extended the message positioning manipulation. On the one hand, it would have been wise to make use of further phrases that trigger consumers’ cognitive and affective associations with moral and selfish motives. That could be labels that appeal to stronger environmental concerns for the moral condition or status motives for the selfish condition. On the other hand, the number and visual variety of labels or messages could be increased. If the respondents would have seen further advertising messages in leaflets or on websites, they might have been more receptive to the manipulation.
7 MANAGERIAL IMPLICATIONS
The discussion on the results of this paper entails some intriguing implications for marketers and firms. First, it is essential to know that green products need a special support directly from scratch. The data has shown that products in the introduction phase have a huge disadvantage over products in the maturity phase with regard to people’s adoption likelihood. Companies promoting sustainable goods, which imply a publicly visible adoption, should spread positive word-of-mouth, advertise heavily and pretend popularity immediately after market launch. It is of high importance to reach the critical mass and therewith the maturity stage as soon as possible. Nevertheless, as Janssen & Jager (2002) have shown, stimulating diffusion of green products needs to be performed with regard to the distinct characteristics of the respective market and its customers. Even though it may be basically good to push the diffusion of a green innovation, there might be products and consumers who demand an initial slower spread in order to maintain their unique selling point or who even call for a product renewal before the maturity stage is reached.
Secondly, the results point out that accompanying a green product with market data that claims a wide product diffusion enforces the consumers’ associations about the good as belonging to the social norm. In other words, advertising one’s sustainable product by communicating its popularity, eventually promotes the company’s turnovers. However, this is not supposed to convey that companies should make use of the truth bias and deceive their customers.
Third, people’s relying on a single current report number demonstrates that marketers do not need to provide potential consumers with extensive market data. Rather, they should prepare short but captivating figures. Late-breaking market data seems to be most convincing when it comes to customer acquisition.
REFERENCES
Albayrak, Tahir, Meltem Caber, Luiz Moutinho, and Ram Herstein (2011), “The influence of skepticism on green purchase behavior”, International Journal of Business and Social
Science, 2 (13), 189-197.
Albert Heijn (2016), “AH Biologisch Groentechips”, (accessed April 15, 2016), [available at http://www.ah.nl/producten/product/wi238822/ah-biologisch-groentechips].
Andersen, Susan M., and Serena Chen (2002), “The Relational Self: An Interpersonal Social-Cognitive Theory”, Psychological Review, 109 (4), 619-645.
Antonetti, Paolo, and Stan Maklan (2014), “Feelings that Make a Difference: How Guilt and Pride Convince Consumers of the Effectiveness of Sustainable Consumption Choices”,
Journal of Business Ethics, 124, 117-134.
Aquino, Karl, and Americus Reed (2002), “The self-importance of moral identity”, Journal of
Personality and Social Psychology, 83 (6), 1423-1440.
Arts, Joep W. C., Ruud T. Frambach, and Tammo H. A. Bijmolt (2011), “Generalizations on consumer innovation adoption: A meta-analysis on drivers of intention and behavior”,
International Journal of Research in Marketing, 28 (2), 134-144.
Bolderdijk, Jan Willem, Claire Brouwer, and Gert Cornelissen (2016), “When do moral pioneers elicit irritation instead of imitation?”, unpublished paper.
Burgoon, Judee K., and Timothy R. Levine (2010), “Advances in Deception Detection”, in
New Directions in Interpersonal Communication Research, Sandi W. Smith and Steven
R. Wilson, eds. Thousand Oaks: SAGE Publications, Inc., 201-220.
Chappells, Heather, and Trentmann, Frank (2015), “Sustainable consumption in history: ideas, resources, practices”, in Handbook of Research on Sustainable Consumption, Lucia A. Reisch and John Thøgersen, eds. Cheltenham: Edward Elgar Publishing Limited, 51-69.
Childers, Terry L., and Akshay R. Rao (1992), “The Influence of Familial and Peer-based Reference Groups on Consumer Decisions”, Journal of Consumer Research, 19 (2), 198-211.
Cialdini, Robert B., and Noah J. Goldstein (2004), “Social influence: Compliance and Conformity”, Annual Review of Psychology, 55, 591-621.
Dahl, Darren W., Rajesh V. Manchanda, and Jennifer J. Argo (2001), “Embarrassment in Consumer Purchase: The Roles of Social Presence and Purchase Familiarity”, Journal
of Consumer Research, 28 (3), 473-481.
East, Robert, Kathy Hammond, and Wendy Lomax (2008), “Measuring the impact of positive and negative word of mouth on brand purchase probability”, International Journal of
Research in Marketing, 25 (3), 215-224.
Feinberg, Matthew, and Robb Willer (2013), “The Moral Roots of Environmental Attitudes”,
Psychological Science, 24, 56-62.
Fischer, Agneta H., and Antony S. R. Manstead (2008), “Social Functions of Emotion”, in
Handbook of Emotions, Michael Lewis, Jeanette M. Haviland-Jones, and Lisa F.
Barrett, eds. New York: The Guilford Press, 456-468.
Furlow, Nancy E. (2010), “Greenwashing in the New Millenium”, The Journal of Applied
Business and Economics, 10 (6), 22-25.
Griskevicius, Vladas, Joshua M. Tybur, and Bram Van den Bergh (2010), “Going Green to Be Seen: Status, Reputation, and Conspicuous Conservation”, Journal of Personality and
Haanpää, Leena (2007), “Consumers’ green commitment: indication of a postmodern lifestyle?”, International Journal of Consumer Studies, 31 (5), 478-486.
Hayes, Andrew F. (2013), Introduction to Mediation, Moderation, and Conditional Process
Analysis: A Regression-Based Approach. New York: Guilford Press.
Iyengar, Raghuram, Christophe Van den Bulte, and Thomas W. Valente (2011), “Opinion Leadership and Social Contagion in New Product Diffusion”, Marketing Science, 30 (2), 195-212.
Janssen, Marco A., and Wander Jager (2002), „Stimulating diffusion of green products – Co-evolution between firms and consumers”, Journal of Evolutionary Economics, 12 (3), 283-306.
Johnstone, Michael-Lee, and Lay P. Tan (2015), “An exploration of
environmentally-conscious consumers and the reasons why they do not buy green products”, Marketing
Intelligence & Planning, 33 (5), 804-825.
Krishna, Aradhna (2011), “Can supporting a cause decrease donations and happiness? The cause marketing paradox”, Journal of Consumer Psychology, 21 (3), 338-345.
Kulviwat, Songpol, Gordon C. Bruner, and Obaid Al-Shuridah (2009), “The role of social influence on adoption of high tech innovations: The moderating effect of public/private consumption”, Journal of Business Research, 62 (7), 706-712.
MacDonald, Geoff, and Mark R. Leary (2005), “Why Does Social Exclusion Hurt? The Relationship between Social and Physical Pain”, Psychological Bulletin, 131 (2), 202-223.
Mahajan, Vijay, Eitan Muller, and Rajendra Kumar Srivastava (1990), “Determinants of Adopter Categories Using Innovation Diffusion Models”, Journal of Marketing
Mazar, Nina, and Chen-Bo Zhong (2010), “Do Green Products Make Us Better People?”,
Psychological Science, 21 (4), 494-498.
Mead, Nicole L., Roy F. Baumeister, Tyler F. Stillman, Catherine D. Rawn, and Kathleen D. Vohs (2011), “Social exclusion causes people to spend and consume strategically in the service of affiliation”, Journal of Consumer Research, 37 (5), 902-919.
Minson, Julia A., and Benoit Monin (2012), “Do-Gooder Derogation: Disparaging Morally Motivated Minorities to Defuse Anticipated Reproach”, Social Psychological and
Personality Science, 3 (2), 200-207.
Monin, Benoit, Pamela J. Sawyer, and Matthew J. Marquez (2008), “The rejection of moral rebels: Resenting those who do the right thing”, Journal of Personality and Social
Psychology, 95, 76-93.
Muster, Viola (2011), “Companies Promoting Sustainable Consumption of Employees”,
Journal of Consumer Policy, 34, 161-174.
Laros, Fleur J. M., and Jan-Benedict E. M. Steenkamp (2004), “Emotions in consumer behavior: a hierarchical approach”, Journal of Business Research, 58 (2005), 1437-1445.
Lee, Kaman (2008), “Opportunities for green marketing: young consumers”, Marketing
Intelligence & Planning, 26 (6), 573-586.
OECD (2008), “Promoting Sustainable Consumption”, OECD Publications, 1-59.
Onkvisit, Sak, and John Shaw (1984), “Self-concept and image congruence: some research and managerial implications”, Journal of Consumer Marketing, 4, 13-23.
Parks, Craig D., and Asako B. Stone (2010), “The Desire to Expel Unselfish Members From the Group”, Journal of Personality and Social Psychology, 99 (2), 303-310.
Peres, Renana, Eitan Muller, and Vijay Mahajan (2010), “Innovation diffusion and new product growth models: A critical review and research directions”, International
Journal of Research in Marketing, 27 (2), 91-106.
Phillips, Diane M., and Hans Baumgartner (2002), “The role of consumption emotions in the satisfaction response”, Journal of Consumer Psychology, 12 (3), 243-252.
Rogers, Everett M. (2003), Diffusion of Innovations, 5th ed., New York: The Free Press.
Salazar, Helen A., Leon Oerlemans, and Saskia van Stroe-Biezen (2013), “Social influence on sustainable consumption: evidence from a behavioural experiment”, International
Journal of Consumer Studies, 37 (2), 172-180.
Smaldino, Paul E., Marco A. Janssen, Vicken Hills, and Jenna Bednar (2015), “Adoption as a Social Marker: The Diffusion of Products in a Multigroup Environment”, eprint arXiv. Täuber, Susanne, and Martijn van Zomeren (2013), “Outrage towards whom? Threats to
moral group status impede striving to improve via out-group-directed outrage”,
European Journal of Social Psychology, 43 (2), 149-159.
Tangney, June P., Jeff Stuewig, and Debra J. Mashek (2007), “Moral emotions and moral behavior”, Annual Review of Psychology, 58, 345-372.
Thøgersen, John (2011), “Green Shopping: For Selfish Reasons or the Common Good?”,
American Behavioral Scientist, 55 (8), 1052-1076.
Vermeir, Iris, and Wim Verbeke (2006), „Sustainable Food Consumption: Exploring the Consumer Attitude-Behavioral Intention Gap”, Journal of Agricultural and
Environmental Ethics, 19 (2), 169-194.
APPENDICES
List of Appendices
Appendix 1: Soft Copy of the Conducted Online Survey ... 30
Appendix 2: Reliability Analysis for Anticipated Social Embarrassment ... 36
Appendix 3: Normality Test for Likelihood to Adopt Green Products ... 38
Appendix 4: Moderated Mediation Analysis (PROCESS for SPSS) ... 39
Appendix 5: Group Statistics on Likelihood to Adopt Green Products ... 40
Appendix 6: Normality Tests for Likelihood to Adopt Green Products (Diffusion Stages) .. 41
Appendix 7: Mann-Whitney U Test ... 42
Appendix 2: Reliability Analysis for Anticipated Social Embarrassment
Case Processing Summary
N %
Cases Valid 142 100,0
Excludeda 0 ,0
Total 142 100,0
a. Listwise deletion based on all variables in the procedure.
Reliability Statistics
Cronbach's Alpha
Cronbach's Alpha Based
on Standardized Items N of Items
,857 ,860 3
Item Statistics
Mean Std. Deviation N
How would you feel about serving these […] Chips to your friends...-Uncomfortable
2,50 1,491 142
How would you feel about serving these […] Chips to your friends...-Awkward
2,18 1,297 142
How would you feel about serving these […] Chips to your friends...-Embarrassed
2,01 1,249 142
Inter-Item Correlation Matrix
How would you feel about serving these […] Chips to your
friends...-Uncomfortable
How would you feel about serving these […]
Chips to your friends...-Awkward
How would you feel about serving these […] Chips to your
friends...-Embarrassed How would you feel about serving these […] Chips to
your friends...-Uncomfortable
1,000 ,708 ,628
How would you feel about serving these […] Chips to your friends...-Awkward
,708 1,000 ,681
How would you feel about serving these […] Chips to your friends...-Embarrassed
Item-Total Statistics Scale Mean if Item Deleted Scale Variance if Item Deleted Corrected Item-Total Correlation Squared Multiple Correlation Cronbach's Alpha if Item Deleted How would you feel about serving
these […] Chips to your friends...-Uncomfortable
4,20 5,450 ,729 ,541 ,810
How would you feel about serving these […] Chips to your friends...-Awkward
4,51 6,124 ,770 ,593 ,764
Appendix 3: Normality Test for Likelihood to Adopt Green Products
Tests of Normality
Kolmogorov-Smirnova Shapiro-Wilk
Statistic df Sig. Statistic df Sig.
How likely are you to choose the chips which you have seen before?
,138 140 ,000 ,937 140 ,000
a. Lilliefors Significance Correction
Descriptives
Statistic Std. Error How likely are you to choose the chips
which you have seen before?
Mean 3,54 ,141
95% Confidence Interval for Mean Lower Bound 3,26
Appendix 4: Moderated Mediation Analysis (PROCESS for SPSS) Model = 7 Y = LAGP X = ID_Stage M = AnSocEmb W = MessPosi Sample size 142 ************************************************************************** Outcome: AnSocEmb Model Summary R R-sq MSE F df1 df2 p ,1093 ,0119 1,4303 ,6511 3,0000 138,0000 ,5836 Model
coeff se t p LLCI ULCI constant 2,2324 ,1019 21,9067 ,0000 2,0309 2,4339 ID_Stage ,0346 ,2141 ,1617 ,8718 -,3887 ,4579 MessPosi -,2300 ,2038 -1,1287 ,2610 -,6330 ,1729 int_1 ,2334 ,4282 ,5452 ,5865 -,6132 1,0801 Product terms key:
int_1 ID_Stage X MessPosi
************************************************************************** Outcome: LAGP Model Summary R R-sq MSE F df1 df2 p ,3987 ,1590 2,3661 15,0467 2,0000 139,0000 ,0000 Model
coeff se t p LLCI ULCI constant 4,3454 ,3364 12,9164 ,0000 3,6802 5,0106 AnSocEmb -,3535 ,1375 -2,5700 ,0112 -,6254 -,0815 ID_Stage 1,0537 ,2670 3,9469 ,0001 ,5259 1,5815 ******************** DIRECT AND INDIRECT EFFECTS ************************* Direct effect of X on Y
Effect SE t p LLCI ULCI 1,0537 ,2670 3,9469 ,0001 ,5259 1,5815 Conditional indirect effect(s) of X on Y at values of the moderator(s): Mediator
MessPosi Effect Boot SE BootLLCI BootULCI AnSocEmb -,5000 ,0290 ,1141 -,1700 ,2872 AnSocEmb ,5000 -,0535 ,1048 -,2657 ,1658
******************** INDEX OF MODERATED MEDIATION ************************ Mediator
Index SE(Boot) BootLLCI BootULCI AnSocEmb -,0825 ,1540 -,4409 ,1939
When the moderator is dichotomous, this is a test of equality of the conditional indirect effects in the two groups.
******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 1000
Level of confidence for all confidence intervals in output: 95,00
NOTE: The following variables were mean centered prior to analysis: ID_Stage MessPosi
Appendix 5: Group Statistics on Likelihood to Adopt Green Products
Case Processing Summary
Cases
Included Excluded Total
N Percent N Percent N Percent
How likely are you to choose the chips which you have seen before? * ID_Stage
142 100,0% 0 0,0% 142 100,0%
Report
How likely are you to choose the chips which you have seen before?
ID_Stage Median Mean Variance Std. Error of Mean Std. Deviation
Introduction 3,00 3,13 2,573 ,175 1,604
Maturity 4,00 4,17 2,461 ,206 1,569
Appendix 8: Mediation Analysis (PROCESS for SPSS) Model = 4 Y = LAGP X = MessPosi M = AnSocEmb Sample size 142 ************************************************************************** Outcome: AnSocEmb Model Summary R R-sq MSE F df1 df2 p ,0970 ,0094 1,4135 1,3104 1,0000 140,0000 ,2543 Model
coeff se t p LLCI ULCI constant 2,3474 ,1458 16,1053 ,0000 2,0593 2,6356 MessPosi -,2300 ,2010 -1,1447 ,2543 -,6274 ,1673 ************************************************************************** Outcome: LAGP Model Summary R R-sq MSE F df1 df2 p ,2484 ,0617 2,6397 2,4957 2,0000 139,0000 ,0861 Model
coeff se t p LLCI ULCI constant 4,3538 ,4059 10,7274 ,0000 3,5514 5,1563 AnSocEmb -,3487 ,1571 -2,2194 ,0281 -,6594 -,0381 MessPosi -,0380 ,2811 -,1351 ,8927 -,5938 ,5178 ******************** DIRECT AND INDIRECT EFFECTS ************************* Direct effect of X on Y
Effect SE t p LLCI ULCI -,0380 ,2811 -,1351 ,8927 -,5938 ,5178 Indirect effect of X on Y
Effect Boot SE BootLLCI BootULCI AnSocEmb ,0802 ,0914 -,0245 ,3612
******************** ANALYSIS NOTES AND WARNINGS ************************* Number of bootstrap samples for bias corrected bootstrap confidence intervals: 1000
Level of confidence for all confidence intervals in output: 95,00
