Elgaaied Leila * Université de Cergy-Pontoise Thema (UMR CNRS 8184),

Transcription

1 MAKING GREEN LOOK MAINSTREAM: TRIGGERING DESCRIPTIVE SOCIAL NORM TO PROMOTE PRO-ENVIRONMENTAL BEHAVIOR. THE ROLE OF AD CREDIBILITY AND ENDORSEMENT Elgaaied Leila * Université de Cergy-Pontoise Thema (UMR CNRS 8184), leila.elgaaied@u-cergy.fr Monnot Elisa Université de Cergy-Pontoise Thema (UMR CNRS 8184), elisa.monnot@u-cergy.fr Reniou Fanny Université Paris-Est Créteil IRG (EA 2354), fanny.reniou@u-pec.fr * Université de Cergy-Pontoise 33 Boulevard du Port, Cergy-Pontoise Cedex, France, Tél : / Fax : Abstract: Based on the observation that descriptive norm appeals are only effective when they are framed positively, our research investigates the efficiency of such norms in a context where the prevalent behavior is not environmentally friendly. Using an informercial promoting purchasing of non-overpackaged products, three experiments served to manipulate the presence and valence of a descriptive social norm, an endorsers presence and his profile. Our findings suggest that triggering a positive descriptive norm might result in the adoption of pro-environmental behavior, even though it does not reflect reality, provided that the ad is perceived as credible. Ad credibility influences intention to avoid overpackaging and is in turn influenced by descriptive norm and celebrity endorser s profile. The presence of a typical-consumer testimonial however influences behavioral intention directly while the profile of this non-famous endorser influences ad credibility and intention through identification. Keywords: Overpackaging ; social norms ; ad credibility FAIRE DE L ECOLOGIE UN STANDARD: ACTIVER UNE NORME SOCIALE DESCRIPTIVE POUR ENCOURAGER LES COMPORTEMENTS ECOLOGIQUES. ROLE DE LA CREDIBILITE DU MESSAGE ET DE L ENDOSSEMENT Résumé : Partant du constat que le recours aux normes descriptives est efficace seulement quand elles sont orientées positivement, notre recherche examine l efficacité de ces normes dans un contexte où le comportement le plus courant n est pas pro-environnemental. A l aide d une publi-information encourageant l achat de produits non-suremballés, trois expérimentations qui manipulent la présence et la valence de la norme sociale descriptive, la présence d un endosseur et son profil ont été conduites. Nos résultats suggèrent que le fait d utiliser une norme descriptive positive peut conduire à l adoption du comportement écologique même si celui-ci ne reflète pas la réalité, à condition que le message soit perçu comme crédible. La crédibilité du message influence l intention d éviter les suremballages et est, à son tour, influencée par la norme descriptive et le profil de la célébrité qui endosse le message. Cependant, la présence du témoignage d un consommateur lambda influence directement l intention de comportement alors que le profil de cet endosseur inconnu influence la crédibilité du message et l intention par le biais d un processus d identification. Mots clef : Suremballages ; normes sociales ; crédibilité du message This research is included in Movida research programme of French Ministry of Sustainable Development.

2 MAKING GREEN LOOK MAINSTREAM: TRIGGERING DESCRIPTIVE SOCIAL NORM TO PROMOTE PRO-ENVIRONMENTAL BEHAVIOR. THE ROLE OF AD CREDIBILITY AND ENDORSEMENT Introduction Among relevant strategies to encourage green behavior, previous research highlights the crucial role of social norms (Schultz, 1999; Goldstein, Cialdini and Griskevicius, 2008; Melnyk et al., 2013). However, in order to reach optimal effects, public service advertising should activate both facets of social norms, namely injunctive and descriptive norms, referring respectively to what most people approve of and to what people typically do (Cialdini, Reno et Kallgren, 1990). While injunctive norms are usually well understood and used appropriately, descriptive norms are sometimes activated inadequately which may lead to counter-productive effects. Cialdini (2003, p.105) particularly deplores advertisers tendency to try to mobilize action against a problem by depicting it as regrettably frequent. In France, for instance, the government has been promoting waste reduction by explaining that every French citizen currently generates an average of 590 kilos of waste every year. Even though these assertions may be accurate, they implicitly send the message that many people are actually adopting these harmful behaviors, indirectly suggesting that it is acceptable to produce a large amount of waste. Based on this observation, Cialdini (2003) concludes that priming descriptive norms is only effective when the pro-environmental behavior at stake is predominant. But paradoxically, public service ads are less critical for those pro-environmental practices that are already adopted by the majority. Governments are logically more prone to communicate in order to raise awareness about environmentally friendly actions that are not yet widespread. Our general proposition is the following: when environmentally harmful behavior is prevalent, the use of a positive descriptive norm will have a negative effect on perceived ad credibility as it does not reflect real behaviors. Nevertheless, this negative effect on ad credibility can be overcome through certain execution techniques reinforcing the weight of the descriptive norm, such as an endorser s presence. In this research, we chose to focus on purchasing of nonoverpackaged products. Whereas recycling is becoming a widespread practice in many European countries, consumers are not sufficiently familiar with pre-cycling strategies to reduce household waste. Conceptual Framework Impact of social norms on pro-environmental behavior. Extant literature shows that social norms have a positive impact on sustainable behavior (Cialdini, Reno and Kallgren, 1990; Goldstein, Cialdini and Griskevicius, 2008). According to Cialdini, Reno and Kallgren (1990), social norms can either be injunctive or descriptive. These two dimensions have distinctive effects in the decision-making process. Injunctive norms have a stronger impact on attitudes while descriptive norms are a stronger predictor of behavior (Melnyk, van Herpen and van Trijp, 2010). This is probably because descriptive norms require a simpler cognitive assessment as they already reflect the behavior of other individuals, which leads consumers to comply more easily with such norms through simple imitation (Cialdini, 2003; Melnyk, van Herpen and van Trijp, 2010). This research mainly focuses on descriptive norms as we study the impact of social norms when the environmentally beneficial behavior is not prevalent and because they are likely to backfire to produce unwanted effects. Previous work highlights the crucial role of ad credibility (i.e., consumers perception that the claims in the ad are true and credible) on attitude formation and behavior (MacKenzie and Lutz, 1989; Cotte, Coulter and Moore, 2005). The impact of ad credibility on persuasion has particularly been investigated in the context of pro-environmental behavior (Manrai et al., 1997; Newell, Goldsmith and Banzhaf, 1998). Given that consumers are increasingly skeptical 1

3 about environmental allegations, ad credibility indeed emerges as an important indicator in this context. An ad perceived as misleading generates a negative attitude and low intention to purchase a green product (Newell, Goldsmith and Banzhaf, 1998). In line with previous works, we expect that ad credibility mediates the effect of descriptive social norms on intention to purchase non-overpackaged products. By stressing that the majority already adopts the promoted behavior, descriptive norms contribute to enhancing ad credibility. According to Cialdini (2003, p.108) norm-based persuasive communications are likely to have their best effects when communicators align descriptive and injunctive normative messages to work in tandem rather than in competition with one another. In other words, to enhance ad effectiveness, injunctive and descriptive norms must both be framed positively, i.e. in favor of the promoted behavior. If the beneficial behavior is not prevalent, we expect that a positive descriptive norm will have a negative effect on ad credibility. Furthermore, we expect that the impact of the descriptive norm on behavioral intention through ad credibility will differ depending on whether the activated descriptive norm is positive or negative. Impact of endorsement on ad credibility and pro-environmental behavior. In order to determine how to strengthen ad perceived credibility while activating a positive social norm, we propose to examine the potential effect of endorsement. According to Melnyk, van Herpen and van Trijp (2010, p.463) norms may be more relevant, and hence more influential, when these come from persons with whom the consumer can easily identify. Identification theory posits that individuals tend to assess their level of similarity with others and to identify with people who are viewed as similar to themselves (Feick and Higie, 1992). The feeling of familiarity induced by this type of endorser has a positive impact on persuasion. We expect a positive effect of a positive descriptive norm relying on an endorser who is perceived as similar to the receiver. The impact of celebrity endorsement has been extensively studied. Several studies showed that the effectiveness of celebrity endorsement relies on the match-up between the endorser and the message (Kamins, 1990; Till and Busler, 2000). In the specific case of a nonprofit context, Wheeler (2009) showed that connection between the celebrity and a non-profit organization is an important variable to consider. A celebrity who fits well with the organization will be considered as more credible than a non-connected celebrity or a connected average person. We expect that the presence of a celebrity connected to the environmental issue will influence ad credibility and intention to purchase non-overpackaged products. Study 1: Impact of Presence and Valence of Descriptive Norm Method. A quasi-experimental design manipulating the presence and the valence of the social norm was conducted. We created an infomercial inspired by existing ads by the French ministry of ecology and sustainable development. At the top of the page, the infomercial displays a few lines priming (positive or negative) descriptive social norm and urges people to stop purchasing overpackaged products. At the left bottom corner of the page, an informative section explains the difference between secondary packaging and overpackaging. The sample consists of 145 respondents equally distributed among the three experimental conditions (absence of social norm, presence of positive social norm, presence of negative social norm). The total sample consists of 58.6% of males and 41.4% of females. The average age is 45.3 (SD=13.4). Ad credibility was measured with three items (Cronbach s alpha=0.934). Intention to avoid overpackaged products was measured with one item. Environmental concern (used as a control variable) was measured with a 6-item scale (Cronbach s alpha=0.898). Results. Manipulation checks showed that participants correctly perceived the presence and valence of the social norm. When the norm is activated, the advertisement was indeed 2

4 more positively associated with a description of the practices and habits of French people in terms of waste management (t=-2.857; p<0.01). As expected, statements relating to the decrease of household waste got significantly higher scores when the ad is displaying a positive (vs. negative) norm (t=-7.862; p<0.001). Our results show that perceived ad credibility varies significantly across our experimental conditions (F=4.31; p=0.015). More specifically, multiple comparison tests show that when the norm is positive, ad credibility is lower than when the norm is negative (mean difference=-0.61; p=0.027), but it is also significantly lower than when the norm is not primed at all (mean difference=-0.58; p=0.039). A moderated mediation test was conducted using the PROCESS macro (model 7). Our results confirm our moderated mediation hypothesis (Index of moderated mediation: b=.14; SE=.07; 95%CI=.027;.314) and reveal that the indirect effect of the descriptive social norm on intention to purchase non-overpackaged products through ad credibility is only significant when the norm is positive (conditional indirect effect when norm is negative: 95%CI= -.057;.082; conditional indirect effect when norm is positive =.048;.301). Study 2: Impact of Endorser and Identification on Ad Credibility and Intention Method. We manipulated the presence (vs. absence) of an endorser in the ad and his similarity (vs. dissimilarity) with the receiver in terms of age. We used the same infomercial as in Study 1 and included a testimonial, with a picture of a person and a quotation saying: I try more and more to be careful about what I buy. Whenever I can, I avoid overpackaged products ). Below the picture, the name and age of the endorser were provided. In order to manipulate endorser s similarity, Study 2 was exclusively conducted among a sample of young adults (n=173). One of the endorsers was presented as a 22-year-old student (similarity condition) and the other as a 65-year-old retired person (dissimilarity condition). The sample consists of 50.3% of females with an average age of 22.1 years (SD=2.13). Ad credibility and intention to purchase non-overpackaged products were measured with the same scales as in Study 1. Identification with the endorser was measured with a 6-item scale (Cronbach s alpha: 0.926). Results. Manipulation checks reveal that the presence of an endorser is perceived correctly (t=-9.58; p<0.000). As expected, the perceived similarity with the endorser is also significantly higher when the latter is a 22 year-old student rather than a 65 year-old retired person (t=-13.02; p<0.000). Our results show that neither the presence of an endorser nor his profile seems to have a direct effect on ad credibility (respectively t=-0.374; p=0.709 and t= ; p=0.869). The presence of an endorser, however, influences positively intention to purchase non-overpackaged products (-2,647; p=0.009) while the profile of the endorser has no direct effect on intention (t=-1,476; p=0.143). An analysis of mediation using PROCESS confirms that the presence of an endorser does not influence intention through ad credibility (indirect effect=0.024; 95%CI=-0.078, 0.187), but rather directly (direct effect=0.615; 95%CI=0.154, 1.077). As for the profile of the endorser (similar vs. dissimilar), a second mediation analysis shows that it does affect ad credibility but only indirectly through the activation of identification with the endorser (indirect effect=0.458; 95%CI=0.230, 0.795). The profile of the endorser also indirectly influences intention through identification (indirect effect=0. 465; 95%CI=0. 164, 0.873). Study 3: Impact of Celebrity Endorsement on Ad Credibility and Intention Method. Using the same ad as in Study 1 and 2, we compared three conditions varying in terms of endorser s profile: (1) non-famous endorser; (2) famous endorser whose image is connected to the environmental issue; (3) famous endorser with no connection to environmental issue. Since data collection took place in a French context, we chose two 3

5 equally popular French public figures: a well-known environmental activist (Nicolas Hulot) who best embodies the values of environmentalism in France and an actor (Jean Dujardin) who also benefits from a very positive image, but has no connection to the environmental issue. Both endorsers were perceived as equally famous (respectively M=6.56 and 6.52; t=-0.187; p=0.852). For the non-famous endorser condition, we chose an ordinary male of approximately the same age as the two celebrities. The total sample consists of 159 participants, 45.9% of them are females. The average age is 44.1 (SD=12.66). Results. Our results show that perceived credibility of the endorser varies strongly among our three experimental conditions (F=26.55; p=0.000). With no surprise, Nicolas Hulot, the famous pro-environmental activist, is perceived as more credible than both the other endorsers (respectively mean difference=1.47; p=0.000 and 1.72; p=0.000). However, we observe no significant difference between Jean Dujardin s credibility and that of our non-famous endorser (mean difference=-0.25; p=0.600). Similarly, ad credibility also varies significantly among the groups (F=3.66; p=0.028). Multiple comparison tests show that the only significant difference is when Nicolas Hulot testifies vs. Jean Dujardin (mean difference: 0.58; p=0.021). Our results also reveal that celebrity endorsement in itself is neither a significant predictor of ad credibility (t=-0.166; p=868) nor of behavioral intention (-0.503; p=0.616). However, as mentioned earlier, the profile of the celebrity does affect ad credibility significantly. A mediation analysis (using as in previous studies age, gender and environmental concern as covariates) shows that the direct impact of the profile of the celebrity on intention to purchase non-overpackaged products is not significant (direct effect=0.026, 95%CI=-0.473; 0.526) while its indirect effect through ad credibility is significant (indirect effect=.326; 95%CI=0.020; 0.680). General Discussion This research raises the question of the efficiency of descriptive norms in a context where the prevalent behavior is not environmentally friendly. Our main findings are the following: (1) the presence of a descriptive norm influences intention to purchase non-overpackaged products indirectly, through ad credibility, and this process is moderated by the valence of the descriptive norm; (2) the presence of an endorser influences intention directly; (3) the profile of the endorser (young vs. old) influences intention indirectly through identification; (4) celebrity endorsement is neither a significant predictor of ad credibility nor of behavioral intention while the profile of the celebrity does affect ad credibility. Our major contribution lies in the connection made between descriptive norm and ad credibility. Testing the effect of endorsement in the case of pro-environmental behaviors is another contribution of this work. Very few studies investigated the efficacy of this advertising technique to promote ecological messages or its effects on ad credibility. In terms of limitations, this research suffers from a lack of external validity as it focuses on the French context and a particular type of behavior, namely purchasing of non-overpackaged products. Future research could replicate our experiments in other cultural contexts and proenvironmental behaviors. Besides, this research does not take into account several potentially relevant variables such as consumers level of susceptibility to normative influence. From a practical viewpoint, public policy makers should first bear in mind that the strategy consisting in deploring the prevalence of a harmful behavior in order to promote the opposite behavior is not efficient. Descriptive norms should always be framed positively in order to be consistent with injunctive norms. Second, campaign designers wishing to trigger descriptive norms in a context where environmentally friendly behavior is not the mainstream should be careful about ad credibility, which can be enhanced through an endorser. However, resorting to a celebrity endorser to encourage pro-environmental behavior is not an end in itself as shown by Wheeler (2009) in a non-profit context. In order to promote green behavior, public 4

6 service announcements can rely on famous endorsers provided that their image is associated with environmental preservation. Using a celebrity who has no connection to environmental issues generates the same results as using an anonymous endorser. Finally, our research raises an important ethical concern. If an environmentally beneficial behavior is not massively adopted, it is morally legitimate to question the approach that consists in suggesting that it is actually a social norm. One may consider that the end justifies the means and that this technique is tolerable as long as it is used to foster environmental preservation. However, one possible solution to activate descriptive norms without resorting to deceitful claims would be to stress on positive aspects that are actually true while avoiding emphasizing the negative one. References Cialdini RB (2003) Crafting normative messages to protect the environment. Current Directions in Psychological Science 12 (4): Cialdini RB, Reno RR and Kallgren CA (1990) A focus theory of normative conduct: recycling the concept of norms to reduce littering in public places. Journal of Personality and Social Psychology 58(2): Cotte J, Coulter RA and Moore M (2005) Enhancing or disrupting guilt: the role of ad credibility and perceived manipulative intent. Journal of Business Research 58(3): Feick L and Higie RA (1992) The effects of preference heterogeneity and source characteristics on ad processing and judgments about endorsers. Journal of Advertising 21(2): Goldstein NJ, Cialdini RB and Griskevicius V (2008) A Room with a viewpoint: using social norms to motivate environmental conservation in hotels. Journal of Consumer Research 35(3): Kamins MA (1990) An investigation into the match-up hypothesis in celebrity advertising: when beauty may be only skin deep. Journal of Advertising 19(1): MacKenzie SB and Lutz RJ (1989) An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. Journal of Marketing 53(2): Manrai LA, Manrai AK, Lascu DN and Ryans JK (1997) How green-claim strength and country disposition affect product evaluation and company image. Psychology and Marketing 14(5): Melnyk V, van Herpen E and van Trijp HCM (2010) The influence of social norms in consumer decision making: a meta-analysis. Advances in Consumer Research 37: Melnyk V, van Herpen E, Fischer ARH and van Trijp HCM (2013) Regulatory fit effects for injunctive versus descriptive social norms: evidence from the promotion of sustainable products. Marketing Letters 24(2): Newell SJ, Goldsmith RE and Banzhaf EJ (1998) The effects of misleading environmental claims on consumer perceptions of advertising. Journal of Marketing Theory and Practice 6(2): Schultz PW (1999) Changing behavior with normative feedback interventions: a field experiment on curbside recycling. Basic and Applied Social Psychology 21(1): Till BD and Busler M (2000) The match-up hypothesis: physical attractiveness, expertise, and the role of fit on brand attitude, purchase intent and brand beliefs. Journal of Advertising 29(3): Wheeler RT (2009) Nonprofit advertising: impact of celebrity connection, involvement and gender on source credibility and intention to volunteer time or donate money. Journal of Nonprofit 5

7 & Public Sector Marketing 21(1):