Comparing the effects of greenwashing claims in environmental airline advertising: perceived greenwashing, brand evaluation, and flight shame

Abstract

Drawing on information process theories, this study investigated whether consumers have the ability to perceive greenwashing in vague and false greenwashing claims, as well as in abstract and concrete compensation greenwashing claims. Moreover, we examined the moderating role of topical environmental knowledge. We also looked at the effects of perceived greenwashing on brand evaluation and flight shame. Findings of an experimental study with a quota-based sample (N = 658) indicate that only concrete compensation claims do not significantly enhance greenwashing perceptions. However, when consumers have a high topical environmental knowledge, they are able to discover greenwashing in concrete compensation claims as well. Once greenwashing perceptions are triggered, they harm brand evaluations and foster flight shame. Implications for research on greenwashing and conclusions for practitioners are discussed.

Keywords:

• Green advertising
• perceived greenwashing
• environmental knowledge
• experiment
• flying
• information processing
• flight shame

As one of the leading figures, the Swedish climate activist Greta Thunberg is stressing the moral responsibility of each individual to approach climate change. Especially, she empathizes the need to restrain from air travel as an individuals’ moral duty (Gössling, Humpe, and Bausch 2020). However, consumers may see part of this responsibility also on the side of big companies. Especially companies offering environmentally inherently unfriendly products or services like airlines have noticed the normative pressure in society, and as a consequence, they have reacted with green advertising (Mayer, Ryley, and Gillingwater 2014). Among various companies offering green products or services, the airline sector represents a special case for advertising due to the ‘flyers’ dilemma’ (Young, Higham, and Reis 2014). Although consumers are becoming more environmentally conscious, flying still remains very desirable (Hibbert et al. 2013). Green airline advertising alludes directly to the dilemma by suggesting that it can be solved.

At the same time, misleading green advertising strategies, so called greenwashing strategies, are increasingly used by airline companies (e.g., Naderer, Schmuck, and Matthes 2017), for instance, when communicating their pro-environmental performance while camouflaging the harmful environmental character of flying (De Freitas Netto et al. 2020). Generally, greenwashing may prevent social change in society that is needed to avert the environmental crisis (Kilbourne et al. 1995). Although consumers are willing to turn to more environmentally friendly consumption behaviors, they have difficulties to identify sustainable products or services because of unclear information (BEUC 2020). Hence, greenwashing could impede true green consumerism (Carlson et al. 1996, 57).

Knowledge about the effects of greenwashing in advertising is limited (Matthes 2019; Naderer, Schmuck, and Matthes 2017). Previous research has shown that consumers’ perceptions of greenwashing vary depending on the advertising claim as well as on consumers’ environmental knowledge (e.g., Parguel, Benoit-Moreau, and Russell 2015; Schmuck, Matthes, and Naderer 2018). So far, studies have only examined false claims, vague claims, and executional greenwashing (e.g., Schmuck, Matthes, and Naderer 2018). Recently, an increasing number of airlines use omission claims. This is done by advertising green compensation measures for flying leaving out important information that is needed for consumers to ‘evaluate its truthfulness or reasonableness’ of the compensation (Kangun, Carlson, and Grove 1991, 51). Compensation claims could be very persuasive as they may evoke or confirm consumers’ compensatory green beliefs (Hope et al. 2018). As a consequence, such claims may confuse consumers as to whether the offered measures can actually compensate for flying (Polonsky, Grau, and Garma 2010, 51).

There are three key research gaps. First, research on compensation claims in the greenwashing context has been neglected so far (Matthes 2019). Second, we generally lack experimental research on greenwashing when it comes to inherently environmentally harmful products and services such as airlines. Third, emotions caused by perceived greenwashing have been largely ignored until now (Koenig-Lewis et al. 2014; but see Szabo and Webster 2021). Thus, in this paper, we experimentally examine consumers’ perceptions of greenwashing in response to vague and false greenwashing claims, as well as in abstract and concrete compensation claims. Moreover, we look at the moderating role of topical environmental knowledge.

Green advertising and greenwashing claims

Greenwashing is defined as a process in which companies stress the environmental friendliness of a service or product, while withholding environmentally unfriendly aspects (e.g., Naderer, Schmuck, and Matthes 2017). While some content analyses have shown that the majority of companies’ green claims contain greenwashing elements (Carlson et al. 1996; Carlson, Grove, and Kangun 1993; Kangun, Carlson, and Grove 1991), others have reported that only less of the half of green ads in magazines and newspapers can be categorized as greenwashing (e.g., Segev, Fernandes, and Hong 2016). However, there is agreement that greenwashing is not uncommon.

According to Carlson, Grove, and Kangun (1993) there are five types of green ads: (1) product, (2) process, (3) image, (4) environmental fact, and a (5) combination. A content analysis showed that claims endorsing a green image of the company were the most used claims followed by claims promoting green attributes of a product or service. Regarding claim deceptiveness, Kangun, Carlson, and Grove (1991) distinguished (1) acceptable claims, (2) vague claims, (3) false claims, (4) claims possessing omissions, and (5) a combination of these claims. Acceptable claims represent substantial advertising messages which are true, do not mislead, and do not deceive. Vague claims are perceived as overly ambiguous and as too unspecific in order to make a well-informed conclusion about the green character of a product or service. Whereas false claims include an outright lie, claims possessing omissions are defined as claims that exclude important information that is needed by consumers to draw conclusions about the ‘truthfulness or reasonableness’ of the claim (Carlson, Grove, and Kangun 1993, 31).

As green airline advertising focuses simultaneously on both, environmentally friendly attributes of the service of flying and an eco-image of the airline, a combination of product and image claims is investigated in this study (e.g., Carlson, Grove, and Kangun 1993). Regarding deceptiveness, vague claims promoting flying as a way to protect the planet (e.g., VietJet 2020), and as ‘the greenest flight’ (e.g., Frontier Airlines 2019) without giving a reason or proposing measures are often used in airline advertising (Carlson, Grove, and Kangun 1993, 31). Further, claims promoting an airline as the company with the lowest carbon emissions, when in reality, this is not correct, or offering a full elimination of flight emissions when flying with the promoted airline, are common examples for false claims (e.g., Ryan Air 2020). Lastly, omission claims promoting environmental protection measures as compensation for the environmentally harmful services they are offering are increasingly used by airlines (e.g., Air New Zealand 2019). In this paper, we refer to these claims as compensation claims (Polonsky, Garma, and Grau 2011). Typically, compensation claims are signaling consumers that the CO2 produced by flying can be compensated by other environmentally friendly measures. However, information about how much carbon one flight produce and how much carbon can be compensated with the promoted measures is not disclosed in these ads. Moreover, information about what the carbon compensation measures cover is often omitted. As a consequence, it is difficult for consumers to make a responsible consumption decision (e.g., Polonsky, Grau, and Garma 2010).

In contrast to vague claims, compensation claims are proposing environmental measures and thus, give reason for airlines to claim to be environmentally friendly. Further, whereas false claims present wrong knowledge such as ouright lies, compensation claims are not per se wrong or outright lies. Instead, they omit important information that is needed for consumers to see the true environmental benefit. To sum it up, compensation claims are neither vague, nor false, rather they are contextualized as omission claims (see Appendix A, Table A1).

Derived from the green airline advertising practice, we distinguish between abstract and concrete compensation claims. Abstract compensation claims are proposing future environmental compensation for the environmental impact of flying, without consumers being able to directly observe the trade-off. For instance, various airlines (e.g., Air New Zealand 2019) offer to invest in tree planting, environmental projects, or environmental research. As the compensation measure is not observable directly on board, but in the future, there exists only an abstract and non-verifiable evidence for the green compensation. Concrete compensation claims are promoting an immediate compensation for the environmental impact of flying. The compensation measure can be directly experienced by consumers and ‘seen’ while using the product or service. Various airlines (e.g., HiFly 2019) state that they are fully plastic free on board or that they recycle all of the used cutlery on bord by recollecting it. This is an observable and verifiable evidence of green compensation, clearly visible to consumers.

Although these claims may be literally true, they could imply a false impression of the possibility of overall sustainable flying and could be easily misunderstood by consumers (Majoras 2008). Compensation claims may not only lead to consumers’ overestimation of an environmental benefit, they also neglect important information. They solely focus on the ecological benefits of flying and thus, may distract consumers from the negative environmental consequences. Therefore, they can be categorized as omission claims and thus, as potentially greenwashing (Majoras 2008; Polonsky, Grau, and Garma 2010).

Additionally, Kangun, Carlson, and Grove (1991) showed that there is also a combination of deceptive claims. Transferred to the flight context, this would mean that airlines make statements that consist of various deceptive elements (e.g., KLM 2019). Finally, image-based emotional appeals, such as pleasant nature imagery, are used to create an environmentally friendly image. This strategy is called executional greenwashing and is widely used in the airline context by showing nature from a bird’s eye view (e.g., KLM 2019).

A rising body of literature has investigated the effects of greenwashing (e.g., Nyilasy, Gangadharbatla, and Paladino 2014; Parguel, Benoit-Moreau, and Russell 2015). These studies, mostly experiments, have shown that the perception of greenwashing is higher when a false claim was presented in comparison to a substantial claim, a vague claim, or an executional greenwashing claim (e.g., Nyilasy, Gangadharbatla, and Paladino 2014; Parguel, Benoit-Moreau, and Russell 2015, Schmuck, Matthes, and Naderer 2018). However, previous research has focused only on a few greenwashing strategies, neglecting other prominent types like omission claims in form of widely used compensation claims (Matthes 2019).

Perceived schema incongruity in green advertising claims for airlines

Schema Incongruity Processing Theory (Mandler 1982) suggests that if new information does not fit in existing mental schemata, attention is directed toward it and the intensity of information processing is heightened (Baker and Petty 1994; Homer and Kahle 1986). For example, independently of individuals’ cognitive effort, their characteristics, and situational influences, unexpected combinations of unrelated objects in advertising motivate individuals to more attentive information processing. Because such ads do not fit into existing mental schemata of consumers, more cognitive effort is used to understand the paradox information (Homer and Kahle 1986).

Due to a change in moral and social norms over the past decades, nowadays, flying is seen as environmentally harmful and socially undesirable (Gössling, Humpe, and Bausch 2020). Hence, when individuals see airline advertising, the environmentally harmful character of flying may be immediately salient. Thus, based on Schema Incongruity Processing Theory, airline advertising promoting the environmental friendliness may be perceived by consumers as unexpected, out-of-context, and inadequate, because it is not congruent with their existing mental structures. Thus, attention is increased and individuals allocate more cognitive resources to the ad. As has been shown, cognitive effort is one of the key preconditions to detect greenwashing (Schmuck, Matthes, and Naderer 2018). Hence, individuals may perceive the companies’ ‘twofolded behavior’ to promote an environmentally friendly image for an inherently environmentally unfriendly service (De Freitas Netto et al. 2020, 6). This mechanism can be theorized to occur for all types of greenwashing claims, because for all claims, the twofolded behavior is clearly apparent. This leads to our first hypothesis:

H1: If (a) an abstract compensation, (b) a concrete compensation, (c) a vague, and (d) a false greenwashing claim is presented in an airline ad, perceived greenwashing will be higher compared to an ad presenting no green claims.

However, the strength of the effect described in our H1 may differ depending on the type of the greenwashing claim. Generally, the use of the compensation logic could be a promising advertising strategy for airlines. In order to achieve a balance between environmentally harmful behavior and environmental protection measures, individuals might use the cognitive strategy of compensatory green beliefs (Hope et al. 2018). Compensatory green beliefs can be defined as individuals’ convictions that pro-environmental behavior can offset harmful environmental behavior. For instance, individuals with strong compensatory green beliefs may believe that if they engage in recycling, it is justified to travel by airplane, because through recycling in the past the CO2 emissions from the flight are neutralized (Hope et al. 2018). For airline advertising, we argue, that green compensation claims may evoke or confirm consumers’ compensatory green beliefs. Even if compensation claims omit important information and thus, consumers are not able to estimate the real compensation that is achieved with the proposed environmental measures, consumers may accept the advertising claim due to compensatory green beliefs (Hope et al. 2018). For vague and false claims, no such mechanisms can be expected. In other words, the incongruity between the claim and the awareness that flying harms the environment should be larger for vague and false claims, as compared to compensation claims. This leads to the following hypotheses:

H2: When (a) a vague or (b) a false claim is presented in an airline ad, perceived greenwashing will be higher compared to abstract and concrete compensation claims.

However, also abstract and concretecompensation claims may differ in terms of perceived greenwashing. Since the prerequisite for compensatory green beliefs is that individuals believe in the effectiveness of the measures, the concept of psychological distance is relevant. Construal-Level Theory of Psychological Distance (Liberman and Trope 1998) states that perceived psychological distance of an event has an effect on individuals’ judgements. For instance, moving into a new apartment next year is connoted as ‘starting a new life’ whereas moving tomorrow is perceived as ‘packing and carrying boxes’ (Liberman and Trope 1998, 8). In other words, events with a high temporal distance are perceived as more abstract than events with a low temporal distance and consequently may have lower psychological influence (Liberman and Trope 1998). In line with this, concrete (green) messages turn out to be more effective in positively influencing consumers’ perceptions, brand attitudes, purchase intentions and environmental behavior compared to abstract messages (e.g., Han et al. 2019). To sum it up, abstract messages with a focus on distant events are less persuasive than concrete messages focusing on immediate events (Zhao et al. 2015). We thus assume that consumers can better imagine the realization of concrete compensation messages due to smaller psychological distance than the realization of abstract compensation claims. Hence, concrete compensation claims may evoke consumers’ compensatory green beliefs to a larger extent than abstract compensation claims (Hope et al. 2018). We assume that, by activating compensatory green beliefs, the motivation to discover greenwashing is lower for concrete compensation claims than for abstract ones. Therefore, we derive our third hypothesis as following:

H3: When an abstract compensation claim is presented in an airline ad, perceived greenwashing will be higher compared to an ad presenting a concrete compensation claim.

The moderating role of environmental knowledge

According to the Elaboration Likelihood Model (ELM; Petty and Cacioppo 1986), involvement, that is the degree of ‘personal relevance of the topic’, determines the motivation to engage in elaboration of media content (O’Keefe 2013, 1). Further, the ELM suggests that attitudes can be formed either via the ‘central route’ in case of high involvement, or, for low involved consumers, via the ‘peripheral route’ (Petty and Cacioppo 1986, 125). When information is processed via the central route, the quality of the arguments used in a persuasive message are closely scrutinized by individuals and other relevant material is included in the assessment of the messages’ credibility. In contrast, when information is processed via the peripheral route, individuals rather focus on peripheral cues.

When it comes to green advertising, the major indicator for consumer involvement is topical environmental knowledge (Parguel, Benoit-Moreau, and Russell 2015). Based on the ELM, topical environmental knowledge about the harmful consequences of flying increases involvement with environmental topics (O’Keefe 2013). Consequently, the motivation to engage in issue-relevant, extensive thinking about green airline advertising claims also increases. In the case of greenwashing, consumers scrutiny due to topical environmental knowledge could uncover the misleading attempt. To sum it up, topical knowledge could help to judge the trustworthiness of advertising and hence could (negatively) affect the ‘persuasive success’ of the green advertising claim (O’Keefe 2013, 3). However, only a few experimental studies have investigated the role of topical environmental knowledge (e.g., Parguel, Benoit-Moreau, and Russell 2015). They have found that topical environmental knowledge helps consumers to perceive greenwashing at least in false claims (Schmuck, Matthes, and Naderer 2018). We therefore assume that consumers with high topical environmental knowledge are better able to detect greenwashing compared to consumers low in topical environmental knowledge. High topical knowledge leads to a ‘thoughtful’ evaluation of the claims (Petty and Cacioppo 1986, 125). Such thoughtful elaboration can be argued to increase the perceived discrepancy between the claim and the fact that flying is environmentally unfriendly. Thus:

H4: When compared to an ad with no green claims, the effect of an (a) abstract compensation claim, (b) concrete compensation claim, (c) vague claim, and (d) false claim on perceived greenwashing is higher for respondents high in topical environmental knowledge as compared to less knowledgeable respondents.

Consumers’ cognitive and affective responses

Generally, green advertising has positive effects on ad attitude (e.g., Matthes, Wonneberger, and Schmuck 2014), brand evaluation (e.g., Hartmann, Apaolaza Ibáñez, and Forcada Sainz 2005), and purchase intention (e.g., Schmuck, Matthes, and Naderer 2018). Therefore, we derive our fifth hypothesis:

H5: If (a) an abstract compensation, (b) a concretecompensation, (c) a vague, and (d) a false greenwashing claim is presented in an airline ad, brand evaluation will be increased compared to an ad presenting no green claims.

Further, a number of studies have found that purchasing green products or services is perceived by consumers as morally important and makes them feel good about themselves (e.g., Meijers et al. 2019). Consumers who purchase green products or services are perceived by individuals as more positive than consumers purchasing non-green products or services (Mazar and Zhong 2010). In the context of airline advertising, this would mean, that green advertising promoting green flights may evoke a ‘warm glow’ and give confirmation to consumers’ views about themselves as being morally good persons when purchasing the advertised service (Hartmann, Apaolaza Ibáñez, and Forcada Sainz 2005, 11). As flying is generally recognized as environmentally harmful and thus, morally reprehensible, green airline advertising suggesting morally good behavior may reduce negative feelings of flight shame. Therefore, we derive our sixth hypothesis:

H6: If (a) an abstract compensation, (b) a concrete compensation, (c) a vague, and (d) a false greenwashing claim is presented in an airline ad, flight shame will be decreased compared to an ad presenting no green claims.

Drawing on the Theory of Reactance (e.g., Brehm 1966), we assume that, if consumers detect the misleading persuasion tactic of greenwashing, it is likely that they react with reactance to protect their freedom to choose a product or service. In the context of persuasive communication, reactance is defined as a psychological state that influences attitudes and behavior (e.g., Dillard and Shen 2005) and is seen as ‘an amalgam of anger and negative cognition’ (Dillard and Shen 2005, 164). Similarly, Rains (2013) showed that reactance can be understood as a ‘latent factor’ with affect and cognition as indicators to measure it (65). If we apply this to the context of greenwashing, where we understand perceptions of greenwashing as a threat to consumers’ freedom, it can be assumed that perceptions of greenwashing induce both, affect and cognitions. In other words, if greenwashing is perceived by consumers, they may react actively against the green advertising message with cognitive re-evaluations as well as affective responses (Koslow 2000). One primarily cognitive effect is a negative evaluation of the brand (Schmuck, Matthes, and Naderer 2018). Therefore:

H7: Perceived greenwashing decreases brand evaluation for the advertised brands.

Greenwashing may also affect emotions toward own behaviors. Such emotional responses were neglected in previous research (Koenig-Lewis et al. 2014; but see Szabo and Webster 2021). In light of current environmental social movements, one particular emotional reaction is especially relevant: flight shame – the negative feeling of ‘unease about the climate implication of air travel’ (Gössling, Humpe, and Bausch 2020, 1). That is, because of widely held pro-environmental beliefs (e.g., Matthes 2019) and the widespread knowledge about the negative environmental impact of flying, individuals find themselves in a flyers’ dilemma (Young, Higham, and Reis 2014). The resulting cognitive dissonance between the desire to travel and the negative impact of flying creates the negative affective reaction of shame. Rephrased, when greenwashing is recognized, it becomes apparent that airline companies want to hide or downplay the negative environmental impact of flying. As argued above, this leads to negative evaluations about airline companies. Yet, the activity of flying also has a positive dimension due to its association with, for instance, vacation and freedom, and therefore, individuals may not want to totally give up the activity of flying. The resulting psychological reaction may be shame. That is, there is a consciousness of a morally wrong behavior, prompted by greenwashing perceptions. In other words, we argue that perceived greenwashing may raise flight shame when individuals think about their own flying behaviors (e.g., Amatulli et al. 2019). Therefore, we suggest the following hypothesis:

H8: Perceived greenwashing increases flight shame.

Our full theoretical model is shown in Figure 1.

Figure 1. The proposed theoretical model.

Material and methods

We conducted an online experiment with a 2 × 5 between-subject design manipulating topical environmental knowledge (i.e., environmental facts about flying vs. non environmental facts) and the type of greenwashing claim (i.e., abstract compensation vs. concrete compensation vs. vague vs. false vs. control) in green ads of three different airlines (e.g., multi message design). To manipulate knowledge, participants of the study (N = 658) were first randomly assigned to either the environmental facts condition or the non-environmental facts condition. In both groups, they viewed a journalistic article from an existing online newspaper. Second, they were randomly assigned to an experimental greenwashing claim condition where we showed them three ads from three existing airline-brands (HiFly, s7 airlines, and SprintAir) or to the control group (i.e., no green claim).

Participants, stimuli, and procedure

We recruited participants (N = 658) from the panel of a professional market research company in December 2019. We worked with a quota-based sample representing German consumers with regard to age (Mage = 41.49, SD = 13.21), gender (50.3% women), and education (29.2% no education or lower-secondary education, 46.4% complete secondary education, 24.5% complete university education).

To manipulate knowledge, participants saw either an article with environmental facts about flying (e.g., environmental consequences of the growing aviation sector) or an article with information about robotics in the working world (control group; e.g., consequences of robots on the job market). While the content of the articles varied across the two conditions, the articles’ length and sentence constellation, as well as the newspapers’ design was identical. In a second step, fifteen versions of an ad were presented in the form of Twitter postings of three different airlines. As a context, Twitter was chosen because it is a frequently used platform for brand communication (e.g., Walasek, Bhatia, and Brown 2018). For each airline, we created five ads and prepared them to fit to one of the four experimental conditions as well as the control group (i.e., non-green claim promoting relaxed holidays). We used a multi message design including three ad versions of three existing but rather unknown airline companies per condition (Reeves, Yeykelis, and Cummings 2016). Since we placed a ‘major emphasis on external validity’ when operationalizing the different greenwashing claims in airline advertising (Slater 1991, 416), we asked European senior scholars (N = 6) in the field of green advertising to assess the different ads and categorize them as abstract compensation, concrete compensation, vague and false claims. We provided scholars with definitions of the claims and asked them to match the 12 ads to these definitions. The ads were presented to them in a random order. Results supported our operationalization of the advertising claims. On average, 8.83 ads out of 12 ads were correctly identified by the experts, suggesting acceptable internal validity of our stimuli.

The abstract compensation condition contained three ads promoting future environmental compensation. The concrete compensation condition included three ads proposing immediate compensation for the environmental impact of flying. Participants in the vague condition viewed three ads with ambiguous or very broad environmental claims (see Schmuck, Matthes, and Naderer 2018). The false condition comprised three ads based on an outright lie. Lastly, in the control condition, no references to environmental or green actions were made, the main focus of the ads was on holidays and relaxation, no green arguments were provided. We created the greenwashed ads drawing on existing ads that contained greenwashed elements. The ads’ layout, colors, and images were identical. In order to avoid strong prior attitudes, potential confounds, and to ensure high external validity, we decided to use small but real airline companies for our stimuli. The stimuli are shown in the Appendix A, Table A2.

Written informed consent was obtained, participants were randomly assigned to the conditions, and the ads were shown in random order followed by the questionnaire and debriefing.

Measures

We assessed topical environmental knowledge with six single-choice items. The items dealt with information about the environmental impact of flying, about flight statistics, and about the Paris climate agreement related to flying. All information to answer the questions correctly was mentioned in the environmental journalistic article. By summing up the correct answers, we created a seven-point scale (0 = no topical environmental knowledge, 6 = high topical environmental knowledge; M = 2.70, SD = 1.25). For all items, see Appendix A, Table A3.

We measured perceived greenwashing with three items adapted from Chen and Chang (2013) on a seven-point Likert scale ranging from ‘strongly disagree’ to ‘strongly agree’ (Cronbach’s α = .79; M = 4.78; SD = 1.61).

For brand evaluation, we built on three adjective pairs for each airline (e.g., Schmuck, Matthes, and Naderer 2018) on a seven-point semantic differential scale. Since (1) there is no theoretical basis to investigate differences between the brands, (2) we are solely interested in the effect across the brands (i.e., the overall effect), (3) a factor analysis using all items yielded one general factor for ‘brand evaluation’ (see Appendix B, Table B1 under OSF; https://osf.io/9wd7p/?view_only=c7afccc4586a469692c7ded6def85df2) and (4) all the brand evaluation variables were highly correlated (see Appendix B, Table B2 under OSF), we created one index consisting of nine items (Cronbach’s α = .97; M = 3.57, SD = 1.42).

Flight shame was measured with seven items, three of them adapted from the Internalized Shame Scale (ISS; Cook 1996). The items were measured on a seven-point scale ranging from ‘strongly disagree’ to ‘strongly agree’ (Cronbach’s α = .91; M = 2.63, SD = 1.49).

Randomization and manipulation check

Randomization and manipulation checks were successful. We detail these findings in Appendix B, Tables B3 and B4 under OSF.

Statistical model

We used PROCESS 3.5, Model 8, involving 5.000 bootstrap samples (Hayes 2018). We treated advertising claim as our independent, multifactorial (multicategorial) variable and used the control condition as our reference condition for analyzing H1, H2, and H4, H5, and H6. To test H3, we used the concrete compensation condition as a reference category.

As mediator, we inserted perceived greenwashing and as moderator, topical environmental knowledge using the topical environmental knowledge control condition as a reference condition. The two dependent variables were examined in two separate models (both Model 8). For the first analysis, we used PROCESS Model 8 (Model 8a), where we defined X as advertising claim, M as perceived greenwashing, W as topical environmental knowledge (dummy-coded), and Y as brand evaluation. For the second analysis (Model 8b), we proceeded in the same way, with the exception that we inserted flight shame as Y. In both models we additionally included the following covariates: age, gender, education, and flying frequency. We included these control variables because, theoretically, these variables could also explain our hypothesized effects as they have been found to be related to various environmental attitudes (Becker et al. 2016; Buttel 1979; Haytko and Matulich 2008; Van Liere and Dunlap 1980; Wright and Lynch 1995).

Additionally, we carried out a variance analysis with repeated measures to check if type of brand has an influence on our findings. Results indicated there is no interaction effect of perceived greenwashing and brand type (F(1.937, 1261.218) = 2.61, p = .075).

Results

We checked the effects of the causal mediation analysis separately for our two models (Model 8a, Model 8b) using the mediation package in R (Tingley et al. 2019). Results showed significant average causal mediated effects (ab = −0.16, 95%CI = [-0.28, −0.07], p < .001; ab’ = 0.18, 95%CI = [0.08, 0.30], p < .001) and mixed total effects (c = .23, 95%CI = [-0.09, 0.57], p = .161; c’ = .45, 95%CI = [0.09, 0.80], p = .012). We continued our moderated mediation analysis with both models justified by Hayes (2018), stating that ‘a single inferential test of the indirect effect is all that is needed’ (116). All effects of the causal mediation analysis can be found in the Appendix B, Table B5 under OSF.

Moderated mediation path a: effect of advertising claim on perceived greenwashing

We found that concrete compensation claims did not lead to a higher level of perceived greenwashing compared to the control condition (b = 0.33, LLCI = −0.21, ULCI = 0.86, t = 1.23, p = .218). However, the abstract compensation claim (b = 1.10, LLCI = 0.56, ULCI = 1.65, t = 3.97, p < .001), the vague claim (b = 1.38, LLCI = 0.84, ULCI = 1.91, t = 5.06, p < .001), and the false claim (b = 1.24, LLCI = 0.71, ULCI = 1.76, t = 4.62, p < .001) had significantly higher levels of perceived greenwashing than the control condition, confirming hypotheses 1a, 1c, and 1d. We found no support for hypothesis 2: Results showed that vague claims (M = 5.25, SD = 1.45) did have a higher level of perceived greenwashing compared to concrete compensation claims (b = 1.04, LLCI = 0.53, ULCI = 1.55, t = 4.01, p < .001; M = 4.55, SD = 1.51), but not compared to abstract compensation claims (b = 0.27, LLCI = −0.25, ULCI = 0.80, t = 1.01, p = .311; M = 4.98, SD = 1.42). Similarly, false claims (b = 0.90, LLCI = 0.40, ULCI = 1.41, t = 3.52, p = .001; M = 5.23, SD = 1.50) did have a higher level of perceived greenwashing compared to concrete compensation (M = 4.55, SD = 1.51), but not compared to abstract compensation (b = 0.13, LLCI = −0.39, ULCI = 0.65, t = 0.50, p = .615; M = 4.98, SD = 1.42). By and large, vague and false claims did not generally lead to more perceived greenwashing compared to compensation claims. In line with H3, when using the concrete compensation condition as a reference category, findings indicated that the abstract compensation claim (M = 4.98, SD = 1.42) significantly increased perceived greenwashing compared to the concrete compensation claim (b = 0.77, LLCI = 0.25, ULCI = 1.29, t = 2.91, p = .004; M = 4.55, SD = 1.51) as well as to the control group (b = 1.10, LLCI = 0.56, ULCI = 1.65, t = 3.97, p < .001; M = 3.86; SD = 1.76). Therefore, H3 is strongly supported. For descriptive statistics, please see Appendix B, Table B6 under OSF.

Moderated mediation path W*a: moderating effect of topical environmental knowledge

Only partly consistent with our assumption (H4), we found an interaction effect of topical environmental knowledge with the concrete compensation condition (b = 0.81, LLCI = 0.06, ULCI = 1.56, t = 2.12, p = .034; M = 4.91, SD = 1.45) on perceived greenwashing. That is, concrete compensation only increased perceived greenwashing when topical environmental knowledge was manipulated. For the other experimental conditions, neither an interaction effect (abstract compensation*topical environmental knowledge: b = 0.05, LLCI = −0.69, ULCI = 0.79, t = 0.14, p = .892; vague*topical environmental knowledge: b = 0.09, LLCI = −0.64, ULCI = 0.83, t = 0.24, p = .804; false*topical environmental knowledge: b = 0.41, LLCI = −0.33, ULCI = 1.16, t = 1.09, p = .277), nor a main effect of topical environmental knowledge on perceived greenwashing (b = −0.08, LLCI = −0.61, ULCI = 0.44, t = −0.31, p = .755) were found. Hence, while hypothesis 4b found support, hypothesis 4a, hypothesis 4c, and hypothesis 4d, did not. The interaction effect is shown in Figure 2.

Figure 2. The interaction effect of the concrete compensation condition and topical environmental knowledge on perceived greenwashing in comparison to the control condition.

Note. Descriptive statistics for all four interaction groups: Concrete compensation condition*Topical environmental knowledge condition (M = 4.91, SD = 1.45); Concrete compensation condition*Topical environmental knowledge control condition (M = 4.25, SD = 1.51); Control condition*Topical environmental knowledge condition (M = 3.82, SD = 1.71); Control condition*Topical environmental knowledge control condition (M = 3.90, SD = 1.83). Interaction effect is significant at the p < .05 level (b = 0.81, LLCI = 0.06, ULCI = 1.56, t = 2.12, p = .034).

Moderated mediation path c: effect of green advertising on brand evaluation

Only partly in line with H5, we found a significant main effect of the concrete compensation condition on brand evaluation (b = 0.55, LLCI = 0.06, ULCI = 1.03, t = 2.23, p = .026) compared to the control group. The other conditions had no effects on brand evaluation (abstract compensation: b = 0.48, LLCI = −0.02, ULCI = 0.98, t = 1.89, p = .061; vague: b = 0.06, LLCI = −0.43, ULCI = 0.58, t = 0.26, p = .797; false: b = 0.39, LLCI = −0.09, ULCI = 0.88, t = 1.59, p = .112).

Moderated mediation path c’: effects of green advertising on flight shame

Contrary to H6, we found no main effects of the experimental ad conditions compared to the ad control condition on flight shame (abstract compensation: b = 0.32, LLCI = −0.21, ULCI = 0.86, t = 1.19, p = .234; concrete compensation: b = 0.41, LLCI = −0.10, ULCI = 0.92, t = 1.58, p = .115; vague: b = 0.37, LLCI = −0.16, ULCI = 0.89, t = 1.37, p = .171; false: b = −0.04, LLCI = −0.55, ULCI = 0.48, t = −0.15, p = .880).

Moderated mediation path b: effects of perceived greenwashing on brand evaluation

In line with H7, there was a negative relation of perceived greenwashing with brand evaluation (b = −0.14, LLCI = −0.21, ULCI = −0.07, t = −4.04, p < .001). Formally, we observed a mediation effect of ad condition on brand evaluation via perceived greenwashing, independent of the topical environmental knowledge manipulation, with respect to the abstract compensation condition (topical environmental knowledge condition: b = −0.17, SE = 0.06, LLCI = −0.31, ULCI = −0.06; topical knowledge control condition; b = −0.17, SE = 0.06, LLCI = −0.30, ULCI = −0.06; index of moderated mediation = −0.01, SE = 0.06, LLCI = −0.13, ULCI = 0.12), the concrete compensation condition (topical environmental knowledge condition: b = −0.16, SE = 0.06, LLCI = −0.30, ULCI = −0.06; topical knowledge control condition; b = −0.05, SE = 0.05, LLCI = −0.15, ULCI = 0.04; index of moderated mediation = −0.12, SE = 0.07, LLCI = −0.27, ULCI = −0.00), the vague condition (topical environmental knowledge condition: b = −0.21, SE = 0.07, LLCI = −0.37, ULCI = −0.08; topical environmental knowledge control condition; b = −0.20, SE = 0.07, LLCI = −0.35, ULCI = −0.08; index of moderated mediation = −0.01, SE = 0.06, LLCI = −0.14, ULCI = −0.10), and the false condition (topical environmental knowledge condition: b = −0.24, SE = 0.08, LLCI = −0.39, ULCI = −0.10; topical environmental knowledge control condition; b = −0.18, SE = 0.07, LLCI = −0.33, ULCI = −0.07; index of moderated mediation = −0.06, SE = 0.06, LLCI = −0.19, ULCI = 0.06), in comparison to the control group.

Moderated mediation path b’: effects of perceived greenwashing on flight shame

In line with H8, we found that perceived greenwashing increases anticipated flight shame (b = 0.16, LLCI = 0.08, ULCI = 0.23, t = 4.17, p < .001). The mediation path via perceived greenwashing was insignificant for the abstract compensation condition (topical environmental knowledge condition: b = 0.18, SE = 0.06, LLCI = 0.08, ULCI = 0.31; topical environmental knowledge control condition; b = 0.18, SE = 0.07, LLCI = 0.06, ULCI = 0.32; index of moderated mediation = 0.01, SE = 0.07, LLCI = −0.12, ULCI = 0.14), the concrete compensation condition (topical environmental knowledge condition: b = 0.18, SE = 0.06, LLCI = 0.07, ULCI = 0.32; topical environmental knowledge control condition: b = 0.05, SE = 0.05, LLCI = −0.04, ULCI = 0.16; index of moderated mediation = 0.13, SE = 0.07, LLCI = −0.00, ULCI = 0.29), the vague condition (topical environmental knowledge condition: b = 0.23, SE = 0.07, LLCI = 0.11, ULCI = 0.39; topical environmental knowledge control condition; b = 0.22, SE = 0.07, LLCI = 0.09, ULCI = 0.38; index of moderated mediation = 0.01, SE = 0.07, LLCI = −0.12, ULCI = 0.14), and the false condition (topical environmental knowledge condition: b = 0.26, SE = 0.08, LLCI = 0.12, ULCI = 0.43; topical environmental knowledge control condition; b = 0.20, SE = 0.07, LLCI = 0.07, ULCI = 0.34; index of moderated mediation = 0.07, SE = 0.07, LLCI = −0.06, ULCI = 0.21), compared to the control group.

For all results, see Table 1. For an overview of the hypotheses and analysis, see Appendix A, Table A4.

Table 1. Summary of the moderated mediation model.

	Perceived Greenwashing		Brand Evaluation		Flight Shame
	b	SE	b	SE	b	SE
^aAbstract Compensation Condition	1.10***	0.28	0.48	0.25	0.32	0.27
^aConcrete Compensation Condition	0.33	0.27	0.55*	0.24	0.41	0.26
^aVague Condition	1.38***	0.27	0.06	0.25	0.37	0.27
^aFalse Condition	1.24***	0.27	0.39	0.25	−0.04	0.26
^bConcrete Compensation Condition	-0.77**	0.27	0.07	0.24	0.09	0.26
^bVague Condition	0.27	0.27	−0.42	0.24	0.04	0.26
^bFalse Condition	0.13	0.26	−0.09	0.24	−0.36	0.26
^cAbstract Compensation Condition	0.77**	0.27	−0.07	0.24	−0.09	0.26
^cVague Condition	1.04**	0.26	-0.48*	0.24	−0.05	0.25
^cFalse Condition	0.90**	0.26	−0.15	0.23	−0.45	0.25
Age	0.01	0.00	−0.01	0.00	-0.01***	0.00
Gender	0.01	0.12	0.11	0.11	−0.11	0.12
Education_high (dummy)	0.27	0.16	-0.38*	0.15	−0.23	0.16
Education_middle (dummy)	0.23	0.14	−0.13	0.13	−0.08	0.14
Flying Frequency	0.22***	0.06	-0.23***	0.06	−0.07	0.06
Topical Environmental Knowledge	−0.08	0.27
Abstract Compensation Condition * Topical Environmental Knowledge	0.05	0.37
Concrete Compensation Condition * Topical Environmental Knowledge	0.81*	0.38
Vague Condition * Topical Environmental Knowledge	0.09	0.37
False Condition * Topical Environmental Knowledge	0.41	0.38
Perceived Greenwashing			-0.14***	0.04	0.16***	0.04
Explained Variance	0.14		0.09		0.06

Note. Macro PROCESS 3.5, Model 8 with 5.000 bootstrap samples; Calculated separately for brand evaluation and flight shame; N = 658.

a Control condition inserted as a reference group; *** p < .001; ** p < .01; * p < .05.

b Abstract compensation condition inserted as a reference group; *** p < .001; ** p < .01; * p < .05.

c Concrete compensation condition inserted as a reference group; *** p < .001; ** p < .01; * p < .05.

Discussion

We showed, for the first time, that while vague claims, false claims, and abstract compensation claims are perceived as greenwashing to a higher degree as non-green(washed) claims by consumers, concrete compensation claims are not. A possible explanation of this finding might be that consumers perceive green airline ads as incongruent with existing mental representations (i.e., flying is environmentally harmful). As a consequence, consumers’ cognitive attention toward the ad may increase, ultimately leading to perceptions of greenwashing. However, not all greenwashing claims may foster this perception. In fact, when we look at the results of H1, we found that consumers accept green claims offering concrete green compensations. This finding may be explained by the fact that, compared to the other greenwashing claims, theoretically speaking, the incongruence between the ad and existing mental representations could be lower. However, contrary to the assumed effects in H2, perceived greenwashing was not higher when a vague or a false claim was presented than when a compensation claim was shown. As hypothesized in H3, perceived greenwashing in concrete compensation claims was only higher compared to abstract compensation. This suggests that consumers may believe in the realization of concrete compensation measures due to smaller psychological distance (Trope and Liebermann 2010). Theoretically, concrete compensation can be immediately observed by consumers, leading to stronger green compensatory beliefs compared to abstract claims (Hope et al. 2018). Compensatory beliefs, in turn, may distract consumers from perceiving greenwashing (Hope et al. 2018). There are, however, other possible explanations.

Further, contrasting to our H4, results showed that the effects of vague, false, and abstract compensation claims were independent from topical environmental knowledge. That is, even consumers with low topical environmental knowledge perceive greenwashing in vague, false, and abstract compensation claims. This finding stands in contrast to prior research, which showed that knowledge can moderate the effects of greenwashed claims on greenwashing perceptions (Schmuck, Matthes, and Naderer 2018). However, we have to keep in mind that airlines offer inherently environmentally unfriendly services. For other services, such as food or convenience goods, knowledge may be necessary to create an incongruence between the ad and pre-existing beliefs, leading to greenwashing perceptions. Yet, our study showed, that when topical environmental knowledge was given, concrete compensation was perceived to a stronger degree as greenwashing than when topical environmental knowledge was not given. One may argue that topical knowledge is necessary to weight the compensation offered in the ad with the actual environmental harm done by flying. For low knowledgeable individuals, the compensation may be perceived as adequate; however, when there is high knowledge, the compensation may appear as a drop in the ocean, compared to the harm of flying. At this point, it needs to be stressed that we conceptualize greenwashing as a perception, not an objective attribute (Seele and Gatti 2017).

Findings only partially support H5 and do not support H6. We found a positive direct effect of the concrete compensation claim on brand evaluation, which was not mediated by perceived greenwashing. This is consistent with findings of other studies suggesting that green advertising with concrete compensation claims has positive effects on brand evaluations. This result may be explained by the fact that concrete compensation may be generally interpreted as a positive, voluntary move of companies, triggering consumers’ positive attitudes. Further investigations are needed to shed light on this. Besides, this study was unable to demonstrate that green advertising reduces flight shame. It seems possible that these results are due to the fact, that the mere exposure to green advertising is too weak to reduce strong feelings like flight shame. Although Mazar and Zhong (2010) showed that mere exposure to green services may trigger ‘social considerations’ (497), future studies that focus on real purchase behavior of green services instead of mere exposure to green advertising are suggested.

Finally, in line with H7 and H8, this study demonstrated that perceived greenwashing has negative consequences. That is, perceived greenwashing decreases consumers’ brand evaluations. Apart from that, we showed, for the first time, that perceived greenwashing evoked flight shame. As shame motivates avoidance tendencies, individuals may tend to avoid the shame-inducing behavior and feel morally responsible to behave more pro-environmentally in the future (e.g., Amatulli et al. 2019). In the context of inherently environmentally unfriendly services like flying, this result is striking. It shows that green airline advertising can lead to a boomerang effect eventually motivating consumers to avoid flying. Obviously, further research is needed to better understand the behavioral consequences of flight shame.

Limitations and future research

There are at least six limitations. First, we only examined greenwashing claims promoting inherently environmentally unfriendly services. Perceptions of greenwashing might be different for other products and services. Second, since we did not directly measure compensatory green beliefs, further research is needed to better understand the effects of compensatory green beliefs as a motivation to accept green claims. Third, we did not investigate long-term effects. Future research should apply non-experimental designs as well as behavioral measures. Fourth, we only measured topical environmental knowledge as a post-test measurement after the experimental knowledge manipulation. Although previous studies have proceeded in a very similar way (Fabrigar et al. 2006; Nelson, Wood, and Paek 2009; Ovrum et al. 2012) it would be methodically more precise to additionally measure pre-existing environmental knowledge of participants (Fernandes, Segev, and Leopold 2020). Fifth, methodically, we used a multi message design that is theoretically strongly guided by observations from advertising practice of airlines (Slater 1991). Although the experts rating of advertising claims suggested acceptable internal validity of our stimuli, we must note, that the results of the experts rating could have turned out more clearly in benefit of our claim categorization. Hence, caution must be applied when interpreting and generalizing the results of this study. Lastly, since there are currently no possibilities to fly sustainably (e.g., Snow 2021), it was not possible for us to create a substantial green claim that is extern valid and serves as a reference group for all our stimuli. In the future, an externally valid substantial green ad should be used as a reference group.

Theoretical, methodological, and practical implications

With regard to theoretical implications, Schema Incongruity Processing Theory (Mandler 1982) allows important predictions. First, greenwashing may be only perceived when existing mental structures do not align with the advertised claim. This means, greenwashing strategies may be detected in the case of airline ads, but could remain unrecognized for other services with less negative mental structures regarding sustainability. Second, the ELM (Petty and Cacioppo 1986) helps us to explain how topical environmental knowledge increases the ability to weight the environmental harms against the environmental benefits promoted in an ad. Third, our study showed for the first time, that consumers’ greenwashing perceptions of hitherto uninvestigated greenwashing claims, namely concrete compensation claims, differ from abstract compensation, vague, and false claims potentially based on different underlying theoretical mechanisms such as perceived psychological distance (Liberman and Trope 1998) or green compensatory beliefs (Hope et al. 2018). Lastly, our study contributes to a better understanding of how perceptions of greenwashing can trigger psychological reactance (Brehm 1966). Besides the negative influence of perceived greenwashing on consumers’ cognitions (brand evaluations), we also observed it on consumers’ affect (flight shame).

Considering methodological implications, no previous study in this context has manipulated topical environmental knowledge until now. We believe that a manipulation of knowledge usefully supplements and extends the previous literature by allowing us to directly test which chunks of knowledge are necessary in order to enhance greenwashing perceptions.

If we turn to practical implications for advertisers, we see several possibilities to avoid greenwashed advertising. First, we recommend that advertisers should focus on truthful, precise, and substantial green claims. Second, although industry guidelines for the use of environmental marketing claims already exist, we believe that they are incomplete. We call for regular updates of the green guides, including abstract and concrete compensation claims (FTC 2012). Third, since it is challenging to communicate complex environmental information concisely and understandable for consumers with different degrees of environmental knowledge, companies need to pay close attention to consumers’ understanding of compensation claims. Here, ‘a uniform, accepted standard’ could help advertisers to avoid greenwashing as well as consumers to get more clarity about the environmental benefit achieved through the compensation measure. Additionally, information about the scope of the compensation, the timing of the compensation, and the quality of the compensation should be mandatory to provide when environmental compensation is offered (Polonsky, Grau, and Garma 2010). Finally, tools that help consumers to detect misleading advertising claims should be implemented. Interactive tools like the technology of Quick Response (QR) codes and barcodes including a traffic light system could make it easier for consumers to interpret various green advertising claims (e.g., Parguel, Benoit-Moreau, and Russell 2015). Especially for compensation claims, carbon labels including ‘product or service footprints’ that provide precise information about the CO2 emissions of the products or services could not only educate, but also motivate consumers to reduce their CO2 emissions (European Commission 2014). Besides, advertising literacy campaigns may empower consumers to question the misleading messages.

Conclusion

Companies are increasingly interested to use green advertising claims to position themselves as environmentally active and friendly. Against this background, greenwashing may appear as an attractive strategy, especially for products and services that are inherently environmentally unfriendly. Our findings on airline ads clearly show that several types of greenwashing claims yield different effects. By and large, however, our study suggests that greenwashing is likely to harm the brand, directly and indirectly: directly, because perceived greenwashing leads to more negative brand evaluations, and indirectly, because greenwashing can foster flight shame, making future airline travel less likely. The only exception are concrete greenwashing claims paired with low consumer topical environmental knowledge. Since environmental knowledge can be expected to rise in the future due to a global increase in environmental concern, we conclude that greenwashing does not pay off for the airline industry. By contrast, it harms more than it benefits.

Notes on Contributors

Ariadne Neureiter is a Ph.D. candidate at the Department of Communication at the University of Vienna. She studied communication science and psychology and focuses on research on strategic (sustainability) communication, effects of green advertising, and digital media. Jörg Matthes (Ph.D., University of Zurich) is a communication science professor and head of the Department of Communication at the University of Vienna. He focuses on research on advertising, digital media, political communication, and quantitative methods.

Author note

Ariadne Neureiter (MA, MSc; phd candidate; [email protected]), Jörg Matthes (Univ.-Prof. Dr; head of department; [email protected]), Advertising and Media Psychology Research Group, Department of Communication, Faculty of Social Sciences, University of Vienna, Währingerstr. Str. 29, 1090 Austria.

Funding statement

To conduct this research, there was no funding involved.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The research data used in this study is freely available in a public data repository of the Open Science Framework (https://osf.io/9wd7p/?view_only=c7afccc4586a469692c7ded6def85df2).]

Appendix A

Table A1. Conceptualization of greenwashing claim types in airline advertising.

Greenwashing Claims	Definition	Example
Omission Claim/Abstract Compensation Claim	Excludes important information that is needed by consumers to assess the environmental benefit of future compensation that cannot be directly observed by consumers	CO2 compensation measures; investment in tree planting, environmental projects or research
Omission Claim/Concrete Compensation Claim	Excludes important information that is needed by consumers to assess the environmental benefit of immediate compensation that can be directly experienced by consumers	CO2 compensation measures; recycling or reducing of used cutlery on bord
Vague Claim	Overly ambiguous and too unspecific to make a well-informed conclusion about the green character of the service	Promotes flying as sustainable without giving a reason or proposing measures
False Claim	Includes an outright lie	Full elimination of flight emissions and consumers’ carbon footprint

Table A2. Examples and description of the used stimuli in the conditions.

Example	Description
	Abstract Compensation Condition	Concrete Compensation Condition	Vague Condition	False Condition	Ad Control Condition
	We want to do everything for our forests! Therefore, we plant a tree for every flight. #hifly; 1 flight = 1 tree.	We want to do everything for our seas! Therefore, we do not use plastic on board of our airplanes. #hifly; 1 flight = 0 plastic packages.	We want to do everything for our environment! Therefore, we make your flight sustainable. #hifly; 1 flight = 1 green journey.	We want to do everything for our environment! Therefore, we make your flight CO2-neutral. #hifly; 1 flight = 0 CO2 emissions.	We want to do everything for our passengers! Therefore, we offer comfortable flights to your vacation. #hifly; 1 flight = 1 swimming vacation.
	Fly with us! We donate a share of the flight revenues and hence support environmental projects. #S7airlines; ecological projects – lots of environment. Flying together. We donate. Supporting environmental projects.	Fly with us! We recycle the on-board waste of our flights and hence protect the environment. #S7airlines; Recycling – lots of environment. Flying together. Recycling of on-board waste. Protecting the environment.	Fly with us! We promote nature conservation and thus protect the environment and nature. #S7airlines; Nature conversation – lots of environment. Flying together. Respecting nature. Protecting the environment.	Fly with us! We fly CO2-neutral around the world with you and hence protect the environment. #S7airlines; CO2 neutral – lots of environment. Flying together. Traveling CO2 neutral. Protecting the environment.	Fly with us! We offer a relaxing flight and hence reduce your traveling stress. #S7airlines; Relaxation – lots of vacation. Flying together. Relaxing on the beach. Enjoying the vacation.
	Innovation for the environment is important to us! We invest a part of the profit in environmental research. #SprintAir; Advance environmental research! Flying with us means supporting environmental research.	Avoiding plastic waste is important to us! We allow your own drinking bottles and are delighted to refill them. #SprintAir; Bring your own drinking bottle! Flying with us means avoiding plastic cups.	Nature protection is important to us! With us, sustainability is the highest priority and is actively lived. #SprintAir; Advance sustainability! Flying with us means living sustainability.	CO2-neutrality is important to us! With us, the CO2 footprint is eliminated. #SprintAir; Reduce the CO2 footprint to 0! Flying with us means eliminating the CO2 footprint.	A relaxing start is important to us! With us, the vacation mood starts already in the airplane. #SprintAir; Take the vacation mood with you! Flying with us means spreading vacation mood.
Example	Description
	Topical Environmental Knowledge Condition	Topical Environmental Knowledge Control Condition
	The news article contains topical environmental facts (see Table A3; measures; topical environmental knowledge).	The article contains the same amount of facts like the article in the topical environmental knowledge condition, but here, they are not environment related. It includes facts such as that the numbers of industrial robots in Austria (2018) show a growth of about 6% compared to the previous year.

Note. This table demonstrates the stimuli for our experimental conditions and control conditions.

Table A3. Description of measures.

Variable	Type of Scale	Items	α	M (SD)
Topical Environmental Knowledge	Single-choice items (seven-point scale; 0 = no topical environmental knowledge, 6 = high topical environmental knowledge)	1. How many grams of greenhouse gas are released into the atmosphere per kilometer travelled by airplane? Answers: About 900 grams of greenhouse gases are released into the atmosphere (f), about 500 grams of greenhouse gases are released into the atmosphere (f), about 200 grams of greenhouse gases are released into the atmosphere (c); 2. What applies to the flying behavior in Austria? Answers: The air travel numbers in Austria (2018) show a growth of about 10% compared to the previous year (c), the air travel numbers in Austria (2018) show a growth of about 3% compared to the previous year (f), the air travel numbers in Austria have stagnated over the last two years (f); 3. What substances does a flying aircraft emit? Answers: Among others carbon dioxide (CO2) and carbon monoxide (c), among others carbon (soot) and glyphosate (f), among others carbon dioxide (CO2) and sulfur monoxide (f); 4. How many square kilometers of Artic sea ice melt each year as a result of air traffic? Answers: 6000 (c), 1000 (f), 300 (f); 5. How many tons of CO2 may each human cause in a year to be in line with the Paris climate convention? Answers: 1,5 tons of CO2 per year (f), 2,3 tons of CO2 per year (c), 0,5 tons of CO2 per year (f); 6. What action can be taken to still meet the Paris climate goals? Answers: If one supplies the apartment with electricity over one year, heats it and takes a return flight from Germany to Tenerife for a vacation (f), if one supplies the apartment with electricity over one year, and takes a return flight from Germany to Tenerife for a vacation (f), if one takes a return flight from Germany to Tenerife for a vacation (c).		2.70 (1.25)
Perceived Greenwashing	Seven-point Likert scale (‘strongly disagree’ to ‘strongly agree’)	1. The advertised measures of the ad portray the airlines as being more useful to the environment than they are; 2. The advertised measures sound more useful for the environment than they actually are; 3. The advertised measures of the ad are misleading.	.79	4.78 (1.61)
Brand Evaluation	Seven-point semantic differential scale	1. Unattractive–attractive; 2. Not likeable–likeable; 3. Uninteresting–interesting.	.97	3.57 (1.42)
Flight Shame	Seven-point Likert scale (‘strongly disagree’ to ‘strongly agree’)	1. When I fly, I have the feeling of having failed morally; 2. When I fly, I have the feeling that others judge me for it; 3. I cannot reconcile with myself to fly; 4. It embarrasses me when others think that I am not environmentally conscious because I travel by air; 5. I prefer telling my acquaintances and my friends about a planned trip by train rather than about a trip planned by plane; 6. On the whole, I am dissatisfied with myself when I use the plane as a means of transport; 7. When I travel by air, I have the feeling that others look down on me.	.91	2.63 (1.49)

Note. For topical environmental knowledge items, the correct answers are displayed as ‘c’, and the wrong answers as ‘f’.

Table A4. Overview of the hypotheses and analyses.

No.	Hypotheses	Analysis	Status
H1	If (a) an abstract compensation, (b) a concrete compensation, (c) a vague, and (d) a false greenwashing claim is presented in an airline ad, perceived greenwashing will be higher compared to an ad presenting no green claims.	Model 8a; path a	Partly accepted
H2	When (a) a vague or (b) a false claim is presented in an airline ad, perceived greenwashing will be higher compared to abstract and concrete compensation claims.	Model 8a; path a	Rejected
H3	When an abstract compensation claim is presented in an airline ad, perceived greenwashing will be higher compared to an ad presenting a concrete compensation claim.	Model 8a; path a	Accepted
H4	When compared to an ad with no green claims, the effect of an (a) abstract compensation claim, (b) concrete compensation claim, (c) vague claim, and (d) false claim on perceived greenwashing is higher for respondents high in topical environmental knowledge as compared to less knowledgeable respondents.	Model 8a; path W*a	Partly accepted
H5	If (a) an abstract compensation, (b) a concrete compensation, (c) a vague, and (d) a false greenwashing claim is presented in an airline ad, brand evaluation will be increased compared to an ad presenting no green claims.	Model 8a; path c	Partly accepted
H6	If (a) an abstract compensation, (b) a concrete compensation, (c) a vague, and (d) a false greenwashing claim is presented in an airline ad, flight shame will be decreased compared to an ad presenting no green claims.	Model 8b; path c’	Rejected
H7	Perceived greenwashing decreases brand evaluation for the advertised brands.	Model 8a; path b	Accepted
H8	Perceived greenwashing increases flight shame.	Model 8b; path b’	Accepted

Appendix B

(under OSF)

Table B1. Pattern matrix of a principal component analysis on mediator and dependent variables using oblimin rotation.

	Rotated Factor Loadings
	1	2	3	M	SD
Perceived greenwashing (α = .79)				4.78	1.61
The advertised measures of the ad portray the airlines as being more useful to the environment than they are.	.87	.07	-.00	4.91	1.96
The advertised measures sound more useful for the environment than they actually are.	.87	.05	-.07	4.47	2.01
The advertised measures of the ad are misleading.	.78	-.12	.10	4.97	1.87
Brand evaluation (α = .97)				3.57	1.42
HiFly: Unattractive–attractive	-.01	.88	-.00	3.63	1.61
HiFly: Not likeable–likeable	-.01	.85	-.03	3.79	1.61
HiFly: Uninteresting–interesting	-.04	.85	.02	3.59	1.69
S7airlines: Unattractive–attractive	.04	.91	.01	3.55	1.55
S7airlines: Not likeable–likeable	.05	.92	.00	3.62	1.54
S7airlines: Uninteresting–interesting	.03	.90	.05	3.53	1.58
SprintAir: Unattractive–attractive	-.01	.89	.01	3.46	1.62
SprintAir: Not likeable–likeable	-.01	.89	.01	3.50	1.58
SprintAir: Uninteresting–interesting	-.03	.89	-.01	3.48	1.62
	Component Loadings
	1	2	3	M	SD
Flight shame (α = .91)				2.63	1.49
When I fly, I have the feeling of having failed morally.	.03	−0.3	.86	2.63	1.80
When I fly, I have the feeling that others judge me for it.	-.08	.05	.86	2.38	1.66
I cannot reconcile with myself to fly.	.03	-.07	.77	3.04	2.03
It embarrasses me when others think that I am not environmentally conscious because I travel by air.	.02	.08	.84	2.58	1.82
I prefer telling my acquaintances and my friends about a planned trip by train rather than about a trip planned by plane.	-.03	.01	.77	2.59	1.91
On the whole, I am dissatisfied with myself when I use the plane as a means of transport.	.13	-.06	.76	2.98	1.96
When I travel by air, I have the feeling that others look down on me.	-.08	.05	.84	2.23	1.64
Note. N = 658; factor Loadings over 0.7 in bold.

Table B2. Correlations between items measuring the dependent variable brand evaluation.

Variables	1	2	3	4	5	6	7	8	9
1. HiFly: Unattractive–attractive	1
2. HiFly: Not likeable–likeable	.87***	1
3. HiFly: Uninteresting–interesting	.85***	.83***	1
4. S7airlines: Unattractive–attractive	.74***	.68***	.71***	1
5. S7airlines: Not likeable–likeable	.76***	.72***	.72***	.91***	1
6. S7airlines: Uninteresting–interesting	.74***	.69***	.72***	.91***	.91***	1
7. SprintAir: Unattractive–attractive	.71***	.66***	.71***	.76***	.76***	.74***	1
8. SprintAir: Not likeable–likeable	.71***	.68***	.68***	.75***	.77***	.74***	.91***	1
9. SprintAir: Uninteresting–interesting	.71***	.65***	.68***	.75***	.75***	.76***	.91***	.90***	1

Note. Pearson correlations for single items of the index brand evaluation. N = 658; *** p < .001; ** p < .01; * p < .05.

Table B3. Randomization check.

Variables
	Topical Environmental Knowledge	Claim Ad Type
Age	F(1, 648) = 0.41, p = .525	F(4, 648) = 1.52, p = .196
Gender	χ2(1) = 0.03, p = .876	χ2(4) = 6.78, p = .148
Education	χ2(2) = 0.35, p = .840	χ2(8) = 4.72, p = .787
Environmental concern	F(1, 648) = 0.51, p = .481	F(4, 648) = 0.76, p = .556
Flight frequency	F(1, 648) = 2.41, p = .122	F(4, 648) = 1.24, p = .556

Note. No systematic differences for age, gender, education, environmental concern and flight frequency between topical environmental knowledge (control) conditions and claim ad type (control) conditions.

Table B4. Manipulation checks.

Condition
Topical Environmental Knowledge	M (SD)					F
Topical Environmental Knowledge	3.07 (1.29)
Topical Environmental Knowledge Control Condition	2.33 (1.09)					F(1, 656) = 63.27, p < .001^a
Claim Ad Type	M (SD)					F
	Investment in environment	Reduction of plastic	General sustainability	Elimination of flight emissions	Holidays
Abstract Compensation	5.91 (1.51)	2.40 (1.92)	4.19 (1.98)	4.34 (2.15)	3.57 (2.06)	F(4, 653) = 155.35, p < .001^b
Concrete Compensation	2.32 (1.66)	6.34 (1.29)	4.18 (2.08)	2.65 (1.98)	2.66 (1.74)	F(4, 653) = 225.34, p < .001^b
Vague	2.56 (1.79)	2.14 (1.59)	5.81 (1.52)	3.87 (2.03)	4.55 (1.84)	F(4, 653) = 49.32, p < .001^b
False	2.17 (1.62)	1.66 (1.31)	5.19 (2.11)	6.00 (1.64)	3.69 (2.02)	F(4, 653) = 92.10, p < .001^b
Ad Control Condition	1.67 (1.30)	1.57 (1.27)	2.68 (1.98)	1.95 (1.43)	6.43 (1.10)	F(4, 653) = 81.28, p < .001^b

Note. Significant differences between experimental conditions and control conditions separately described for topical environmental knowledge and claim type. ^aCompared to the topical environmental knowledge condition. ^bCompared to all other four ad conditions. Significantly highest means (M) with standard deviations (SD)for claim type topics are depicted in bold.

Table B5. Overview of effects of the moderated mediation model.

	Estimate	95% CI lower	95% CI upper	p
Brand Evaluation
ACME	−0.16	−0.28	−0.07	.001***
ADE	0.41	0.07	0.74	.020*
Total Effect	0.23	−0.09	0.57	.160
Proportion Mediated	−0.71	−6.72	5.88	.161
Flight Shame
ACME	0.18	0.08	0.30	< .001***
ADE	0.27	−0.09	0.62	.131
Total Effect	0.45	0.09	0.80	.012*
Proportion Mediated	0.40	0.15	1.56	.012*

Note. R package ‘mediation’ with 5.000 bootstrap samples.Calculated separately for brand evaluation and flight shame. N = 658; *** p < .001; ** p < .01; * p < .05.

Table B6. Means and standard deviations of mediator and dependent variables.

		Perceived Greenwashing		Brand Evaluation		Flight Shame
Condition/Claim Ad Type	n	Mean	SD	Mean	SD	Mean	SD
Control Condition	129	3.86	1.76	3.42	1.32	2.39	1.34
Abstract Compensation Condition	136	4.98	1.42	3.66	1.38	2.88	1.61
Concrete Compensation Condition	127	4.55	1.51	3.79	1.50	2.59	1.39
Vague Condition	135	5.25	1.45	3.31	1.41	2.77	1.61
False Condition	131	5.23	1.50	3.68	1.44	2.52	1.44
TOTAL	658	4.78	1.61	3.57	1.42	2.63	1.49