Paint it green: Effects of green claims and nature imagery in green
advertisements on brand attitude and a brand’s ecological image.
Master Thesis
Graduate School of Communication – University of Amsterdam
Master Persuasive Communication (MSc)
Author: Sandra Leinsle, Student-ID: 10968075 Supervisor: Dr. Ester de Waal
Abstract
In recent times, there has been a shift towards green advertising of companies caused by the rising demand for green products in today’s society. Companies want to meet consumer’s needs of environmentally friendly products. Thereby not every company is truly green but advertises that its products are which can be referred to as greenwashing. This study aims at identifying greenwashing techniques that companies might use in their advertisements by researching effects of green advertising claims and green imagery on brand attitude and brand’s ecological image of the consumer. Thereby a special focus on vague claims and nature imagery was set because these tools have the potential to easily mislead consumers. 277 participants were exposed to an online experiment (2x2x2 between subjects design) and randomly assigned to one of four different green advertisement conditions. Results showed that specific claims led to more positive brand attitudes and brand’s ecological image compared to vague claims. Furthermore, this study could not identify positive influences of nature imagery on brand attitude or brand’s ecological image. Green involvement was supposed to moderate effects between claims and imagery on the dependent variables. Against expectations, this study solely found a main effect of green involvement on brand attitude when consumers were presented with a specific claim. This effect was caused by attitude towards green products and green purchase behavior, sub-concepts of green
involvement as measured in this study. In summary, vague claims and nature imagery were not found to be efficient tools to mislead consumers with greenwashed advertisements and green involvement did not play a moderating role, opposed to what has been expected.
Keywords: green branding, claims, vagueness, nature imagery, brand attitude, brand’s ecological image, green involvement
Green claims and nature imagery in advertising
Around 96% of Europeans think that environmental protection is very important (Finisterra do Paco & Reis, 2012). Thus more and more brands/companies come up with green
products/services to meet the growing need for environmentally friendly products (Parguel, Benoit-Moreau & Russell, 2015).This results in green branding and green advertising of those companies and the fact that nowadays almost everything is green: car companies, airlines, buildings, products, food, etc. This is seemingly positive however it is negative when green advertising is used to increase sales by promoting products as conscious that are in fact not environmentally friendly. The general function of green advertising is to inform (prospect) customers about environmental aspects of a brand or its products and services (Finisterra do Paco & Reis, 2012). Along with this, a phenomena called greenwashing developed.
Companies use greenwashing to artificially state environmental concern of the company in order to appeal to more potential customers (Delmas & Burbano, 2011). The neologism ‘greenwashing’ has emerged in 1986 and can be defined as “the marketing tactic of
misleading consumers about a product or service’s environmental friendliness” (Terrachoice, 2010, p.1). Results of a report of over 5,200 products about greenwashing in the USA and Canada showed that 95% of these products can be considered as greenwashed according to the researcher’s definitions (Terrachoice, 2010). Delmas and Burbano (2011) describe two behaviors of companies that are characteristically for greenwashing. A poor environmental performance on the one hand but a communication about positive environmental performance on the other hand. Greenwashing has been subject to some studies in the past years. Drivers of greenwashing (Delmas & Burbano, 2011), an analysis about the state of greenwashing in advertisements in India (Fernando, Sivakumaran & Suganthi, 2014), the influence of greenwashing on green trust and green perceived risk of Taiwanese consumers (Chen &
Chang, 2013) and greenwashing effects caused by nature imagery in advertisements (Parguel et al., 2015) have been subjects of research.
In the current research field of green advertising a general shortage of knowledge about consumer responses to ecologically themed advertising appeals is documented in the literature (Tucker, Rifon, Lee & Reece, 2012). There is a limited body of studies testing experimentally what influence green environmental information had on brand or product attitude formations (Spack, Board, Crighton, Kostka & Ivory, 2012). On the one hand green ad claims have been researched extensively especially with regards to its characteristics such as e.g. the formulation (Obermiller, 1995), promotional vs. preventional focus (e.g. Bickart & Ruth, 2012) and its strength (Chang, 2011; Tucker et al., 2012). Furthermore, green ad claims have been researched with a focus on consumer responses e.g. its effects on ad credibility (Carlson, Grove & Kangun, 1993; Tucker et al., 2012), green purchase intention (Tucker et al., 2012) and attitude towards brand and advertisement (e.g. Matthes, Wonneberger & Schmuck, 2014). Studies which specifically addressed greenwashing practices in green advertising also mainly focused on the text elements of an advertisement, its claims (e.g. Tucker et al., 2012; Manrai, Manrai, Lascu & Ryans Jr., 1997). On the other hand only a few studies addressed green ad imagery and its influence on consumer responses like brand
attitude formation (Hartmann, Apaolaza Ibanez & Sainz, 2005; Hartmann & Apaolaza-Ibanez, 2009; Matthes et al., 2014), brand memory (Hartmann, Apaolaza & Alija, 2013) and a brand’s ecological image and ways of how to work against misleading greenwashing effects caused by nature evoking imagery (Parguel et al., 2015). The lack of research of ecological imagery in advertising leads to real-life limitations as well. Green advertising regulations did not focus on executional elements as much as on advertising claims. This can be attributed to the fact that there is a lack of research that proves greenwashing effects of executional ad elements in green advertising. Therefore it is important to deliver further evidence of greenwashing
effects caused by nature imagery as well as of greenwashing by using misleading claims (Parguel et al., 2015) and the combinational effects of both. Recent studies have already addressed both of these elements and their reciprocal effects on brand attitude and attitude towards the ad (Matthes et al., 2014) or the influence on a brand’s ecological image (Parguel et al., 2015). What the current body of research is lacking is further proof of potential
greenwashing techniques especially by vague green ad claims in combination with nature imagery compared to a combination with no nature imagery. The dependent variable brand’s ecological image is a relatively new concept and has not been researched in a lot of studies concerning green advertising. Thus, further scientific proof of ad factors that influence a brand’s ecological image is needed.
Parguel et al. (2015) examined what they call the ‘executional greenwashing’ effect. This term refers to “the use of nature-evoking elements in advertisements to artificially enhance a brand’s ecological image” (Parguel et al., 2015, p.107). Thereby Parguel and colleagues (2015) suggest that executional advertising elements depicting nature can be as misleading for consumers as greenwashed claims. Based on the argumentation above, this study will make a difference between claim greenwashing and executional greenwashing effects as suggested by Parguel et al. (2015). These two concepts are combined in this study and the reciprocal influence is being examined. We aim at researching the combinational effects of text elements and nature imagery of a green advertisement on consumer responses (brand attitude and brand’s ecological image). The overarching aim of this research is to deliver further proof of potential greenwashing effects of certain combinations between claims and nature themed advertisement imagery. Therefore the following research question is stated:
RQ: What effects do (vague) claims of green advertisements in combination with (nature) imagery have on consumer’s brand’s ecological image and brand attitude?
Claim greenwashing: vague green advertising claims
A vague verbal claim can be used as a greenwashing technique by companies to mislead consumers about environmental benefits of the advertised product or the brand in general. That means a confusing product claim or a specifically manufactured claim with the goal to appeal to green consumers without being sincerely environmentally friendly (Finisterra do Paco & Reis, 2012). For the definition of a misleading green claim this study will focus on a research report of the before mentioned Canadian green marketing agency Terrachoice. In 2010 they published their last research on greenwashing in the USA and Canada and for this purpose defined ‘7 sins of greenwashing’ (Terrachoice, 2010). These seven sins are hidden tradeoff, no proof, vagueness, irrelevance, lesser of two evils, fibbing and worshipping false labels. A content analysis with the aim to identify greenwashing in American and British magazines conducted by Baum (2012) used these definitions as well. The ‘greenwashing technique’ this study wants to focus on is vagueness as this was the most frequent
greenwashing practice that Baum (2012) identified in her content analysis. An example of a vague claim is “all natural” whilst there are substances that occur naturally but are toxic which makes a product described as “all natural” definitely not green. Other examples can be “green”, “environmentally friendly” or “eco-conscious” because without an explanation they have no meaning and are vague. In line with this Kangun, Carlson and Grove (1991) describe that vague or ambiguous terms can be characterized as being transported through uncertainty in the text message.
Consumers can actually identify vague textual claims and know the difference between specific and vague claims (Spack et al., 2013 as in: Davis, 1993). Hartmann and Apaolaza-Ibanez (2009) noted in one of their studies about green advertising that specified environmental claims are perceived as more positive and persuasive compared to vague claims which leads to positive effects on a consumer’s brand attitude. Furthermore, studies which examined argument strength in advertisement’s textual claims concluded that strong
arguments are more persuasive than weak ones (Tucker et al., 2012). Remembering the fact that vague claims are meaningless they can be classified as weak arguments.
Because consumers can identify differences between vague and specific claims a different evaluation of the advertiser’s image can result (Davis, 1993). Vague claims lead to a negative image, more specific ones to a positive image. Davis (1993) stated as well that linking a brand with vague claims can even cause brand image damages. Brand’s ecological image can be seen as a subset of overall brand image (Ng, Butt, Khong & Ong, 2014). Chen (2010) defines green brand image as “a set of perceptions of a brand in a consumer’s mind that is linked to environmental commitments and environmental concern” (Chen, 2010, p.309). Thus it can be assumed that the negative effects of vague claims on brand image are also valid for a brand’s ecological image. Hypothesis 1 is formulated as follows:
H1: Vague green claims lead to negative effects on brand attitude and brand’s ecological image compared to specific green advertising claims which lead to a positive
brand attitude and brand’s ecological image.
But not only textual advertising elements such as claims influence consumer responses to ads. Also executional elements like images can play a prominent role in predicting
consumer’s responses to advertisements (Spack et al., 2012). Nature imagery
Greenwashing effects can also occur via nature imagery in advertisements (Hartmann & Apaolaza-Ibanez, 2009; Hartmann, Apaolaza & Ibanez, 2013). Early studies found evidence that nature imagery might be ineffective because it can be understood as vague claims (Hartmann & Apaolaza-Ibanez, 2009). In contrast to that some researchers claim that nature imagery cannot be treated as vague verbal claims and might not lead to negative persuasive outcomes compared to vagueness in verbal green claims because the persuasion process is
different (Hartmann & Apaolaza-Ibanez, 2009). With regards to the ELM (Petty & Cacioppo, 1986) it can be argued that nature imagery is processed non-cognitively
(emotionally/affectively) on the peripheral route compared to verbal claims mainly processed cognitively on the central route, depending on its strength (a further elaboration of the
processing of claims will follow later in the text). The description above states a somehow mixed research field regarding the effectiveness of nature imagery in advertising. This disparity could be explained through the use of different images of those studies and the different effects these images cause. A study by Hartmann et al. (2009) demonstrated that different nature biome can cause different affective reactions. Even though this is a relevant issue for further research it should not be of relevance for this study.
An early study by Carlson et al. (1996) found that consumers perceive a product or company as more favorable when the advertisement uses associative claims (= images) related or depicting environmental facts such as nature imagery than using no such imagery. In the field of nature imagery in advertising it is essential to name a series of studies
conducted by Hartmann and colleagues (Hartmann, Apaolaza-Ibanez & Sainz, 2005; Hartmann & Apaolaza-Ibanez, 2009; Hartmann & Apaolaza-Ibanez, 2010; Hartmann,
Apaolaza & Alija, 2013). Three of the studies (Hartmann et al., 2005; Hartmann & Apaolaza-Ibanez, 2009; Hartmann et al., 2013) examined the influence of advertising imagery depicting nature on brand attitude. Results showed that nature imagery indeed has a positive influence on brand attitude. This is because images of nature can evoke the same emotions as real nature which in turn leads to more positive attitudes towards the brand (Hartmann & Apaolaza-Ibanez, 2009; Hartmann et al., 2013). Research also showed that nature-evoking elements like nature imagery have an influence on the consumer’s perception of a brand’s ecological image (Parguel et al., 2015).
For this study it is expected that nature imagery has a positive influence on the evaluation of the advertisement. More precise, ads depicting nature imagery will lead to a
positive brand attitude and brand’s ecological image compared to advertisements with no nature imagery. Therefore the second Hypothesis of this research is:
H2: Green advertisements depicting nature imagery will lead to a positive brand attitude and brand’s ecological image compared to ads with no nature image.
Effects of green claims in combination with nature imagery
Not only the separate effects of claims and imagery are important to be detected but also and more importantly the combinational effects of both of them on consumer’s evaluation of a brand. Research showed that images in advertisements can serve as support and reinforcement of stated claims (Joireman, Liu & Kareklar, 2016) which leads to more positive outcomes. A combination of emotional nature imagery and a verbal claim which appeals to environmental consciousness lead to stronger attitude formations than using only imagery or claim. This can be assigned to the interaction of cognitive and emotional processes when forming attitudes towards green products (Hartmann & Apaolaza-Ibanez, 2005). Furthermore, ads containing environmental claims in combination with nature imagery show a higher degree on brand attitude than ads with no nature imagery (Hartmann & Apaolaza-Ibanez, 2009). In the focus of this study it is expected that nature imagery can serve as a justification of the vague claim (Joireman et al., 2016) and reduce the before mentioned potential negative effects on
consumer’s brand attitude and brand’s ecological image.
The aim of this research is to find results with regards to the reciprocal influences of green advertising claims and nature imagery. In other words that means that the effect of greenwashing of the vague claim is higher (more positive) which would result in a more positive brand’s ecological image and brand attitude when presenting a vague claim with nature imagery than presenting a vague claim without such an imagery.
H3: The combination ‘vague claim and nature imagery’ leads to a more positive brand attitude and brand’s ecological image compared to ‘vague claim and no nature imagery’.
The role of green involvement of the consumer
In general, consumers differ in their responses to advertising appeals. The individual level of (green) involvement is a core variable in advertising research and is also important when studying specifically green advertisements (Matthes et al., 2014). Involvement also plays an essential role in the Elaboration Likelihood Model (ELM) (Petty & Cacioppo, 1986). Thus green involvement is therefore expected to play a moderating role in this research. In the current literature a difference in the operationalization of the concept of green involvement is manifested. Green involvement can e.g. be defined as past environmental behaviors and perceived consumer efficacy (Tucker et al., 2013) or consumer knowledge about
environmental issues (Parguel et al., 2015). This study draws from the operationalization of Matthes et al. (2014) who emphasized three essential conceptualizations to define and measure green involvement. They claimed that this complex variable consists of three sub-concepts: environmental concern, (previous) green purchase behavior and attitude towards green products (Matthes et al., 2014).
The ELM differs between high and low involved (green) consumers. People elaborate differently on message information. Due to their ability and motivation high involved people process on the so called central route of persuasion which means forming attitudes based on arguments presented. Low involved people on the other hand follow a peripheral route of persuasion which means relying more on heuristics presented such as e.g. imagery (Petty & Cacioppo, 1986). With regards to specific green claims the ELM suggests an attitude formation based on the presented arguments for high involved consumers which means following a central route elaboration. Low involved people are lacking the ability and/or motivation to process centrally (Petty & Cacioppo, 1986). Thus they are not able to process
specific green claims and rely on heuristics presented in the advertisement such as imagery. As stated before, nature imagery has a positive effect compared to no nature imagery. For green ads with specific claims we therefore hypothesize:
H4a: For low involved people the combination ‘specific claim and nature imagery’ leads to a more positive brand attitude and brand’s ecological image than ‘specific claim and no nature
imagery’, whereas for high involved people there is no difference.
Outcomes that the ELM suggests with regards to vague green advertising claims are that high involved consumers are more likely to identify the vague green claim based on the central processing. Due to high involved consumer’s ability and motivation to process centrally they experience a mismatch between their high available processing resources and resources actually needed to process a vague claim and therefore form a negative ad judgment (Meyers-Levy & Malaviya, 1999). Thus, when presented with a vague claim they are
suspected to form a more negative brand’s ecological image and brand attitude than low involved consumers. Furthermore, the ELM also suggests that messages that are overly vague can cause a message elaboration on the peripheral route for high involved people as well. These very vague claims can reduce a subject’s ability to process issue-relevant arguments (Petty & Cacioppo, 1986). Thus very vague claims can drive high involved people to a peripheral message elaboration.
Low involved consumers will most likely rely on heuristics of the advertisement presented. They do not have motivation and ability to process centrally. It can be assumed that for them a vague claim can serve as a heuristic cue and therefore lead to a more positive outcome than for highly involved people. Low involved consumers who process mainly via the peripheral route and rely more on heuristics than on arguments presented in claims are positively influenced by (nature) imagery and additionally the vague claim.
In conclusion that means that low involved people rely on heuristics such as the vague claim and nature imagery together. High involved people on the other hand are motivated to process presented arguments centrally but caused by the vague claim they rely on the peripheral route as well and form judgments based on the presented imagery. In contrast to low involved people only the presented imagery will serve as a heuristic cue for them leading to a more negative brand attitude and brand’s ecological image compared to low involved people. Thereby a presentation of a vague claim with nature imagery is still more persuasive than with no nature imagery.
This research aims to find out if executional elements of a greenwashed ad such as imagery evoking nature can weaken the negative effect of vague claims so that also high involved consumers can be ‘tricked’ by greenwashed advertising. Hypothesis 4 of this research is formulated as follows:
H4b: The combination ‘vague claim and nature imagery’ leads to a more positive brand attitude and brand’s ecological image compared to ‘vague claim and no nature imagery’.
Low involved people thereby hold a more positive brand attitude and brand’s ecological image compared to high involved people.
Methods
Design
This study featured a 2 (claim: vague vs. specific) x 2 (nature image: present vs. absent) x 2 (green involvement: low vs. high) between-subjects factorial design with brand attitude and brand’s ecological image as the dependent variables. Claim type and image type were the manipulated independent variables, green involvement the moderator of this study. The rejection level for all statistical analyses was anchored at p = .05.
Participants
412 people participated in the online experiment. After deleting 115 participants due to missing values the sample consisted of 297 participants. Outliers were conducted and the final sample size consisted of 277 participants, 114 male (41%) and 163 female (59%). The average age was M = 28 (SD = 8.05). Using a convenience sampling method the participants were recruited through the researcher’s social media network of Facebook, LinkedIn and Twitter as well as via the offline network. The participants showed demographic differences regarding country of origin and educational level. 59.20 % were German, 18.10 % Dutch, 3.2% Austrian, 3.2% American, 2.5% Swiss and 13.7% from other countries. The majority of the participants were high educated and held either a Bachelor degree (36.1%), a Master degree (40.1%) or a Doctorate degree (2.5%). No compensation for participation was offered.
Stimulus Materials
This research focuses on potential greenwashing techniques/tools with regards to a company’s advertisements. A focus on the product was selected because product-oriented green claims are most common according to Leonidou et al. (2011) who analyzed green claims that appeared in The Economist between 1988 and 2007. The American brand Seventh
Generation was chosen to be the stimulus brand. With regards to the product the researchers
decided to present laundry detergent because this is a product everyone needs and potentially also buys, it is a low involvement product because of its low price and the product category often uses green advertising appeals and has environmental relevance which support realism of the stimulus material (Schuhwerk & Lefkoff-Haguis, 1995; Tucker et al., 2012).
Respondents were expected to be mainly from the EU region thus the researchers chose for the American brand Seventh Generation with regards to keeping brand familiarity low but at the same time a relatively real “look and feel” of the product. All created stimuli
Generation. The position of the product in the ad was kept constant. The stimulus ads differed
with regards to the advertising claim (vague vs. specific) and the background image (nature image vs. no nature image). This resulted in four different advertisement conditions. The ads were created by the researcher with Photoshop (see Appendix B for the stimuli). The nature imagery was chosen because lush green vegetation and clear water leads to the most favorable emotional responses (Matthes et al., 2014). For the no nature imagery conditions we decided for a green background because this research aims at finding differences in the effectiveness of green advertisements and does not strive to compare green and non-green appeals.
Therefore it was important to present all stimuli in a green appeal.
As mentioned before this study will focus on the greenwashing technique of vagueness transported through imagery and vague green claims. The specific/strong product claim was taken from a previous study and consisted of arguments that were actually used in real laundry detergent advertisements (Matthes et al., 2014). This claim can be considered as specific or strong since it delivers evidence why the product is environmentally friendly (Tucker et al., 2012). Please see Appendix B for the claims. When choosing for the vague claim the decision was based on former study results. For the vague claim we chose for “Seventh Generation. 100% environmentally friendly”. Environmentally friendly was rated as a weak argument based on a pretest in a previous study (Spack et al., 2012) and weak
arguments can be seen as vague because they are lacking explanation of the claim.
Procedure
The experiment was conducted online. Participants had to click on a link to the Qualtrics survey where they first had to declare their consent to the participation in the experiment. After that they were presented with one of the four manipulations of this study (see Appendix B). The survey with the measurement of the main concepts of this research followed. Demographics were positioned at the end of the survey to avoid potential influences
on responses. After completing the survey the participants were thanked and the experiment was finished.
Measures
Brand attitude. The dependent variable brand attitude was measured with the help of a five-item semantic differential scale. An example item was 1 ‘bad’ – 7 ‘good’ (Matthes et al., 2014). See Appendix A for all items. A principal components factor analysis showed that the scale loaded uni-dimensional with only one component with an eigenvalue above 1 (eigenvalue 3.53), explaining 71% of the variance. The scale showed a very good reliability indicated through a Cronbach’s Alpha of .89 (M = 5.39, SD = .91).
Brand’s ecological image. The dependent variable brand’s ecological image was assessed using a three-item scale of a previous study (Parguel et al., 2015). An example item was “The brand Seventh Generation is concerned with respect for the environment” (1 ‘totally disagree’ – 7 ‘totally agree’). See Appendix A for all items. The explanatory factor analysis indicated that the scale was uni-dimensional and only one component has an eigenvalue above 1 (eigenvalue 2.16), explaining 71.84% of the variance. The Cronbach’s Alpha of .80 indicated a good reliability of the three-item scale (M = 5.80, SD = .85).
Green involvement. This study’s moderator is green involvement. As suggested by Matthes et al. (2014) green involvement has three dimensions: environmental concern, attitudes towards green products and green purchase behavior, each of them measured with a four-item scale ranging from 1 ‘totally disagree’ to 7 ‘totally agree’. See Appendix A for all items. Through measuring these three subconcepts an overall green involvement score for each participant was calculated, resulting in low and high green involvement. Thus factor analysis and reliability test were conducted on all items of the three subconcepts together. The explanatory factor analysis indicated that two items of attitude towards green products loaded
double. Both items had an eigenvalue above 1 (eigenvalue 5.95 and eigenvalue 1.36) explaining 49.55% and 11.03% of the variance. Thus they have been excluded from the overall scale. That left ten items for the calculation of the green involvement scale. For the final items please see Appendix A (deleted items are marked). The Cronbach’s Alpha of .90 (M=5.29,SD = .93) indicated a very good reliability of the overall scale of green
involvement. Lastly a median split was performed in order to create two groups of
participants (low green involved people vs. high green involved people). This was necessary to test Hypothesis 4 on potential moderation effects of green involvement.
Manipulation checks
Manipulation checks were included in the survey questions to check whether
participants perceived the claims (vague vs. specific) as intended to. It was decided against a manipulation check for the manipulated factor imagery since it was considered as too obvious and could have confused participants. To check for the claims two independent samples t-test were performed. Results indicated a significant difference between the two conditions of TypeClaim on both independent variables. The mean of the “manipulation check variable specific claim” of the specific claim (M =5.53, SD = 1.03) was significantly different to the vague claim condition (M = 3.72, SD = 1.60, t (275) = 11.32, p =.00). Secondly the mean of the “manipulation check variable vague claim” of the specific claim condition (M =3.04, SD =1.67) was significantly different to the vague claim condition (M =5.60, SD =1.46, t (275) = -13.58, p =.00). Thus the manipulation of the experiment was successful and vague claims were indeed perceived as vague and specific claims as specific.
Results
Randomization Check
To test the distribution of age and education between the conditions two-way ANOVA were conducted. The test revealed that none of the conditions had a significant effect on education. Neither type of image (nature image vs. neutral green image) F(1, 273) = .00, p =.98, η2 = .00 nor type of claim F(1, 273) = .23, p =.63, η2 = .00 nor the interaction of both factors F(1, 273) = .01, p =.94, η2 = .00 showed a significantly difference on mean education. There was no main effect of TypeImage on age F(1, 273) = .00, p =.98, η2 = .00 nor of TypeClaim F(1, 273) = .22, p =.63, η2 = .00 on age. To test for random assignment of gender two Chi-Square tests were conducted. Results showed that 79 females (82.40%) saw a neutral green image, 84 females (80.60%) a nature image. No significant difference resulted between the two
percentages within female participants, (2) =1.55, p =.460. Type Image condition was also equally distributed among male participants. 60 males (57.1%) were exposed to neutral green imagery, 53 males (55.9%) to nature imagery, with no significant difference between the conditions (2) =1.55, p =.460. Thus, the randomization of gender, age and education was successful and none of the tested variables above needed to be included as a covariable in the main analysis.
Testing of Hypotheses
Before analyzing the Hypotheses outliers were excluded which resulted in a sample size of N = 277. To test the first two Hypotheses independent sample t-tests were conducted, two per Hypothesis, one for each DV (brand attitude and brand’s ecological image) with claim type (vague green claim vs. specific green claim) and type of imagery (nature imagery vs. no nature imagery) as the independent variables. Hypothesis 1 claimed that vague claims have a negative effect on the brand attitude and brand’s ecological image compared to
specific claims. Results indicate a significant difference between the two conditions. The mean brand attitude of subjects who saw the specific green claim condition (M = 5.54, SD = .95) was significantly different to subjects who were exposed to the vague green claim condition (M = 5.23, SD = .85, t (275) = 2.85, p = .005). Additionally, results showed a significant difference between the two conditions on brand’s ecological image as well. The mean brand’s ecological image of subjects who saw the specific green claim (M = 5.97, SD = .80) was significantly different to the mean of subjects who were exposed to the vague green claim conditions (M = 5.60, SD = .87, t (270) = 3.83, p = .00). Hypothesis 1 is accepted.
Hypothesis 2 claimed that nature imagery in green advertisements lead to a positive brand attitude and brand’s ecological image compared to no nature imagery. Results of the independent samples t-test indicate no significant difference between the two conditions. No nature image (M = 5.29, SD =.86) showed no significant difference on the mean brand attitude compared to the nature imagery (M = 5.49, SD = .95, t (275) = -1.77, p = .08). The mean brand’s ecological image was not significantly different between participants that saw no nature imagery (M = 5.81, SD = .84) and participants that were exposed to nature imagery (M = 5.79, SD = .88, t (275) = .28, p =.78). Hypothesis 2 is therefore declined.
Hypothesis 3 claimed that a combination of a vague green claim with nature imagery leads to a higher brand attitude and brand’s ecological image of the participants compared to a presentation of vague claim and no nature imagery. Two two-way ANOVA with TypeImage and TypeClaim as the independent variables and brand attitude or brand’s ecological image as the dependent variable were conducted. Results indicate a main effect of TypeClaim on brand attitude F(1, 273) = 7.73, p =.006, η2 = .03. Neither the main effect of TypeImage F(1, 273) = 3.00, p = .084, η2 = .01 nor the interaction effect of TypeClaim and TypeImage was
significant F(1, 273) = .86, p =.35, η2 = .00. With regards to the dependent variable brand’s ecological image results are similar. TypeClaim showed a significant main effect on brand’s
ecological image F(1, 273) = 14.85, p =.00, η2 = .05. Neither the main effect of TypeImage
F(1, 273) = .20, p =.66, η2 = .00 nor the interaction effect of TypeClaim and TypeImage were
significant F(1, 273) = 1.01, p =.32, η2 = .00. Thus Hypothesis 3 is rejected.
Hypothesis 4a deals with specific claims and the effects of imagery and green involvement on brand attitude and brand’s ecological image in these conditions. Before running the analysis for Hypothesis 4a we selected cases that were exposed to the specific claim conditions (specific claim and nature image vs. specific claim with no nature image). Thus a sample size of n =143 resulted, 81 female (56.6%) and 62 male (43.4%). Two two-way ANOVA with TypeImage and green involvement as independent variables and brand attitude and brand’s ecological image as dependent variables were run. The analysis of brand attitude revealed no main effect of TypeImage F(1, 139) = .29, p =.59, η2 = .00 nor an
interaction of TypeImage and green involvement F(1, 139) = 1.49, p =.23, η2 = .01. But there was a significant main effect of green involvement on brand attitude F(1, 139) = 15.90, p =.00, η2 = .10. The effect size of η2 = .10 indicates a medium effect size. The second two-way ANOVA with brand’s ecological image as the dependent variable revealed no significant main effect of TypeImage F(1, 139) = .17, p =.68, η2 = .00 or of green involvement F(1, 139) = .37, p =.55, η2 = .00 nor an interaction effect of both variables F(1, 139) = .46, p =.50, η2 = .00.
Due to the significant main effect of green involvement on brand attitude further independent sample t-tests of the three sub-concepts of green involvement (environmental concern, attitude towards green products and green purchase behavior) were conducted to reveal deeper insights in what might have caused this effect. Results of environmental
concern revealed no significant difference in the mean brand attitude for low (M = 5.42, SD = .85) and high environmental concern (M = 5.66, SD = 1.03, t (141) = -1.48, p = .14). On the other hand there was a significant difference in the mean brand attitude between low attitude
towards green products (M = 5.08, SD =.86) and high attitude towards green products (M = 5.93, SD = .84, t (141) = -6.00, p = .00). Furthermore there was a significant difference in mean brand attitude between low green purchase behavior (M =5.30, SD =.92) and high green purchase behavior (M = 5.78, SD = .91, t (141) = -3.09, p = .00).
Hypothesis 4b stated that the effect between vague claims and imagery (nature image vs. no nature image) is moderated by the level of green involvement of the participant. Before running analysis for Hypothesis 4b we only selected cases that were exposed to a condition with a vague claim (vague claim and nature imagery vs. vague claim and neutral green imagery). Thus, the sample size was decreased to n = 134. Two two-way ANOVA with TypeImage and green involvement as the independent variables and brand attitude or brand’s ecological image as the dependent variable were conducted. The analysis of brand attitude revealed no significant main effect of TypeImage F(1, 130) = 2.72, p =.10, η2 = .02 or green involvement F(1, 130) = 3.48, p =.06, η2 = .02 and no significant interaction effect of both factors F(1, 130) = 2.28, p =.13, η2 = .02. The second two-way ANOVA with brand’s ecological image as the dependent variable revealed no significant main effect of TypeImage
F(1, 130) = 1.45, p =.23, η2 = .01 or of green involvement F(1, 130) = 2.04, p =.16, η2 = .02.
Furthermore, there was no significant interaction effect of the independent variables F(1, 130) = .79, p =.38, η2 = .01 on brand’s ecological image. H4b is therefore rejected.
Because the green involvement moderator of this study consisted of three sub-concepts (environmental concern, attitude towards green products and green purchase behavior) we decided to undertake further analysis to investigate the separate effects of each sub-concept solely and in interaction with TypeImage on brand attitude and brand’s
ecological image. Only cases which were exposed to the vague claim conditions were selected. Furthermore, it is important to mention that the previously extracted two items of attitude towards green products have not been excluded for the further analysis of green
involvement since we are interested in the separate effects of the sub-concepts. All items of attitude towards green products loaded uni-dimensional with only one item loading above 1 (eigenvalue 3.09) and explaining 77.45% of the variance.The reliability of the attitude towards green products scale with a Cronbach’s Alpha of .90 can be considered as high (M = 5.70, SD = 1.06).
First effects on the dependent variable brand attitude were analyzed. We conducted three two-way ANOVA, one for each sub-concept of green involvement. Analysis revealed a significant interaction effect of green purchase behavior and TypeImage on brand attitude
F(1, 130) = 5.36, p =.02, η2 = .04. There were no main effects of TypeImage or of green
purchase behavior. Simple effect analysis on TypeImage revealed that in the nature imagery condition no significant differences between low and high green purchase behavior existed (all p > .34). In the no nature image condition a significant difference between low (M = 4.91,
SD = .71) and high (M = 5.39, SD = .83, p = .02) green purchase behavior was detected. There
was a significantly different mean brand attitude between low and high purchase behavior participants. A second two-way ANOVA of TypeImage and attitude towards green products on brand attitude was conducted. Results indicated no significant interaction effect of attitude towards green products and TypeImage on brand attitude. There was no significant main effect of TypeImage but there was a significant effect of attitude towards green products F(1, 130) = 18.80, p =.00, η2 = .13 on brand attitude. The third two-way ANOVA of TypeImage and environmental concern revealed no main effects on brand attitude, nor an interaction effect of the independent variables (see Appendix C, Table 1 for all ANOVA results).
A two-way ANOVA of TypeImage and green purchase behavior on brand’s ecological image was conducted. Results indicated no significant interaction effect, nor main effects of green purchase behavior and TypeImage on brand’s ecological image. Furthermore, there was no main effect of TypeImage F(1, 130) = .56, p =.46, η2 = .00 but of attitude towards green
products F(1, 130) = , p = .03, η2 = .04. Results indicated no significant interaction effect. With regards to environmental concern, no main effects or interaction effect of environmental concern and TypeImage were indicated (see Appendix C, Table 2 for all ANOVA results).
To sum up, the more elaborate inspection of the moderator green involvement revealed that brand attitude differs with regards to green purchase behavior and advertising imagery. Furthermore, a significant main effect of attitude towards green products on a brand’s ecological image was revealed. Please note that these results are valid in the vague claim condition.
Discussion & Conclusion
The goal of this study was to analyze effects of green claims (vague vs. specific) and imagery (nature vs. no nature) of green advertisements on consumer’s brand attitude and brand’s ecological image and if potential effects differ depending on the level of green involvement of the consumer (low vs. high). As the study strived to identify greenwashing techniques that could be used in advertisements to persuade people, a special focus was set on advertising imagery as well as on vague claims.
As expected, results indicated that vague claims lead to a more negative brand attitude and brand’s ecological image compared to specific claims. This finding is in line with the current knowledge about effectiveness of specific claims in green advertisements (e.g. Tucker et al., 2012; Hartmann & Apaolaza-Ibanez, 2009). Furthermore, this supports the statement that specific claims are more persuasive than vague claims and delivers more evidence of the persuasiveness of specific claims as manifested in the current literature. It can be argued that vague claims are more difficult to process and visualize and might appear as unclear
compared to specific claims (Joireman, Liu & Karklar, 2016). Furthermore, research showed that recipients of advertisements are more persuaded when advertisements deliver evidence of
the claimed advantages of e.g. a product (Hoch & Ha, 1986). This research shows that consumers might not be as easy to trick and capable to identify vague misleading claims.
Secondly, and in contrast to what was expected, there was no difference in people’s brand attitude and brand’s ecological image between the conditions nature imagery and no nature imagery. Previous studies found that nature imagery indeed has a positive influence on brand attitude (Hartmann et al., 2005; Hartmann & Apaolaza-Ibanez, 2009; Hartmann et al., 2013). Results of this study are not in line with previous findings and could not identify differences. Thus imagery in green advertisements did not influence a consumer’s judgment about brand attitude and brand’s ecological image of the advertised brand. An explanation for this can be found in the presented stimuli. Previous studies used nature imagery in contrast to no nature imagery in the sense that they used images that cannot be considered as green at all (e.g. a white background or urban sceneries) (Hartmann et al., 2005; Hartmann & Apaolaza-Ibanez, 2009; Hartmann et al., 2013). In other words differences between nature imagery and no nature imagery of previous studies might be attributed to the fact that these studies
compared green advertisements to non-green advertisements. Thus the detected positive effects might have been triggered by the overall green appeal and not by the use of nature imagery per se. This study compared nature images to green images because the aim was to find differences in between green advertisement’s executional elements. Thus green appeals in all conditions were necessary to draw adequate conclusions. The findings of claim and image effectiveness suggest that firms should focus more on stating specific and substantial claims and less on advertising imagery.
Thirdly, also the claimed positive effects of nature imagery in combination with vague claims were insignificant. It was not found that the negative outcomes of vague claims (Davis, 1993) could be overshadowed by nature imagery. This could be attributed to the fact that nature images might be indeed seen as vague claims as suggested by previous studies
(Hartmann & Apaolaza-Ibanez, 2009) and be ineffective instruments in persuading the consumer of environmental friendliness of a brand. This also underlines the findings that claims or arguments are in general more persuasive than images (Tucker et al., 2012).
With regards to the role of green involvement a main effect on brand attitude was observed but only when people were exposed to the specific claim. People with high green involvement showed a higher brand attitude compared to people with low green involvement. A look closer identified that higher attitudes towards green products or a high green purchase behavior were the factors leading to higher brand attitudes. Environmental concern did not play a role. This supports findings of previous research of Matthes et al. (2014): In functional ads (simultaneous to the specific claim conditions of this research) Matthes et al. (2014) found that brand attitude differed depending on green purchase behavior and attitude towards green products. Furthermore, the finding is in line with what the ELM expects. Highly involved people follow a central processing in which they form brand attitudes based on the specific claim. Thus their brand attitude is more positive compared to low involved people who most likely did not have the motivation and/or ability to process centrally and thus formed their attitudes not based on the specific claims but on heuristics like imagery (Petty & Cacioppo, 1986). Thereby it does not matter if the participants saw a nature image or no nature image, no effects of imagery on brand attitude or a brand’s ecological image were found in this study.
But more of interest for this study were vague claims. In contrast to the findings of specific claims, overall green involvement did not have an influence on the effectiveness of vague claims and images on brand attitude and brand’s ecological image. A possible
explanation is that this study did not measure green brand attitude specifically, but the general brand attitude of participants (see Appendix A for items). While the moderator and sub-concepts measured green involvement with environmental issues it could be possible that it
did not influence general brand attitudes but green brand attitudes. Future research is advised to measure green brand attitude in specific to detect potential moderation roles of green involvement. A further explanation could be that the mean overall green involvement (M= 5.29,SD =.93) as well as the mean green involvement of participants who saw a vague claim
(M =5.31, SD=.90) was rather high. This supports research findings that documented a shift to an overall high involvement in environmental issues in today’s society (Chen, 2010).
Considering that this study did not find any differences in mean brand attitude and brand’s ecological image between low and high involved consumers indicates that they are equally persuaded by vague claim conditions.
Taking into account that the mean green involvement of participants was rather high it raises the question if it is still relevant at all to research moderating roles of green
involvement of consumers or if research should e.g. focus more on skepticism among highly involved consumers since skepticism towards green advertising was found to be high among people who are interested in buying green (Hartmann & Apaolaza-Ibanez, 2009). This is a highly relevant topic since skepticism lessens the positive effects of communication on consumers (Finisterra do Paco & Reis, 2012).
With regards to further analysis of the green involvement sub-concepts this study found evidence that the brand’s ecological image of people who were exposed to a vague claim was influenced by their attitudes towards green products. It can be assumed that people who hold positive attitudes towards green products are not as negatively influenced by vague claims as people who hold more negative attitudes towards green products. For the latter one vague claims would rather justify their negative attitudes and in turn lead to a more negative brand attitude. Furthermore, green purchase behavior played a role. In combination with no nature imagery a high green purchase behavior had a positive effect on brand attitude. It can be assumed that people who have bought green products before hold positive brand attitudes
towards a green brand. The findings that attitudes towards green products and green purchase behavior are the variables that influence brand attitude and brand’s ecological image are in line with previous research (Matthes et al., 2014). As Matthes et al. (2014) this study also did not find any influence of environmental concern.
To sum up, with regards to practice this study could give valuable insights into green advertising execution. The focus on claims and images was relevant to research since
companies make frequent use of images that can support their stated claims (Joireman, Liu & Kareklas, 2016). Nevertheless, it is advised that marketers should focus on advertising claims and a specific formulation of those since they were proven to be more persuasive compared to vague claims which is in line with previous research results. Nature imagery on the other hand was not found to influence a consumer’s brand evaluation. Thus this study does not support the assumption that nature imagery might also have misleading effects in the sense that they mislead consumers to evaluate a brand more positively compared to no nature imagery. The same is valid for vague claims. Specific claims were more persuasive than vague ones. Thus also the interaction of nature image and vague claims did not have the potential to mislead consumer. In the introduction of this paper it was stated that green advertising regulations mainly focused on regulations concerning advertising claims. This study delivers evidence that justifies this focus and does not back up some voices expressing a need for imagery regulations. Furthermore, expected differences between low and high involved green consumer were only found in specific claim conditions.
Limitations
Besides contribution to the existing body of research of green advertising this study was executed with some limitations. One limitation of this study is the poor generalizability to other populations due to the sample which consisted mainly of Bachelor and Master students (76.2%) and German people (59.20%). In addition, demographics of participants of this study
are typical for people who show high green involvement: young, well-educated and urban (Mainieri et al., 1997). This is a potential factor that might have influenced results of this study concerning the moderating influence of green involvement as explicated in the
discussion of this paper. Furthermore, this study was not conducted in a real life setting which excluded potential influences affecting advertising effects. The medium through which an ad is transported or the context were proven to be important factors that influence advertisement processing, just to name two (Meyers-Levy & Malaviya, 1999). Furthermore, the calculation of the green involvement scale can be seen as a limitation. It was constructed of three sub-concepts (attitudes towards green products, green purchase behavior and environmental concern). Hereby it should be mentioned that green purchase behavior measured an actual (previous) behavior in contrast to attitudes measured by attitude towards green products and environmental concern. Nevertheless, after excluding two items of attitudes towards green products the overall scale of green involvement loaded uni-dimensional. Furthermore, this research was limited to a low involvement product category, laundry detergent. Future research is advised to focus on different product categories e.g. high involvement products such as cars.
References
Baum, L.M. (2012). It’s not easy being green…Or is it? A content analysis of environmental claims in magazine of the United States and the United Kingdom. Environmental
Communication, 6(4), 423-440.
Bickart, B.A. & Ruth, J.A. (2012). Green Eco-Seals and Advertising Persuasion. Journal of
Advertising, 41(4), 51-67.
Carlson, L., Grove, S.J., Kangun, N. & Polonsky, M.J. (1996). An international comparison of environmental advertising: substantive versus associative claims. Journal of
Marketing, Fall, 57-68.
Chang, C. (2011). Feeling ambivalent about going green. Implications for green advertising processing. Journal of Advertising, 40(4), 19-31.
Chen, Y.S. (2010). The Drivers of Green Brand Equity: Green Brand Image, Green Satisfaction and Green Trust. Journal of Business Ethics, 93(2), 307-319. Chen & Chang (2013). Greenwash and Green Trust: The Mediation Effects of Green
Consumer Confusion and Green Perceived Risk, Journal of Business Ethics, 114(3), 489-500.
Davis, J.J. (1993). Strategies for environmental advertising. Journal of Consumer Marketing, 10(2), 19–36.
Delmas, M.A. & Burbano, V.C. (2011). The drivers of greenwashing. California
Management Review, 54(1),64-87.
Fernando, A.G., Sivakumaran, B. & Suganthi, L. (2014). Nature of green advertisements in India: Are they greenwashed?, Asian Journal f Communication, 24(3), 222-241. Finisterra do Paco, A.M. & Reis, R. (2012). Factors Affecting Skepticism toward Green
Advertising, Journal of Advertising, 41(4), 147-155.
versus emotional positioning strategies. Marketing Intelligence & Planning, 23(1), 9-29.
Hartmann, P. & Apaolaza-Ibanez, V. (2009). Green advertising revisited: Conditioning virtual nature experiences. International Journal of Advertising, 28(4), 715-739.
Hartmann, P. & Apaolaza-Ibanez, V. (2010). Beyond savanna: An evolutionary and
environmentally psychology approach to behavioral effects of nature scenery in green advertising, International Journal of Advertising, 30, 119-12.
Hartmann, P., Apaolaza, V. & Alija, (2013). Nature imagery in advertising. International
Journal of Advertising, 32(2), 183-210.
Hoch, S.J. & Ha, Y.-W. (1986). Consumer Learning: Advertising and the Ambiguity of Product Experience. Journal of Consumer Research, 13(2), 221-233.
Joireman, J., Liu, R.L. & Kareklas, I. (2016). Images paired with concrete claims improve skeptical consumer’s responses to advertising promoting a firm’s good deeds. Journal
of Marketing Communications, 2016, 1-20.
Kangun, N., Carlson L. & Grove, S.J. (1991). Enviromental advertising claims. A preliminary investigation. Journal of Public Policy and Marketing, 10(2), 47-58.
Leonidou, L. C., Leonidou, C.N., Palihawadana, D. & Hultman, M. (2011). Evaluating the Green Advertising Practices of International Firms: A Trend Analysis. International
Marketing Review, 28 (1), 6–33.
Mainieri, T., Barnett, E.G., Valdero, T.R., Unipan, J.B. & Oskamp, S. (1997). Green Buying: The Influence of Environmental Concern on Consumer Behavior. The Journal of
Social Psychology, 137(2), 189-204.
Manrai, L., Manrai, A., Lascu, D.-N. & Ryans, J. (1997). How Green-Claim Strength and Country Disposition Affect Product Evaluation and Company Image, Psychology and
Marketing, 14(5), 511-537.
persuasive effects of emotional versus functional ads. Journal of Business Research, 67, 1885-1893.
Meyers-Levy, J. & Malaviya, P. (1999). Consumers’ Processing of Persuasive
Advertisements: An Integrative Framework of Persuasion Theories. Journal of
Marketing, 63, 45-60.
Ng, P.F., Butt, M.M., Khong, K.W. & Ong, F.S. (2014). Antecedents of Green Brand Equity: An Integrated Approach. Journal of Business Ethics, 121, 203-215.
Obermiller, C. (1995). The baby is sick / the baby is well: A test of environmental communication appeal. Journal of Advertising, 24(2), 55-70.
Parguel, B., Benoit-Moreau, F. & Russell, C.A. (2015). Can evoking nature in advertising mislead consumers? The power of ‘executional greenwashing’. International Journal
of Advertising, 34(1), 107-134.
Petty, R.E. & Cacioppo (1986). The Elaboration Likelihood Model of Persuasion. In: Petty, R.E. & Cacioppo, J.T.. Communication and Persuasion. New York: Springer. Schuhwerk, M.E. & Lefkoff-Hagius, R. (1995). Green or Non-Green? Does Appeal Matter
When Advertising a Green Product?, Journal of Advertising,
Spack, J.A., Board, V.E., Crighton, L.M., Kostka, P.M. & Ivory, J.D. (2012). It’s Easy Being Green: The Effects of Argument and Imagery on Consumer Responses to Green Product Packaging. Environmental Communication, 6(4), 441-458.
Terrachoice (2010). The sins of greenwashing. Home and family edition. Retrieved from http://sinsofgreenwashing.com/index35c6.pdf.
Tucker, E.M., Rifon, N.J., Lee, E.M. & Reece, B.B. (2012). Consumer receptivity to Green Ads. Journal of Advertising, 41(4), 9-23.
Appendix A Scales
Brand attitude
bad – good, unattractive – attractive, negative –positive, not likeable – likeable and not recommendable – recommendable.
Brand’s ecological image
The brand XYZ is concerned with respect for the environment.
I have the impression that the brand XYZ tries to respect the environment. XYZ is environmentally friendly.
Green involvement
Environmental concern
I am concerned about the environment.
The condition of the environment affects the quality of my life. I am willing to make sacrifices to protect the environment. My actions impact the environment.
Attitude towards green products
I like green products.
I feel positive towards green products.
Green products are good for the environment. (excluded item for green involvement scale) I feel proud when I buy/use green products. (excluded item for green involvement scale)
Green purchase behavior
I make a special effort to buy precuts in biodegradable packages.
I would switch from my usual brands and buy environmentally safe cleaning products, even if I have to give up some cleaning effectiveness.
I have switched from products for ecological reasons.
When I have a choice between two equal products, I purchase the one less harmful for the environment.
Appendix B
Stimuli
Specific claim and no nature image
Specific claim text: Seventh Generation.
- plant-based ingredients from sustainable sources - phosphate-free, no optical brighteners or chlorine - not tested on animals
- 100% recyclable PlantPlastic bottles from sugarcane - Biodegrades quickly and completely after use - Minimum impact on aquatic life
Vague claim and no nature image
Appendix C Table 1.
Two-way ANOVA results of green involvement sub-concepts on brand attitude.
Sum of Squares
df Mean
Square
F p η2
Two-way ANOVA TypeImage and green purchase behavior
TypeImage 2.02 1 2.02 2.99 .09 .02 Green purchase behavior .59 1 .59 .86 .35 .01 TypeImage* Green purchase behavior 3.68 1 3.68 5.36 0.22 .04 Error 87.94 130 .68 Total 95.03 134
Two-way ANOVA TypeImage and attitude towards green products
TypeImage 2.46 1 2.50 3.98 .05 .03 Attitude towards green products 11.62 1 11.62 18.80 .00 .13 TypeImage* Attitude towards green products .17 1 .17 .28 .60 .00 Error 80.32 130 .62 Total 95.03 134
Two-way ANOVA TypeImage and environmental concern
TypeImage 2.80 1 2.80 3.98 .05 .03 Environmental concern .79 1 .79 1.13 .29 .01 TypeImage* Environmental concern .00 1 .00 .00 .97 .00 Error 91.48 130 .70 Total 95.03 134
Table 2.
Two-way ANOVA results of green involvement sub-concepts on brand’s ecological image.
Sum of Squares df Mean Square F p η2 Two-way ANOVA TypeImage and green purchase behavior
TypeImage .76 1 .76 .99 .32 .01 Green purchase behavior .04 1 .04 .05 .82 .00 TypeImage* Green purchase behavior .10 1 .10 .14 .71 .00 Error 98.81 130 .76 Total 99.66 134
Two-way ANOVA TypeImage and attitude towards green products
TypeImage .41 1 .41 .56 .46 .00 Attitude towards green products 3.50 1 3.50 4.79 .03 .04 TypeImage* Attitude towards green products .63 1 .63 .86 .35 .01 Error 94.96 130 .73 Total 99.66 134
Two-way ANOVA TypeImage and environmental concern
TypeImage .70 1 .70 .92 .34 .01 Environmental concern .48 1 .48 .64 .43 .01 TypeImage* Environmental concern .02 1 .02 .02 .89 .00 Error 98.45 130 .76 Total 99.66 134
