Master Thesis
The Effects of CSR Labeling Certifications on Critical Stakeholder Engagement: A Consumer’s Perspective
Katarzyna Jaskiewicz
Student ID: 10626948
Master Program Corporate Communication Graduate School of Communication
University of Amsterdam
Supervisor: Anker Wonneberger 27/06/2014
Abstract
This study aims to address the attitudes of consumers towards CSR labeled products and how these translate into attitudes towards the brands using ethical marketing claims in general. Furthermore, this paper moves away from the widely researched topic of purchase intentions and instead focuses on exploring CSR labels, Fairtrade International (FLO) in particular, as triggers of curiosity. Therefore, the main purpose of this paper is to demonstrate the relationship between ethical marketing and information seeking – and to identify the partial mediation effects of trust and skepticism. A quantitative experiment was conduced with a sample of 95 respondents to explore the relationship between consumers’ attitudes towards brands using ethical marketing claims supported by a CSR label certification (FLO) and consumers’ information seeking intentions and behaviors. The results indicate a significant difference in consumers’ attitudes towards brands using CSR certifications. In the presence of a CSR certification on the product the trust increases and skepticism decreases. However, no significant effect has been found on information seeking. Nevertheless this study offers a point of departure for further research on the effects of labeling on information seeking in the complex environment of ethical marketing. The managerial implications of this research most importantly include a recommendation for companies to make use of external endorsements such as CSR labeling schemes as a part of their long-term marketing strategy in order to increase trust of the consumers in the brand.
Introduction
Corporate Social Responsibility (CSR) has become a globally renowned concept and since a few decades is constantly gaining momentum worldwide. Furthermore, as CSR has become a recognized source of competitive advantage (Chen and Chang, 2012), an increasing number of companies from various business sectors are implementing CSR policies and certifications in hopes of obtaining superior economic performance. For the purpose of this paper a broad definition of Corporate Social Responsibility is adopted, where CSR is conceptualized as company’s role in, and impact upon the economic, environmental and social framework in which it is embedded (Crane, Matten and Moon, 2004). This perspective is also reflected in the ‘triple bottom line’ approach including: people, planet and profit suggesting a threefold nature of CSR activities. Those include healthy social, ecological and financial accounts. Thus, the key to understanding CSR lies in the idea that no company can afford to act in opposition to, or in isolation of, issues in society and alienated from its stakeholders (Matten and Moon, 2005).
Nowadays social environmental responsibility has become a vital management function and the last decade has witnessed a radical change in consumer preferences towards responsible products (D’Souza, Taghian and Lamb, 2006). Hence, different tools for CSR product branding have risen. This paper will focus specifically on CSR labeling certifications. According to Hartlieb and Jones (2009) CSR certified labeling has a two-fold nature. First of all, CSR labels set guidelines for brands in order to reach a superior social and/or
environmental performance. From this perspective, CSR certifications aim at fostering
responsibility backwards on the supply chain (Carrero and Valor, 2012). Secondly, consumers use CSR labels in order to simplify and drastically reduce the costs associated with
information seeking and processing (Chakrabarty and Grote, 2009; Drichoutis, Lazaridis and Nayga, 2006). Therefore, CSR labeling certifications carry a significant potential in helping
consumers to identify brands that score high on the environmental and/or social performance dimension and thus could lead to encouraging responsible practices across the value chain.
Furthermore, currently a vast proliferation of CSR certified products entering the market can be noticed worldwide. This trend has been especially prevalent in the Nordic Countries, where there are ecolabels for 55 product groups and 2800 products (Rex and Baumanm, 2006). Similarly in Japan 64 product groups have criteria established for ecolabels and more than 5000 products have been accepted. Nevertheless, considering the overall market shares of CSR labeled products an alternative picture arises. Today ecolabeled food products in Europe still make only 5% of sales across the continent (Rex and Baumanm, 2006).
However, the food sector has been recognized as the branch where the green market continues to grow when in the other sectors the growth has been discontinued. Thus, a question that rises here is whether this modest market share of CSR labeled products means that the green and/or responsible dimension of the product is an insufficient incentive to choose a certified product over an uncertified one? A vast number of studies addressed the above mentioned question in order to investigate the effect of CSR labeling on purchase intentions of customers. However, hardly any attention has been paid to the effects of CSR labeling on information seeking of customers. Investigating the effects of CSR labels on information seeking, as opposed to the purchase intentions of customers, allows for theorizing on how attitudes translate into critical stakeholder engagement independent of monetary investments. Therefore, this paper moves away from the widely researched topic of purchase intentions and instead focuses on CSR labels as triggers of curiosity. Hence, the main aim of this research is to investigate the relationship between attitudes towards brands triggered by CSR labeled products and the needs of customers to find out more about the processes behind
the products. In order to do so this study takes on an experimental research design in attempt to examine the effects of CSR labels on critical consumer engagement.
Theoretical framework Credibility of CSR labels
In Europe 240 different CSR labels are being used. However, according to Teisl, Roe and Levy (1999) ecolabels do not provide an objective basis for comparing marketing claims. These claims often tend to be vague, impossible to verify and misleading for consumers. This misuse of ecolabels should not be taken trivially. Davies and Wright (1994) also conclude on the basis of their research on UK food labels, that manufacturers should ensure clearer
labeling to correctly inform the customer. They stated that even though consumers are able to locate and process label information, they find it difficult to assess label claims. Szykman, Bloom and Levy (1997) take a step further in theorizing that lack of knowledge and
skepticism towards ethical labels has a significant negative effect on sales of nutrition labeled products. Thus, the main issue concerning the credibility of CSR labels is the asymmetric nature of allocation of information between the sellers and the consumers. Product attributes can be generally divided into search, experience and credence attributes. The consumer cannot judge the later neither before nor after purchasing a product, thus the ethical quality of a product is an example of credence attribute. However, a credible CSR label has the potential to change the allocation of information and thus convert the credence attribute into a search attribute, making it possible for consumers to judge the product before the purchase (Pant and Sammer, 2003). Therefore, increasing the amount of certification proof could lead to raising the credibility of the label by for instance raising the credibility of the label issuer.
Many authors have proven that labels should be easily understandable and contribute to lower information cost by reducing the time to find an ethical product. Thus, they argue
that attracting attention is more significant than conveying information (Singh and Cole, 1993; Zadek, Lingayah and Forstater, 1998). According to Pant and Sammer (2003) too much information may also be counter-productive. Thus, the information provided by the label should be clear and concise. Interestingly enough Mielants, De Pelsmacker and Janssens (2003) found that 50 per cent of consumer in the UK wanted more information on labels and have difficulty in finding information they are looking for. On the other, hand Wansink (2003) theorizes that there is an optimal amount on information on ethical claims on products. Thus, too much information can confuse consumers while too little information has the potential to mislead them. Wansink (2003) studied whether products should only have a label on the front of the package or alternatively add more information at the back of the package. His findings suggest that both are required. Hence, extra information on one hand could be perceived as information overload leading to consumer confusion and skepticism, but on the other hand it could be perceived as a significant attribute leading to consumer trust and confidence in the product.
When discussing CSR labels it is important to distinguish the criteria along which they are classified. These criteria are established on two premises. One is the issue included: environmental, social justice and animal welfare. Second premise is the quality set by the International Organization for Standardization (ISO), which divides the labels into three types. Type I CSR labels are process-based, third-party labeling schemes. Type II CSR labels are self-declarations not backed by a third party. Type III CSR labels are performance-based, third-party labeling schemes (D’Souza, Taghian, Lamb and Peretiatko, 2007). As already mentioned above previous research has shown consumers appreciation for CSR labeled products (D’Souza et al., 2007). However, an attitude-behavior gap is widely recognized by the scholars, where consumers appreciate labels but do not choose labeled-products over non- labeled products at the selling point.
Effects of CSR labels on information seeking
The increased global proliferation of CSR labels and certificates across sectors leads to a paradoxical effect that is embedded in the nature of CSR communication. According to Morsing, Schultz and Nielsen (2008) companies are regularly encouraged to engage in CSR, at the same time being discouraged to communicate about this engagement. Thus, in the context of reputation, being perceived as socially responsible is definitely a challenge for many organizations today. “The challenge is for companies to be perceived as socially
responsible across stakeholders. The difficulty is in how to make it known, and acknowledged by stakeholders, that the company is dedicated to a path of social responsibility, and further, to what extent and how the company deliberately should communicate it.” (Morsing, et al., 2008: 98). Therefore, CSR communication in a form of branding through established labeling certifications on one hand enhances the trust of customers in the product, but on the other hand often leads to consumer skepticism. Thus, the CSR labeling can be understood as a “double edged sword” simultaneously enhancing consumers’ confidence in the brand as well as leaving it vulnerable to critique.
There are multiple reasons underlying the importance of understanding the effects of CSR labeling certifications. However, CSR labeling certifications represent a new and promising trend in the market as they are trying to satisfy the environmental and social expectations of their customers (Blengini and Shields, 2010), at the same time according to Pahl (2003) there is a growing skepticism among consumers about the validity of green product claims. This can be partly explained by the fact that CSR labeling certifications range from policy level government-sponsored programs to claims based on an assessments made by an outside group to corporate self-assessment (Harris, 2007). This paper suggests that CSR labeling both increases the trust of customers in the product as well as their skepticism
triggered by the CSR labels. CSR labeling schemes reduce the knowledge seeking behaviors by customers in the case of trust in the certification (Chakrabarty and Grote, 2009).
Additionally, this paper suggests that in case of skepticism triggered by the CSR label the knowledge seeking behaviors by customers increase. Thus, the central research question of this study reads as follows: What are the effects of CSR labeling certifications on knowledge seeking?
Consumer trust
It is generally accepted that consumers are interested in the social and environmental conditions under which their purchases are produced. However, there are strengths and weaknesses embedded in the CSR labeling schemes as information tools. Labels can be defined as “any words, particulars, trademarks, brand names, pictorial matter or symbols on any packaging, document, notice, board or collar accompanying or referring to a product (Fliess, Lee, Dubreuil and Agatiello, 2007: 20). The main strength of the CSR labeling lies in its visibility and simplicity (Fliess et al., 2007). They have been proven particularly useful and effective in communication about product characteristics that meet consumers’ social and environmental needs. Furthermore, labels appeal especially to consumers who do not have sufficient amount of time to search for information about how the goods and services have been produced and delivered if they are not readily visible (Fliess et al., 2007). These consumers often place their trust in the labeling certifications. When analyzing CSR labels from a macro perspective, they remain the most widely accepted way for companies to communicate CSR credentials.
However, they are only credible if accompanied by reliable standards and certification systems. Thus, labeling schemes as information tools also bear weaknesses. First of all, they are still available for a very limited number of consumer goods. Secondly, they are not easily
adaptable for intermediate goods. Thirdly, it has been noted that the growing number of CSR labeling schemes will likely result in consumer confusion and possible erosion of credibility.
Nevertheless, based on the heuristic-systematic model of information processing (Chaiken, 1980), CSR certifications increase trust in the brand, because due to their nature that requires minimal effort on the part of the recipient of the information, the message is processed along the lines of the heuristic processing. This effect is amplified in the case of external expert endorsements such as CSR labels due to the fact that is allows the recipients to avoid fully processing the semantic content of the message. Hence, the increase in trust and persuasion triggered by the CSR label could in turn lead to a decreased need for additional information. Therefore, hypothesis 1 till 3 read as follows:
H1: CSR labeling certifications increase the trust of consumers in the brand.
H2: Trust in the brand decreases a) information seeking intentions and b) information seeking behaviors about the brand.
H3: Trust in the brand mediates the effect of CSR certifications on a) information seeking intentions and b) information seeking behaviors.
Consumer skepticism
Furthermore, experience shows that labeling can be subject to fraud and misrepresentation (Fliess et al., 2007). Thus, an increased number of cases in which controversies about CSR labeling certification echoed in the public debate resulted in consumer skepticism towards the CSR labels. This can be partly explained by the fact that while consumer awareness has risen substantially over the last decade business acceptance of CSR practices has lagged behind. Additionally, CSR standards been criticized for imposing an unnecessary financial burden for producers in developing countries. Therefore, CSR labels can lead on one hand to
Stolle, 2004), while on the other causing information overload and confusion resulting in consumer skepticism.
While there are many studies suggesting that CSR labels contribute to trust creation (Chen and Chang, 2012), there are a few scholars proposing that CSR labels contribute to an increase in skepticism. According to Annunziata, Ianuario and Pascale (2011) due to a vast number of labels and limited promotion consumers have difficulties recognizing the symbols on the packaging. Additionally, consumers report having difficulties in understanding labels (D’Souza et al., 2007). Furthermore, consumers find it problematic to distinguish what kind of standards the company complied with, who has awarded the label and what it is their
credibility. This is amplified by the fact that many labels look alike and use the same visual language although prerequisites for obtaining them vary to a significant extent. Thus, consumers also are reported to distrust CSR labels (De Pelsmacker, Janssens, Sterckx, and Mielants, 2005). Type I and III labels are seen as more credible than self-awarded
certifications. However, some consumers also do not trust the label even when there are good reasons to deem it credible. Thus, research has demonstrated that endorsement of a label by a third party such as Non Governmental Organization (NGO) or the government is most
effective in reducing skepticism (D’Souza et al., 2007). Moreover, important for this research is that organic and fair trade labels are recognized as most trusted (Fliess et al., 2007).
However, it does not come as a surprise that customers distrust CSR labels. It is often the case that brands try to pass self-declaration certifications as third-party declarations using for instance similar logos. Therefore, hypothesis 4 till 6 read as follows:
H4: CSR labeling certifications increase the skepticism of consumers towards the brand. H5: Skepticism towards the brand decreases a) information seeking intentions and b) information seeking behaviors about the brand.
H6: Skepticism towards the brand mediates the effect of CSR certifications on a) information seeking intentions and b) information seeking behaviors.
The visual representation of the hypotheses in presented in the theoretical model below.
Figure 1. Theoretical Model
Methods Experimental Design
In order to test the hypothesis and answer the research question, this study took a between subjects experimental design. The data was gathered in a span of three weeks time from 15th of May 2014 till 4th of June 2014. The participants were recruited though means of an online questionnaire by convenience sampling method. Therefore, the participants mainly included students from various institutional environments, such as the University of Amsterdam in the Netherlands and Erasmus University Rotterdam, Netherlands. The questionnaire was in English. In total 139 respondents participated in the experiment, however only 95 respondents (N=95) finished the questionnaire. Thus, the drop out rate was 32%.
The experiment comprised of two parts. Firstly, the participants were randomly assigned to two groups. One group was exposed to a visual stimulus depicting a product with
a Fairtrade (FLO) certificate. The second group of participants was exposed to a visual stimulus depicting the same products without a Fairtrade (FLO) certificate. The image presented a package of Starbucks coffee, which included a claim about the fair trade
conditions under which the coffee was produced. The claim red: “Guaranteeing a better deal for Third World Producers” (see Appendix). In total 53% of the participants saw the visual stimulus without the certification seal (n=50) and 47% of the participants saw the visual stimulus with the certification seal (n=45).
Secondly, both groups were asked to answer questions aimed at measuring their attitudes towards the brand depicted in the stimulus material. The specific attitudes measured were trust and skepticism towards the brand claims in regards to supporting fair trade
conditions. Subsequently, the participants were asked to answer questions about their
information seeking intentions about the brand and its value chain. The last part of the online experiment aimed at measuring information seeking behaviors by providing the participants with a link to an open source text on the subject of global trade of coffee and fair trade in specific. Several control variables were includes such as: age, gender, education level, nationality, brand familiarity and general brand attitude. Additionally, a manipulation check was performed in order to control weather the respondents saw the certification.
Demographics
The sample (N=95) consisted of 41% of male participants (n=39) and 59% of female
participants (n =56). The age of the participants ranged from 20 to 56 years (M=28.16 years,
SD=9.26 years). The largest nationality group was Polish, 35% (n=33), followed by Dutch
(n=16) and British (n=12). Other nationalities constituted 36% of the sample (n=34) and included participants from Germany (n=7), USA (n=5), Argentina (n=3), Belgium (n=3), France (n=3), Italy (n=2), Portugal (n=2), Bulgaria (n=1), Croatia (n=1), Indonesia (n=1),
Ireland (n=1), Hong Kong (n=1), Hungary (n=1), Peru (n=1), Romania (n=1), and South Korea (n=1). Most of the participants have obtained higher education, with a Bachelors degree constituting 51% of the sample (n=48). 34% of the participants held a Masters degree (n=32). Doctorate degree was held by 4 participants (n=4), similarly to high school graduates (n=4) and college without a degree graduates (n=4). While Post-doctorate degree was the least common and was held by 3 respondents (n=3).
Measurements
Independent variable. Visual stimulus material.
This research analyzes the effects of the Fairtrade International (FLO) label on consumers in the specific case of fair trade coffee. The Fairtrade label can be categorized as a social justice type III performance-based, third-party certification. FLO has been initiated in the
Netherlands in 1989 and in 1997, 17 national initiatives established an umbrella organization Fairtrade Labeling Organization (FLO) International. Today there are 531 producer
organizations certifies by FLO, representing over one million farmers and workers in over 50 countries (Fliess et al., 2007). Furthermore, 667 traders have been registered including exporters, importers, processors and manufacturers. Between 2003 and 2004 the fair trade sales across the world grew by 56%. While in the North America and the Pacific region alone the 2004 retail value of fair trade increased by 29% for a total of 376.42 million USD (Fliess et al., 2007). For the visual material presented to the participants see Appendix.
Manipulation check.
A manipulation check was included in the questionnaire in order to check if the manipulation worked as intended. At the end of the survey the participants were asked to answer a question: “Did you see a Fairtrade logo on the product presented to you at the beginning of the
survey?”. The answers possible were: “Yes” and “No”. In order to test if the manipulation worked as intended a One-Way ANOVA was conducted with the answer to the manipulation check question (1=yes, 0=no) as a dependent variable and the experimental condition as the independent variable (1=logo, 0=no logo). The analysis yielded a significant effect of the treatment on the manipulation check question (F (1,93) =13.24, p<.001). Participants in the treatment condition (1=logo) experienced a higher score on the manipulation check question (M=.93, SD=.25) then the participants who did not undergo treatment (0=no logo) (M=.64, SD=.49). Therefore, it can be concluded that the manipulation worked successfully as intended.
Mediators. Trust.
The questions that were used to measure consumer trust were adapted from Erdem and Swait (2004). The scale used to measure consumer trust consisted of four items: (1) “This brand delivers what it promises in supporting fair trade conditions.”; (2) “This brand’s product claims about supporting fair trade conditions are believable.”; (3) “This brand has a name you can trust in supporting fair trade conditions.”; and (4) “This brand doesn’t pretend to be something it isn’t in terms of supporting fair trade conditions.” These items were measured using a seven-point scale (1=strongly disagree to 7=strongly agree).
Skepticism.
The questions that were used to measure consumer skepticism were adapted from Mohr, Eroglu and Ellen (1998). The scale used to measure consumer skepticism consisted of four items: (1) “Most statements made by this brand on package labels about supporting fair trade conditions are true.”; (2) “Most statements made by this brand on package labels about supporting fair trade conditions are intended to mislead rather than to inform the consumer.”; (3) “Because most of the statements made this brand which say that they support fair trade conditions are not true, consumers would be better off if such statements were eliminated from package labels.”; and (4) “I do not believe most statements regarding support fair trade conditions on package labels.” These items were measured using a seven-point scale
(1=strongly disagree to 7=strongly agree).
Dependent variables. Information seeking.
Intentions.
The scale used to measure consumer information seeking intentions consisted of four questions: (1) “How likely are you to seek further information about this brand?”; (2) “How likely are you to seek further information about how this product was manufactured?”; (3) “How likely are you to seek further information about supporting fair trade conditions by this brand?”; and (4) “How likely are you to seek further information about the global coffee trade?” These items were measured using a seven-point scale (1=very unlikely to 7=very likely).
Behavior.
Information seeking behavior was measured by providing the opportunity for the participants to find out more about the subject of fair trade coffee by including a link to an online report published by SOMO – Centre for research on Multinational Corporations titled “A fair share
for smallholders: a value chain analysis of the coffee sector”. According to their website
SOMO is an independent research and network organization focusing on investigating multinational corporations and the consequences of their activities for people and the
environment around the world (SOMO, 2014). SOMO was established in 1973 in Amsterdam and is known for providing reliable and well-informed reports. 52% of the participants were not interested in getting the link to the report (n=49), while 48% were redirected to the report (n=46). The link to the report can be found in the Appendix.
Control variables. Brand attitude.
The questions that were used to measure brand attitude were adapted from Sengupta and Johar (2002). The scale used to measure brand attitude consisted of three items: (1) “I think Starbucks is a very good coffee brand.”; (2) “I think Starbuck is a very convenient
coffeehouse chain.”; and (3) “My opinion of Starbuck is very favorable.” These items were measured using a seven-point scale (1=strongly disagree to 7=strongly agree).
Brand familiarity.
The questions that were used to measure brand familiarity were adapted from Martin and Stewart (2001). The scale used to measure brand familiarity consisted of four items: (1) “How familiar are you with Starbucks?”; (2) “How familiar are you with Starbucks coffee
uses?” and (4) “How familiar are you with coffee products in general?”. These items were measured using a five-point scale (1=not at all familiar, 5=very familiar).
Results Principle Component Analysis
The first step conducted in the statistical analysis was to check for missing values in the data set. In order to do that descriptive statistics was run for each item per variable. None were found, thus, the second step involved factor analysis to identify underlying factors of the variables. Finally, the internal consistency of the variables was measured with the internal- reliability method Cronbach’s Alpha. The reliability of the chosen items for each variable was checked by means of a reliability analysis (Cronbach’s alpha ≥ .60, total item correlation ≥ .30, & Alpha if item deleted ≤ Cronbach’s alpha) (George and Mallery, 2003). According to George and Mallery (2003) the following rules apply for values of Cronbach’s Alpha: _ >.90 – Excellent, _ >.80 – Good, _ >.70 – Acceptable, _> .60 – Questionable, _ > .50 – Poor, and _ < .50 – Unacceptable (George and Mallery, 2003). Subsequently the items were computed in order to establish new variables. The new variables were computed by creating the mean of all the items that measure the variable.
The results of principal component analysis (PCA) showed that all the variables have one underlying component. PCA showed that the four items for the mediating variable “trust” form a single uni-dimensional scale: only one component has an eigenvalue above 1
(eigenvalue 2.98) and there is a clear point of inflexion after this component in the scree plot. All items correlate positively with the first component. The total variance accounted for by the four factors was 74.40%. Reliability of the scale is good, Cronbach's alpha =.88. On average respondents scored 3.98 (SD=1.31) on the trust scale.
The four items for the mediating variable “skepticism” form a single uni-dimensional scale, only one component has an eigenvalue above 1 (eigenvalue 1.966) and there is a clear point of inflexion after this component in the scree plot. Only three items correlate positively with the first component. The total variance accounted for by the four factors was 49.14%. The reliability of the scale was unacceptable, Cronbach's alpha =. 31. Therefore the first item was deleted from the scale. The reliability of the scale after deleting the first item was
questionable, but still acceptable Cronbach's alpha =.64. The total variance accounted for by the three factors was 59.15%. On average respondents scored 4.27 (SD=1.24) on the
skepticism scale.
The four items for the dependent variable “information seeking intentions” form a single uni-dimensional scale, only one component has an eigenvalue above 1 (eigenvalue 2.64) and there is a clear point of inflexion after this component in the scree plot. All items correlate positively with the first component. The total variance accounted for by the four factors was 66.08%. Reliability of the scale is good Cronbach's alpha =.83. On average respondents scored 3.77 (SD=1.48) on the information seeking intentions scale.
For the control variable “brand attitude” three items form a form a single uni-
dimensional scale, only one component has an eigenvalue above 1 (eigenvalue 2.55) and there is a clear point of inflexion after this component in the scree plot. All items correlate
positively with the first component. The total variance accounted for by the six factors was 85.03%. Reliability of the scale is excellent, Cronbach's alpha =.91. On average respondents scored 3.83 (SD=1.75) on the brand attitude scale.
For the control variable “brand familiarity” four items form a single uni-dimensional scale, only one component has an eigenvalue above 1 (eigenvalue 2.25) and there is a clear point of inflexion after this component in the scree plot. All items correlate positively with the
first component. The total variance accounted for by the six factors was 56.24%. Reliability of the scale is acceptable, Cronbach's alpha =.73. On average respondents scored 3.24 (SD=.84) on the brand attitude scale. Thus, on the basis of the principal component analysis five variables were computed.
Regression Analysis
In order to test direct effect of the independent dichotomous variable as well as the mediating variables (trust and skepticism) on the two dependent variables (information seeking
intentions and information seeking behaviors) a series of linear and logistic regressions was run. Control variables included in the regression models were: age, gender, education and brand familiarity. Brand attitude was not included in the analysis due to an overlapping conceptual basis with trust in a brand and skepticism towards the brand.
The first regression model with trust in the brand as the dependent variable and exposure to the visual stimulus with the certification seal (treatment), age, gender, education and brand familiarity as independent variables was statistically significant F (5,89)=2.83, p<.05. The regression model can therefore be used to predict the trust in the brand, but the strength of the model is low. Only 14 per cent of the variation in trust can be predicted on the basis of the treatment and gender (R2 =.14). Nevertheless, the treatment has a significant effect on trust, b*=.27, t=2.70, p<.01, 95% CI [.19, 1.24]. Additionally, gender has a significant effect on trust, b*=.20, t=2.00 95% CI [.01, 1.07]. Men score on average lower (M=3.64, SD=1.37) on the trust in the product scale then women (M=4.21, SD=1.23). The results show that age, education and brand familiarity have no significant effect on trust (p>.05).
The regression model with skepticism towards the brand as the dependent variable and treatment, age, gender, education and brand familiarity the independent variables was not
statistically significant F (5,89)=2.01, p>.05. The regression model cannot therefore be used to predict skepticism towards the brand. However, two variables namely the treatment and age have a significant effect on skepticism towards the brand. The treatment has a significant negative effect on skepticism, b*= -.27, t=2.61, p<.05, 95% CI [-1.16, -.16]. Additionally, age has a significant effect on skepticism, b*= -.24, t= -2.06 95% CI [-.06, - .001]. Per year of age the predicted skepticism decreased by 0.24. Gender, education and brand familiarity however, have no significant effect on skepticism (p>.05).
The regression model with the dependent variable being information seeking
intentions and treatment, trust, skepticism, age, gender, education and brand familiarity as the independent variables was not statistically significant F (7,87) =1.63, p>.05. The regression model cannot therefore be used to predict information seeking intentions. However, brand familiarity has a significant effect on information seeking intentions, b*=.22, t=2.16, p<.05, 95% CI [.03, .75]. Trust, skepticism, age, gender, and education however, have no significant effect on information seeking intentions (p>.05).
Lastly, a logistic regression was performed in order to predict the effects of the treatment, trust, skepticism, age, gender, education and brand familiarity on the dependent variable information seeking behaviors. The logistic model was not significant Chi2 =6.46, p>.05, df =8. The regression model therefore cannot be used to predict information seeking behaviors. Furthermore, the results show that none of the independent variables have a significant effect on information seeking behaviors. The results of the four regression models are presented in Table 1. and Table 2. on the following page.
Table 1. Regression models to predict trust and skepticism (N = 95).
Note. * p <.05. ** p <.01. *** p <.001.
Mediation Analysis
In order to confirm the result of the regression analysis, Preacher and Hayes (2008) method for testing mediation was followed. Preacher and Hayes (2008) provide the necessary tools to run mediation analysis with multiple mediators as well to estimate models with dichotomous outcome variable. Following the recommendations two models were tested for each
dependent variable separately. Therefore, the first model included treatment as the
independent variable, trust and skepticism as mediators and information seeking intentions as dependent variable. The second model included treatment as the independent variable, trust and skepticism as mediators and information seeking behaviors as dependent variable. In both cases control variables were age, gender, education and brand familiarity. The terminology used is as follows: a-path reflects the relationship between the independent variable and the mediator(s), the b-path is the relationship between the mediator(s) and the dependent variable, the c-path is the total effect of the independent on the dependent variable, and the c’-path is the direct effect of the independent on the dependent variable when the mediators are not included. It was conducted using a bootstrap of 5000 and a confidence interval of 95%.
The results of Preacher and Hayes (2008) analysis tool in the case of the first model show that the a-path from the treatment to trust is significant (b=.71, p<.01). The a-path from the treatment to skepticism is also significant (b=-.66, p<.05). However, the b-paths from trust and skepticism to information seeking intentions do not show significance (p> .05). Similarly the c-path and c’-path are not statistically significant (p>.05). Therefore the model is not significant (p>.05).
The results of Preacher and Hayes (2008) analysis tool in the case of the second model confirm that the a-paths from the treatment to trust and skepticism are significant. However, the b-paths from trust and skepticism to information seeking behaviors do not show
significance (p>.05). Similarly the c-path and c’-path are not statistically significant (p>.05). Therefore, the model is not significant (p>.05).
In both cases the results show no significant effect of age, gender or education on the dependent variables (p>.05). A significant effect of brand familiarity was found on information seeking intentions (b=.39, p<.05), however no significant effect of brand familiarity was found on information seeking behaviors (p>.05). In sum, the results of the macro mediation analysis confirmed the significant results from the linear regression analysis for the effect of treatment on trust and skepticism.
Therefore, based on the results from the regression analyses H1: CSR labeling
certifications increase the trust of consumers in the brand has been confirmed. H2: Trust in the brand decreases a) information seeking intentions and b) information seeking behaviors about the brand has been rejected. Further H4: CSR labeling certifications increase the skepticism of consumers towards the brand has been rejected. Also H5: Skepticism towards the brand decreases a) information seeking intentions and b) information seeking behaviors about the brand has been rejected.
Based on the results from the mediation analysis H3: Trust in the brand mediates the
effect of CSR certifications on a) information seeking intentions and b) information seeking behaviors and H6: Skepticism towards the brand mediates the effect of CSR certifications on a) information seeking intentions and b) information seeking behaviors have been rejected.
Figure 2. Revised Model.
Note. * p <.05. ** p <.01., b* values reported from regression analyses, n.s = p >.05.
Discussion and Conclusions
Ethical labeling initiatives set out to function as market based instruments to translate value- based assumptions on better ways of doing business into business practices (Hartlib and Jones, 2009). Ultimately they seek to spread more humane business models on a larger scale by incorporating these practices into the mainstream business environment (Hartlib and Jones, 2009). This study suggested that CSR certifications could play a crucial role in this process by leading to information seeking and acquiring knowledge by consumers about more ethical trade relations. This could in turn result in changing purchase behaviors on a larger scale.
Nevertheless, this paper moves away from the widely research theme of purchase intentions and instead focuses on investigating CSR labels (FLO in particular) as triggers of curiosity. Even though, nowadays a wide proliferation of ethical labels might have led to consumers’ fatigue with these issues, on the contrary many studies suggest an increasing interest and demand from consumers for transparent information into the business practices of their favorite brands (De Pelsmacker et al., 2005). Therefore, this paper extends the research
of consumer attitudes, in specific trust and skepticism into the field of ethical marketing. The main purpose of this paper was to demonstrate the relationships between ethical marketing and information seeking – and to identify the partial mediation effects of trust and skepticism.
The results of this research have shown that there is enough evidence to state that products certified with a CSR label have an effect on consumers’ attitudes towards the brand. Interestingly enough, the regression analysis and the mediation analysis both confirm that ethically certified products increase trust in the brand and reduce skepticism towards the brand. Therefore, as hypothesized consumers trust the brands with CSR certifications present on their products more as opposed to brands that do not include CSR certifications. This effect has been previously found in many studies focusing on for instance green trust (Chen and Chang, 2012). Noteworthy is also the effect of gender on trust in the brand. The results show that on average women trust more in the brands than men. Furthermore, education, age or brand familiarity had no effect on trust, which is surprising especially in the case of
education. This could be explained due to the nature of the sample, which mostly consisted of respondent with a high educational background.
However, unpredictably but consequently consumers’ skepticism towards the brand seems to decrease in the incidence of a CSR certification present on the product. Therefore, the hypothesized effect of the ‘double-edge’ sword of CSR communication (Morsing, et al., 2008) has not been confirmed by this study. An interesting relationship has been found however, between age and skepticism towards the brand. With age the skepticism decreases suggesting that younger consumers are more critical and thus, could be more engaged then older consumers. However, in the case of skepticism gender did not play a significant role, neither did education or brand familiarity.
Nevertheless, it is important to note that the product presented to the participants included a fair trade claim on the packaging. The only difference was that in one case the claim was supported by an ethical certification (FLO) and in the second case it was not. Therefore, the results of this study are only applicable when comparing brands using different ethical marketing tools as opposed to brands refraining from addressing the ethical dimension of their products. Thus, it can be concluded that consumers deem ethical claims supported by a CSR certification more credible than ethical claims that are not supported by a certification.
This finding is interesting to investigate further in regards to different types of CSR certification schemes. As previously mentioned FLO is a social justice type III performance- based, third-party certification (D’Souza et al., 2007). Therefore, further research could examine the effects of type I and type II certifications. Various degrees of skepticism towards the brand could be possibly noticed here. Additionally, the fact that FLO is a widely
recognized credible certification might explain the negative relationship found in this study between the labeled product and skepticism towards the brand. Hence, it seems that even though the Fairtrade label has been a subject of a number of controversies in the public debate (Jaffee, 2007), the overall reputation of the certification has not suffered substantially.
However, the public opinion about the Fairtrade certification scheme has not been the subject of this study it offers a point of departure for further research into this topic.
Although, prior studies widely discussed the effects of CSR labels on purchase intentions (Harris, 2007; Chen and Chang, 2012; D’Souza et al., 2007) this paper aimed at filling the research gap in regards to the effects on information seeking. Therefore, in order to address this under-researched subject this study suggested that, knowledge seeking would decrease in the case of consumer trust in the brand triggered by a CSR certification. This effect could be explained based on the heuristic processing on information (Chaiken, 1980).
Nevertheless, the results of the regressions as well as the mediation analysis suggest that there is no significant relationship between consumers’ trust in the brand and information seeking intentions or behaviors.
On the other hand, this study suggested that simultaneously CSR labels trigger skepticism leading to an increase in information seeking. Based on the elaboration – likelihood model of persuasion (ELM), in the case of consumers that care about ethical business conducts information about CSR certifications would be processed along the central path leaving a long lasting persuasive effect. However, if the recipient of the message is unfavorably inclined towards fair trade and skeptical towards the CSR certification schemes, then a boomerang effect is likely to occur, resulting in increased information seeking (Petty and Cacioppo, 1986). Nevertheless, the results show no significant effect of skepticism on knowledge seeking intentions or behaviors.
Surprisingly, the results do not show any significant effects of gender, age or
education on information seeking. This is remarkable especially for the level of education. In accordance to prior research (Annunziata et al., 2011) the level of education has been proven to be a significant predictor of critical engagement in ethical consumption issues. However, better brand familiarity increases information seeking intentions of consumers, which is in line with research suggesting consumers’ demand for information about brands that are recognizable.
Interestingly enough, when not accounted for the effects of age, gender, education or brand familiarity the results of single logistic regressions show a tendency of data that seems to support the hypotheses. It has been found that trust indeed decreased information seeking behaviors and skepticism-increased information seeking behaviors. This finding was not confirmed for information seeking intentions. Also not significant direct effect of CSR
certifications has been found on information seeking intentions or behaviors. However, single predictor models are usually not reliable indicators and cannot be used to generalize, this finding offers an interesting starting point for further research. Therefore, further research could focus on improving the methodology by for instance using different stimulus material. It could be possible that the results would be significant when incorporating several different ethical labels or various product types. Also interesting to explore could be the difference between brands that do not use ethical marketing strategies compared to brands that do. In this case the stimulus material used would vary more than in this study leading to stronger effects.
There are several limitations of this study that need to be addressed. Firstly, of all due to the nature of the stimulus material the questionnaire might have seemed more relevant to coffee drinkers as opposed to non-coffee drinkers. Some respondents have included in their feedback comments addressing this particular issue. Therefore, it is recommendable that further research includes different types of products. Nevertheless, coffee was chosen for this study due to the fact that it is a highly contested product in the fair trade debate. Furthermore, this study has not included a differentiation between heavy and non-heavy coffee drinkers as present in other studies (De Pelsmacker et al., 2005). It might be so that heavy coffee drinkers have more interest in finding out about the ethical business conducts of their coffee brand due to high daily consumption and their perceived influence on global coffee trade. Secondly, this study did not include controlling for prior knowledge and awareness about fair trade in general, fair trade products or fair trade campaigns. Also, personal ethical values and economic and political orientation could have been included as control variables. Thirdly, social desirability bias might have to be accounted for in measuring information seeking. Lastly, the measurement of information seeking behaviors was not precise enough due to the fact that it did not actually measure how many respondents clicked on the report and red it after they got redirected to it.
There are three practical implications of this study. Firstly, this study confirms that CSR certifications supporting ethical claims of certain products increase consumers’ trust in the brand. Thus, if companies would like to increase consumers’ trust and brand reputation they should make use of external endorsements such as CSR labeling schemes as a part of their long-term marketing strategy. Secondly, in a more contested and complex marketing contexts such as ethical marketing companies should invest in acquiring performance based third party certifications in order to decrease consumers’ skepticism towards the ethical claims and the brand in general. Thirdly, since ethical marketing can become a way of differentiating and positioning nowadays, companies should explore this opportunity.
However, in doing so a great dose of caution is recommendable. Therefore, even though some firms manage to develop long-term strategies based on ethical marketing, the true challenge is how to incorporate their ethical mission into business models rather than to only promote their products.
In conclusion, it is possible to state that the ethical label, as an extrinsic quality indicator, contributes on one hand to an increase in consumers’ trust in a brand and consequently on the other hand to a decrease in consumers’ skepticism towards a brand. Furthermore, it can be stated that companies’ ethical claims should be supported by CSR labels. Hence, it becomes fundamental for firms to improve the credibility of their ethical claims by including a third part performance based endorsement on their products.
References
Annunziata, A., Ianuario, S., & Pascale, P. (2011). Consumers' attitudes toward labelling of ethical products: The case of organic and Fair Trade products. Journal of Food
Products Marketing, 17(5), 518-535.
Blengini, G. A., & Shields, D. J. (2010). Green labels and sustainability reporting: Overview of the building products supply chain in Italy. Management of Environmental Quality:
An International Journal, 21(4), 477-493.
Carrero, I., & Valor, C. (2012). CSR-labelled products in retailers' assortment: A comparative study of British and Spanish retailers. International Journal of Retail & Distribution
Management, 40(8), 629-652.
Chakrabarty, S., & Grote, U. (2009). Child labor in carpet weaving: impact of social labeling in India and Nepal. World Development, 37(10), 1683-1693.
Chaiken, S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of personality and social psychology,
39(5), 752.
Chen, Y. S., & Chang, C. H. (2012). Enhance green purchase intentions: The roles of green perceived value, green perceived risk, and green trust. Management Decision, 50(3), 502-520.
Crane, A., Matten, D. & Moon, J. (2004), Stakeholders as citizens? Rethinking rights, participation, and democracy. Journal of Business Ethics, 53(1), 107-22.
Davies, M. A., & Wright, L. T. (1994). The importance of labelling examined in food marketing. European Journal of Marketing, 28(2), 57-67.
De Pelsmacker, P., Janssens, W., Sterckx, E., & Mielants, C. (2005). Consumer preferences for the marketing of ethically labelled coffee. International Marketing Review, 22(5), 512-530.
Drichoutis, A.,C., Lazaridis, P., & Nayga, R.,M. (2006). Consumers’ use of nutritional labels: a review of research studies and issues. Academy of Marketing Science Review, 9. D'Souza, C., Taghian, M., & Lamb, P. (2006). An empirical study on the influence of
environmental labels on consumers. Corporate Communications: An International
Journal, 11(2), 162-173.
D’Souza, C., Taghian, M., Lamb, P., & Peretiatko, R. (2007). Green decisions: demographics and consumer understanding of environmental labels. International Journal of
Consumer Studies, 31(4), 371-376.
Erdem, T., & Swait, J. (2004). Brand credibility, brand consideration, and choice. Journal of
Consumer Research, 31(1), 191-198.
Fliess, B., Lee, H. J., Dubreuil, O. L., & Agatiello, O. R. (2007). CSR and trade: informing consumers about social and environmental conditions of globalised production: part I
George, D., & Mallery, P. (2003). SPSS for window Step by Step.
Harris, S. M. (2007). Does sustainability sell? Market responses to sustainability certification.
Management of Environmental Quality: An International Journal, 18(1), 50-60.
Hartlieb, S., & Jones, B. (2009). Humanising business through ethical labelling: Progress and paradoxes in the UK. Journal of Business Ethics, 88(3), 583-600.
Jaffee, D. (2007). Brewing justice: Fair trade coffee, sustainability, and survival. Univ of California Press.
Martin, I. M., & Stewart, D. W. (2001). The differential impact of goal congruency on attitudes, intentions, and the transfer of brand equity. Journal of Marketing Research,
38(4), 471-484.
Matten, D. & Moon, J. (2005). Corporate Social Responsibility, Journal of Business Ethics,
54 (4), 323-37.
Micheletti, M., & Stolle, D. (Eds.). (2004). Politics, products, and markets: Exploring political consumerism past and present. transaction publishers.
Mielants, C., De Pelsmacker, P. & Janssens, W. (2003). Knowledge, attitude and behaviour of Belgians with respect to fair trade products. Conclusions of four focus groups, working
paper for DWTC-PODO II-project, Antwerp.
Mohr, L. A., Eroglu, D., & Ellen, S. P. (1998). The development and testing of a measure of skepticism toward environment claims in marketers' communications. The Journal of
Consumer Affairs, 32(1), 30-55.
Morsing, M., Schultz, M., & Nielsen, K. U. (2008). The ‘Catch 22’of communicating CSR: Findings from a danish study. Journal of Marketing Communications, 14(2), 97 111. Pahl, L. (2003). EcoRange: Market-Oriented Environmental Certification for Rangeland
Pastoral Industries. 1. Project Overview, 2003.
Pant, R. & Sammer, K. (2003). Labeling as an appropriate strategy and instrument for sustainable consumption, paper presented at the 6th Nordic Conference on Environmental Social Sciences.
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion (pp. 1-
24). Springer New York.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research
Methods, 40, 879-891.
Rex, E., & Baumann, H. (2007). Beyond ecolabels: what green marketing can learn from conventional marketing. Journal of cleaner production, 15(6), 567-576.
Sengupta, J., & Johar, G. V. (2002). Effects of inconsistent attribute information on the predictive value of product attitudes: Toward a resolution of opposing perspectives.
Singh, S. N., & Cole, C. A. (1993). The effects of length, content, and repetition on television commercial effectiveness. Journal of Marketing Research 30(1).
SOMO. (2014). About SOMO. Retrieved from http://www.somo.nl/about-somo.
Szykman, L. R., Bloom, P. N., & Levy, A. S. (1997). A proposed model of the use of package claims and nutrition labels. Journal of Public Policy & Marketing, 228-241.
Teisl, M. F., Roe, B., & Levy, A. S. (1999). Ecocertification: Why it may not be a “field of dreams”. American Journal of Agricultural Economics, 81(5), 1066-1071.
Wansink, B. (2003). How do front and back package labels influence beliefs about health claims?. Journal of Consumer Affairs, 37(2), 305-316.
Zadek, S., Lingayah, S., & Forstater, M. (1998). Social labels: Tools for ethical trade. Report
prepared by the New Economics Foundation for the European Commission, DG Trade–www. ilo. org/dyn/basi/docs F, 126.
Appendix Figure 1A. The questionnaire
Dear participant!
As part of my master thesis at the University of Amsterdam I am conducting a research on fair trade labels. I would be very grateful if you could spare 5 minutes of your time to help me by filling out this short survey. The survey involves looking at a product supporting fair trade conditions and answering a few questions about it. Participation is voluntary, there are no good or bad answers, so please be honest, the answers will be anonymous and used only for the purpose of this study. On the next page you will be asked to give your informed consent. Please read the provided information and check the box to continue to the survey.
Thank you for your time and participation!
I hereby declare that I have been informed in a clear manner about the nature and method of the research, as described on the previous page. My questions have been answered satisfactorily. I agree, fully and voluntarily, to participate in this research study. With this, I retain the right to withdraw my consent, without having to give a reason for doing so. I am aware that I may halt my participation in the experiment at any time. If my research results are used in scientific publications or are made public in another way, this will be done such a way that my anonymity is completely safeguarded. My personal data will not be passed on to third parties without my express permission. If I wish to receive more information about the research, either now or in future, I can contact Anke Wonneberger (a.wonneberger@uva.nl, Kloveniersburgwal 48, 1012 CX Amsterdam). Should I have any complaints about this research, I can contact the designated member of the Ethics Committee representing the ASCoR, at the following address: ASCoR secretariat, Ethics Committee, University of Amsterdam,
Kloveniersburgwal 48, 1012 CX Amsterdam; 020-525 3680; ascor-secr-fmg@uva.nl. o I declare that I have read these statements and I agree to take part in this study. Please take a close look at this product. (Randomized)
Please indicate how strongly you agree or disagree with the statements below. Strongly Disagree Disagree Somewhat Disagree Neither Agree nor Disagree Somewhat Agree Agree Strongly Agree
This brand delivers what it promises in supporting fair trade conditions.
This brand’s product claims about supporting fair trade conditions are believable. This brand has a name you can trust in supporting fair trade conditions.
This brand doesn’t pretend to be something it isn’t in terms of supporting fair trade conditions. mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm
Please indicate how strongly you agree or disagree with the statements below. Strongly Disagree Disagree Somewhat Disagree Neither Agree nor Disagree Somewhat Agree Agree Strongly Agree
Most statements made by this brand on package labels about supporting fair trade conditions are true.
Most statements made by this brand on package labels about supporting fair trade conditions intended to mislead rather than to inform the consumer.
Because most of the statements made by this brand which say that they support fair trade conditions are not true, consumers would be better off if such statements were eliminated from package labels.
I do not believe most statements regarding fair trade conditions on package labels. mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm (Continued)
Please indicate how strongly you agree or disagree with the statements below. Strongly Disagree Disagree Somewhat Disagree Neither Agree nor Disagree Somewhat Agree Agree Strongly Agree I think Starbucks is a very good coffee brand. I think Starbuck is a very convenient coffeehouse chain. My opinion of Starbuck is very favorable. mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm Please indicate how familiar you are with the following.
Not at all familiar Slightly familiar Somewhat familiar Moderately familiar Very familiar How familiar are you with
Starbucks?
How familiar are you with Starbucks coffee products?
How familiar are you with the type of advertising that Starbucks currently uses?
How familiar are you with coffee products in general? mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm Please indicate how likely you are to do the following.
Very Unlikely Unlikely Somewhat Unlikely Undecided Somewhat Likely Likely Very Likely How likely are you to
seek further information about this brand? How likely are you to seek further information about how this product was manufactured? How likely are you to seek further information about supporting fair trade conditions by this brand?
How likely are you to seek further information about the global coffee trade? mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm mm (Continued)
Would you like to find out more about the fair trade terms and conditions in the case of coffee products? mm Yes. Please redirect me to an online report.
mm No. I want to proceed to the next question.
The link below provides more information about the global coffee trade. It is a report published by SOMO – Centre for research on Multinational Corporations titled “A fair share for smallholders: a value chain analysis of the coffee sector”.
Please click on it if you are interested to find out more on this subject. http://somo.nl/publications-en/Publication_2539/at_download/fullfile
Thanks for your help so far! Finally, I would like you to provide some information about yourself. It is very important that you answer these questions as honest and as detailed as possible. Be assured that the data is confidential and that information will be solely used for this research project. Please indicate your age. Please specify your gender.
mm Male mm Female
What is the highest level of education you have completed? mm Did not finish high school
mm High school graduate - high school diploma or equivalent mm Some collage but no degree
mm Bachelors degree (For example: BA, BS) mm Masters degree (For example: MA, MS, MBA) mm Doctorate degree (For example: PhD)
mm Post-doctorate degree (Prof) What is your nationality?
Did you see a Fairtrade label on the product presented to you in the beginning of the survey? mm Yes
mm No
Thank you for your time and participation!
If you would like to know more about the study or the findings of the study, feel free to contact me (katarzyna.jaskiewicz@student.uva.nl). Note that the information presented to you about the brand and the product was fictitious and will be used only for the purpose of this study.
If you should have any remarks, comments or other feedback about the survey, please state them in the box below and continue to the next page to finish the survey.
