Factors influencing consumers’ willingness to pay for
organic food
Bachelor thesis
Julia Popma
Amsterdam, July 30, 2014
Student number: 10277404 Thesis seminar: Business Studies Supervisor: Bram Kuijken Academic year: 2013 – 2014 Semester 2, block 3Abstract
The primary reasons why consumers buy organic food are “quality and security”- and “trust”. Previous research on the predictors of attitudes towards organic food has examined the role of health consciousness, food safety concern and ethical self-identity. Food safety was a recurrent term and the effect of the provision of such information on consumers’ willingness to pay is being investigated. Furthermore, the effect of the presence of a well-known eco-label is being examined. Finally, it is investigated if such relationships are moderated by consumers’ previous knowledge. The data is obtained by means of an innovative online second-price sealed bid auction experiment Veylinx. The hypotheses are tested using a sample consisting of 366 participants. Surprisingly, not any significant evidence is found, suggesting that neither an eco-label and food safety information nor consumers’ knowledge have an effect on willingness to pay for organic food under Dutch consumers.
Table of content
Abstract……….2 Foreword………...5 1. Introduction………..6 2. Literature review 2.1 Signaling theory………..…….8 2.2 Effect of eco-label………92.3 Attitudes toward organic food………10
2.4 Influence of knowledge on ecologically behavior……….11
2.5 Influence of socio-demographic factors……….13
2.6 Conclusion………..13
3. Conceptual framework 3.1 Eco-label………..……...14
3.2 Food safety information……….14
3.3 Influence of consumers’ knowledge………..15
4. Methodology 4.1 Research design………..16 4.2 Research method………....16 4.3 EKO-Quality Mark………....17 4.4 Product choice………....19 5. Results 5.1 Descriptive statistics………...19 5.1.1 Sample characteristics………19
5.1.2 Data characteristics (auction bid data)……….….21
5.1.3 Testing for normality………...22
5.2 Correlations………..…..24
5.3 Hypothesis testing 5.3.1 Eco-label……….24
5.3.2 Food safety information………..26
5.3.3 Knowledge………...26
6. Discussion 6.1 Key findings………...28
6.2 Contribution to theory and practice………29
6.3 Managerial implications....……….………...……….30
6.4 Limitations and further research………30
7. Conclusion………...31
Bibliography………...33
Appendix A……….36
Appendix B……….39
Foreword
First, I would like to thank Hannah Berkers for assigning me to this topic and all her efforts to allocate Bram Kuijken as my supervisor. Second, I would like to thank my sister Andrea for her willingness when I was having difficulties seeing my vision. Third, I would like to thank Anouar El Haji for his personal help with some statistical matters. Fourth, I would like to thank my friend Morven for revising my English. Most of all, I would like to thank my supervisor Bram Kuijken for his openness to ideas and always adequately responding to my e-mails and questions.
1. Introduction
Concern for one’s environment and health have become increasingly important for both consumers and producers, affecting a product’s lifecycle from production to distribution to disposal (Chan, 2000). Owusua and Anifori (2013) examined whether consumers are willing to pay a premium price for organic watermelons from Ghana, compared to conventional watermelons. Their empirical findings show that
socioeconomic characteristics, product freshness and cleanliness are having positive effects on consumers’ willingness to pay a premium price for an organic watermelon. Similar research on consumers’ willingness to pay for organic food is found in Nigeria (Phillip & Dipeolu, 2010) and Spain (Gil, Gracia & Sanchez, 2000). According to Philip and Dipeolu (2010), consumers’ age, work experience and household size significantly influence the perception that conventional vegetables are less healthy than organic vegetables. Gil, Gracia and Sanchez (2000) segmented the different customers in the market based on lifestyle and socio-economic
characteristics, age, gender, education level, family size, and income.
With regards to the socio-demographic factors of age, gender and education, there is conflicting evidence stating that in the EU, female consumers in younger age groups, with children, of higher income and higher education groups are most likely to buy organic food (Krystallis, Fotopoulos & Zotos, 2006).
There are several motives underlying the purchase of organic food, such as health consciousness, food safety concern and ethical self-identity. The last two motives were found to be the most important predictors of attitude towards organic products (Michaelidou & Hassan, 2008). With regards to the purchasing decisions of Greeks, Krystallis & Chryssohoidis (2005) identified several factors that influence this decision; “quality and security”, “trust”, “senses”, “price sensitivity”,
“convenience”, and “brand-name value”. ”Quality and security” and “trust” are considered as the most important and influential to the WTP for most organic food categories (Krystallis & Chryssohoidis, 2005). Better quality, security and trust are the perceptions consumers have towards organic products in contrast to conventional products, and therefore they are willing to pay a premium for this type of food. However, this means that if consumers are not entirely sure that the product is
organic, they are unwilling to pay this premium. Whether a consumer’s willingness to pay increases when there is more elaborated information provided about the organic
product with regard to food safety is a view that has not yet to be investigated, and therefore a focus of this research. This has not been investigated yet.
Furthermore, eco labels could be a determining factor as well. According to Janssen and Hamm (2012) consumers have different preferences and willingness to pay for different organic certifications. Since July 2010, all organic products produced and sold in the EU must consist of a mandatory EU logo. Suppliers have the
opportunity to place other organic certification logos owned by other organizations. For marketers it is of great interest to understand which brand labels are valued most by consumers in order to create the most effective packages for the product (Janssen & Hamm, 2012). It has been recommended for suppliers to use an additional label for organic products, alongside to the mandatory logo.
Consumers’ willingness to pay for organic products also depends on the product category. Krystallis and Chryssohoidis (2005) hypothesized that Greek consumers are willing to pay more for the organic alternative if the product is something that they frequently buy (basic products) vs. products they buy
occasionally (non basic products). The results of their research are not entirely valid since their results reveal that the willingness to pay depends on different product categories. Therefore, this factor is disregarded in this research.
In conclusion, it has not been investigated whether a consumer’s willingness to pay increases when there is more elaborated information provided about the organic product with regard to food safety. In addition, the direct effect of an eco-label on consumers’ willingness to pay for organic food has not been researched yet. This research aims to fill this gap.
Altogether, this lead to the following research question: ‘How do the presence of an eco-label, the provision of food safety information and the consumer’s
knowledge influence consumers’ willingness to pay for organic products?’ The research question adds further understanding to the psychological processes driving consumer behavior. It is an extension of previous research by combining the effect of both an eco-label and food safety information. Furthermore, specifying which variables increase consumers’ willingness to pay for the products is of interest to managers. This information should be adapted into marketing strategies
accordingly. In section two, existing literature related to the willingness to pay for organic food is discussed. Section three is dedicated to the hypotheses, which are based on the literature. The methodology is discussed in section four, using Veylinx
as a research tool. The results are presented in section five in order to test the
hypotheses. The key findings, contributions, managerial implications and limitations are covered in section six. Finally, a conclusion is drawn is section seven.
2. Literature review
Organic food is defined as food grown and processed using no synthetic fertilizers or pesticides. Global demand for the organic market has increased significantly. The global market for organic food has tremendously been expanding by over five billion US Dollars a year in 2007 (Sahota, 2009). The highest global revenues in that year were distributed by Europe and North America (Sahota, 2009). Ever since a
significant amount of research regarding consumers’ willingness to pay for organic food has taken place. In this section, the existing literature concerning factors influencing consumers’ willingness to pay for organic food and the signaling theory will be discussed in order to obtain an overview and to define the topic of this research.
2.1 Signaling theory
Signaling theory is used to define behavior when two parties (the sender and the receiver) have access to different information (Connelly, Ireland, Certo & Reutzel, 2011). The theory is concerned with how the sender chooses how to communicate information, and how the receiver chooses how to interpret that information. Asymmetric information occurs when one of two parties possesses different
information compared to the other party (Stiglitz, 2002). Stiglitz (2002) provides an example of asymmetric information; ‘the owner of a car knows more about the car than potential buyers (p. 470)’. The same holds for an organic producer selling organic food; the producer knows more than the potential consumer.
Due to asymmetric information, one of the best instruments for firm to communicate product information is through advertising. Advertisements have the ability to increase consumer perceived quality and alter consumers’ willingness to pay (Caswell & Padberg, 1992). Advertising is crucial in establishing or maintaining product value in the food market. The reason for this is that the quality information in this type of market is usually asymmetric allowing competition among sellers by the use of advertising instead of price rivalry (Caswell & Padberg, 1992). Therefore
different advertisements are used in an online auction in order to research the constructs that this research is aiming to explain. This auction tool will be discussed in section four; the methodology.
2.2 Effect of eco-label
The eco-label is part of the signaling theory in this research. According to Caswell and Padberg (1992) consumers rely on an information set in making product
selections. An information set consists of labels, prior experience, media advertising, word-of-mouth information, and dietary education programs provided either by the government, health professionals, or private groups (Caswell & Padberg, 1992). The food label adds to consumers’ information base and assists the consumer with some guidance for the buying process.
As mentioned above, brand value is a factor influencing the purchasing decision of organic food (Krystallis & Chryssohoidis, 2005). Organic labels have an ‘emotional resonance in personal wellbeing and health’ (MINTEL, 2000). The value-added quality of organic or GMO-free labeling as an effective marketing mechanism seems to depend on the type of product that is being marketed (Loureiro & Hine, 2002). Loureiro and Hine (2002) discovered that the organic or GMO-free labeling was an effective mechanism for vegetables and diary products, but it could not be generalized to the potato sector. Similarly, earlier research from Wessels et al. (1999) indicated that the eco-labeled certification seems to be working better for some fish species compared to others.
Furthermore, Janssen and Hamm (2012) emphasize the importance of a well known and trusted organic certification logos. According to Janssen and Hamm (2012) consumers have different preferences and willingness to pay for certain organic certifications because consumer perceptions of organic certifications are considered as being of subjective nature and not based on objective knowledge. In their experiment the highest price premiums where obtained for labels with cognized strict organic standards and a strict control system (Janssen & Hamm, 2012). On some occasions it was advisable to use an additional organic label next to the most common one.
According to MINTEL (2000) it turned out that the best-know brand in the UK is perhaps just the word “organic” itself rather than explicit certification labels. However, the result of Mintel (2000) is in contrast to the findings of McCluskey
(2000) who emphasizes the importance of a credible organic label, supported by a third-party monitoring organization rather than a company that claims to be organic without a certified eco-label.
A research conducted by MarketResponse Nederland BV (2012) completed 1105 interviews in order to investigate the value of the EKO-Quality Mark under Dutch consumers compared to other labels such as Demeter, Ik kies Bewust, and the EU-label. They reported that the majority of the participants knew the EKO label. 2.3 Attitudes towards organic food
Research focusing on the predictors of attitudes towards organic food has examined the role of health consciousness, food safety concern and ethical self-identity (Michaelidou & Hassan, 2008).
Health consciousness
Health-conscious consumers are living a lifestyle characterized by nutrition, the environment, stress and fitness (Kraft & Goodell, 1992). According to Magnusson, Hursti, Åberg and Sjödén (2003), health is a strong predictor of the attitudes toward the purchase intention and purchase frequency of organic food. This is in line with the findings of Schifferstein and Oude Ophuis (1998), who researched a sample of Dutch consumers, suggesting that health aspects were more salient that environmental aspects for the purchase of organic food.
Ethical self-identity
The consumption of organic food is also an expression of ethical beliefs. Consumers who are concerned with ethical self-identity are characterized by having concerns about several ethical issues such as fair trade, animal welfare and the environment (Harper & Makatouni, 2002). Severe ecological problems are important drivers for this factor (Laroche et al., 2001). According to Harper and Makatouni (2002), the standards of animal welfare are used as indicators of the safety and health conscious of food.
Food safety
There are different associations with the term organic, ranging from “healthy products, less contaminated, good taste, unpackaged, no chemicals/pesticides” to an
“earthy feel” (Padel & Foster, 2005). Wholesomeness and the absence of chemicals are primary reasons to buy organic foods (Schifferstein & Oude Ophuis, 1998). Furthermore, organic food might be viewed as safe, since it has a status of being chemical-free, locally produced, and/or free of genetically modified ingredients (Lockie et al., 2002). Among food safety, ethical self-identity and health
consciousness as motives underpinning the consumption of organic behavior, the first two are reported to be the most important predictors of attitude towards organic produce (Michaelidou et al., 2008). This is in contrast to a research from McEachern and McClean (2002) who concluded that food safety and health are shown to be the primary motives for buying organic dairy products. Hence, food safety seems
repeatedly to be the most important determinant for the attitude towards organic food. This research will therefore only investigate the effect of food safety on consumers’ WTP for organic food. Food safety will be communicated to the consumers by providing information in an advertisement.
2.4 Influence of knowledge on ecologically behavior
The impact of knowledge factors is an important but neglected topic in the study of environmental attitude (Vining & Ebreo, 1990). Vining and Ebreo’s (1990) research focused on the impact of knowledge of recyclers and non-recyclers. Their results indicate that recyclers are overall better informed about the recycle process that non-recyclers. Implications include increasing educational efforts, which successively will increase awareness of recycling. These findings are also supported by Chan (2000); according to his research green consumers possessed better knowledge about green consumption. This might also have an effect of the attitudes and purchase behavior of organic food.
Johnson and Russo (1984) investigated whether product familiarity improved shoppers’ ability to learn new product information. They first analyzed an earlier study that concluded that greater familiarity boosted learning during a new purchase decision. Their own research confirmed this and demonstrated that consumers, who are experienced and have prior knowledge, use this information in the purchasing process. The prior knowledge enables them more efficient information processing in order to focus on the information that is of interest to the task that needs to be executed (Johnson & Russo, 1984).
Rao and Sieben (1992) elaborate on this by examining whether experienced consumers with prior knowledge have a different limit of price acceptability and to what extent this type of consumer uses different types of information. The acceptable price range consists of an upper and lower limit. The upper limits defines the price above which consumers would believe the product would be too expensive, whereas the lower limit classifies the price below which consumers would be cautious about the product’s quality (Rao & Sieben, 1992). The type of information is divided into intrinsic versus extrinsic information that is related to intrinsic and extrinsic
motivation. Intrinsic motivation is defined as ‘doing an activity for its inherent
satisfactions rather than for some separable consequence’ (Ryan & Deci, 2000, p. 56). Hence, doing an activity simply for the enjoyment, rather than its instrumental value. Conversely, extrinsic motivation is ‘a construct that pertains whenever an activity is done in order to attain some separable outcome’ (Ryan & Deci, 2000, p. 60). Intrinsic motivated people use intrinsic information in contrast to extrinsic motivated people who use extrinsic information such as price (Rao & Sieben, 1992). It is found that as knowledge increases, the upper and lower acceptable price range also increases until a certain point whereafter it levels off. Furthermore, high knowledgeable consumers tend to rely on intrinsic information patterns in contrast to low knowledgeable consumers who are unable to interpret intrinsic information and thus solely rely on extrinsic information (Rao & Sieben, 1992).
Xue, Mainville, You, and Nayga (2009), stated that consumers’ prior knowledge about nutrition affects their attitudes towards food, which will
subsequently be translated into their willingness to pay. They tested their hypotheses by conducting in-store experiments using pasture-fed beef. Their results supported their expectation and revealed that nutrition knowledge significantly influences consumers’ willingness to pay. The subjective knowledge is even found to moderate the relationship between attitudes and choice behavior for a durable product (Berger, Ratchford & Haines, 1994). The results support the predictions that product
knowledge plays an important role. Subjective knowledge influences choice
probabilities both directly and in interaction with product attitudes (Berger, Ratchford & Haines, 1994).
2.5 Influence of socio-demographic factors
Different socio-demographic factors such as gender, age, income, marital status, family size, and employment status seem to influence consumers’ willingness to pay. Krystallis, Fotopoulos and Zotos (2006) analyzed the willingness to pay of Greek consumers for a several organic products (olive oil, bread, raisins, oranges etc.). It turned out that, female consumers in younger age groups, with children, with a higher income and higher education groups are most likely to buy organic food. Consistent with these findings, Lockie et al. (2002) highlighted a clear gender dimension to organic consumption; women consumed more organic food than their male counterparts. However other studies did not find clear gender differences and a specific organic consumer’s profile. Therefore, nothing will be hypothesized about social-demographic factors in the conceptual framework.
2.6 Conclusion
In conclusion, different attitudes toward organic food influence consumers’ willingness to pay for organic product; in particular food safety seems to be an important determinant. Furthermore, transferring the message that product is organic appears to be an important factor. However, it is not very clear yet what is the most effective way to implement this; by using a credible organic label, supported by a third-party monitoring organization versus just claiming to be organic without a certified eco-label. Finally, nutrition knowledge affects consumers’ willingness to pay for some products. The subjective knowledge is found to moderate the relationship between attitudes and choice behavior for a durable product. The effect of all these factors together on the consumer’s willingness to pay (WTP) for organic food has not yet been investigated. The current research aims to fill this gap by examining the following research question: How do the provision of an eco-label and food safety information and the consumer’s knowledge influence consumers’ willingness to pay for organic products?
3. Conceptual framework
After analyzing all the previous research concerning organic food in the literature review, the next section proposes the framework and the hypotheses that emerged in this research.
3.1 Eco-label
Based on the findings of the literature described above, it is of great importance for producers to use well-known trustworthy organic labels (Caswell & Padberg, 1992; Krystallis & Chryssohoidis, 2005; Loureiro & Hine, 2002; Janssen & Hamm, 2012; McCluskey, 2000). By displaying this label on the product consumers might be willing to pay a premium for the product they are seeking to purchase. This leads to the following hypothesis:
H1: The presence of a well know trustworthy organic eco label will increase consumers’ WTP for organic food
3.2 Food safety information
Based on the literature, food safety is considered as one of the most important determinants of consumers’ purchase intentions of organic food. (McEachern & McClean, 2002; Michaelidou et al., 2008). This research will examine how this predictor influences consumers’ WTP for organic food. The view of organic food as chemical-free, and/or free of genetically modified ingredients is one of the primary reasons to buy organic foods (Schifferstein &Oude Ophuis, 1998). The focus will be lying on the effect of displayed information on the advertisement of the organic product, which will be auctioned. It is expected that consumers’ willingness to pay will increase when the food safety information is clarified in the advertisement since the additional information is expected to reduce asymmetric information as suggested in the signaling theory. Food safety includes signaling that the product is pesticide, residue and chemical free. This leads to the following hypothesis:
H2: The provision of food safety information within the product description will lead to an increase in consumers’ WTP for organic food
After predicting outcomes using these two independent variables separately, the combination of both the presence of an eco-label and the provision of food safety information within the product description is also expected to increase consumers’ WTP.
H3: The presence of an eco-label and the provision of food safety information within the product description will lead to an increase in consumers’ WTP
3.3 Influence of consumers’ knowledge: moderating effect
Consumers’ knowledge seems to have a positive influence on the attitude toward organic food. As mentioned in section two, awareness seems to be an important factor (Vining & Ebreo, 1990). Furthermore, product familiarity and prior knowledge
improve shoppers’ ability to learn new product information (Johnson & Russo, 1984). Higher price acceptability limits are found for high knowledgeable consumers
suggesting that high knowledgeable consumers are willing to pay more for a certain product (Rao & Sieben, 1992). Berger, Ratchford, and Haines (1994) also highlighted the importance of subjective knowledge and showed that knowledge influences choice probabilities both directly and in interaction with product attitudes (Berger, Ratchford & Haines, 1994). In light of these findings, it is expected that consumers’ knowledge in this research will perform as a moderator (see figure 1). This will be tested by asking participants about their familiarity of the eco-label and whether they think to have knowledge about organic food. These answers will be linked to their WTP for the organic product. This sets the following and last hypothesis:
H4a: Knowledge of organic food will moderate the relationship between the eco-label and consumers’ WTP for organic food
H4b: Knowledge of organic food will moderate the relationship between food safety and consumers’ WTP for organic food
H1 H3 + + H2 + + + H4a H4b Eco-label
Food safety information
Consumers’ WTP
Knowledge
Figure 1.
4. Methodology
In this section several aspects from the methodology will be outlined. First, the research design will be exemplified. Second, the use of Veylinx as the research method and the different treatments for the experiment will be discussed. Third, the product open for auction will be illustrated. Whether the sample in this research is representative will be discussed in section 5.1.1 under sample characteristics. 4.1 Research design
In order to carry out this explanatory research, a suitable research method must be selected. In this research three factors influencing WTP for organic food are being investigated. The most effective and suitable tool to study this is by conducting experiments (Saunders, Lewis & Thornhill, 2012). Experiments allow for
manipulations of independent variables in a controlled setting. Hence employing experiments is the most obvious research method.
The experiment is based on a second price sealed-bid Vickrey auction (Vickrey, 1961). During this auction, participants bid on the product offered without knowing about the bids of other participants. The highest bidder of the auction wins and will receive the product, not paying his own bid price, but the second-highest bid. In this way, truthful bidding is being stimulated which leads to the highest expected outcome (Vickrey, 1961). Each individual is uncertain about the other’s reservation values, believing that each party will bid the maximum amount he is willing to (Riley & Samuelson, 1981).
4.2 Research method
The second price sealed-bid auction tool Veylinx will be carried out for this research. Veylinx (www.veylinx.com) is founded by two PhD students at the UvA. The website is designed for research purposes only, and its goal is not to make profit on the
products being auctioned. Currently, the website has 1500 active subscribers, and subscriptions are open to everyone without any entry fee. A limitation is that the
platform is only available in the Netherlands so far. All the auctions are in Dutch, so this might limit the generalization of the findings.
4.3 EKO Quality Mark
To test the third and fourth hypotheses a well-know trustworthy eco label must be found. In order to achieve this a pilot study was conducted where several organic labels, such as Demeter, and Bio, were shown to participants. Hereafter the
participants were asked whether they were familiar with one of them and which one seems to be the trust worthiest to them. The EKO-Quality Mark was named mostly and is therefore considered as a well-know trustworthy organic label for organic food and used in this research. This is in line with the findings of MarketResponse
Nederland BV (2012), which also indicated that the EKO Quality mark was known by the majority of the participants. The Foundation EKO Quality mark is the legal owner of the collective EKO Quality mark. The Foundation is aiming to construct a
certification, perceived as trustworthy, for products and services that support the organic branch. The four principles (ecology, health, fairness and care) formulated by IFOAM (the platform of European organic organizations), are embedded in the EKO Quality mark. The EKO Quality Mark guarantees the features of organic farming. The provision of such characteristics might be important for consumers’ willingness to pay. Organic farming is ‘an agricultural method with respect for nature and
characterized by the utilization of the natural life cycles’ (Foundation Quality Mark). Organic farming is characterized by seven features:
1. For an efficient use of natural resources at the spot, an ample crop rotation scheme is necessary.
2. Very strict restraints at the use of chemical synthetic pesticides and fertilizers, antibiotics for livestock, food additives, technical aid substances and other additives.
3. It is strictly prohibited to use any genetically modified organisms.
4. Local resources, such as organic manure of animal feedstuff produced at the farm, are used for the production.
5. Plant varieties and animal species adapted to local conditions and those are tolerant or resistant against pests and diseases conditions, are used.
6. The livestock need to have access outdoors and they should be fed with organic forage.
7. Complying the farming with the natural needs of the different livestock European legislation provides the European organic label; a green leaf with stars. This label ensures that the product carries out the following values: ecological, healthy, fair and care. As mentioned in the introduction, there are other additional organic labels that emphasize different important aspects from organic farming, such as the use of green energy, and that the entrepreneur should be transparent about sustainability toward its buyers and consumers according to the signaling theory. Foundation Quality mark is as such an example of an organization that is committed to these additional values and the seven features mentioned above and hence they created the EKO-quality label. The effect on consumers’ willingness to pay of such a quality mark is questioned and the focus of this research.
Altogether, the sample will be divided up in four treatments during the experiment (an overview is given in table 4.2):
1) Treatment one (baseline): No organic EKO-Quality Mark is displayed and no product safety information is displayed.
2) Treatment two: Only product safety information about the product is displayed (no use of pesticides and chemical, free of genetically modified ingredients). 3) Treatment three: Only organic EKO-Quality Mark is displayed
4) Treatment four: Both the organic EKO-Quality Mark and the complementary product safety information about the product is being displayed
After the participants finished their bid on the product, they are asked to fill out some multiple-choice questions (see appendix A). These questions test if the consumers are acquainted with the EKO-label and their knowledge about organic food. The answers will enable inferences with regard to the potential moderating effect of knowledge on consumers’ WTP.
Table 4.2 four different treatments
1 2 3 4
Control X
4.4 Product choice
For this research a jar (450 gram) of honey from organic producer de Traay will be auctioned (appendix A). De Traay is a unique company that strives to produce nature products. They own a bee farm for over 30 years and their bees collect their nectar freely in the well-preserved nature. De Traay doesn’t use any pesticides and their sites are minimally seven km from the industry or traffic. Daily contact with their bee population ensures high quality and constitutes to the basis of their traditional
methods in producing the honey. Furthermore, de Traay places high quality demands on its processing and its products. All of their honeys are established along EKO guidelines and possess the EKO-Quality mark and are therefore well suited for this research.
5. Results
The results of this research will be analyzed in this section. First, some descriptive statistics will be provided. Second, the auction bid data will be tested for normality. Third, correlations between the variables will be discussed. Fourth, the reliability of the sample outcome will be tested. Finally, the hypotheses will be tested using different statistical tests. No tests will be carried out to measure reliability with Cronbach’s alpha. Cronbach’s alpha is an estimate of the internal consistency
associated with the different scores that can be derived from a scale. In this research, the bid amount is the only item that measures participant’s willingness to pay. Hence, a reliability test is superfluous.
5.1 Descriptive Statistics
The next section will focus on the descriptive statistics of the data with regards to sample- and auction bid data characteristics.
5.1.1 Sample characteristics
In this experiment, 366 subscribers of Veylinx participated. Of these 366 subscribers, 187 participants were male, and179 participants were female (Table 5.1.1a). Resulting in 51.1 and 48.9 percent respectively. Compared to the gender distribution of the Dutch population, which consists of approximately 49.5% men and 50.5% women
(CBS, 2014), men are slightly overrepresented and women are slightly underrepresented in the sample.
As shown in table 5.1.1b, there are some interesting differences between the Dutch population and the sample with regard to the age distribution. The sample contains only 2.7% of individuals under 20 years old as opposite to the Dutch population of 22.9% (CBS, 2014). Only 0.3% of the sample participants were older than 80 years, contradictory to 4.3% of the Dutch population (CBS, 2014). Hence, the age groups under 20 and older than 80 years are underrepresented in this sample. A possible explanation could be the recruitment strategy. In order have access to the Veylinx auction tool, students were obligated to recruit at least 100 subscribers. Since users of the Veylinx tool so far are mostly students, it is obvious that these students use their own network to recruit new subscribers and would hence approach friends, family and acquaintances through for example social network. The students might not know many people falling into these two underrepresented age groups. Finally the age groups of 40-65 and 65-80 years are both slightly overrepresented in the sample.
When looking at the education level of the sample (table 5.1.1c) it is notable that the data of 93 (25.4%) participants was missing. Most of the remaining
participants indicated that they are being in or having obtained scientific education. Again, a possible explanation for this relatively high percentage is the recruitment strategy. Most of the users who recruited were students at the UvA itself and used their own network to recruit subscribers. It is likely that their network encompasses a relatively high share of highly educated individuals. It is difficult to compare these education level results of the sample to the Dutch population, since the CBS only focuses on the obtained education of the labor force; individuals between 15-65 year old. Men were found to be willing to pay the most in the control group compared to women, whereas women were willing to pay more than men when an eco-label was presented or food safety information was provided.
Table 5.1.1a Gender distribution Dutch population compared to gender distribution sample (N=366)
Men Women
Dutch population 49.5 50.5
5.1.2 Data characteristics (auction bid data)
Table 5.1.2a summarizes the descriptive statistics for the original auction bid data set. The mean bids for treatment 1, 2, 3 and 4 are 207.59, 207.02, 191.01, and 185.50 eurocents respectively. The corresponding Standard Deviations of these treatments are 218.819, 167.575, 152.815, and 139.909 respectively. The mean bid and the associated SD for treatment one turned out to be the highest compared to the other treatments.
However, the standard deviations are relatively high for all treatments
indicating that the bid amount data are spread out over a large range of values, which could raise questions. The maximum bids for all treatments are relatively high compared to the mean bids. Treatment one (control group) has for example the highest bid of all treatments with 1600 eurocents. That value seems to be an extreme outlier when analyzing the box plots in appendix B. Another possible explanation could be found while studying that the minimum bid for all treatment was zero, as shown in table 5.1.2a. People who bid zero euros indicated that they were not Table 5.1.1b Age distribution Dutch population compared to age distribution sample (N=366)
<20 20-40 40-65 65-80 >80
Dutch population 22.9% 24.5% 35.3% 13.1% 4.3% Dutch population Cumulative percent 22.9% 47.4% 82.7% 95.8% 100% Sample (N= 366) percent 2.7% 42.9% 47.3% 6.8% 0.3% Sample (N=366) Cumulative percent 2.7% 45.6% 92.9% 99.7% 100%
Table 5.1.1c Education level sample (N=366)
Primary Education Secondary Education Middle-level applied education (MBO) Higher professional education (HBO) Scientific Education (WO) Sample (N=366) 1.6% 7.9% 20.2% 21.9% 23.0%
interested in the product that was offered and therefore their willingness to pay is zero. The best option to visually approach a normal distribution, only the top 50 percentile and LN bid amounts are used for further statistical analysis.
5.1.3 Testing for normality
This section will test for the normality of the auction bid data using different test statistics. After all, in order for the significance test to be accurate, the sampling distribution is required to be normally distributed (Field, 2013). According to the central limit theory, the sample distribution can be approximated by a normal distribution if the sample size for each treatment consists of at least 30 participants, (Field, 2013). Although, the sample size for each of the four treatments consists of more than 30 participants, this approximation turns out not to be useful in this research. First, the normality will be tested visually by means of histograms and boxplots. Secondly, the normality will also be tested numerically, by means of both the Kolmogorov-Smirnov and Shapiro-Wilk test.
Looking at the histograms of the frequency of the bids under each treatment, it can be noted that the auction bids in the original dataset are skewed to the right and are therefore considered as positively skewed (appendix B). The same can be
Table 5.1.2a Descriptive statistics auction bid data in eurocents (N=366)
Treatment Mean Std. Dev. Min. value Max. value
1 (N=97) 207.59 218.819 0 1600 2 (N=92) 207.02 167.575 0 753 3 (N=85) 191.01 152.815 0 720 4 (N=92) 185.50 139.909 0 500
Table 5.1.2b Descriptive statistics auction bid data top 50 percentile in eurocents (N=177) Treatment Mean Std. Dev. Min. value Max value 1 (N=47) 355.96 222.779 200 1600 2 (N=44) 351.91 110.460 200 753 3 (N=41) 316.80 108.809 200 720 4 (N=45) 300.51 88.062 200 500
observed for the auction bids in the dataset where only the top 50 percentile is used (appendix B). Similarly, this phenomenon is also observed in both box plots,
suggesting that the data for both datasets are non-normally distributed. However, the data set of the top 50 percentile approaches a normal distribution the best.
From the Kolmogorov-Smirnov statistics, it is deduced that all the outcome statistics are significant (p<0.05) indicating a deviation from normality (Field, 2013) and thus suggesting violation of the assumption of normality. The Shapiro-Wilk statistics confirm this finding. This also reflects the skew in the histograms for the entire auction bid data.
For the statistical analysis of this research, the variable bid amount is recoded into the natural logarithm (LN bid amount) to partly correct for the non-normally distribution. In addition, only the top 50 percentile of the dataset is used for the statistical analyses; both One-way ANOVA and regression.
Table 5.2.1. Tests for normality of Auction bid data
Treatment a= original data set b= top 50 percentile
Kolmogorov-Smirnov Shapiro-Wilk
Statistic df Sig. Statistics df Sig. 1a (N=97) 0.171 97 0.000 0.756 97 0.000 1b (N=47) 0.242 47 0.000 0.600 47 0.000 2a (N=92) 0.162 92 0.000 0.925 92 0.000 2b (N=44) 0.226 33 0.000 0.889 44 0.000 3a (N=85) 0.130 85 0.001 0.927 85 0.000 3b (N=41) 0.220 41 0.000 0.854 41 0.000 4a (N=92) 0.147 92 0.000 0.929 92 0.000 4b (N=45) 0.137 45 0.033 0.901 45 0.001
5.2 Correlations
In appendix C, a table concerning correlations is provided. According to Field (2013), correlations values of ±0.1 represent a small effect, ±0.0.3 a medium effect, and ±0.5 a large effect. Several significant correlations are found. First, a small positive correlation of r(364)= 0.104 , p<0.01 between familiar with eco-label and gender is found. When calculating frequencies, 82.1 percent of women- and 75.9 percent of men indicated to be familiar with the eco-label; suggesting that more women are familiar with the EKO-quality mark.
Another significant correlation of r(364)=0.25, p<0.05 is found between knowledgeable and bid amount. Overall, the mean bid amount of participants considered as knowledgeable, are higher compared to the bid amount of non-knowledgeable individuals. Section 5.3.3 elaborates on this finding.
In addition, there is a significant positive correlation of r(364)=0.259, p<0.05 between knowledgeable and familiarity with eco-label. This suggests that people scoring high on knowledge participants be more likely be familiar with the eco-label.
Finally, there is a significant negative correlation between above 30 years and education, indicating that people above 30 score lower on education levels. This is not of interest for this research though, so no further implications about this finding will be discussed.
5.3 Hypotheses testing 5.3.1 Eco-label
Hypothesis one stated that the presence of a well know trustworthy eco-label will increase consumers’ WTP for organic food. An eco-label was presented in both treatment 3 and 4. To test this prediction, a One-way ANOVA was carried out. The p-value of the ANOVA willingness to pay between groups is not found to be significant (0.151<0.05). Table 5.3.1a shows the mean of the bid values for all four treatments. A striking finding is that the mean of both treatment 3 and 4 is lower than the mean of treatment 1. This suggests that the presence of an eco-label provokes a
counterproductive effect and negatively influences willingness to pay. This
corresponds to the correlation of r(364)=-0.055 between the variable eco-label and bid amount in section 5.2. However, this correlation is not found to be significant (appendix C). In addition, when looking at the results of the LSD test in table 5.3.1c, the mean difference between treatment 1 and 3 is not found to be significant. Neither
the mean difference between treatment 1 and 4 is found to be significant. This result was already expected since the One-way ANOVA was not found to be significant. Hence, hypothesis 1 is rejected.
Table 5.3.1a Means of LN bid amount for all treatments for top 50 percentile bid amount (N=177)
Treatment N Mean SD Minimum Maximum
1 91 5.77 0.428 5 7
2 86 5.82 0.293 5 7
3 41 5.71 0.311 5 7
4 45 5.67 0.292 5 6
Total 177 5.74 0.340 5 7
Table 5.3.1b ANOVA LN bid amount for all treatments for top 50 percentile bid amount (N=177)
Sum of squares Df Mean square F Sign. Between groups 0.611 3 0.204 1.787 0.151
Within groups 19.730 173 0.114
Total 20.341 176
Table 5.3.1c LN bid value mean differences between treatments
1 2 3 4 1 -0.054 0.057 0.102 2 0.054 0.111 0.156* 3 -0.057 -0.111 0.045 4 -0.102 -0.156* -0.045 Note. N=176, *p<0.05
5.3.2 Food safety information
Hypothesis 2 stated that the provision of food safety information would increase consumers’ WTP for organic food. Both in treatment 2 and 4 food safety information is provided. After carrying out the One-way ANOVA test, it is found that the mean of the bid amount for treatment 2 is higher than treatment 1. However, the mean bid amount for treatment 4 is lower than treatment 1. This would suggest that only the provision of food safety information would increase willingness to pay. The LSD rest results show that the bid value mean differences between treatment 1 and 2 are not significant. This was actually already expected since the One-way ANOVA was not found to be significant. Altogether, the results suggest that it is not supported that the provision of information will lead to an increase in willingness to pay. Hence,
hypothesis 2 is rejected. Also hypothesis 3 is rejected. 5.3.3 Knowledge
For the moderation analysis, a new variable has to be created. The new variable ‘knowledgeable’ is defined as the average of ‘having knowledge’ and ‘liking to immerse’. People who scored between 1 and 3 on this scale are considered as knowledgeable, i.e. having knowledge about organic food. Again, variables recoded as 0 are not considered as knowledgeable, whereas variables recoded as 1 are seen as knowledgeable. The data of 11individuals (3 percent) of the sample is missing. From the remaining group, 63.9 percent of the simple indicated to be knowledgeable (table 5.3.3).
In addition, two dummy variables were computed. The first one ‘eco-label’ is constructed by recoding the treatments where an eco-label was provided. Treatment one and 2 where no eco-label is provided is recoded to 0, whereas both treatment 3 and 4 are recoded to 1 since an eco-label was used in both advertisement. The second one ‘food safety information label’ is constructed by recoding the treatments where food safety information was presented. Treatment 1 and 3, where no food safety information was presented, are coded as 0. Treat 2 and 4 are coded as 1 since both of them included a provision of food safety information. To test for moderation, there should be a significant main effect between the variable ‘knowledgeable’ and ‘bid amount’. Section 5.2 already revealed that there is a significant positive correlation between knowledgeable and bid amount (r(364)=0.256, p<0.01). This indicates that when people score high on knowledgeable, their willingness to pay is also higher.
Figure 5.3.3 shows this effect graphically by means of bar charts. For all treatments, participants who indicated to be knowledgeable, have higher mean bid amount than participants who indicated not be knowledgeable.
An interesting emergence is that willingness to pay for knowledgeable people decreases when an eco-label is present. This is a counter intuitive finding. In contrast, not knowledgeable people were willing to pay more with the presence of an eco-label. This suggests that knowledge has a negative influence on the relationship between eco-label and willingness to pay.
Another interesting emergence is the willingness to pay for knowledgeable people increases as soon as food safety information is provided. In contrast, the willingness to pay of not knowledgeable people decreases with the provision of food safety information. Most likely, the knowledgeable people understand and value the information since they like to immerse about organic food and are thus willing to pay more. This would suggest that knowledge moderates the relationship between food safety information and willingness to pay.
In order to test for moderation, a regression analysis is carried out. The regression analysis resulted in a p-value of 0.122 (>0.05) for the model and is not significant. Therefore, there is no evidence that the effect of knowledge as a moderator is significant. Hence, both hypothesis 4a and 4b are rejected.
Table 5.3.3 Descriptive statistics sample variable knowledge N=355
Men Women Total
Knowledgeable 45.1 54.9 100% Non-knowledgeable 54.9 45.1 100%
6. Discussion
The research examined factors influencing willingness to pay (WTP) for organic food. First, it investigated whether the presence of an eco-label would increase the willingness to pay for organic products. Second, the study explored whether the provision of food safety information would have a positive effect on willingness to pay. Third, during the research it is analyzed whether a combination of the two factors would lead to an even higher increase in willingness to pay. Lastly, it was examined if these factors would have an even greater effect if participants would have knowledge about organic food. Altogether, the study aimed to answer the following research question: ‘How do the provision of an eco-label and food safety information and the consumer’s knowledge influence consumers’ willingness to pay for organic
products?’ This section reflects on the key findings of the results, implications to theory, and the limitations & suggestions for further research.
6.1 Key findings
None of the proposed hypotheses are significantly supported. The first hypothesis, which stated that the presence of an-eco label would increase consumers’ WTP, is rejected. Perhaps, the eco-label did not have a significant effect since the
advertisements’ titles already stated that the honey jar was organic. Mintel (2000) found that the word ‘organic’ itself was already sufficient to increase consumers’ willingness to pay It was necessary to do so, since participants otherwise did not know that it concerned an organic product. If the word ‘organic’ was not stated in the title, the effect of conventional vs. organic product would be researched and that was not the focus of this research. The second hypothesis, which stated that the provision of food safety information would increase consumers’ WTP, is also rejected. The third hypothesis, which was a combination of the first two, expected that the presence- and provision of both an eco-label and food safety information is also
rejected. Finally, both hypotheses four and five, which investigated the potential moderating effect of knowledge on the relationship between eco-label and WTP and the relationship food safety information and WTP, are rejected.
6.2 Contribution to theory and practice
The eco-label was found to have a counter intuitive effect, decreasing (instead of increasing) consumers’ WTP. The statistical One-way ANOVA test resulted in no significance, so this finding is not considered as valid. The possibility that the eco-label might not have a positive influence on WTP for organic food may be due to the honey jar that was auctioned. According to previous research, the value-added quality of organic labeling seems to depend on the type of product that is being marketed (Loureiro & Hine, 2002). The possible negative effect of the eco-label is in contrast with the findings of Janssen and Hamm (2012) where the highest price premiums were obtained for labels with cognized strict standards.
Food safety was found to be one of the most important determinants in purchasing organic food (Michaelidou et. Al, 2008). The mean bid amount for the food safety treatment was higher than the mean bid amount for the control group, indicating that food safety would have a positive influence on WTP. However, this finding was not significantly supported and is therefore rejected.
Knowledge is found to be an interesting variable. The mean bid amounts for all treatments were found to be higher when participants indicated to be
knowledgeable. This outcome is in line with the findings of the research of Rao and Sieben (1992) who found that as knowledge increases, the upper and lower acceptable price range increases. Knowledge seemed visually to have an influence on the
relationship of eco-label and WTP suggesting that knowledgeable people would be willing to pay more for the product compared to non-knowledgeable people. Most likely, knowledgeable people understand and value the food safety information since they like to immerse about organic food and are thus willing to pay more. This would be in line with the results of Jonhnson and Russo (1984), who found that product familiarity improved shoppers’ ability to learn new product information. In addition, this was also in line with the findings of Xue, Mainville, You and Nayga (2009) who revealed that nutrition knowledge significantly influenced consumers’ WTP.
Knowledge seemed to have a counter intuitive effect on the relationship between an eco-label and willingness to pay. Knowledgeable people were willing to
pay less when an eco-label was provided, compared to non-knowledgeable people. Perhaps the knowledgeable people would be skeptical about the underlying reasons of presenting an eco-label. They might accuse the company of using the eco-label just to create a better reputation for example.
However, it is difficult to define the exact contributions to theory since none of these findings were significantly tested.
6.3 Managerial implications
Although no significant evidence is found, companies targeting knowledgeable people when selling organic products should put effort in providing the consumer food safety information. This potentially offers an increase in WTP, which should be taking into consideration when computing a selling price. On the other hand, when a company aim to target a broader audience including non-knowledgeable people, food safety information could decrease consumers’ WTP, perhaps because the information scare potential customers off.
The provision of an eco-label could have a counterproductive effect. Hence, it is recommended to first investigate what eco-label your consumers value the most before including these in marketing practices. An eco-label might only slightly
increase WTP of non-knowledgeable consumers whereas is might lead to a substantial decrease in WTP of knowledgeable consumers. Surprisingly, men were found to be willing to pay more for organic food compared to women in the control group. On the other hand, women were more influenced by the presence of an eco-label and food safety information. Hence, what would be the best strategy depends on the gender of the consumers the company is aiming to target with the marketing strategy.
6.4 Limitations and suggestions future research
The results of this research also involve limitations. One of the limitations concerns the participants the sample consists of. All participants were Dutch, sharing the same culture. Dutch people are generally known for being down to earth. Dutch people might be skeptical about organic farming and might avoid non-sense. Further research should examine whether the same results hold for different populations and different types of culture. American people for example, might respond differently since the American purchasing behavior might be more influential through advertisements.
Another limitation is that the hypotheses were tested by auctioning a jar of organic honey. The influence of the factors might depend upon the type of product that is being auctioned. Further research should investigate whether these findings are also found when auctioning other organic food products such as vegetables and diary product. People might respond differently to these types of organic food since they generally have a shorter shelf life and the ‘freshness’ for these types of products might be more important. The influence of the factors might also depend on the frequency of purchasing a product. Krystallis and Chryssohoidis (2005) already hypothesized that Greek consumers were willing to pay more for organic alternatives if the product was something that they frequently buy (basic products) vs. products they buy occasionally (non-basic products). However, these results were not valid. Further research should replicate this finding and research whether the factors influencing consumers’ willingness to pay depend on the purchasing frequency of a product, for example milk (basic product) vs. olive oil (non-basic product).
In addition, some limitations might be found in the after auction
self-administrated questionnaire. A certain amount of questions were included, in order to prevent participants to experience the questionnaire as tedious. More questions should be asked to test the potential moderating effect of knowledge. Perhaps a real
knowledge scale could be developed and tested.
Furthermore, the EKO quality mark is a Dutch eco-label. The label is not found to significantly increase consumers’ WTP. Further research could investigate whether other Dutch labels would have a significant influence on WTP for organic food such as Demeter and Bio. The effect of foreign labels such as Bioswiss and USDA organic could be tested as well.
7. Conclusion
Nowadays, the concern for the environment has become increasingly important for both consumers and producers (Chan, 2000). Global demand for the organic market has increased significantly (Sahota, 2009). Ever since a lot of research is been done concerning consumers’ willingness to pay for organic products. Several recurring identified factors influencing the purchase decision are “quality and security”, “trust”, “senses”, “price sensitivity”, “convenience”, and “brand-name value” (Greeks,
eco-label, food safety information and knowledge have been done. This study investigated how the presence of an eco-label, the provision of food safety information and the consumer’s knowledge influenced consumers’ willingness to pay for organic products. Results were obtained by auctioning an organic product in four different ways by means of an online auction platform Veylinx. Surprisingly, despite of the interesting correlations found, results turned out not to be significant. Consequently, all 4 hypotheses were rejected.
Bibliography
Berger, I. E., Ratchford, B. T., & Haines Jr, G. H. (1994). Subjective product
knowledge as a moderator of the relationship between attitudes and purchase intentions for a durable product. Journal of Economic Psychology, 15(2), 301-314.
Broekert, T. De., & Rooijen, I. Van (2012). Meerderheid kent het EKO-keurmerk. Onderzoek naar de waarde van het EKO keurmerk onder Nederlandse booschappers. MarketResponse Nederland BV.
Caswell, J. A., & Padberg, D. I. (1992). Toward a more comprehensive theory of food labels. American Journal of Agricultural Economics, 74(2), 460-468.
CBS (2014). Bevolking; kerncijfers. Accessed on 23 July 22, 2014.
(http://statline.cbs.nl/StatWeb/publication/?VW=T&DM=SLNL&PA=37296n
ed&D1=a&D2=0,10,20,30,40,50,60,(l-1),l&HD=130605-0924&HDR=G1&STB=T)
Chan, K. (2000). Market segmentation of green consumers in Hong Kong. Journal of International Consumer Marketing, 12(2), 7-24.
Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011). Signaling theory: A review and assessment. Journal of Management, 37(1), 39-67. Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage. 4thed.
Chicago
Foundation Quality Mark. What does organic mean? Retrieved from:
http://www.eko-keurmerk.nl/organic
Gil, J. M., Gracia, A., & Sanchez, M. (2000). Market segmentation and willingness to pay for organic products in Spain. The International Food and Agribusiness Management Review, 3(2), 207-226.
Harper, G. C., & Makatouni, A. (2002). Consumer perception of organic food production and farm animal welfare. British Food Journal, 104(3/4/5), 287-299.
Janssen, M., & Hamm, U. (2012). Product labeling in the market for organic food: Consumer preferences and willingness-to-pay for different organic
certification logos. Food Quality and Preference, 25(1), 9-22. Johnson, E. J., & Russo, J. E. (1984). Product familiarity and learning new
Kraft, F. B., & Goodell, P. W. (1992). Identifying the health conscious consumer. Journal of Health Care Marketing, 13(3), 18-25.
Krystallis, A., & Chryssohoidis, G. (2005). Consumers' willingness to pay for organic food: factors that affect it and variation per organic product type. British Food Journal, 107(5), 320-343.
Krystallis, A., Fotopoulos, C., & Zotos, Y. (2006). Organic consumers' profile and their willingness to pay (WTP) for selected organic food products in
Greece. Journal of International Consumer Marketing, 19(1), 81-106. Laroche, M., Bergeron, J., &Barbaro-Forleo, G. (2001). Targeting consumers who
are willing to pay more for environmentally friendly products. Journal of consumer marketing, 18(6), 503-520.
Lockie, S., Lyons, K., Lawrence, G., & Mummery, K. (2002). Eating ‘green’: motivations behind organic food consumption in Australia. Sociologiaruralis, 42(1), 23-40.
Loureiro, M. L., & Hine, S. (2002). Discovering niche markets: A comparison of consumer willingness to pay for local (Colorado grown), organic, and GMO-free products. Journal of Agricultural and Applied Economics, 34(3), 477-488 Magnusson, M. K., Arvola, A., Hursti, U. K. K., Åberg, L., & Sjödén, P. O. (2003).
Choice of organic foods is related to perceived consequences for human health and to environmentally friendly behaviour. Appetite, 40(2), 109-117.
McCluskey, J. (2000). A game theoretic approach to organic foods: An analysis of asymmetric information and policy. Agricultural and Resource Economics Review, 29(1), 1-9.
McEachern, M.G. & McClean, P. (2002) Organic purchasing motivations and attitudes: are they ethical? International Journal of Consumer Studies, 26, 85– 92.
Michaelidou, N., & Hassan, L. M. (2008). The role of health consciousness, food safety concern and ethical identity on attitudes and intentions towards organic food. International Journal of Consumer Studies, 32(2), 163-170.
MINTEL (2000). Organic Food and Drink Retailing. Market Intelligence Unit of the UK Economic Intelligence Unit, London.
Owusua, V., & Anifori, M. O. (2013). Consumer Willingness to Pay a Premium for Organic Fruit and Vegetable in Ghana. International Food and Agribusiness Management Review, 16(1).
Padel, S., & Foster, C. (2005). Exploring the gap between attitudes and behaviour: Understanding why consumers buy or do not buy organic food. British Food Journal, 107(8), 606-625.
Phillip, B., & Dipeolu, A. O. (2010). Willingness to pay for organic vegetables in Abeokuta, South West Nigeria. African Journal of Food, Agriculture, Nutrition and Development, 10(11).
Rao, A. R., & Sieben, W. A. (1992). The effect of prior knowledge on price acceptability and the type of information examined. Journal of Consumer Research, 256-270.
Riley, J. G., & Samuelson, W. F. (1981). Optimal auctions. The American Economic Review, 381-392.
Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary educational psychology, 25(1), 54-67.
Sahota, A. (2009). The global market for organic food & drink. The World of Organic Agriculture, Statistics and Emerging Trends.
Saunders, M., Lewis, P., & Thornhill, A. (2012). Research methods for business students, 6th edition, Pearson Education India.
Schifferstein, H. N., & Oude Ophuis, P. A. (1998). Health-related determinants of organic food consumption in the Netherlands. Food quality and Preference, 9(3), 119-133.
Stiglitz, J. E. (2002). Information and the Change in the Paradigm in Economics. American Economic Review, 460-501.
Vickrey W.(1961) Counterspeculation, Auctions, and Competitive Sealed Tenders. Journal of Finance, 16(1), 8-3.
Vining, J., & Ebreo, A. (1990). What makes a recycler? A comparison of recyclers and nonrecyclers. Environment and behavior, 22(1), 55-73.
Xue, H., Mainville, D. Y., You, W., & Nayga Jr, R. M. (2009). Nutrition Knowledge, Sensory Characteristics and Consumers’ Willingness to Pay for Pasture-Fed Beef. In 2009 Annual Meeting, July 26-28, 2009, Milwaukee, Wisconsin (No. 49277). Agricultural and Applied Economics Association.
APPENDIX A
A1 The auctioned product
A2 Treatment 1 (baseline)
A3 Treatment 2 A4 Treatment 3
A5 Treatment 4
A5 Survey questions after the auction (in Dutch)
1)Ben je bekend met het EKO-‐keurmerk? o Ja
o Nee
2) Ik verdiepmijgraag in biologischeproducten
Likert scale: helemaalmeeeens 1 -‐2 -‐3 – 4 – 5 helemaalmeeoneens
3) Ik hebveelkennis van biologischeproducten
Likert scale: helemaalmeeeens 1 – 2 – 3 – 4 – 5 helemaalmeeoneens
A6 English translation of survey questions after the auction
1) Are you familiar with the eco quality trademark? o Yes
o No
2) I like to immerse about organic products
Likert scale: Totally agree 1 – 2 – 3 – 4 – 5 Totally disagree
3) I have a lot of knowledge of organic food
Likert scale: Totally agree 1 – 2 – 3 – 4 – 5 Totally disagree
APPENDIX B
1) Histograms and box plots auction data bid amount
2) Histograms and box plots auction data bid amount top 50 percentile
APPENDIX C
Correlation Bid
amount Treatment Gender Education Familiarwith eco-‐ label Knowl edgeab le Above 30 year Label treatm ents Food safetyt reamte nts Bid amount 1 Treatment -‐0.054 1 Gender -‐0.026 0.024 1 Education 0.023 0.027 -‐0.020 1 Familiar with eco-‐ label -‐0.029 0.009 0.104* 0.111 1 Knowledgea ble 0.256** 0.017 0.142** 0.078 0.259** 1 Above_30 -‐0.074 0.008 -‐0.038 -‐ 0.360** 0.085 -‐0.015 1 Label treatments -‐0.055 0.897** -‐0.006 0.057 0.030 0.018 0.012 1 Food safety treatments -‐0.010 0.471** 0.066 -‐0.053 -‐0.041 0.001 -‐0.008 0.033 1 Note. *p< 0.01 **p<0.05
