R ED , W HITE OR G REEN W INE ?

Master Thesis – Maarten Kwakman – University of Groningen 1

R

ED

,

W

HITE OR

G

REEN

W

INE

?

O

D

:

T

B

,

Master Thesis – Maarten Kwakman – University of Groningen 2

R

ED

,

W

HITE OR

G

REEN

W

INE

?

O

D

:

T

B

,

University of Groningen Faculty of Economic and Business Master Thesis Marketing Management

Master Thesis July 28, 2013 Maarten Kwakman Rode Weeshuisstraat 5 9712 ET Groningen m.kwakman.1@student.rug.nl Tel: +31653921701 Student number: 1701940

Thesis supervisor: Dr. W. Jager Co-assessor: Dr. D. Trampe

Master Thesis – Maarten Kwakman – University of Groningen 3

Abstract

The present study attempts to offer more insight into the Dutch consumers’ perception of organic wine. Despite growing attention to sustainable farming practices and organic food, the research on organic wine has been limited. With use of quantitative data, we found evidence that wine characteristics are differently evaluated for organic wine in comparison to conventional wine. Information about the production method is of more importance for organic wine than it is for conventional wine. Secondly, we found significant positive effects of normative and informative influence on the purchase intention of organic wine. This effect is moderated by the amount of organic knowledge. Furthermore this paper underlines the importance of label attractiveness and gender differences in the preference for organic wine description’s. Finally, consumers’ health-consciousness is positively related to organic wine purchase intention.

Master Thesis – Maarten Kwakman – University of Groningen 4

Table of Contents

2.1 Theory of Planned Behaviour ... 12

2.2 Attitude ... 13

2.3 Social influence ... 16

2.4 Perceived behavioural control... 19

2.4.1 Perceived product availability... 20

2.4.2. Perceived Consumer Effectiveness ... 20

2.4.3 Perceived Financial Resources ... 21

2.5 Consumer Characteristics ... 21 2.5.1 Health Consciousness ... 22 2.5.2 Quality Consciousness. ... 23 2.5.3 Price Consciousness. ... 23 2.6 Product Knowledge ... 24 2.7 Purchase Intention ... 25 2.8 Self-Reported Behaviour ... 25 2.9 Conceptual model: ... 27 3. Methodology ... 28 3.1 Research method. ... 28 3.1.1 The interview ... 29 3.1.2 The questionnaire ... 30 3.3 Plan of analysis... 34 4. Results ... 36 4.1 Attribute importance ... 37 4.2 Social influence ... 43

4.2.1 Normative social influence ... 43

4.2.2 Informative social influence ... 45

4.3 Perceived behavioural control... 46

4.3.1 Perceived financial resources (PFR) ... 47

4.3.2 Perceived consumer effectiveness (PCE) ... 48

4.3.3 Perceived product availability (PPA)... 48

4.4 Customer characteristics ... 48

4.5 Self-reported Behaviour ... 49

4.6 Quality vs. Sustainable wine description ... 51

5. Discussion ... 53

5.1 Wine Attributes ... 54

5.1.1 Organic vs. Conventional Wine ... 55

5.2 Social Influence ... 57

5.2.1 Knowledge about organic products ... 58

5.3 Perceived behavioural control... 58

Master Thesis – Maarten Kwakman – University of Groningen 5

5.5 Self-reported behaviour. ... 59

5.6 Quality or Sustainable description? ... 60

5.7 Gender ... 60

5.8 Limitations ... 61

5 Conclusion and Recommendations ... 61

6 References ... 63

7 Appendix 1 ... 69

The questionnaire ... 69

8 Appendix 2 ... 74

Master Thesis – Maarten Kwakman – University of Groningen 6

“As far as I’m concerned, if there is a supreme being then He chose

organic evolution as a way of bringing into existence the natural

world” – David Attenborough

1. Introduction

Master Thesis – Maarten Kwakman – University of Groningen 7 importance of sustainable food had led to an increased amount of research on organic consumers’ behaviour, especially on food (Hugner et al., 2007; Chryssohoidis et al., 2005; van Doorn & Verhoef, work in progress; Padel & Foster, 2005). But less is known about the organic wine consumption. Also the concepts of sustainable farming and organic farming have been around for a long time, but organic wine farming is quite new and not fully accepted. As a respondent in the Remaud et al. (2008) research put it aptly: “Organic? I think food, I don’t think wine’. Nevertheless, there is a strong increase in the organic wine trade observed. The increase between countries varied a lot from 5% to almost 90%, with a total average of about 18% for all European countries combined (Jonis et al., 2008). According to our business expert, the organic wine market is less than 1% of the total Dutch wine market. Nonetheless he identified an immense growth within all segments (retail, hotel and catering industry). Furthermore did he state that organic wine has outgrowth the alternative circuit and is becoming more mainstream.

According to Remaud et al. (2008) do consumption trend of organic wine differ from those of the conventional wine market. The latter remains stable and even decreases in the main European producing countries like France, Italy and Spain. This implicates that there could be opportunities in the organic wine market and that further exploration of this topic is necessary. As mister Bijleveld (General Manager of Dutch wine importer Intercaves B.V.) stated during an interview: “There is not much concrete information on organic wine in the Netherlands, moreover reason for you to sink your teeth into it”.

Master Thesis – Maarten Kwakman – University of Groningen 8 Europe. This could be explained by the fact that there is no univocal definition of organic wine in Europe and the US (Winespectator1). Until recently, organic wine in Europe was defined as “wine made of organic grapes”, but since 2012 new rules on organic wine are defined in the Regulation No 203/2012 EC of 8 March. The “Organic logo of the EU” (See Figure 1) may be used provided that the wine-making process complies with Chapter 3a of Title II of this regulation. In sum, organic vines are grown without the use of pesticides, fungicides and herbicides. Also organic wines do not contain sulphur dioxide (SO2). All these synthetic additives are believed to pollute the air, drink water and the agricultural land and are perceived as bad for the health. Often ‘organic’ and ‘sustainable’ wines are used as synonyms, however there are some differences we would like to stress out. In contrast to the organic certification, there are no set of standards for sustainable wineries. Wineries that take the ecology of the vineyard into account and try to minimize chemical treatments and energy use are called sustainable (Organic Wine Journal, 2008). Sustainability also refers to arrange of practices that are not only ecologically sound, but also economically viable and social responsible (winespectator). A third designation of wine is that of ‘biodynamic’. Similar to sustainable wine, biodynamic wine does not have to meet a set of standards. Biodynamic wines are often made without following the strict list of rules in winemaking. In this paper we will focus solely on organic wine and thus wines that are entitled to the European organic logo. International news agency Reuters (April 17th, 2012) stated that organic produce is seeping into the wine industry as more and more producers adopt green practices. In France for instance, the number of organic wineries had doubled in three years and organic wines now makes up 6% of the sales (The Connexion, 2012). But also in other countries (e.g. Australia) the amount of organic wineries is growing.

1

Master Thesis – Maarten Kwakman – University of Groningen 9

Figure 2: Use of pesticides in organic vs. conventional wine (source: CVUA Stuttgart2₎

Organic viticulture is often viewed as a more sustainable practice than conventional viticulture. One of the main objectives of organic farming is to give more importance to soil biological functioning in order to improve its physical (affecting the circulation of water, aeration), chemical (affecting the availability of nutrients) and biological (affecting the biodiversity and fate of organic matter) properties (Coll et al., 2011). In wine production, the soil is considered, together with climate as a key component of terroir (Van Leeuwen et al., 2004). There is some evidence that organic is a more sustainable system and can produce higher quality (Wheeler and Crisp, 2009). For example a significant difference in residue load between conventional and organic wines is observed. As shown in figure 2, the amount of pesticides found in conventional wine is much higher than the amount found in organic wine (CVUAS3). The use of pesticides in agriculture is often criticized. Not only are pesticides dangerous to the environment, but they are also hazardous to a person’s health. Pesticides are stored in your colon, where they slowly but surely poison the body. For example, if you are eating a non-organic apple, you are also eating over 30 different pesticides that have been sprayed on the apple (Nature, 20114). As you can see some of the chemicals used in conventional viticulture could be harmful for consumers. However, the long-term differences between organic and conventional farming on for example the soil quality are difficult to assess (Coll et al., 2011).

Master Thesis – Maarten Kwakman – University of Groningen 10 (Fotopoulus et al., 2003). One of the outcomes is that consumers with a healthy life style are those willing to pay a higher price for an organic wine (Brugarolas Mollá-Bauzá, 2005). On average the price premium that the Spanish consumers are willing to pay for organic wine is between +16.29% and +16.92%. Moreover, around 14% of the population is willing to pay a premium price of +25% for an organic wine (Brugarolas Mollá-Bauzá, 2005). Results from Skuras and Vakrou (2002) indicate that wine consumers’ willingness to pay varies only according to social and demographic characteristics. Also Spanish wine consumers were segmented based on age, income, family size and region (Gil and Sanchez, 1997) Besides consumer segments with a healthy lifestyle, consumers concerned about the environment and product quality can also be considered as a market opportunity for organic products (Grugarolas Mollá-Bauzá, 2005). Besides these papers, the research on organic wine consumption has been limited. Furthermore, the effect of social influence in organic wine purchases remains underexposed.

Master Thesis – Maarten Kwakman – University of Groningen 11 However, some modifications are done to the model. First of all we replaced ‘subjective norms’ by Burnkrant and Cousineau‘s (1975) Normative and Informative social influence. Secondly we used ‘attitude’ towards the object in a different way and measured the relative importance of wine attributes and their influence on the purchase Becauseintention. Because there is a growing interest and production of organic wine, it is necessary to inquire the purchase motives or deterrents for organic wine. There are mixed results about the acceptance of environmentally friendly consumer products. Gupta and Ogden (2009) stated that it is quite a challenge for marketer to identify green consumer and effectively market green products. This research should provide more insight in how to target organic wines in the Dutch market. Hence the research question: What makes Dutch consumers want to buy organic wine?

Procedures

Master Thesis – Maarten Kwakman – University of Groningen 12

2. Literature Review

2.1 Theory of Planned Behaviour

Since 1990’s research concerning the determinants of organic food consumption has gained momentum. In the first references, organic food consumption was mainly approached by applying the values theory of Rokeach (1973) and Schwartz (1992). More recently, researchers also tested the theory of planned behaviour. Research in social psychology has extensively referenced and used Fishbein and Ajzen’s Theory of Reasoned Action (TRA) to predict and understand motivational influences on behaviour. The TRA was developed to better understand relationships between attitudes, intentions and behaviour (Fishbein, 1967; Fishbein and Ajzen, 1975; Ajzen & Fishbein, 1980). Ajzen has expended this theory by including perceived behavioural control as additional predictor of intentions and behaviour which is known as the Theory of Planned Behaviour (TPB, see figure 3) (Ajzen, 1985; Ajzen, 1991; Madden et al., 1992).

Figure 3: Theory of Planned Behaviour - Icek Ajzen (1991)

Master Thesis – Maarten Kwakman – University of Groningen 13 as anticipated impediments and obstacles (Ajzen, 1991). Results have shown that theory of planned behaviour is superior to better than theory of reasoned action in predicting unethical behaviour (Chang, 1998). Aertsen et al. (2009) have found that the value theory and the theory of planned behaviour are relevant theories for better understanding consumers’ choice for organic food. In our study a modified The TPB will be used as a guiding principle for the investigation ofng the Dutch consumers’ intentions to buy organic wine.

2.2 Attitude

Attitude is the first predictor of behavioural intentions according to the TPB. It refers to the degree to which a person has a favourable or unfavourable evaluation or appraisal of the behaviour in question (Ajzen, 1991). Attitudes are evaluative responses, directed towards an attitudinal object. They result from the multiplication of beliefs with their evaluations. In their research on organic fruit, Saba and Messina (2003) found that attitude was a significant predictor of intention of eating organic fruits and vegetables. Also studies from Chen (2007), Thogersen (2007) and Dean et al. (2008) report a significant positive relation between consumers’ intention to purchase organic food and their attitude to organic food purchase, subjective norm and perceived behavioural control (PBC) (Aertsen et al., 2009).

Table 1: Important wine attributes (Fotopoulus et al., 2003)

Glass bottle Paper Packing

Size Label

Taste Aroma

Colour Price

Wine specific info:

 No. of bottles per harvest year  Grape variety  Alcoholic degree Dates written:  Bottling  Harvest  Best before Origin:  Country  Area of production Quality assurance:

 Appellation d’origine contrôlée (AOC) label

 Organic label  ISO/HACCP

Brand name Producer’s name

Information:

 Text with information about the production method

 Text with information about the area of production

 Preservation instructions

Wine bottle with different than the usual shape

Master Thesis – Maarten Kwakman – University of Groningen 14  Consumption instructions

Wine is an experiential consumer product that is difficult for a consumer to judge by looking at it (Barber et al., 2009). Wine consumers are generally facing a wide variety of wines with different qualities, prices and other related attributes. Choosing a specific wine, therefore, is a complicated task for consumers (Lowengart and Cohen, 2006). That’s why we don’t think that the choice between organic or conventional wine is as clear-cut as that and beside an organic claim depends on many other characteristics. In this paper we will therefore not focus on the attitude in general towards organic wine, like TPB would suggest. All the more because organic wine has a low market share and is not fully accepted yet by the Dutch consumers we are interested in what characteristics organic wine should have to become a substitute of conventional wine. To gain insight in why consumers would buy organic wine, we would like to understand how consumers select and choose a bottle of (organic) wine. We will therefore focus on the evaluation of wine characteristics; the Dutch consumers perceived importance of different wine attributes; instead of a general attitude towards organic wine.

Fotopoulos et al. (2003) identified an attribute list on which Greek consumer value organic wine (see table 1). They found differences in preferences between organic buyers and non-organic buyers. Organic buyers for instance prefer a large variety of different wine attributes (e.g. value-for-money, traditional image of the label, region of production, its AOC sign, number of bottles per year, harvest date, cylindrical bottle and clear colour) because they perceive that these attributes satisfy their search for high quality. The organic buyers mostly prefer extrinsic quality cues found on the wine label, because their purchasing motives are ‘high quality’, ‘healthiness/purity’, and ‘traditional image/authenticity’, to a greater extent than those of organic buyers. The non-organic buyers, in turn, appreciate appearance and organoleptic wine attributes because their purchasing motives are food’s ‘attractiveness’, ‘valuable information’, ‘good taste’, ‘distinctiveness’ and ‘control’, to a greater extent than those of the buyers Overall, there are five main motivational benefits of wine purchase: Healthiness, quality, information, attractiveness and good taste (Fotopoulos et al., 2003). The attributes that matter in the

Master Thesis – Maarten Kwakman – University of Groningen 15 consumers buying decision, according to our business expert, are price, wine house, attractiveness of the bottle, a screw top (only for organic wines) and label attractiveness.

Results from a research conducted in New Zealand found that consumers have an increasingly preference for wines that have been produced using environmentally sustainable practices and are labelled as such. A majority of consumers wished to see information about production practices on the labels of environmentally sustainable wines (Forbes et al., 2009). In a study under Italian consumers (Corduas et al., 2013), the following wine attributes were identified as key drivers of wine choice: grape variety, region of origin, protected geographical status, producer, wine complexity or taste, aroma/ bouquet, alcoholic degrees, colour, food-pairing, wine features described by the label information, bottle shape, brand name and label aspects and the quality-price ratio.

The quality of a wine is unknown until the bottle is opened and the wine is tasted. At the moment that the consumer has to make a decision, in the absence of information about the product, the extrinsic attributes are often used as heuristic cues to evaluate the quality of the wines and facilitate decision-making (Speed, 1998). Extrinsic attributes can be modified without changing the quality of the product, such as price, brand, packaging and shelf location. Intrinsic attributes are directly related to the product, such as the grapes variety, alcohol content and wine’s style (Lockshin & Hall, 2003). The most important attributes found in recent research were price, the origin of the wine region. The label of the wine bottle has proven to be a very powerful tool in attraction and seduction of customers (Jennings and Wood, 1994; Chaney, 2000; Fernandes Ferreira Madureira, 2013). Besides a front label there is often a back label on wine bottles with a sensory description that provides information on the taste of the wine. The wine maker has to make sure that this taste description is understandable, credible and relevant to the consumer, and reflects his or her likely perception of the wine (Mueller et al.5). We will therefore conduct an experiment with two types of organic wine descriptions (quality- and sustainability focussed) to measure what kind of description is more relevant to consumers. We conduct this experiment because consumers concerned about the

5

Master Thesis – Maarten Kwakman – University of Groningen 16 environment and product quality can be considered as a market opportunity for organic products (Grugarolas Mollá-Bauzá, 2005),

Organic food consumers tend to perceive organic products as having particular intrinsic characteristics (Yiridoe et al. 2005). The result is that consumer choice for wine is more complex than the choice for many other products (Lockshin & Hall, 2000). It is therefore interesting to examine what attributes are important for customer to buy organic wine, because previous research lacked a thorough examination of wine attributes in consumers buying decisions. Moreover we want to investigate whether or not the importance of wine attributes differs between conventional and organic wines. A second objective is Furthermore we would like to know measuring to what extent individualthe attitude towards product attributes influence the purchase intention of organic wine. According to the TPB, the purchase intention is affected by the attitude a consumer holds about the object and the behaviour, hence the first hypothesis: We expect that the importance of wine attributes in the purchase decision will differ between organic and conventional wine. Besides, we think that a wine attribute that is perceived as important will have a significant effect on the purchase intention.

H1: A positive attitude towards organic wine attributes will positively affect the intention to buy.

H1a: The perceived importance of wine attributes will differ between organic and conventional wine

H1b: Highly perceived important wine attributes will have a positive effect on the organic wine purchase intention.

2.3 Social influence

Master Thesis – Maarten Kwakman – University of Groningen 17 (e.g. expert opinion). Normative social influence is referred to as the tendency to conform the expectations of other persons or groups. The greatest normative influence is usually exerted within primary reference groups such as the immediate family or friends (Cooley, 1962). Under either informational or normative conditions, the uniformity of information provided by others regarding the quality of a product should have a direct bearing on consumers’ evaluations (Cohen and Golden, 1972). The data also suggested that even for familiar product whose taste was the sole criterion for evaluations, individual judgments may be modifiable by the perceived evaluations of others. In line with the research of Cohen and Golden, Pincus and Waters (1977) also found that informational social influence in the form of others’ evaluation does operate in judgments of product quality.

The effect of social influence on moral behaviour (Vallerand et al., 1992) and organic food buying behaviour has been recognized (Tarkiainen et al., 2005). However, within the research on wine purchase behaviour the effect of social influence has not been examined very thoroughly. Wine is a product where the elements of risk during the decision making process are significant for the vast majority (Batt and Dean, 2000; Lockshin and Halstead, 2005 ).It has been suggested that wine can be intimidating due to a range of factors, including its complexity (many tastes, manufacture methods, regions or origin, brands, prices etc.); the 100.000’s labels available for purchase; a general lack of wine product knowledge by consumers; the perceived formality of wine and the variability of style, vintage and variety (Batt and Dean, 2000; Lockshin and Halstead, 2005). Furthermore, the quality of the wine is difficult to assess before it is consumed. These are all factors that could motivate a consumer to rely on social influence in their buying decision. Vallerand et al. (1992) found that subjective norms influence moral behaviour. A shopper research conducted in the Netherlands also found evidence that Dutch consumers often rely on the opinion of others when purchasing wine (Rapport Shopperonderzoek Wijn, 20106). These opinions came from friends, families, wine magazines and wine clubs. Consumers however differ in the degree to which they are influenced by referents while making product and brand consumption decisions (Childers and Rao, 1992). In this paper we would like to examine to what extent people are socially

6

Master Thesis – Maarten Kwakman – University of Groningen 18 influence in the purchase intention of organic wine. We are going to focus on two

distinctive types of social influence, namely normative and information influence. The hypotheses are:

H2a: Normative influence will affect the intention to buy organic wine H2b: Informative influence will affect the intention to buy organic wine

2.3.1 Product Knowledge

Susceptibility to interpersonal influence is a general trait that varies across persons and that a person’s relative influenceability in one situation tends to have a range of other social situations. Susceptibility to influence by others is related to other personal characteristics (e.g. self-esteem, intelligence) (McGuire, 1968; p. 1132). Informational influence operates through the process of internalization, which occurs if information from others increases the individual’s knowledge about some aspects of the environment (Bearden et al., 1989). Involvement with a product and environmental issues are

important constructs when assessing consumers willing to purchase, with previous research emphasizing the importance of the relationship between product involvement and product knowledge. An integration of product use as a measure of familiarity in addition to self-reported knowledge is evidenced in Bettman and Park (1980). They refer to prior knowledge and experience as factors which influence consumers’ information search strategies. Product knowledge and involvement often correlate with each other (Bettman and Park, 1980). However Sujan (1983) argued that knowledge has an independent effect on information processing, as involvement could not explain the information processing patterns. Because knowledge influences the information processing, we expect knowledge to affect the amount of interpersonal influence that is present. We therefore hypothesize that product knowledge would moderate the effect of social influence on the purchase intention.

H3a: Product knowledge will have a moderating effect on the relation between Normative Influence and Purchase Intention

Master Thesis – Maarten Kwakman – University of Groningen 19 2.4 Perceived behavioural control

The third determinant of purchase intention which also influences the actual behaviour according to the theory of planned behaviour is perceived behavioural control (PBC). PBC is referred to as the ease or difficulty of performing the behaviour, or “the perceived control over performance of behaviour” (Ajzen, 1991, 2001). This construct was added in an attempt to deal with situations in which people may lack complete volitional control over the behaviour of interest (Ajzen, 2002). People intend to engage in behaviours of which they feel they are capable (Conner & Armitage, 1998). Behaviour may be said to be completely under a person’s control if the person can decide at will to perform it or not to perform it. Conversely, the more that performance of the behaviour is contingent on the presence of appropriate opportunities or on possession of adequate resources (time, money, skills, cooperation of other people, etc.) (Ajzen & Madden, 1986)

Research on organic food has identified two important determinants of organic buying behaviour. Both availability and price have been proven to have a significant effect on the perceived behavioural control (van Doorn & Verhoef, work in progress). The cost and availability of organic products has also been found to be a barrier to consumption in other studies (Fotopoulos and Krystallis, 2002; McEachern and McClean, 2002; Lea and Worsley, 2005). The study by Lea and Worsley (2005) has found that people would purchase more organic food if it were less expensive and more available.

Master Thesis – Maarten Kwakman – University of Groningen 20 H2a1: Product knowledge will have a moderating effect on the relation between Normative Influence and Purchase Intention

H2b1: Product knowledge will have a moderating effect on the relation between Informative Influence and Purchase Intention.

2.4.1 Perceived product availability

The availability of organic products varies substantially across product categories (Bezawada and Pauwels 2010), which could be an important obstacle for organic purchases (Vermeir and Verbeke 2006). The term ‘Perceived availability’ indicates if a consumer feels he/she can easily obtain or consume a certain product. Although the motivation of consumers to buy sustainable products can be high, it is possible that the intention to purchase sustainable products is hampered due to a low real or perceive availability of the good (Vermeir and Verbeke, 2006). According to Shepherd et al. (2005) the availability and the offering of organic foods for sale has increased in several European countries during recent years. However, in some European countries (e.g. Greece and Spain) consumers still perceive limited availability to be a major obstacle (Shephard et al., 2005). There is not much known about the perceived availability of organic wine, especially not in the Dutch market. Therefore we would like to examine to what extent Dutch consumers feel organic wine is available and what consequences the perceived availability will have for the purchase intention. The third hypothesis is: H3a1: Low perceived product availability will have a negative effect on the purchase intention of organic wine

H3a2: Low perceived product availability will have a negative effect on the buying behaviour of organic wine

2.4.2. Perceived Consumer Effectiveness

Master Thesis – Maarten Kwakman – University of Groningen 21 their purchase contributes to something good and have an impact on the environment. That brings us to our next hypotheses:

H3b1: Low perceived consumer effectiveness will negatively affect consumers’ purchases intention of organic wine

H3b2: Low perceived consumer effectiveness will have a negative effect on the buying behaviour of organic wine

2.4.3 Perceived Financial Resources

Fotopoulos & Krystallis, 2001 identified that price is a crucial factor of the organic wine marketing mix but possibly not of the conventional wine. Organic products that do not have a price above the normal products are likely to achieve a very small market share (Verhoef, 2005), assuming that an organic product means high quality for its buyers and that high quality demands high price and vice versa (Fotopoulos et al., 2003). In the literature on organic consumption, income is often tried to relate to organic purchase behaviour. The proportion of people consuming organic food has been found to rise with an increase in income (Lockie et al., 2002; McEachern and McClean, 2002; Storstad and Bjorkhaug, 2003). Contradicting these outcomes, van Doorn and Verhoef (work in progress) did not found a significant effect on income on organic purchasing behaviour. However they did conclude that larger households are less likely to purchase organic, which implies that the availability of sufficient financial resources is an important boundary condition (van Doorn & Verhoef, work in progress). This leads us to the next hypothesis:

H3c1: Low perceived financial resources will have a negative effect on the purchase intention of organic wine

H3c2: Low perceived financial resources will have a negative effect on the buying behaviour of organic wine

2.5 Consumer Characteristics

Master Thesis – Maarten Kwakman – University of Groningen 22 organic food has been found to rise with an increase in income and they tend to be more highly educated than non-organic consumers (Lockie et al., 2002; Lea and Worsley, 2005). Some authors even managed to define consumer segments (e.g. Chryssohoidis and Krystallis, 2005; Cowan and Roddy, 1996). The choice for organic food is often related to environmental or health concerns. For example Lockie et al. (2002) had found that health, taste and environmental benefits have been found to be important considerations in the choice of organic foods by consumers. Health appears to be a more important influence on purchasing than the environment (Shepherd et al., 2005). These outcomes contradict with other research. Van Doorn and Verhoef (work in progress) found that quality and health motives appear unrelated to organic consumption. A person’s values determine the extent to which he/she weighs individual interests, such as money and convenience, against collective interests, such as a better environment or animal welfare. Stern and Dietz (1994) argue that attitudes of concern about environmental issues are based on a person’s more general set of values. That is attitudes about environmental issues are based on the relative importance that a person places on themselves, other people, or plants and animals which Stern and Dietz (1994) labelled egoistic, social-altruistic, and biospheric (three-factor model) (Schultz, 2001). Buying organic products depends on both individual- (i.e., higher quality, healthfulness) and other-oriented motives (i.e., prosocial benefits) (van Doorn and Verhoef, 2005). In this paper we will assess if individual-oriented customer characteristics will affect the relationship between a consumers attitude towards organic wine and their the organic wine purchase intention. These self-oriented characteristics are health consciousness, quality consciousness and price consciousness.

2.5.1 Health Consciousness

Master Thesis – Maarten Kwakman – University of Groningen 23 women tend to be more concerned about health than men (Lea & Worsley, 2005). Because the overwhelming majority of studies fined ‘health’ to be the primary reason consumers buy organic foods (e.g. Hutchins and Greenhalgh, 1995; Shifferstein and Ophuis, 1998; Zanoli and Naspettie, 2002; Hughner et al., 2007), we would like to examine if health-consciousness influences the purchase intention of organic wine. We therefore hypothesize:

H4a: The more health-conscious a consumer is, the higher the purchase intention will be for organic wine.

2.5.2 Quality Consciousness.

Several studies have found ‘taste’ to be among the most important criteria in organic food purchases (Roddy et al., 1996; Schifferstein and Ophuis, 1998; Magnusson et al., 2001; Hughner et al., 2007). This perception of better quality of organic food is mostly because the price premium that has to be paid for organic products. People tend to believe that higher price means better quality, which is a dominant motive for food choice (Vermeir & Verbeke, 2006). Conversely, van Doorn and Verhoef (work in progress) did not found support for this claim and indicate that quality considerations are not a sufficient reason for consumers to purchase organic. Some people might doubt the quality and functionality of sustainable products (van Doorn & Verhoef, 2011; Luchs et al., 2010). However, to the best of our knowledge there hasn’t been a research that examined the effect of quality consciousness in relation to the purchase intention of organic wine.

H4b: The more quality-conscious a consumer is, the higher the purchase intention will be for organic wine.

2.5.3 Price Consciousness.

Master Thesis – Maarten Kwakman – University of Groningen 24 than for conventional food (van Doorn and Verhoef, 2011; Remaud et al., 2008) In this paper we would like to examine the moderating effect of consumer’s price consciousness on the relation between attitude and purchase intention of organic wine. Because organic products are often more expensive, we hypothesize:

H4: Consumers characteristics will have a moderating effect on the relation between attitude and purchase intention

H4c: The more price-conscious a consumer is, the lower the purchase intention will be for organic wine.

2.6 Product Knowledge

Susceptibility to interpersonal influence is a general trait that varies across persons and that a person’s relative influenceability in one situation tends to have a range of other social situations. Susceptibility to influence by others is related to other personal characteristics (e.g. self-esteem, intelligence) (McGuire, 1968; p. 1132). Informational influence operates through the process of internalization, which occurs if information from others increases the individual’s knowledge about some aspects of the environment (Bearden et al., 1989). Involvement with a product and environmental issues are

important constructs when assessing consumers willing to purchase, with previous research emphasizing the importance of the relationship between product involvement and product knowledge. An integration of product use as a measure of familiarity in addition to self-reported knowledge is evidenced in Bettman and Park (1980). They refer to prior knowledge and experience as factors which influence consumers’ information search strategies. Product knowledge and involvement often correlate with each other (Bettman and Park, 1980). However Sujan (1983) argued that knowledge has an independent effect on information processing, as involvement could not explain the information processing patterns. Because knowledge influences the information processing, we expect knowledge to affect the amount of interpersonal influence that is present. We therefore hypothesize that the more knowledge a consumer has about the product, the less he/she will be affected by social influence.

H5a: The more knowledgeable a consumer is, the lower the effect of Normative influence will be on the purchase intentions.

Master Thesis – Maarten Kwakman – University of Groningen 25 2.7 Purchase Intention

“Behavioural intentions are instructions that people give to themselves to behave in certain ways” (Triandis, 1980). They are people’s decision to perform particular actions. Intentions can be inferred from participants’ responses that have the form, “I intend to do X”, “I plan to do X”, or “I will do X” (Sheeran, 2002). The behavioural intention (i.e. purchase intention), based on the TPB is basically determined by three factors discussed earlier (Attitudes, subjective norms/social influence, and perceived behavioural control). We adapted the model and measure the effect of attribute evaluation, normative and informative influence, the perceived behavioural control and customer characteristics on the purchase intention. Furthermore we add product knowledge to the model as a moderator on the effect of social influence on purchase intention (as shown in the conceptual model in figure 4). Our first objective is to measure the effect of attitudes, social influence and perceived behavioural control on the consumers purchase intention for organic wine. Secondly, this study attempt to ascertain whether the consumer purchase intentions of organic wine are influenced by different degrees of product knowledge or customer characteristics.

2.8 Self-Reported Behaviour

Although public interest in sustainability increases and consumer attitudes are mainly positive, behavioural patterns are not univocally consistent with the attitudes (Vermeir & Verbeke, 2006). The TRA and TPB have been used extensively in the prediction of a wide range of behaviours (e.g., Armitage and Conner, 2001; Sheppard et al., 1988). In general, research shows that whereas these theories capture the variance in behavioural intentions very well, the explained variance in behaviour is much lower (Armitage and Conner, 2001; Sutton, 1998). This is also referred to as the “intentions-behaviour gap”. In this research we want to examine whether or not there is a gap between the intention and the purchase behaviour of organic wine consumers and what implication this may have for marketers. We hypothesize that in line with the TPB, the purchase intention will have a positive effect on the actual behaviour.

Master Thesis – Maarten Kwakman – University of Groningen 26 The overall problem statement of this paper is:

Master Thesis – Maarten Kwakman – University of Groningen 27 2.9 Conceptual model:

Master Thesis – Maarten Kwakman – University of Groningen 28

3. Methodology

3.1 Research method.

To provide an answer to our problem statement, we collected quantitative data from a sample of Dutch consumers. A self-administrated questionnaire was used to provide this data. The advantage of this type of questionnaire is that it is cheap; it preserves confidentiality and can be completed at the respondent’s convenience without time pressure and the influence of the interviewer. The downside of self-administered questionnaires is result in a low response rate and unfinished questionnaires. The web-based questionnaire was created using Qualtrics, a software tool that facilitates the development and administration of online surveys. A convenience sampling method was used since the survey was distributed among acquaintances. A disadvantage of this method is that the research sample could become less diversified. The respondents were assured that the survey was purely conducted for academic purposes and all responses to the survey would be confidential and anonymous. The link to the online questionnaire is distributed among acquaintances. Facebook is also used to find respondents for the survey.

Preparatory to our quantitative research, an interview was taken in order to find information about organic wine from someone with expertise within the Dutch wine industry. Both open and closed questions about organic wine were put in front of the respondent. The main purpose of this interview was to collect information about the influence of organic wines on the Dutch wine market. Secondly, we wanted to find out whether or not people within the business think social influence affects (organic) wine purchases. Finally, this interview gave us the opportunity to test if our wine attributes are indeed relevant in consumer’s buying decisions. The problem with only interviewing one person within the wine business is of course the generalizability of the findings. However, it should gain us some extra insights from a ‘business perspective’.

Master Thesis – Maarten Kwakman – University of Groningen 29

3.1.1 The interview

We found Rudolf Th. Bijleveld willing to participate in an interview. Mr. Bijleveld is the General Manager of Intercaves BV, a Dutch import- and distribution organization of wine. In 2009 they were declared ‘Wine buyer of the year’ by Hubrecht Duijker, the most famous Dutch wine reviewer. Recently Intercaves started ‘The Green Wine Company’, a subdivision that is only involved with organic winemakers.

The open questions we asked to Mr. Bijleveld ware about the size of the Dutch organic market and the growth rate. Furthermore we asked him about the supply side of organic wines. We used closed questions on a 7-point Likert scale to measure the importance of wine attributes in purchase decisions (1: not important; 7: extremely important) and to what extent social influence is present in buying decisions of Dutch consumers (1: Not at all; 7: Very much). And finally, we asked if he thinks the assortment or price of organic wines should be adapted in order to increase organic sales (1: not at all; 7: very much). The interview questions and answers can be found in appendix 2.

The interviewee estimated that the total organic wine market is less than 1% of the total wine market. Although it is still has a relatively small market share, he detects an enormous growth and it predicts that organic wine will become mainstream and has already outgrown the ‘alternative’ circuit. To the question if the cultivation of organic wines is geographically bounded he answered: “no”. The organic wines in the Netherlands are imported from ‘new countries’ (e.g.South-Afrika and Australia) as well as from the classical wine countries (e.g. France and Spain). The questions if the opinions of experts, friends and families matter in the purchase decisions where neutrally answered. And according to Mr. Bijleveld, organic wines would sell more if the organic assortment will be enlarged within stores. Besides that, organic wines should get more prominent places within the wine shelves in order to increase sales.

Master Thesis – Maarten Kwakman – University of Groningen 30 attractiveness and screw top (organic wine) and 2) Price and wine house. The least important attributes are region and country of origin.

3.1.2 The questionnaire

The survey used for this research was developed using the academic literature in the field of organic- and wine consumption. The (web-based) online version of the survey was created using Qualtrics, a software tool that facilitates the development and administration of online surveys. In order to collect the data for this study, a convenience sampling method was used since the survey was distributed among acquaintances, which were easier to approach. The respondents were assured that the survey was purely conducted for academic purposes and all responses to the survey would be anonymous.

3.1.2.1 Measures

Attitude.Perceived aPerceived aAttribute importance. We defined attribute importance as: ‘the extent to which an attribute is used in the evaluation of a bottle/ carton package of wine”. To measure the attitude towardsattribute importance for organic and conventional wine, we used direct measurement with both Likert scale (1: not important, 7: very important)- and semantic differential scaling. This section of the research was more qualitative of nature and was used to find out how the Dutch consumer rate the different product attributes of both conventional and organic wine. The product attributes of the two types of wine were measured by means of seven attributes on a 7-point Likert scale. Secondly, the relative importance of eight different types of product information was measured on a 7-point Likert scale. Furthermore, the respondents were asked to make a five trade-offs between two wine attributes (e.g. Screw cap vs. cork). These trade-offs were asked on a semantic differential 5-point scale.

Informational social influence. Informative influence is defined as “the tendency to

accept information from others as evidence of reality”. Informational social influence was measured by four statements which were rated on a 7-points scale. Question examples are ‘I seek information from those who work at the store where I purchase my wine’ and ‘I often consult other people to help me choose a bottle of wine’. The articles of Park &

Master Thesis – Maarten Kwakman – University of Groningen 31

Lessig (1977) and Bearden et al. (1989) are used as guidance in developing the questions. These statements were derived from the questionnaire of Park & Lessig (1977).

Normative social influence. To measure the normative influence, the articles of Park &

Lessig (1977) and Bearden et al. (1989) are used as guidance in developing the questions.

Normative influence is defined as “the tendency to conform to the expectations of other persons or groups”. Normative influence waswas and was measured via five statements that where rated on a 7-points scale (1: totally disagree; 7: totally agree). Examples of the questions are ‘I would buy organic wine because people who are close to me think that I should buy organic wine’ and I would buy organic wine because I think others will approve it’. These statements were also derived from the questionnaire of Park & Lessig (1977).

Informational social influence. Informational social influence was measured by four

statements (e.g. I seek information from those who work in the store where I buy my wine) which were rated on a 7-points scale. These statements were also derived from the questionnaire of Park & Lessig (1977).

Perceived behaviour control. This construct is build-up of three items, namely perceived

consumer effectiveness’, perceived financial resources’ and perceived availability which

were respectively measured by two, two and three statements. The respondents had to rate these statements on a 7-poins Likert scale (1: Totally disagree; 7: Totally agree).

We defined perceived financial resources as “the extent to which the consumer has the financial resources to purchase organic wine’. The questions ‘I would buy organic wine if it were less expensive’ and ‘I cannot effort to pay the price premiums for organic wine’ were used to measure this construct. Perceived consumer effectiveness is defined as ‘the extent to which the consumer believes that buying organic wine can contribute to a better environment”. It is measured by the questions ‘I believe organic wine would contribute to a better environment’ and ‘consuming organic is amongst the action I can take to protect the environment’. Questions like ‘Organic wine is always sufficiently available’ and ‘I would buy more organic wine if it were more available’ measured the perceived

Met opmaak: Links

Master Thesis – Maarten Kwakman – University of Groningen 32

availability construct. It is defined as “the extent to which the consumer feels organic wine is present in the shelves’

Product knowledge. The respondents were asked how they perceive their knowledge of conventional wine, organic wine and other organic products. The items werewas

measured on a 7-points scale (1: very low; 2: very high). A question example: ‘What is your level of knowledge about organic products?’.

Customer characteristics. Three different types of customer characteristics were

investigated. The health-, quality- and price-consciousness were measured on respectively 1, 2 and 2 statements. These items were measured on a a 7-point Likert scale, anchored at (1) totally disagree and (7) totally agree. Health consciousness is defined as “The extent to which consumers care about health consequences during their purchase decisions”. We measured this construct by the questions ‘I think of myself as a health-conscious consumer’ and ‘I often think about health issues’. The definition of quality consciousness is: “The extent to which consumers care about the quality of the products they purchase”. This construct is measured by the question ‘I care about the quality of products I buy’. Finally, we defined price-consciousness as “The extent to which consumers care about the price of products during their purchase decisions”. The questions ‘The price of a product is very important to me’ and ‘I always buy the cheapest product available’ were used to measure this construct.

Purchase intention. The respondents purchase intention is measured by means of two

statements. These statements are ‘I intend to buy organic wine in the near future’ and ‘I would buy organic wine if I would find it in my supermarket’. These statements were rated on a 7-points scale (1: totally disagree; 7: totally agree).

Self-reported behaviour. Three different questions were given to measure the

Master Thesis – Maarten Kwakman – University of Groningen 33 bought organic wine (yes/no). The third questions was asked on an ordinal scale ranging from ‘never’ to ‘every day’ and measured the amount of organic wine purchases. Because we used a questionnaire to measure the purchase behaviour it could be that our measurement is not very accurate. People tend to give socially acceptable answers when it comes to behaviour which could lead to biased information about the purchase behaviour. Because we do not have a database with actual purchases, we rely on self-reports. Such measures are common because they are relatively easy to obtain and are often a feasible way to assess constructs of interest (Donaldson, 2002).

Experiment. Within the online survey we did a small experiment. Its purpose was to find

out what kind of organic wine description is more appealing to consumers: a quality focused or an environmental/sustainable focused description. The first description put emphasis on the increase in wine quality by organic farming. The second description underlined that the sustainable way of wine production preserves the nature and the farming lands.

Description 1: Quality

De wijnmakers van Chateau Dumoulin hebben een nieuw soort kwaliteitswijn weten te maken. Door hun biologische werkwijze zijn de smaken nog intenser en is deze wijn werkelijk een genot om te drinken. Doordat het gebruik van besproeiingsmiddelen vermeden is hebben de druiven hun pure smaak behouden wat heeft geresulteerd in een wijn van uitstekende kwaliteit.

Description 2: Sustainability

Deze heerlijke wijn uit een van de mooiste streken van Frankrijk behoort tot de biologische topwijnen die op de Nederlandse markt te vinden zijn. De wijnmakers van Chateau Moncoutié hebben door middel van biologisch telen, het beste uit het vruchtbare grond weten te halen. Deze wijn is met respect voor de mens en natuur gemaakt en dat vertaalt zich terug in de smaak van wijn. Bij deze wijn proeft u werkelijk de natuur.

Master Thesis – Maarten Kwakman – University of Groningen 34

Control variables. In this study we used socio-demographics like Age, Gender, Income,

Education, Occupation and number of children in the household as control variables. In past research, these variables has proven to be of use in describing the organic shopper (e.g. Hines et al., 1987; Thompson, 1998; Padel and Foster, 2005; Ngobo, 2011)

3.3 Plan of analysis

The purchase intention of organic wine was the major dependent variable of this study. The influences of the independent variables on the purchase intention where tested by regression analysis. All analyses were conducted by SPSS, statistical software. Frequency of participants’ responses to the questions was measured, and cross-tabulations were performed with the control variables (e.g. age, gender etc.). After conducting a reliability analysis, we combinedperformed a factor analysis to merge individual questions and statements into one scaleitem (e.g. normative social influence). The outcomes of the

factor- and reliability analyses can be found in table 2. A Varimax rotated principle component factor analysis was performed with the KMO and Bartlett’s test of sphericity. For the experiment we performed a Pearson Chi-square contingency table analysis. This analysis can measure if there is a significant relation between the choice for wine description and independent variables (e.g. control variables, organic buyer yes/no). We used this model instead of a simple Pearson Chi-Square model because we use multiple predictors and our predictors are both categorical and continuous. This test measures if our observed frequencies are significantly different from the expected frequencies. In other words we measure if ‘wine description decision’ depend upon other variables (control variables and buyer/ non-buyer).

Table 2: Results reliability analysis and Items per construct.

Scale Survey Questions

(see appendix 1)

Cronbach’s Alpha (α)

Informational influence 5.1 / 5.2 / 5.3 .558 Normative ‘Other-oriented’ 6.1 / 6.2 / 6.3 .794 Normative ‘Self-oriented’ 6.4 / 6.5 .806 Perceived behavioural control:

Financial 7.1 / 7.2 .675

Effectiveness 7.3 / 7.4 .859

Availability 7.5 / 7.6 / 7.7 .675 Customer characteristics:

Master Thesis – Maarten Kwakman – University of Groningen 35 Quality conscious 9.3

Price conscious 9.4 / 9.5 .509

Master Thesis – Maarten Kwakman – University of Groningen 36

4. Results

This research focuses on a sample of Dutch consumers from age 16 and higher, the legal age in the Netherlands. A total of 172 consumers filled in the questionnaire of which 74 were male and 98 female. Around 69% of the respondents are between 16 en 25 years old, 16% between 26 and 35, 10 persons are between ages 36-45, six respondents are between 46 and 45 and 5.8% is 55 and older. Of the sample, almost 61% is student. 56 respondents are employed of which 8 are self-employed. Ten persons are unemployed, of which four are retired. Only 9% of the respondents have children in their household. Seven of them have one child, another 7 respondents have 2 children and 3 respondents live with 3 children in their household. As you can see, the people in our sample are very young which could be harmful for the results. In previous research, age has shown to be of influence in the purchase behaviour of organic products. However, normative and informative influence tent to be more present at younger age. So in light of social influence, this sample is quite interesting.

Table 3: Frequency of gender, age, occupation and number of children in household

Total N Gender: N Male 43% Female 47% 172 Age: N 16-25 69.1% 26-35 15.6% 36-45 5.8% 46-55 3.4% >55 5.8% 172 Occupation: N Student 61.6% Employed 27.9% Unemployed and searching 3.5% Self-employed 4.7% Others 2.3% 172 #of children in household: N None 90.6% 1 4.1% 2 4.1% 3 1.2% 172 Organic buyer N Yes 59.9% No 40.1% 172

Master Thesis – Maarten Kwakman – University of Groningen 37 Only 3% has an annual household income of more than €100.000. 80% of the respondents claimed to have tasted organic wine at least once in their live. The amount of

Table 4: Percentage of organic buyers and amount of organic purchases Total N:

Organic wine buyer: Yes 59.9% No 40.1% 172 Amount of organic wine purchases: Never 25.6% <1 a month 44.2% 1x per month 19.2% 2-3 per month 8.7% 1x per week 1.7% 1-3x per week 0% Daily .6% 172

people that have ever bought organic wine is less (N=103) but still around 60% of the sample. Only one person claimed to buy organic wine once a day and three respondents purchase wine once a week. The vast majority however answered that they never consumes organic wine (25.6%) or not more than once per month (44.2%). The remaining 15 respondents purchase organic wine 2-3 times a month. An overview of the control variables can be found in table 3, 4 and 5.

Table 5: Frequency table Education- and Income level

Education level Frequency Income level Frequency

Primary School 2.9% < €10.000 (%) 59.9% High School 0.6% €10.000 - €19.999 9.3% Intermediate vocational Education 4.7% €20.000 - €29.999 7.6% Higher Vocational Education 23.8% €30.000 - €39.999 2.3% University 68% €40.000 - €49.999 4.7% €50.000 - €59.999 2.3% €60.000 - €69.999 2.3% €70.000 - €79.999 5.2% €80.000 - €89.999 3.5% €90.000 - €99.999 0 % €100.000 < 2.9% Total N 172 172 4.1 Attribute importance

The first part of this research concerns the investigation of the relative importance of different wine attributes and wine-specific information, for both conventional- and organic wine. We would like to find out if some wine characteristics are more important for organic wine than they are for conventional wine and vice versa. We therefore hypothesized: The perceived importance of wine attributes will differ between organic

Master Thesis – Maarten Kwakman – University of Groningen 38 whether there were significant differences between the attribute means of conventional and organic wine. The analysis showed a significant different mean for country of origin, grape type, label attractiveness, price and taste (see table 5). These attributes are more important in the evaluation of conventional wine than they are for organic wine, because the mean for conventional wine is higher.

In contrast to the difference in wine attributes, the wine-specific information seems more relevant in valuation of organic wine than for conventional wine. We found significant differences for information about the product method, preservation, the number of bottles produced, alcohol percentage and the quality insurance. The consumption information and awards where not valued any different between organic and conventional wine. The wine characteristic with the largest difference in mean is ‘information about production method’. The importance of production method info for organic wine scores 1.2 points (on a 7-points scale) higher than for conventional wine. This large difference in mean really stresses out the importance of information about the production method on a wine label. The second largest mean difference is taste from experience (+.36). This indicates that taste is more important in the evaluation of conventional wine (a score of 6.24 out of 7), than it is for organic wine (a score of 5.88 out of 7).

Table 6: Conventional VS. Organic Wine. Items where measured on a 7-point scale

Attributes: Mean

conventional:

Mean Organic: Difference: (Mc –Mo) Sig (2-tailed) Country of origin 4.32 4.12 +.20 .005*** Region of origin 3.50 3.51 NS Winehouse 3.37 3.28 NS Harvest date 3.15 3.22 NS Grape type 4.37 4.16 +.21 .019** Label attractiveness 4.44 4.15 +.29 .003*** Price 5.41 5.19 +.22 .020**

Taste (from experience) 6.24 5.88 +.36 .000***

Wine-specific information:

Master Thesis – Maarten Kwakman – University of Groningen 39 Preservation 2.78 3.01 -.23 .017** Consumption 2.94 2.95 NS # of bottles produced 2.30 2.50 -.20 .002*** Alcohol % 3.13 3.35 -.22 .004*** Awards 4.17 4.22 NS Quality assurance 4.14 4.34 -.20 .025** Trade-offs: Screw vs. Cork 3.19 3.03 +.16 .093*

Glass vs. Recycled glass 3.01 3.35 -.34 .000***

Bottle vs. Carton package 1.31 1.71 -.30 .000***

Traditional label vs. Modern label 2.44 2.79 -.35 .000*** Traditional bottle vs. Modern bottle 1.98 2.28 -.30 .000*** ***Significant at 0.01 level

**Significant at 0.05 level *Significant at 0.1 level

Green: These scores are above 3.5 and thus on average perceived as ‘important” in buying decisions

Master Thesis – Maarten Kwakman – University of Groningen 40 also of more importance when it comes to organic wine. The mean difference of Traditional vs. Modern label is the second largest (Mc–Mo=-.35) after Label attractiveness.

We furthermore hypothesized that important wine attributes will have a positive effect

on the organic wine purchase intention. a positive attitude towards organic wine would positively affect the organic purchase intention. We performed a linear regression with purchase intention as dependent variables and all product attributes and control variables as independent variables (IV’s). The model was overall significant (α=0.000) and 11.8% of the variance of purchase intention is explained by the IV’s (Adj. R²= 0.118). Because multicollinearity could be a problem in an analysis with this much variables, we generated the VIF scores for the variables. The VIF scores were lower than 5 which indicate that we can assume there is no multicollinearity problem. Table 7 provides an overview of the regression outcome. Only one attribute significantly affects the purchase intention. A unit increase of Label attractiveness adds .129 to the purchase intention. It’s the third largest predictor (Standardized β=.197) of purchase intention (PI) in this model. Gender is most influential in predicting PI (Standardized β=.300). With every unit increase in gender, the purchase intention heightens the PI. Furthermore, the regression analysis shows us that three out of five trade-offs significantly influence the purchase intention. These are, from most influential to least influential, glass versus recycled glass, bottle vs. Carton and the trade-off between traditional- and modern label. A unit increase in these three trade-offs respectively change PI with .155, -.165 and .156. With these results we can say that our hypothesis that important wine attributes positively influence the PI is only accepted for Label attractiveness. Furthermore, we found evidence that some wine characteristics differ in their importance to the evaluation of organic vs. conventional wine.

Table 7: Outcome linear regression with Wine Attributes on Purchase Intention

Variable: Unstandardized coefficient Standardized coefficient Sig. Control variables: Gender .604 .300 .000***

Master Thesis – Maarten Kwakman – University of Groningen 41 Age .008 .112 NS Education .014 .012 NS OcupationOccupation -.008 -.014 NS Household income .031 .085 NS # children in household -.222 -.118 NS Wine attributes: Country of origin -.074 -.125 NS Region -.034 -.058 NS Winehouse -.008 -.013 NS Harvest date .030 .046 NS Grape type .067 .126 NS Label attractiveness .129 .197 .034** Price -.016 -.022 NS

Taste (from experience) .001 .001 NS

Wine-specific information: Info_Production method .061 .114 NS Info_Preservation .064 .108 NS Info_Consumption .023 .037 NS Info_# of bottles -.044 -.065 NS Info_Alcohol % -.089 -.148 NS Info_Awards .027 .046 NS Info_Quality indication .035 .064 NS Trade-offs: Screw VS. Cork .062 .091 NS

Glass VS. Recycled Glass .155 .210 .014**

Bottle VS. Carton -.165 -.185 .042**

Traditional label VS. Modern label .159 .178 .050**

Traditional bottle VS. Modern bottle -.087 -.090 NS

***Significant at 0.01 level **Significant at 0.05 level

4.1.1 Moderating effect of customer characteristics: