Country & organic labels

1

Country & Organic Labels

Thesis

Guus Notermans

11421061

MSc Business Administration – Marketing Track

Supervisor: Mr. H. Güngör

23-6-2017

Statement of originality

This document is written by Guus Notermans who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Abstract

A lot of studies have investigated the consumer preferences for products displaying certain country labels on the one hand and organic labels on the other hand. This is one of the first researches that through an online survey studies the difference that the presence of the labels separately or combined have on Quality Perception and Intention to Buy. It was found that the degree to which a product fits the country where it comes from is an important predictor of Quality Perception for products with only a country label, especially if the product is foreign. The degree to which someone is ethnocentric was found to influence the Quality Perception and Intention to Buy of domestic products. For products with only an organic label,

environmental involvement was found to be a predictor of Quality Perception and Intention to Buy, just like the degree to which someone trusts the organic label. When both a country and organic label are displayed on a product, the degree to which a product fits a country and the degree to which the organic label is trusted simultaneously influence the Quality Perception of the product, together with the degree to which someone is environmentally involved.

Differences were found when looking at foreign and domestic organic products. Further it was found that the Quality Perception of foreign products was a stronger predictor for Intention to Buy than the Quality Perception of domestic products was.

Keywords: Country labels, Country-Of-Origin Effect, Ethnocentrism, Organic labels,

Table of Contents

Introduction………..……….………….…………..……..……p.9 1. Literature review……..…….……….………...……….………..………p.11

1.1 Intrinsic and extrinsic cues………..……...……….………..p.11 1.2 Country-of-Origin effects……….………....………p.12 1.2.1 Product-Country match………...…….……...………...……p.12 1.2.2 Ethnocentrism….………..…….………p.13 1.3 Organic food products………...………….……….……….……….p.15

1.3.1 Environmental Involvement………..………...………..…..……..p.15 1.3.2 Organic Label trust………..……….……..……...….p.17 1.4 Country & Organic Labels……….………….………..………p.18 1.5 This research………...……..………..………..p.19

2. Method……….…….p.22

2.1 Design and participants……….………p.22 2.2 Procedure……….………...………..p.23 2.3 Product profiles……….………p.24 2.4 Measurements……….………..p.25 3. Results…..……….p.30 3.1 Descriptive Statistics……….p.30 3.2 Correlations……….……….p.33 3.3 Visualisation……….………p.38 3.4 Hypotheses Testing………..p.43 3.5 Explorative Analysis……….………p.53

4. Conclusion………p.63 4.1 Summary of the Results and Implications………..p.65

5. Discussion………..………p.71 5.1 Research Limitations and Future Research………p.71 6. Reference list……….………..……….……….p.73 7. Appendix…….………..p.79

9

Introduction

In the last decade, organic food has become one of the fastest growing segments of agriculture worldwide. The public interest in sustainable production and consumption practices has increased in all levels of the food chain (Vermeir &Verbeke, 2006). In the Netherlands, the organic food consumption of 2016 increased by 12 percent compared to the consumption in 20141. Part of the reason for this increase in interest is that the ethical standards of food products are becoming important choice criteria for many consumers (Zander & Hamm, 2010).

Organic products are obtained by practices involving a careful use of natural resources and reduction of pollution caused by chemical fertilizers (Browne et al., 2000). The reduction of pollution has become an important topic within the scientific community, as there has been a broad agreement that climate change is real and primarily due to the human use of fossil fuels, like oil and coal, that release greenhouse gases into the air, having harmful effects on ecosystems. The increasing demand for sustainable products shows that people more and more feel a sense of responsibility in assuring a liveable planet for future generations. In many countries the increasing demand for organic food is growing substantially faster than the domestic production and supply (Thøgersen, 2017). This supply deficit has led to high import shares for many organic food products (Willer & Schaack, 2016). As a result, domestic consumers are presented with a variety of organic products from foreign Country-of-Origins (COOs), and presumably consider and develop preferences based on this

characteristic. Although there are a lot of studies that have investigated the consumer preferences for COO on the one hand and organic food on the other hand, there is a lack of research on COO effects within the context of organic food (Xie et al., 2015). Or in other

10

words, there is a need for research on how consumers evaluate foreign or domestic organic food products. Thøgersen (2017) also finds that there is a need for research that investigates the combined effects of COO and organic labelling on consumer food product preferences and choices. Altogether, this has led to the following research question, which will be central in the current paper.

RQ1. How is a consumer’s Intention to Buy predicted by the presence of either a country or

organic label, and what is their combined predictive power?

11

1. Literature Review

In this chapter, the concepts that are part of the current research will be explained. First of all, the distinction between intrinsic and extrinsic cues is made and the concept of Origin effect will be explained, after which the cognitive processes that underlie Country-of-Origin effects will be introduced. Hereafter organic food products will be discussed, followed by the attitude-behaviour relationship. After this the possible combined effect of Country-of-Origin labelling and organic labelling will be discussed. All the concepts will lead to a number of hypotheses. At the end of this chapter the reader can find an overview of the hypotheses, which is visualised in the conceptual framework.

1.1 Intrinsic and extrinsic cues

Caswell and Mojduszka (1996) define product quality as the bundle of attributes that

determine the performance of the product. The Total Food Quality Model that Grunert et al. (1996) developed says that people’s buying intention of a food product before purchase depends on the cost cues and quality cues. There are two types of cues that predict quality in the mind of the consumer in the model; intrinsic and extrinsic cues. Intrinsic quality cues refer to physical properties of the product, such as taste, colour and aroma. On the other hand, extrinsic cues refer to everything else (Olson and Jacoby, 1972), and are the characteristics of the product that cannot be experienced from the product itself. Examples of the extrinsic cues are the brand; price; store reputation; the country where the product is produced (country-of-origin); and whether it’s organic or not. Altogether, intrinsic and extrinsic quality cues are used by consumers as an indication of the quality of food and help them in making buying decisions, as they form expectations of the product. In the current research the focus will lie

12

on two of those extrinsic cues that may potentially influence the Quality Perception of a product, namely organic and country labels.

1.2 Country-of-origin effects

According to the cue utilization theory, consumers rely more heavily on extrinsic cues when intrinsic cues are difficult to judge or assess, or consumer expertise is low (Maheswaran, 1994, Zeithaml et al., 1988; as cited in Thøgersen, 2016, p.5). When intrinsic cues are missing or cannot easily be assessed, consumers tend to rely more on extrinsic cues (Jacoby et al., 1977); this is often the case for low-involvement products, since the cost of searching for intrinsic cues to aid consumers in product evaluation far exceed the benefits (Zeithaml, 1988). So, consumers do not always spend a great deal of time or cognitive effort in making purchase decisions (Keller, 1993). They often try to minimize complexity in the decision-making process by relying on simple heuristics, and Country-of-Origin can be one of those. The impact of Country-of-Origin is one of the oldest and most persistent topics in international marketing (Dekhili & Achabou, 2014). Herz and Diamantopolous (2013) define Country-of-Origin effects as: “any influence or bias on product evaluation, risk perception,

buying intention, etc. resulting from Country-of-Origin information”. So, in other words, a

product’s Country-of Origin-can influence the consumers’ evaluative judgements of the product (Pharr, 2005). There are a couple of factors that influence the effectiveness of the Country-of-Origin effect.

1.2.1 Product-country match

Insch and Florek (2009) suggest why people might take a country label into consideration in the buying process. They say that when a country has a certain status in producing certain

13

products, a country label may function as a sign of quality for a product. For instance, despite potentially large differences in price, consumers are likely to prefer French to Austrian wine; Italian to Finnish fashion; German to Chinese cars; and Japanese to Mexican electronics (Chattalas et al., 2008). Consumers process the information of a country label based on their perception of that country and the congruence between the product category and the country image (Roth & Romeo, 1992). So, the match of product and country image would be

favourable when the feature of the specific product category is important to the consumers and the country is positively perceived on that feature. On the other hand, if the country is perceived to be weak on the important feature of that product category, the Product-Country Match would be unfavourable. The current research will look at whether the Product-Country Match has effect on the Quality Perception of products with a country label. The expectation that a high Product-Country Match positively influences the Quality Perception has led to the first hypothesis:

H1: The higher the Product-Country Match the higher a consumer perceives the quality of

products with a country label.

1.2.2 Ethnocentrism

One of the most researched variables moderating the Country-of-Origin effect is a consumer’s level of Ethnocentrism (Shimp and Sharma, 1987). Shimp and Sharma (1987), define

consumer Ethnocentrism as “the beliefs held by [...] consumers about the appropriateness,

indeed morality, of purchasing foreign-made products”. Highly ethnocentric consumers may

also tend to or avoid the purchase of products from certain countries based on their social and personal values. These consumers see their buying behaviour as some sort of “vote” in favour or against the producing country’s policies and practices of its government. So, the

Country-14

of-Origin needs to be politically in line with a person’s normative view or interest in order to gain a positive attitude towards its products.

Consumers may also systematically prefer domestic above imported products as the purchase of the latter may be perceived as unpatriotic or socially undesirable (Ahmed et al., 2004, Shimp and Sharma, 1987, as cited in Thøgersen, 2017). They find it a moral action to purchase the products manufactured or grown in one’s own country, because this represents a supporting behaviour of the country’s economy.

In highly ethnocentric consumers, the country label has a relatively larger effect on Quality Perceptions and purchase intentions (Chattalas et al., 2008). The literature suggests that consumers high in Ethnocentrism generally pay more attention to the Country-of-Origin cue, often perceiving the consumption of imported products as socially undesirable and unpatriotic, while finding the opposite of domestic products. Ethnocentrism will therefore be tested in this research, as it’s expected to influence both the consumer’s Quality Perception of a product with a Country-of-Origin label and the Intention to Buy. Further, there will be looked at whether differences can be observed for domestic and foreign products

H2a: Ethnocentrism is positively related to the Quality Perception of domestic products. H2b: Ethnocentrism is positively related to the Intention to Buy domestic products. H2c: Ethnocentrism is negatively related to the Quality Perception of foreign products H2d: Ethnocentrism is negatively related to the Intention to Buy domestic products.

15

1.3 Organic food products

Another cue that consumers use as an indication of the quality of food, which also helps them in making buying decisions, as it forms expectations of the product, is an organic cue. A product can only be regarded

organic when its certified and allowed by the law to display an organic label. A product can only be certified when the product (and its entire product chain) is under control. Stichting Skal certifies organic products in the Netherlands2. It is the control organization for organic products and is committed to demonstrate its reliability. Every European country has at least one certification organization (like Skal in the Netherlands) and they certify products using the same European legislation. When products meet the requirements for being regarded as organic, they are allowed to display the European organic logo, the ‘Euro Leaf’ (see figure 1). Foods may be labelled "organic" only if at least 95% of their agricultural ingredients meet the necessary standards. Organic production outlaws the use of genetically modified organisms and derived products3_{. There are a couple of factors that influence the effectiveness of an}

organic label.

1.3.1 Environmental Involvement

The growing demand of organic food has been linked to an increased consumer awareness about human health and environmental issues (Cicia, 1994). Consumers buying organic food are therefore relatively highly involved in the buying process (Zanoli & Naspetti, 2002).

2_{https://www.skal.nl/assets/Infobladen/Infoblad-Vervaardigen-en-verhandelen-van-biologische-producten.pdf} 3_{https://ec.europa.eu/agriculture/organic/eu-policy/eu-legislation/brief-overview_nl}

16

Consumers that are highly involved in a buying decision are assumed to follow a high-effort path, spending time to process information in order to make a weighed and well-considered choice of the highly differentiated product alternatives (Hoyer et al., 2013). As a result, their attitudes are more elaborate and stable. For example, Thøgersen et al. (2010) explored consumer responses to ecolabels by means of a mall-intercept survey. They found that consumers with high environmental motivations were also highly involved in the purchase of eco-labelled products, including acquiring a higher amount of relevant knowledge to make an informed decision (as cited in, Thøgersen, 2017).

Remarkably, despite this high involvement, it seems the consumer often doesn’t know exactly what it means that a product is organic. Research shows that individuals interpret the term ‘organic’ in a multitude of ways depending on the context (Thøgersen, 2017). Many studies find that consumers associate organic food with environmental protection, animal welfare and social aspects such as local farming (Aertsens et al., 2011, Harper and Makatouni, 2002, Padel and Foster, 2005). It is also often found that consumers infer health benefits from consuming organic food (Aerstsens et al., 2011). In addition, consumers that buy organic food believe that it tastes better than ‘normal’ (Marian and Thøgersen, 2013).

Zhou et al. (2013) found an important moderator of the attitude-behaviour relationship with regard to organic food. They found that consumers’ value priorities moderate the

relationship between consumer attitudes and intentions regarding buying organic food. This means that consumers are more likely to actually purchase organic food the more important they find it. The effect that Environmental Involvement has on the Quality Perception and Intention to Buy organic products will be investigated using the following hypothesis:

17 H3a: The higher a consumer’s Environmental Involvement, the higher the Quality Perception

of organic products is.

H3b: The higher a consumer’s Environmental Involvement, the higher the Intention to Buy

organic products is

1.3.2 Organic Label trust

Although it might not be clear to every consumer what organic exactly means, it is clear that organic food evokes favourable attitudes (Hughner et al, 2007). However, it is common to find a gap between the attitudes and behaviour towards buying organic food (Ascheman-Witzel and Niebuhr Aagaard, 2014). For many people an important obstacle of buying organic food is its insufficient availability, high prices and the scepticism towards organic food labelling (Hughner et al., 2007). In the market for organic food, the degree to which consumers trust the organic label is a crucial issue, as the consumers are not able to verify whether a product is an organic product, not even after consumption. The organic label is an extrinsic cue of which some consumers tend to be sceptical about the integrity of the

production process of the products. Therefore the consumer’s trust in the product integrity is of crucial importance, as it might potentially prevent consumers from buying more organic food. Consumers often distrust the certifications of organic products, as they do not know the exact process that farmers have to go through in order to get their products certified (Essoussi et al., 2009). Mistrust towards an organic label diminishes the consumer’s Intention to Buy organic food. Although consumer perceptions of what stands behind an organic certification logo are of subjective nature and in many cases not based on objective facts, what the

consumer thinks is very important and might influence the consumer’s Quality Perception of a product, which in the end influences the buying decision (Janssen and Hamm, 2012). The

18

effect that Organic Label Trust has on the Quality Perception of organic products will be investigated using the following hypothesis:

H4: The more the consumer trusts the organic label, the higher a consumer perceives the

quality of an organic product.

1.4 Country & Organic labels

As seen in the previous sections, many studies have investigated the effect of Country-of-Origin and organic labels separately. There are only a few studies that have investigated the combined effect of country and an organic labels. Dekhili and Achabou (2014) investigated whether a country’s ecological image affects the evaluation of an organic product. This study found that, even if products met the requirements for the same ecolabel, it led to a significant decrease in purchase intention when a Country-of-Origin was displayed on the product with a negative environmental image.

Most of the studies on the effect of COO for organic food products focused on the preferences for domestic versus imported organic foods (Dransfield et al., 2005, Xie et al., 2015, as cited in Thøgersen, 2017, p. 12). These studies found a domestic country bias, meaning that people consistently like domestic organic products more than foreign ones. Hempel and Hamm (2016) confirm this trend, as they found that consumers are increasingly demanding locally produced food, seemingly using “local” as a quality indicator. Adams and Salois (2010) explored this development and found that consumers have more positive

attitudes towards local food and in many cases even prefer it over organically produced food. A few studies found that the disadvantage of being imported is smaller for organic food products. This means that there is a positive interaction between organic and foreign Country-of-Origin after controlling for the negative direct effect of foreign countries

19

(Onozaka and McFadden, 2011, Xie et al., 2015). In such cases, country and organic labels seem to be perceived by consumers as supplementary information about quality of a product (Onozaka and Mcfadden, 2011, as cited in Thøgersen, 2017).

For the current research it is expected that organic makes less of a difference for Quality Perception for domestic products. While on the other hand, for foreign products it is expected that the organic label is more important for the consumer. This will be tested using the following hypothesis:

H5: Organic Label Trust and Product-Country Match simultaneously positively influence

the Quality Perception of products with a country and organic label.

1.5 This research

After analysing the available literature on country and organic labelling separately and together, it may be concluded that there is a need for more research on the interaction of the these two topics. Although a lot of studies investigated the consumer preferences for Country-of-Origin on the one hand and organic food on the other hand, there is a need for more

research on Country-of-Origin effects within the context of organic food (Xie et al., 2015). Thøgersen (2017) also finds that there is a need for research that investigates the possible combined effect of country and organic labelling on consumer food product preferences and choices.

The effects that both country and organic information have on Quality Perception will be tested with the following hypothesis:

20 H6: There is a positive relationship between Quality Perception and Intention to Buy,

meaning the higher the Quality Perception is, the higher the intention to Buy will be

The conceptual model (see figure 2) gives an overview on what constructs discussed in the literature review are expected to influence Quality Perception, and finally the

consumers’ Intention to Buy for products with country and/or organic labels to see what the attitude-behaviour gap is.

21

Figure 2: Conceptual Model A: Products with Country Labels

B: Products with Organic Labels

22

2. Method

The current research will look at how country and organic labels influence the Quality Perception and Intention to Buy a product. It will be tested whether Ethnocentrism and Environmental Involvement moderates this effect. This chapter will first shed light on the research design and on who participated in this research. Then the procedure will be discussed. Finally, attention will be paid to how the variables of this research have been measured.

2.1 Design and participants

The current research consists of an online questionnaire that the participants had to fill in. The data for this research have been collected between the 11th of May and the 6th of June 2017. The respondents that participated were gathered within the direct circle of the researcher’s network via Facebook and Whatsapp. On Facebook and Whatsapp, messages were distributed to recruit people to participate in the research. In the messages, participants were kindly asked to fill in the survey. The request was accompanied by a link to the online survey at

Qualtrics.com, a website where online questionnaires can be compiled and conducted. Further, the researcher posted the survey on FOK! Forum, SwapSurvey and SurveyStudent. A total of 295 people participated in the survey, of which 254 (86%) completed the entire questionnaire. Thus, 41 participants did not complete the entire questionnaire, which means that these answers cannot be used in the current research. The majority of the

respondents are between 20-29 years old (59,1%); followed by 30-39 years old (13,5%), 50-59 years old (8,3%), 60+ years old (7,9%) , 40-49 years old (6,3%) and below 20 (4,8%).

23

There were slightly more male respondents (51,6%) than female respondents (48,4%).

2.2 Procedure

The data of this research have been collected using an online questionnaire. The respondents received a request to participate including a link to the online survey. First of all, the

participants were thanked for their participation, after which they were informed that their answers were used for a master thesis project for Business Administration at the University of Amsterdam. In addition, the participants were assured that all their answers are completely anonymous and cannot be traced back to them as individual.

The first questions from the questionnaire allowed the respondent to indicate their age and gender. Hereafter he or she were firstly randomly assigned to one of the 5 olive oil conditions and secondly to one of the 5 yogurt conditions. For both the olive oil and yogurt condition the respondents had to answer questions about their evaluation of the olive oil and yogurt. Questions had to be answered concerning the product-country match (if a country-of-origin cue was present in the condition), the degree to which the organic label was trusted (if present in the condition), their Quality Perception of the product and their Intention to Buy it. Hereafter, the respondents were asked to indicate to which extent they agree with statements about buying foreign products. These statements were used to measure how ethnocentric the respondents were. In the end the participants were asked about how frequent they consume olive oil and yogurt, after which they have to reply to a couple of statements that measure the participant’s Environmental Involvement. After answering the questions, the respondents were thanked for their cooperation.

24

2.3 Product profiles

The products that were selected for this research were olive oil and yogurt. It was assumed that for olive oil, Italy would have a favourable Product-Country Match, as it is one of the largest producers of olive oil in the world. For yogurt, on the other hand, it was assumed that The Netherlands would have a favourable Product-Country Match. Namely, the Netherlands is famous for its dairy production and should be familiar to the respondents. While on the other hand, it was assumed that Dutch olive oil would have an ambiguous or unfavourable Product-Country Match, just like Italian yogurt, as there’s no clear link between these products and countries.

Olive oil and yogurt were selected as these two products both complied with several criteria. Namely, both olive oil and yogurt were assumed to be familiar and frequently bought by the majority of the respondents. Besides that, both products are known for having organic production methods and can be produced solely in one country. Both products can also be easily marketed unbranded in order to avoid strong impact on Quality Perception from brand cues. Another important factor is that foods are seen as low-involvement products, for which consumers usually seek the most convenient ways to form their product perception and purchase decision, therefore the participants have to base their opinion about the products on the country and organic label. Figure 3 shows the selected product images and the country and organic labels that accompanied them in different combinations. The product pictures that were selected are supposed to be neutral representations of both olive oil and yogurt, so that the respondents can only make a quality judgment based on the country and/or organic label(s) that they are exposed to. The absence of intrinsic cues should make the respondents focus on the extrinsic cues.

25

2.4 Measurements

All constructs were measured using a seven point Likert-type scale. To ensure construct validity, scales from previous studies were adapted wherever possible.

Independent variables

The independent variables in this research were the product-country match and organic label trust. Respondents had to reply to statements about first olive oil and then yogurt pictures. To measure the perceived product-country match, 4 items from Roth & Romeo (1992) were used. If a country-of-origin label was present in the condition that the respondent was assigned to, they had to indicate to what extent they agreed to 4 statements using a 7-point Likert scale, where 1 meant that the participant strongly disagreed with the statement, and 7 that they strongly agreed. One of the 4 statements about the product-country match was as follows: ‘This country is well-known to me for this product’ (M= 4.05, SD= 1.50, α= .901).

To find out the degree to which the respondents trust the organic label, 5 items where

26

used from Janssen & Hamm (2012b). The statements were about the organic label’s credibility, standards and control. If an organic cue was present in the conditions that the respondents were assigned to, they had to indicate on a 7-point Likert scale to what degree they agree with them, where 1 meant that the participant strongly disagreed with the statement, and 7 that they strongly agreed. One of the 4 statements about the organic label trust was as follows: ‘I think the control and inspection system behind this organic label is

very strict’ (M= 4.53, SD= 1.32, α= .908). An overview of the means and standard deviations

of the various constructs for domestic, foreign and organic products that were measured in the current research can be found in Table 2.

The respondents did not always have to answer the same questions in different conditions. When the product that the respondent was exposed to did not have an organic label, it would have no use to ask them questions regarding the Organic Label Trust, just like respondents did not have to answer questions regarding the Product-Country Match when their product did not include a country label. Table 1 gives insight in what conditions what questions were asked about the Product-Country Match and the Organic Label Trust. The condition with neither a country nor organic label was left out of this research, as it would not have any contribution to the purpose of the current research.

27

Dependent variables

The dependent variables in this research were the Quality Perception and intention to Buy. Respondents had to reply to 3 statements about the Quality Perception on a 7-point Likert scale, where 1 meant that the participant strongly disagreed with the statement, and 7 that they strongly agreed. One of the 5 statements about the Quality Perception of the product was as follows: ‘This product is of high quality’ (M= 4.93, SD= 1.06, α= .898).

For measuring the respondents’ Intention to Buy the product, they had to reply to 2

statements on a 7-point Likert scale, where 1 meant that the participant strongly disagreed with the statement, and 7 that they strongly agreed (M= 4.42, SD= 1.27, α= .797).

Table 2: Means and standard deviations Table 1: Different conditions

28

Moderators

The moderators in the current research are the variables Ethnocentrism’ and Environmental Involvement. For measuring how ethnocentric the participants are, they had to reply to

statements (8 in total) that were based on the CETSCALE (Shimp & Sharma, 1987) (M= 3.72,

SD= 1.20, α= .869), using a 7-point Likert scale, where 1 meant that the participant strongly

disagreed with the statement, and 7 that they strongly agreed. One of the statements about the participants’ Ethnocentrism was as follows: ‘You should primarily buy products from the

country you live in’.

To find out how environmentally involved the participants were they had to reply to statements from the self-identification as a green consumer scale (Tarkiainen & Sundqvist, 2009) (M= 4.57, SD= 1.14, α= .821), using a 7-point Likert scale, where 1 meant that the participant strongly disagreed with the statement, and 7 that they strongly agreed. One of the statements was as follows: ‘I always buy products that are friendly to the environment’.

Control variables

The control variables age and gender were measured in the beginning of the research. The respondents could select their age on a 6-point scale with the following options: below 20 to 20-29; 30-39; 40-49; 50-59; and 60+ (M= 2.78, SD= 1.36), after which they could click their gender: male/female (M=1.48, SD=.50).

To find out whether the participants actually consume yogurt or olive oil 2 questions were included, on which the respondents could answer on a 5-point scale, varying from ‘never’ to ‘(almost) daily’. The question for olive oil looked as follows: ‘How often do you

consume olive oil’? (M= 3.93, SD= .98); and for yogurt: ‘How often do you eat yogurt’? (M=

29

For the entire questionnaire see Appendix. In Table 1 an overview of the scale means, standard deviations and Cronbach’s Alpha’s can be found.

30

3. Results

This chapter will discuss the results of the current research. First, the descriptive statistics of the variables will be discussed. After that, by using a correlation matrix, the relationships between the variables will be looked at. Finally, with the help of regression analyses light will be shed on whether hypotheses can be accepted or rejected.

3.1 Descriptive Statistics

All the mean scores in the following section are on a 7-point scale. Independent Variables

As expected in the Olive Oil Conditions, the Italian Olive Oil (so either with or without an organic label) scored the highest average on Product-Country Match (M=5,61, SD= 1,19), this means that off all conditions the respondents found the highest fit between the product olive oil and the country Italy. The average Product-Country Match of Dutch Olive Oil (either with or without organic label) was the lowest (M= 2,52, SD= 1,19). In the Yogurt Condition the Dutch Yogurt (either with or without organic label) scored the highest on Product-Country Match (M= 5.37, SD= 1.12), while the Italian Yogurt (either with or without organic label) scored the lowest (M= 2.69, SD= 1.18). An overview of the means and standard deviations can be found in table 4.

The average Organic Label Trust was highest in the Dutch Organic Yogurt condition (M= 5.02, SD= 1.26) and lowest in the Organic Olive Oil condition (M= 4.34, SD= 1.50). Overall, the respondents seem to trust the organic label (M= 4.53, SD= 1.32).

31

Dependent Variables

The people that participated in this research tend to think of the quality of the Olive Oil and Yogurt condition that they were assigned to, to be quite high (M= 4.93, SD= 1.06). The respondents scored about equally high on Quality Perception for the following conditions; Organic Italian Olive Oil (M= 5.71, SD= .89); Italian Non-Organic Olive Oil (M= 5.60, SD= 1.10); Dutch Organic Yogurt (M= 5.68, SD= .93); and Dutch Non-Organic Yogurt (M= 5.68,

SD= .87). On the other hand, there was only one condition that scored the lowest on Quality

Perception, which was the Italian Non-Organic Yogurt condition (M= 3.84, SD= 1.36). It was already expected that Italian Non-Organic Yogurt would score low on Quality Perception, as the purpose of this condition was that it would have a low Product-Country Match. The addition of the organic label in the Italian Organic Yogurt condition raised the Quality Perception (M= 4.47, SD= 1,18).

However, this was also the purpose of the Dutch Olive Oil conditions. Strikingly, Dutch Non-Organic Olive Oil already scores quite high on Quality Perception (M= 4.44, SD= 1.33). The addition of the organic label raised the Quality Perception even more (M= 4,68,

SD= 1.25). In hypothesis 5 it will be tested whether both Organic Label Trust and

Product-Country Match positively predict Quality Perception simultaneously.

32

The Intention to Buy was highest in the Italian Organic Olive Oil condition, which was slightly higher than the Italian Non-Organic Olive Oil and the Dutch Organic/Non-Organic Yogurt. Interestingly, the two non-organic products with unfavourable product country match, Non-Organic Italian Yogurt and Non-Organic Dutch Olive Oil, scored the lowest on buying intention.

Further, it is remarkable that in both the Olive Oil and Yogurt conditions, the

respondents perceive a product with only a favourable country label to be from higher quality than a product with only an organic label.

An overview of the scores on the dependent and independent variables for all conditions can be found in Table 5.

33

Moderators

For the variable Ethnocentrism, it can be said that the respondents on average aren’t very ethnocentric (M= 3.72, SD= 1.20). From the respondents that participated, females (M= 3.79,

SD= 1.24) and people in the age group of 40 to 49 (M= 4.79, SD= 1.39) tend to be the most

ethnocentric, while the people in the age group of 20 to 29 were the least ethnocentric (M= 3.48, SD= 1.08). Note that the age group of 20 to 29 was a lot larger (N= 149) than the age group of 40 to 49 (N= 16), which means that extreme answers have a bigger effect on the average of the age group.

Overall, when looking at the answers for the Environmental Involvement variable, the respondents that participated in the current research tend to be quite environmentally involved (M= 4.58, SD= 1.13). The average of the age group above 60 was the highest (M=5.19,

SD= .91) and the respondents that participated in the age group 40 to 49 scored the lowest

(M= 4.36, SD= 1.49). The female participants (M= 4.81, SD= 1.11) scored on average almost half a point higher on Ethnocentrism than the male participants (M= 4.37, SD= 1.11).

3.2 Correlations

The relationships between the variables are now investigated using a bivariate correlation matrix. The matrix was used to give insight in the relationships between the variables in the 10 olive oil and yogurt conditions. The variables that were included in the matrix are the independent-, dependent-, moderator- and control variables, being Product-Country Match; Organic Label Trust; Quality Perception; Intention to Buy; Environmental Involvement; Ethnocentrism; Age; Gender. Further, Quality Perception and Intention to Buy were split up in Organic/Dutch/Italian, in order to get better insights in what the relation is with Product-Country Match and Organic Label Trust. In the correlation matrix 66 significant Pearson

34

correlations were found between the 14 variables. First, the significant correlations between variables that are relevant for the current research will be discussed. Hereafter, some other significant correlations will be discussed that have not been hypothesized, which might also be interesting to further investigate. The correlation matrix can be found in Table 6.

Correlations for variables hypotheses

First, the significant correlations with Product-Country Match will be discussed. Product-Country Match is one of the independent variables in the current research and it highly correlates with Quality Perception (r =.621, p <.01). Hypothesis 1 predicts a linear relation between these two variables, this is an indication that there might indeed be a relation between the two variables. Further, Product-Country Match highly correlates with Intention to Buy (r=.560, p <.01). Interestingly, when looking at whether the Product-Country Match has the highest effect on Foreign or Domestic Products’ Quality Perception, although both highly correlate, it can be seen that the correlation is stronger with foreign products (foreign: r=.739,

p <.01; domestic: r=.542, p <.001) . Further, it can be seen that the correlation of Foreign

Quality Perception is about equally high as the Foreign Intention to Buy (r=.741, p <.01). This in contrast with the correlation Domestic Quality Perception and Intention to Buy, as the correlation with Intention to Buy is lower for Domestic (r=.453, p <.01).

As expected, Product-Country Match moderately correlates with Organic Label Trust (r=.249, p <.01); and Environmental Involvement (r=.165, p <.01), as the Product-Country Match is about products with a country label and Organic Label Trust and Environmental Involvement about products with an organic label. Interestingly, Ethnocentrism, a variable that just like Product-Country Match is expected to influence the perception about products with a country label does not correlate with Product-Country Match (r=.035).

35

Further, Ethnocentrism does also not significantly correlate with the Quality

Perception (r=-.022) and Intention to Buy (r=-.020) the foreign products that were used. This has led to the rejection of hypotheses h2c and h2d, that claimed that the higher a respondent scores on Ethnocentrism, the more negative the Quality Perception and Intention to Buy foreign products will be.

The other independent variable of the current research, Organic Label Trust,

significantly correlates with, besides the already mentioned relationship with Product-Country Match, positively with Quality Perception (r=.526, p <.01); Quality Perception organic products (r=.661, p <.01); Intention to Buy (r=.498, p <.01); Environmental Involvement (r=.241, p <.01); and, Ethnocentrism (r=.234, p <.01). The relationship with Quality Perception and with Quality Perception of an organic product is especially interesting for hypothesis 4 of the current research, which predicts a linear relationship between Organic Label Trust and Quality Perception. These correlations give an indication that there might indeed be a linear relationship between the variables.

The two dependent variables of this research, Quality Perception and Intention to Buy significantly and highly and positively correlate with one another (r=.772, p <.01), which means there’s no reason to reject hypothesis 6 at this point. Beside significantly correlating with Intention to Buy, Quality Perception further does moderately correlate with the moderators of the current research Environmental Involvement (r=.241, p <.01) and

Ethnocentrism (r=.126, p <.01) which gives an indication that there might be a relationship between the variables, as predicted in hypotheses 3 and 4. Just like there are also significant and positive relationships between Intention to Buy and the moderators (Ethnocentrism:

r=.201, p <.01; Environmental Involvement: r=.367, p <.01), as also predicted in hypotheses

36

Further, Environmental Involvement does moderately correlate positively with Age (r=.164, p <.01). The same goes for Ethnocentrism (r=.179, p <.01), as this variable also positively correlates with age.

The other control variable Gender positively correlates with Organic Label Trust (r=.190, p <.01) and Environmental Involvement (r=.195, p <.01).

37 Table 6: Pearson Correlations all variables and all conditions

**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).

( ). Cronbach’s alpha’s in parentheses

Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1. Product-Country Match 4,05 1,50 _{1 (,901)}

2. Organic Label Trust 4,53 1,32 _,249** _{1 (,908)}

3. Quality Perception (QP) 4,93 1,07 _,621** _,526** _{1 (,898)}

4. Intention to Buy (ITB) 4,42 1,27 _,560** _,498** _,772** _{1 (,797)}

5. Environmental involvement 4.57 1.14 _,165** _,241** _,250** _,367** _{1 (,821)} 6. Ethnocentrism 3.72 1.20 _{,035 ,234}** _,126* _,201** _{,118 1 (,869)} 7. Organic QP 5,10 1,10 _,364** _,661** _,797** _,634** _,370** _,186** ₁ 8. Organic ITB 4,57 1,50 _,307** _,564** _,524** _,845** _,496** _,236** _,718** ₁ 9. Foreign QP 4,75 1,42 _,739** _,262** _,881** _,703** _,221** _{-,022 ,517}** _,351** ₁ 10. Foreign ITB 4,33 1,55 _,741** _,313** _,780** _,855** _,216** _{-,020 ,490}** _,525** _,863** ₁ 11. Domestic QP 5,06 1,14 _,542** _,475** _,880** _,682** _,240** _,219** _,739** _,471** _,475** _,475** ₁ 12. Domestic ITB 4,49 1,35 _,453** _,433** _,673** _,876** _,292** _,341** _,561** _,767** _,390** _,463** _,739** ₁ 13. Age 2,78 1,36 _{,002 -,055 ,006 ,037 ,164}** _,179** _{,023 ,047 -,035 -,004 ,022 ,050 1} 14. Gender 1,48 ,50 _{-,001 ,190}** _{,087 ,086 ,195}** _{,063 ,155}* _,190** _{-,010 -,077 ,132 ,091 ,001 1}

38

3.3 Visualisation

Before testing the hypotheses, the relationships between certain variables will be visualised in a diagram with the use of ‘scatterplot’. The scatterplot plots the scores on one variable against their score on another. The aim of doing this is because it will tell us several things about the data, namely whether there seems to be a relationship between the variables, but also what kind of relationship it is. If the scatterplot suggests that there could indeed be a relation between the two variables, it will give reason to test the hypothesis that they represent. The first hypothesis of this research predicts that there is a linear relationship between Product-Country Match and Quality Perception of products with a country label. The

scatterplot shows us that the Quality Perception increases the higher the Product-Country Match is. The explained variance of the model that tested the relationship between the two variables was (R²=) .385). This gives reason to further investigate the hypothesis. For the scatterplot of hypothesis h1 see table 7.

___________________________________________________________________________

___________________________________________________________________________

39

The second hypothesis predicts a linear relationship between Ethnocentrism and Quality Perception (h2a) and Intention to Buy (h2b) for domestic products. Further, it is predicted that the higher the respondents score on Ethnocentrism the lower the Quality Perception (h2c) and Intention to Buy foreign products will be (h2d). No significant correlations were found between these variables in the correlation matrix, therefore

hypotheses h2c and h2d were rejected. See Table 8 for the scatterplots of the hypotheses that will be further researched, h2a and h2b.

___________________________________________________________________________

___________________________________________________________________________

The scatterplot shows us that the more ethnocentric the respondents are, the higher the Quality Perception of domestic products is, although this relationship does not explain a lot of the variance (R²=.048), but there seems to be an upward trend. The scatterplot of Ethnocentrism and Intention to Buy domestic products also shows an upward trend. The explained variance (R²=.116) is more than twice as high as for Quality Perception domestic products, which indicates that Ethnocentrism is better in predicting Intention to Buy than Quality Perception domestic products. Further testing has to proof whether these relations are significant.

40

The third hypothesis says that the higher the respondents score on Environmental

Involvement, the higher Quality Perception (h3a) and Intention to Buy (h3b) for organic products will be. The scatterplot with the variables Environmental Involvement and Quality Perception organic products shows that there seems to be an upward trend between the two variables. Although, the explained variance of the two variables is not very high (R²=.137).

The scatterplot with the variables Environmental Involvement and Intention to Buy organic products also shows an upward trend with an almost doubled explained variance (R²=.246). Further testing has to proof whether Environmental Involvement can indeed predict values of Quality Perception and Intention to Buy for organic products, so that the hypotheses can be accepted or rejected. For the scatterplots of h3a and h3b see table 9.

___________________________________________________________________________

___________________________________________________________________________

41

The fourth hypothesis (h4) of this research expects a linear relationship between Organic Label Trust and Quality Perception organic products. The scatterplot (see Table 10) clearly shows that Quality Perception organic products increases at higher levels of Organic Label Trust. The explained variance of the model that tested the relationship between the two variables was (R²=) .437. This gives sufficient reason to further investigate hypothesis 4.

___________________________________________________________________________

___________________________________________________________________________

No scatterplot was made for the fifth hypothesis, because it predicts that two variables, namely Product-Country Match and Organic Label Trust, simultaneously influence Quality Perception. The scatterplots have given reason to further investigate hypothesis 1 and 4. Further testing will tell whether Product-Country Match and Organic Label Trust form a better model in predicting Quality Perception when they are combined or separated and will see whether the organic label has a bigger effect on Quality Perception for foreign than domestic products.

42

The sixth hypothesis of this research predicts a linear relationship between Quality Perception and Intention to Buy. In the scatterplots (see Table 11), it can be seen that the Intention to Buy indeed seems to be going up at higher levels of Quality Perception. The explained variance of the model that tested the relationship between the two variables was (R²=) .595. This gives sufficient reason to further investigate hypothesis 6.

___________________________________________________________________________

___________________________________________________________________________

43

3.4 Hypotheses Testing

In this section the hypotheses of the current research will be tested. The descriptive statistics, correlations and visualisations of the relationships between variables have given some insights on whether it would be meaningful to further investigate the relationships between several variables of the current research. Except for parts of hypothesis 2, the previous steps in analysing the variables has not yet given any reason to reject any other hypothesis.

Hypotheses h2c and h2d were rejected with the use of the correlation matrix, as it showed that there is no relationship between Ethnocentrism and the Quality Perception and Intention to Buy foreign products in the current research. The next sections will one by one discuss the results of the hypotheses testing.

3.4.1. Product-Country Match and Quality Perception

For hypothesis 1, it was expected that Product-Country Match can predict values of Quality Perception. Hierarchical regression was performed to investigate the ability of Product-Country Match to predict values of Quality Perception.

In the first step of the regression, the independent variable Product-Country Match and the dependent variable Quality Perception was entered. This model was statistically

significant F (1, 242); p <.001 and explained 38.5% of the variance in Quality Perception. Product-Country Match recorded a high beta value of (β =) .62 p < .001). In other words, if the standardized value of Product-Country Match increases by one, the standardized Quality Perception will increase by .62. Thus, Product-Country Match and Quality Perception are positively related. For this reason, h1 is accepted.

44

__________________________________________________________________________

Table 12: Hierarchical Regression Model for Hypothesis 1

Dependent variable: Quality Perception

R² F- Value t p B(SEb) β Step 1 Constant 19.98 .00 3.12 (.16) Product-Country Match 12.31 .00 .44 (.04) .62 .38 151.64*** _________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001. _________________________________________________________________________

3.4.2 Ethnocentrism, Quality Perception and Intention to Buy domestic products.

For hypothesis 2 it was predicted that Ethnocentrism can predict values of Quality Perception and Intention to Buy for domestic (h2a and h2b) and foreign products (h2c and h2d).

Hypotheses h2c and h2d have already been rejected, as no significant correlations were found between Intention to Buy foreign products and Ethnocentrism. For hypothesis h2a, again a regression analysis was performed to investigate the ability of Ethnocentrism to predict values of Quality Perception domestic products.

In the first step of hierarchical regression, the independent variable Ethnocentrism was entered. The dependent variable in the regression was Quality Perception domestic. This model was statistically significant F (1, 161); p < .005 and explained 0.05% of the variance in Quality Perception domestic. Ethnocentrism recorded a beta value of (β =) .22 p < .005).

45

the ability of Ethnocentrism to predict values of Intention to Buy domestic products.

In the first step, the independent variable Ethnocentrism was entered. The dependent variable in the regression was Intention to Buy domestic. This model was statistically significant (1, 161); p < .001 and explained 11.6% of the variance in Intention to Buy domestic. Ethnocentrism recorded a beta value of (β =) .34 p < .001. The constant for both Quality Perception in hypothesis h2a is higher than for Intention to Buy. The constant for Quality Perception is above average, which means that if someone would score a 1 out of 7 on Ethnocentrism, that he would still perceive the quality of the domestic product to be quite high. For Intention to Buy, on the other hand, the constant scores below average, which means that if someone would score the lowest on Ethnocentrism, that the Intention to Buy for this person would be quite low. Although higher scores on Ethnocentrism have bigger effects on Intention to Buy domestic products than for Quality Perception.

In both cases, higher scores on Ethnocentrism result in higher scores on Quality Perception and Intention to Buy domestic products. Therefore, both hypotheses h2a and h2b are accepted. For the regression models of hypothesis 2, see Table 13a and 13b.

46 ________________________________________________________________________ Table 13a: Hierarchical Regression Model for Hypothesis 2a

Dependent variable: Quality Perception domestic products

R² F- Value t p B(SEb) β Step 1 Constant 15.39 .00 4.30 (.28) Ethnocentrism 2.85 .00 .20 (.07) .22 .05 8.13** _________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001. _________________________________________________________________________ __________________________________________________________________________ Table 13b: Hierarchical Regression Model for Hypothesis 2b

Dependent variable: Intention to Buy domestic products

R² F- Value t p B(SEb) β Step 1 Constant 9.73 .00 3.10 (.32) Ethnocentrism 4.60 .00 .37 (.08) .34 .12 21.21*** _________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001. __________________________________________________________________________3.

47 4.3 Environmental Involvement, Quality Perception and Intention to Buy organic

products

For hypothesis 3 it was predicted that Environmental Involvement can predict levels of Quality Perception (h3a) and Intention to Buy (h3b) for organic products.

For hypothesis h3a the control variable that significantly correlated (as seen in the correlation matrix) with Quality Perception organic, Gender, was added in step 1. In step 2, the independent variable Environmental Involvement was added. Gender turned out to not significantly correlate with Quality Perception organic and was therefore not added in a new regression with only Environmental Involvement and Quality Perception organic. The new model was statistically significant (1, 207); p <.001 and explained 13.3% of the variance in Quality Perception organic. Environmental Involvement record a beta value of (β =) .34

p <.001. As higher levels of Environmental Involvement result in higher levels of Quality

Perception of organic products, hypothesis h3a was accepted.

For hypothesis h3b, the independent variable Environmental Involvement and the dependent variable Intention to Buy organic products was added. The model was statistically significant (1, 208); p < .001; p <.001 and explained 24.6% of the variance in Intention to Buy organic products. Environmental Involvement recorded a beta value of (β =) .50 p <.001. As higher scores on Environmental Involvement predict higher scores on Intention to Buy organic products, hypothesis h3b was also accepted.

When looking at the constants for both models, the constant of Quality Perception organic products is very close to neutral (a score of 3.47 on a 7-points scale). Therefore, even a person that is not Environmentally Involved at all (a score of 1 on a 7-points scale) would still find the quality of an organic product to be not bad, this while the more Environmentally Involved someone is, the higher the Quality Perception of the product will be. When looking

48

at the regression of Intention to Buy organic products, the constant is far lower. People therefore have to be way more Environmentally Involved in order to score high on Intention to Buy.

For the regression models of hypothesis 3, see Table 14a and 14b.

_________________________________________________________________________ Table 14a: Hierarchical Regression Model for Hypothesis 3a

Dependent variable: Quality Perception organic products

R² F- Value t p B(SEb) β Step 1 Constant 11.93 .00 3.47 (.29) Environmental Involvement 5.74 .00 .35 (.06) .37 .14 32.95*** _________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001. _________________________________________________________________________

49 _________________________________________________________________________ Table 14b: Hierarchical Regression Model for Hypothesis 3b

Dependent variable: Intention to Buy organic products

R² F- Value t p B(SEb) β Step 1 Constant 4.35 .00 1.61 (.37) Environmental Involvement 8.23 .00 .64 (.08) .50 .25 67.77*** ________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001. ________________________________________________________________________

3.4.4 Organic Label Trust and Quality Perception Organic Product

Hypothesis 4 predicts a linear relationship between Organic Label Trust and Quality Perception organic product. For hypothesis h4, hierarchical multiple regression was

performed to investigate the ability of Organic Label Trust to influence the Quality Perception of organic products.

In step 1, the control variable that significantly correlated in the correlation matrix with Quality Perception organic product, Gender, was added. No significant results were found for Gender, therefore a new regression was run with only Organic Label Trust as independent variable and Quality Perception organic products as dependent. This model was statistically significant F (1, 208); p < .001 and explained 43.7% of the variance in Quality Perception organic products. Organic Label Trust recorded a high beta value (β = .66, p < .001). Higher scores on Organic Label Trust result in higher Quality Perception of organic

50

products. Therefore, hypothesis h4 is accepted. The constant of the regression model is low, meaning that if someone does not trust the organic label, this person will find the quality of the organic product to be quite low (2.61 on a 7-point scale). Only persons that score high on Organic Label Trust will reach moderately high scores on Quality Perception organic

products.

For the regression model of hypothesis 4, see Table 15.

___________________________________________________________________________

Table 15: Hierarchical Regression Model for Hypothesis 4

Dependent variable: Quality Perception organic products

R² F- Value t p B(SEb) β

Step 1

Constant 12.76 .00 2.61 (.20)

Organic Label Trust 12.70 .00 .55 (.04) .66 .44 161.27***

_________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001.

_________________________________________________________________________

3.4.5 Product-Country Match, Organic Label Trust and Quality Perception

Hypotheses h1 and h4 have shown that Product-Country Match and Organic Label Trust both separately have effect on Quality Perception of products with either a country or organic label. Hypothesis h5 predicts that when a product has both a country label and an organic label, that Product-Country Match and Organic Label Trust both simultaneously positively

51

influence the Quality Perception of the product. For h5 hierarchical regression was used. In step 1, Organic Label Trust and Product-Country Match were added as independent variables. The dependent variable in this regression is Quality Perception of products with both a country and organic label. Ethnocentrism did not significantly predict values for Quality Perception and was therefore deleted from the regression analysis. In step 2, Product-Country Match and Organic Label Trust were added. This model was statistically significant (3, 158); p <.001 and explained 55,3% of the variance in Quality Perception of products with both a country and organic label. Environmental Involvement recorded beta value of (β=) .22,

p <.001; Product-Country Match β = .34 p <.001; and Organic Label Trust had the highest

beta value (β=) .48 p <.001. Higher scores on the three independent variables result in higher scores on Quality Perception of products with both a country and organic label, therefore, hypothesis 5a is accepted. The constant of Quality Perception of products with both labels in the regression model is very low, but this is what someone would score if a person would score the lowest on Environmental Involvement, Product-Country Match and Organic Label Trust. Higher scores on these three variables will impact the Quality Perception of products with both labels a lot.

52

___________________________________________________________________________

Table 16 Hierarchical Regression Model for Hypothesis 5

Dependent variable: Quality Perception products with country and organic label

R² F- Value t p B(SEb) β Step 1 Constant 9.50 .00 3.34 (.35) Environmental Involvement 5.20 .00 .38 (.07) .38 .15 27.01*** Step 2 Constant 3.30 .00 1.06 (.32) Environmental Involvement 3.98 .00 .22 (.05) .22 Product-Country Match 6.32 .00 .27 (.04) .35 Organic Label Trust 8.62 .00 .42 (.05) .48 .55 64.46***

_________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001.

_________________________________________________________________________

3.4.6 Quality Perception and Intention to Buy

Hypothesis 6 predicts a linear relationship between Quality Perception and Intention to Buy. To find out whether Quality Perception has the ability to predict values for Intention to Buy, hierarchical regression was used.

In step 1, the independent variable Quality Perception and the dependent variable Intention to Buy was added. The model was statistically significant F (1, 252); p < .001 and

53

explained 59.5% of the variance. Quality Perception recorded a very high beta value (β = .77,

p <.001). Higher levels of Quality Perception will therefore result in higher levels of Intention

to Buy. Therefore, hypothesis h6 is accepted. No statements can be made about the constant of the regression, as it was not significant.

For the regression model of hypothesis 6, see Table 17.

___________________________________________________________________________ Table 17 Hierarchical Regression Model for Hypothesis 6

Dependent variable: Intention to Buy products with country and organic label

R² F- Value t p B(SEb) β Step 1 Constant -.45 .65 -.109 (.24) Quality Perception 19.26 .00 .92 (.05) .77 .59 370.96*** _________________________________________________________________________ Note. *p <.05, **p <.01, ***p < .001. _________________________________________________________________________

3.5 Explorative Analysis

3.5.1 Exploration Hypothesis 1; Product-Country Match and Quality Perception

Except for hypotheses h2c and h2d all the other hypotheses were supported by the results that were found. H2c and h2d were rejected as it did not find evidence that Ethnocentrism has different effects on the Quality Perception and Intention Buy of foreign products. This is in