The influence of a fair trade label on the perception of quality

(1)

The influence of a fair trade label on

the perception of quality

Master Thesis Marketing

Johan J. Walda

(2)

1

The influence of a fair trade label on the perception of quality

Name Student: Johan J. Walda

Student number: s1780336

E-mail: johanwalda@hotmail.com

Address: Ceramstraat 32a, 9715JP Groningen

Telephone: +31 (6) 29562593

University: University of Groningen

Faculty: Faculty of Economics and Businesses

Education: Marketing Management

Module: Master thesis Marketing Management

1st_Supervisor: _{drs. M. Keizer}

2nd_Supervisor: _{Prof. dr. K. van Ittersum}

(3)

2

Management Summary

This study examines the effect of a fair trade label on the perception of quality. In that, this research uses pictures of a product of coffee beans. Perception of quality is stated in three different aspects, namely, taste, price-quality ratio, and overall quality. Besides that, this research also uses factor perceived quality in which the three different aspects are combined. In this way, this research can see which aspect is mostly influenced by a fair trade label and the effect on the combined factor. Besides testing for the effects of a fair trade label, this research also tests for the influence of price and altruism. From literature, the expectation is extracted that the higher the price, the higher the expectations of quality. This research is especially interested in the effects of a price discount. Since a higher price leads to higher expectations of quality, a price discount is believed to lower perceived quality. For altruism holds that it is closely linked to fair trade since altruism is about caring for others and fair trade is a way in which the consumer can care for the producers of the product. That is why this research expected that a fair trade label has a stronger effect on altruistic people.

The research is conducted with an online survey in a 2x2 between-subjects research design. The main part of the survey was a picture of an advertisement for coffee beans. Participants needed to rate the product on the three quality aspects, respectively, taste, price-quality ratio and overall quality. There were four conditions, in which two variables were altered. Firstly, two pictures contained a fair trade label on the package of the product and two did not have a fair trade label. Secondly, in two out of four conditions the price was displayed as a discount, in the other two there was a regular price visible. Next to that, the participants needed to answer general questions (e.g. age, gender), questions about fair trade habits (e.g. frequency of buying fair trade products), and lastly questions about the importance of certain life values in their lives (e.g. equality, power).

(4)

3

when a fair trade label is on the product, the overall quality is perceived higher than in a situation without a fair trade label.

Additionally, this research also tested the combination of the three quality aspects, called factor perceived quality. Also here a fair trade label and price are significant predictors of perceived quality. By this, this research can conclude that it is important for fair trade products that the label is easy to recognize and easily visible for consumers. This because it has a positive influence on the perception of quality of that product. Furthermore, price was a surprising result. This was surprising is the sense that the reverse effect occurred. The expectation was that the lower the price, the lower consumers expect quality to be. In fact, this research found that a discount would enhance the perception of quality. Even though there is a strong effect of price-quality ratio on this negative sign, overall quality also showed, although not significant, a negative relationship. This finding can be interesting for future research. In addition, the control variables age and frequency of buying fair trade products were also significant. It could be that because fair trade products are linked with something positive, a person who has a lot of experience with these positive products will transfer this feeling to general products as well. Although this is a topic for future research, evidence for this conclusion would result in practical implications for marketers. It expects that it helps to make positive attributes of fair trade salient to the consumer. Positive information about fair trade in the shopping environment could make retrieving this positive image easier from memory for consumers who buy fair trade products frequently. Besides that, consumers who do not buy fair trade products frequently could make a positive image by seeing this information. This could ultimately lead to higher expected quality of products in general.

The results of this research can be used by fair trade organization to stimulate the demand for fair trade products or to raise the perception of quality for fair trade products. Firstly, the label is an important asset in improving the perception of quality. Secondly, displaying the price as a discount can stimulate the demand, but this needs to be further researched. Lastly, the positive image of fair trade products can be transferred to all products in general. This is also a point of further research but could be important in understanding the effects of fair trade.

(5)

4

Preface

This research is the final assignment for my master in Marketing Management. The subject of this research is related to ‘the irrational consumer’. By this, I mean the fact that consumers do not make conscious, rational decisions all of the time. Most of our behavior is automatic and a result of cues in the environment. The environment in this case can be several things, for example, the background colors of a web page or a certain brand name. These things affect our preferences and therefore our behavior. In light of this, this research investigated the effects of a fair trade label on the preference of consumers, specifically, the perceived quality. The production of fair trade products does not differ from conventional products, so a fair trade label only means that working conditions are good and the source of the products gets a fair price. That means that quality should not be different whether products carry a fair trade label or not. Hence, the mind of the irrational consumer forms expectations about the product when faced with such a label. That is also why I find this topic interesting. Psychology and marketing are very interesting topics and, in light of that, it is interesting to see how consumers are affected by marketing actions.

This master thesis is the last assignment before I graduate. This means that this is an end of a life period. I am really excited to apply all the knowledge and qualities I learned as a student in a future career.

In my time as a student I had a lot of support from several persons, which I want to thank. First of all, I would like to thank my parents, because they always support me. Secondly, a special thanks to my first supervisor of this thesis, drs. Martijn Keizer. Normally you do not have a lot of personal time with a professor or supervisor of the university, but this thesis is something different. The personal feedback was very helpful and I enjoyed working together. Last but not least, I would like to thank all the respondents who made my research possible.

(6)

5

Introduction

Through the ease of globalization in recent years, companies are more and more focused on profits and margins. By outsourcing some parts of production to lower wage countries, the western world is able to achieve these higher margins. Despite this advantage for the outsourcing company, producers in low wage countries are in most cases faced with bad working conditions and unfair payments. Besides that, child labor and discrimination are also a result of bad living conditions (Stratton and Werner 2013). By counteracting this trend, an ‘anti-globalization movement’ (Klein 2001) emerged in which companies have a different business model with an explicit goal to transform traditionally exploitative global production and trade relations (Murray and Raynolds 2000). This transformation entails that products need to be produced in a fair way, in other words, fair trade. Fair trade involves a fair payment for producers in low wage countries that ensure better working and living conditions, access to medical care, education, and prohibition of child labor and discrimination (Stratton and Werner 2013; Langen 2011).

In order to show consumers that a certain product is produced according to the fair trade principle, a fair trade label is shown on the package of the product. In this way consumers know that, for example, the producers of a labelled product got a fair price that guarantees better living conditions. Additionally, the question arises what kind of expectations this label elicits in the mind of the consumer surrounding the products’ quality for example.

In this research, the focus will be mainly on the effects of a fair trade label on the perceived quality of a product. Because the production process of fair trade products does not differ from conventional products, a fair trade label only ensures that working conditions are good and the producers got a fair price (Pelsmacker et al. 2006). Therefore, the label does not indicate superior quality or taste. However, people can only process a limited amount of information (Simons & Chabris 1999). Consumers do not have the time to evaluate every aspect of quality for a given product and rely therefore most of time on heuristics in their decision making. In those situations, a fair trade label functions as a heuristic (Jacoby et al. 1971). This means that differences found in quality are perceived differences and not actual differences. That is why this research focuses on the effects of a fair trade label on perceived quality.

(9)

8

lower in quality than when the product has a regular price. Additionally, this discounting effect may be smaller for individuals who care about fair trade and the effect of a fair trade label may be stronger for those individuals.

The next chapter consists of a theoretical framework with existing literature on this topic and hypotheses that will cover the effect of a fair trade label on perceived quality and which effects price and altruism can have. Secondly, the chapter research design will go further in depth about the type of research and plan of analysis. Thirdly, the analysis of the research will be discussed in the results chapter. After that, the discussion chapter will compare the results with theory and practical implications of the results will be discussed. Lastly, this article will consist of the chapters ‘limitations and future research’ and ‘conclusion’. In the former, limitations of this research and possible directions of future research will be discussed. The latter will briefly summarize the main findings of this research.

Theory and Hypotheses

The first initiative in fair trade, called Ten Thousand Villages, was back in 1946 concerning the United States buying needlework from Puerto Rico. Soon after this initiative, other parts in the world also began fair trade movements. In 1967 the first Fair Trade Organization was a fact and in 1988 the coffee producer ‘Max Havelaar’ from the Netherlands introduced the first fair trade label (International Trade Forum 2006). This first label certified for the consumer that growers of coffee beans got a fair price that guarantees better living conditions. The resulting premium price paid by consumers for these fair trade products reached the growers and not the middlemen involved (Renard 2003).

(10)

9

of fair trade (EFTA 2001). It has three objectives: market development, fair trade monitoring, and advocacy (International Trade Forum 2006).

When producers want to use a fair trade label, certain conditions need to be met. First of all, producers need to purchase directly from the source and have to handle a price that covers the cost of production and a social premium to improve living and working conditions. Secondly, they have to pay in advance to prevent the source from falling in debt. Lastly, the fair trade relationship has to be official by contract that allows long-term production planning and sustainable production practices (Renard 2003).

Next to that, also the source or farmers need to meet certain conditions in order to grant certification for fair trade. Firstly, child labor is absolutely not done. Secondly, there are certain actions which will improve and maybe guarantee better living conditions. One of them is participation in a democratic organization. Another option is participation in trade union activities which will deliver a good salary, decent housing, health, and safety standards.

The Positive Relationship between a Fair Trade Label and Perceived Quality

When a fair trade relationship is official, products can wear the label ‘fair trade’. In this way consumers can identify these products. The question then arises what the attitude of consumers is towards fair trade products and what kind of expectations a certain label brings to mind.

Several studies have shown, in light of social product labels, that a fair trade label will create specific expectations about the labelled product in the mind of the consumer (Deliza and MacFie 1996; Schifferstein, Kole and Mojet 1999). With these expectations, this research means what the consumer expects of the product in terms of quality, performance, visual aspects, and so forth. Expectations of higher quality as a result of a fair trade label on a product are of greatest interest for this research.

Deliza and MacFie (1996) found that packaging, advertising and information generate sensory expectations. These expectations can subsequently affect a consumer’s choice. This can mean that a fair trade label, which functions as packaging c.q. information, affects expectations about the product and therefore affects product choice.

(11)

10

when the Coca Cola brand was visible to participants. This shows again that information can manipulate the perception of consumers.

Furthermore, Schifferstein, Kole and Mojet (1999) investigated the effects of expectations created by a product description on the perception of that product. They found that the judgments of consumers were linked to the evaluation of performance of the product. In other words, expectations of, for example, quality, are affected by packaging.

One aspect of packaging for fair trade products is the fair trade label. Prior research on the effect of a label is, among others, done by Luchs et al. (2010) and Poelman et al. (2008). Luchs et al (2010) found that that the presence of an ethical attribute, in this case a fair trade label, can influence consumer perceptions of other valued attributes, which in turn can affect product preference. This can mean that the label itself is expected to increase the perceived quality of a product.

According to Poelman et al. (2008) consumers expect that organic (or fair trade) foods tastes better, which influences their acceptability of such products. A fair trade label will therefore elicit a sort of halo-effect. This phenomenon holds that if a product is judged to be superior on one attribute, the other attributes will also be perceived more favorably (Asch 1946; Thorndike 1920). That can mean in this case that when a product holds a fair trade label, not only the taste is perceived as better but also other aspects of the product like the overall quality.

Similarly, consumers evaluate products generally on extrinsic product cues (e.g., price, label), rather than intrinsic cues like nutritional value (Pohl 2004). That means that an extrinsic cue, like a fair trade label, can influence the perception of quality of the product at hand. According to Lotz et al. (2013) are products that have a fair trade label rated higher on quality than unlabeled products. Therefore this research expects, first of all, that a fair trade label on a product will enhance the perceived quality of that product.

Despite this finding, there is also evidence that such an effect may not exist. According to Didier and Lucie (2008), a fair trade label serves as a credence attribute. This means that consumers cannot observe the qualities of a label, which makes it difficult to measure its utility. The same article argues that a credence attribute like a fair trade label will be overlooked by the consumer when determining the (perceived) quality of a product. This contrasting evidence about the effect of a fair trade label on perceptions ensures that there is still room for further research on this topic.

(12)

11

different than ‘regular’ products (Bourn and Prescott 2002; Woese et al. 2007). According to Bourn and Prescott (2002) do organically grown products not differ on nutritional value, sensory quality, or food safety. Although this prior research is more focused on organic foods, this research expects that the same holds for fair trade production versus conventional production. This expectation comes from the fact that the price paid for these products cover the production costs and a social premium. The extra payment will therefore foremost improve the welfare of the farmers and not the products itself.

Also in the research of Woese et al. (2007) they found that taste is not better or different than with conventional products. They investigated this by reviewing more than 150 investigations comparing the quality between conventional and organic foods.

Even though the research by Woese et al. (2007) and Bourn and Prescott (2002) are in the field of organic foods and not specific fair trade products, this research believes that these studies are applicable to fair trade and, therefore, this research expects that the actual product quality does not differ between fair trade and conventional products. Therefore, differences in expectations about the product are believed to be caused by the label. This results in the following hypothesis:

H1: Consumers will rate a product with a fair trade label higher on perceived quality than

when the product has no label.

The Influence of Price

As stated in the hypothesis above, this research expects that the sole effect of a fair trade label on a product will be that perceived quality enhances. That point is strengthened by the article of Renard (2003), in which a label functions as a distinctive sign of product quality. The perceived quality therefore increases when a product has a fair trade label according to this article.

Next to that, there are more signs that a consumer uses when making expectations about the quality of a product. One of such signs can be the price. According to Combris (1997) prices serve as a quality sign in a situation of asymmetric information. When a consumer does not have full information about the quality of a product (asymmetric information), the price can be used as a heuristic to assign certain quality expectations.

(13)

12

competitive supply and demand. A high price will therefore be a result of high demand and high demand must be high quality in the minds of a consumer.

In an earlier research, Dodson et al. (1978) also found that consumers see a discount as an expression of poorer quality. Additionally, Rao and Monroe (1989) showed empirical evidence that consumers often perceive lower-priced products and services to be of lower quality, especially if they have no simple alternative way to assess quality. This also strengthens the previous point that a higher price is linked with higher quality and vice versa a lower price, or discount, with lower quality.

Thus, in a situation where the consumer does not have full information and therefore relies on heuristics, the following holds: the higher the price, the higher the perception of quality. This leads to the second hypothesis of this research, namely:

H2: A price discount will lead to a lower perceived quality of a product.

With the price, this research means the reference price of a product. This is a price which the consumer has in mind for a given product and is an expression of how much the product is worth, subjectively. This can be formed by considering an average market price (Biswas and Blair 1991) or an expectation about the price (Winer 1986). Consequently, when a product is on discount and this is made clear by stating also the original price or a certain percentage by which the original price is lowered, the internal reference price will also be lower (Grewal et al. 1998; Grewal and Compeau 1992; Rajendran and Tellis 1994). A discount will, in this light, lower the internal reference price and may therefore decrease the perception of quality (Blattberg and Neslin 1990).

On the other hand, like mentioned before, a fair trade label may increase the perception of quality. Therefore a fair trade product on discount can create incongruence in the mind of consumers. This may happen because consumers do not expect fair trade products to be on discount. The premium price of fair trade product is due to the fact this premium reaches the producers and not the middlemen involved (Renard 2003). That means that the regular, premium price is believed to be the right price for fair trade. A discount could then result in incongruence because how can fair trade be retained with a lower price? This research believes therefore that discounting will result in lower sales for fair trade products than in a regular price situation. Also, the perceived quality of a fair trade product may be damaged by putting it on discount.

(14)

13

other information that maybe relate to this effect. One of them is the research by Grewal et al. (1998), where they found that a negative effect of price discounts are counter-balanced by the positive effects of a brand name and a brand’s perceived quality. This can mean that if a brand name can counter-balance the negative effects of a discount, a fair trade label, as a sort of brand, may also counter-balance this effect. Therefore, the resulting perceived quality would be the same as in the case of a regular product (no fair trade label) with a regular price. Thus, this research proposes the following hypothesis:

H3: A fair trade label will counter-balance the effect of a price discount on the perceived

quality of a product.

The Influence of Altruism

Despite the evidence surrounding the first two hypotheses, there is also research that did not find a significant effect of a label on the perception of that product (Poelman et al. 2008). In this prior research they gave subjects a piece of pineapple and varied information about organic production and fair trade. In general, the results showed that fair trade information does not seem to influence the perception of the product. Despite this, when these subjects were grouped according their attitude towards fair trade, they did find a link between fair trade information and perception. People with a positive attitude towards fair trade perceived the products to have a higher sensory impact and therefore perceived the products to have more positive characteristics. Therefore this research can conclude that even though there are mixed results from research in the field of fair trade and perception, adding moderators to this relationship can result in ´more pieces of the puzzle´. That is why Poelman et al. (2008), among others, points out that there is still a lot of room for future research on this topic.

(15)

14

interesting to see whether altruistic consumers, who are concerned about the welfare of others, will react differently to fair trade labels than non-altruistic consumers.

By altruism, this research means beliefs or behavior of concern for the welfare of others. According to Kareklas (2014) are certain values consistent with altruistic motives, namely ecology, universalism, benevolence, and equality. These values predict organic and fair trade food consumption according to this article.

As a result, this research expects that altruism is also linked to a more positive attitude towards fair trade. A label of fair trade can give the consumer the idea that what they are buying and eating is somehow better (Getz 2006), in the sense of product quality as in ethicality. Therefore, the following hypothesis holds:

H4: For more altruistic consumers, the effect of a fair trade label on perceived quality of a

product is stronger.

Furthermore, this research also expects that altruistic consumers react differently to fair trade products on discount. As Poelman et al. (2008) pointed out, people with a positive attitude towards fair trade perceive these products to have a high sensory impact and therefore perceive fair trade products to have more positive characteristics. This means that for more altruistic consumers, the fair trade label serves as a sort of halo-effect. The label can positively affect the way altruistic consumers see other aspects of the product, including the price. Consequently, this could mean that a fair trade product on discount is seen less negative by altruistic consumers than non-altruistic consumers do. This relationship is not yet investigated in other studies and relies therefore mainly on expectations. The following hypothesis relates to this:

H5: For altruistic consumers, the positive effect of a fair trade label is stronger than the

negative effect of a price discount on perceived quality.

(16)

15

Research design

In this chapter, this article will outline the methods used regarding the research and data collection.

Data Collection

In order to test the above hypotheses, this research conducted an online survey. The survey is conducted in a 2x2 between-subjects research design. This means that there are four different conditions. First of all there are two levels in the independent variable fair trade label. This variable is tested in the form of an advertisement, one with a fair trade label and one without. Secondly, the independent variable price discount also consisted of two levels, namely regular price versus discount price.

(17)

16

the survey is put on social media in order to attract from a big source of possible respondents. Additionally, each participant is randomly assigned to one of the four different conditions. In that way each condition will have almost the same number of participants.

Questionnaire Design

Like stated above, the questionnaire is in the form of an online survey. For this research I used the software on the website ‘qualtrics.com’. The survey is set up in Dutch since all the participants are from the Netherlands. In this way the questions are easier to read and easier to interpret for participants then when the survey was conducted in English. In the subsequent chapters this research will refer to the questions and constructs used in the survey by an English translation but the original text will be enclosed by brackets.

When participants used the link to the survey they first saw a welcome message in which was stated that the goal of this research is to see what kind of expectations a new product on the market will bring to mind. Hence, participants will not know, before they start, that the survey is about the effects of a fair trade label. This approach is chosen because it is believed that when the participants know beforehand that the research is about the effects of a fair trade label, this may affect their subsequent responses. This can mean that participants rate a product higher on the different quality aspects when a fair trade label is on it because they know this research tests for that. Whether this would happen or not, this research chose to minimize the risk by not telling the true goal of this research.

After the welcome message, participants had to fill in some general question about their gender, age, education level, and how much coffee they drink per week. This last question is added here because this could be a moderator. The more frequently people drink coffee, the more familiar they are with the product used in this research. This could mean that when participants do not drink coffee frequently, it is harder to rate the product. Thus, the results of the hypotheses in this research could differ on the basis of frequency of drinking coffee. Next to that, this question is believed not to influence subsequent questions. After the general questions, participants saw a picture of a coffee beans package and they had to give a rating between 1 (low) and 10 (high) on three different quality measures. These measures are taste (‘smaak’), price-quality ratio (‘prijs-kwaliteit verhouding’) and

(18)

17

two show that the product is on discount. Below, in figure 2, the different pictures are visually displayed.

In this way, this research can see whether the pictures with a fair trade label score higher on the different quality measures. According to the ‘ceteris paribus’-principle, this particular research design results in the fact that differences found in the scores between pictures are due to label or price since the rest is held constant.

(19)

18

After that, the participants got a list of values and they had to give a rating between -1 (against my life principles) and 7 (extremely important) to what extent this value is important in their lives (De Groot and Steg 2008). A few examples of such life values:

equality (‘gelijkheid’), prevent pollution of the environment (‘milieuvervuiling voorkomen’),

wealth (‘rijkdom’), ambitious (‘ambitieus’), helpfulness (‘behulpzaamheid’). With this list of values, this research tries to scale each participant on the importance of altruism in their lives.

In appendix A, the whole questionnaire is displayed.

Sample Characteristics

After data collection, 160 people participated in this research. Subsequently, 42 participants were taken out of the research, because the list with life values was not filled in correctly in the eyes of this research. For this

exclusion, this research sets clear boundaries. First of all, when participants had 4 or more values stated as ‘extremely important’, although the survey told the participant that normally a person has only 2 extremely important values in life, the record was deleted. Secondly, when participants filled in all the values in a range of 3 different points of importance, then this record was also excluded from further analysis.

This results in a total of 118 participants for this research. Table 1 shows an overview of the general descriptive statistics. When looking at this table, more males (59,3%) than females (40,7%) participated in this research. The mean age is 36,14 years old with a standard deviation of 12,8. The youngest

person is 20 years old and the oldest person 63 years old. This means that there is a relatively large difference in age, except the fact that 41,5% is between 21 and 30 years old.

Table 1:

Descriptive Statistics (N=118)

Variable Count Percentage

Gender # %

Male 70 59,3

Female 48 40,7

Age (in years) # %

(20)

19

This is due to the fact that this research used Facebook to attract respondents and this research believes that the network used consisted mostly of young people.

Besides the statistics in table 1, this research also asked some questions about fair trade and coffee. First of all, there is a question about the coffee drinking habits of the participant. What is interesting about this is that 22 participants (18,6%) indicated that they drink less than 1 cup of coffee in a week against 49 participants (41,5%) who drink more than 15 cups of coffee in a week. The remaining 47 participants (39,9%) are in between that. It seems that either you drink very little/never coffee or you drink a lot (more than 15 times a week).

Secondly, the survey also asked whether the participant knows the concept of fair trade and if they answer yes, how often they buy fair trade products and if they buy it consciously. On the former question almost every participant answered ‘yes’ (97,5%), so almost everyone know the concept of fair trade. On the subsequent question about how often they buy fair trade product the answers are more divided. About a quarter (25,2%) indicated that they have no idea how often they buy fair trade products. While, besides that, almost another quarter of the participants (23,5%) indicated they never buy fair trade products, around 40% buys it once per quarter of a year (17,4%) or monthly (21,7%). On the question whether they buy fair trade products consciously answered 48 (42,9%) ‘yes’ against 64 (57,1%) ‘no’.

Variables

Some of the variables in this research are measured by multiple items (or questions). Those variables are altruism and perceived quality. In the next sections, this research will go further in depth about the factor analysis of both variables.

Factor Analysis - Altruism

(21)

20

In order to judge whether the outcome of the factor analysis is appropriate, several criteria are used (Hair et al. 2006). First of all, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy needs to be higher than .60 and the Bartlett’s test of sphericity needs to be significant. For this research the KMO is .724 and the Bartlett’s test of sphericity is significant. That means that the first part is approved.

Secondly, all the communalities need to be .40 or higher. In the communalities table all the items are above .40 except the variable a peaceful world: free of war and conflict

(‘Een vreedzame wereld: vrij van oorlog en conflict’) which scores .382. According to this rule, this research should exclude this item from the factor analysis and do it all over again, but because it is nearly .40 and this research believes that this item is measuring a piece of altruism, it is not excluded.

Furthermore, to assign a certain item to an underlying factor, this item cannot have large cross-loadings. The item is excluded when the difference between the loadings on the several factors is below .30. In other words, when it is not clear to which factor the item belongs, the item is excluded from the factor analysis. This does not hold for the current factor analysis, because the differences between cross-loadings are not below .30.

The conclusion from the above stated criteria is that the factor analysis is appropriate. The next step is to see how many underlying factors the 16 items have. For the determination of this, a three step process is used (Field 2000; Rietveld & van Hout 1993). First, the eigenvalue of the factors must be larger than 1. Secondly, use the number of factors which account for minimal 60 percent of the variance. And thirdly, the number of factors before the breaking point or elbow of the scree-plot must be kept. In all the three steps, four factors are the result.

For this research only the factor or construct altruism is important for further analysis. When looking at the four factors, one factor clearly measures a sort of altruism. In table 2, on the next page, the items belonging to this construct are displayed. Also the communalities and the Cronbach’s alpha is shown. Cronbach’s alpha is used to measure the internal consistency of the factor or construct. Normally the Cronbach’s alpha needs to be higher than .60 (Malhotra 2007). Since this is .720, this research can state that the construct

(22)

21 Factor Analysis – Perceived Quality

Besides altruism, perceived quality is also measured by multiple items. Perceived quality is a very important dependent variable in this research. In the survey, respondents had to rate a pack of coffee beans on several quality measures. These consisted of taste (‘smaak’), price-quality ratio (‘prijs-kwaliteit verhouding’), and overall quality (‘algemene kwaliteit’). These quality measures are, firstly, analyzed in separate regression models, because these quality measures each measure a different aspect of quality. This research wants to see which aspect of (perceived) quality is mostly influenced by the independent variables. After analyzing these quality measures separately, this research also looks at the hypotheses when the three quality measures are combined to one factor of quality. This research chose this approach because the quality measures each measure a different aspect of quality and it is interesting to see which aspect is most influenced by the various independent variables.

Firstly, to analyze whether the three quality measures can be combined into one factor of quality, factor analysis is conducted. For this analysis the same criteria were used concerning validity and reliability of the factor. From the factor analysis, this research concludes that the three measures correlate enough with each other (KMO: .628, sig.: .000). Besides that, the factor is internally consistent with a Cronbach’s Alpha of .807. That means

Table 2: Factor Analysis Results

Construct Items Communalities Cronbach’s

Alpha (α) Altruism Equality: everybody has equal chances

(‘Gelijkheid: gelijke kansen voor iedereen’)

.743 .720

A peaceful world: free of war and conflict (‘Een vreedzame wereld: vrij van oorlog en conflict’)

.526

Social righteousness: restore injustice, care for the weak

(‘Sociale rechtvaardigheid: herstel van onrecht, zorg voor zwakken’)

.819

Helpfulness: care for the welbeing of others (‘Behulpzaamheid: werken voor het welzijn van anderen’)

(23)

22

that this research can use the combination of the three perceived quality measures, called

factor perceived quality, as one factor in subsequent analysis.

Statistical Measurement Techniques

The statistical analyses are carried out with SPSS (version 20.0.0.2, 2011). To see whether the expected relationships in the conceptual model are significant, this research uses linear regression. Although analysis of variance (ANOVA) also measures the difference that changes in an independent variable makes in a dependent variable, the null hypotheses are different. While regression focuses more on whether there is an effect at all, ANOVA focuses more on if the effect is different across categories. This research is more interested in relations (e.g. whether a fair trade label has an effect) than group differences. That is why regression is used to analyze the data.

Stepwise regression is used in which more and more variables will be added to the regression model. In this way, this research can control for the fact that the more independent variables in the model, the more the variance must be divided over the predicting variables. This could otherwise by a problem, since the more variables in a regression model, the smaller the chance that a single predicting variable is significant.

Subsequent analysis will, firstly, only conduct the direct effects on the criterion variable. In latter stages other variables and moderators are added to the model. The equations used in multiple regression analysis are stated in the following section: econometric models.

Econometric Models

(24)

23 (1) T = α + β1 FTL + β2 Pr

(2) T = α + β1 FTL + β2 Pr + β3 FTL*Pr

(3) T = α + β1 FTL + β2 Pr + β3 FTL*Pr + β4 Alt + β5 FTL*Alt + β6 Alt*FTL*Pr

(4) T = α + β1 FTL + β2 Pr + β3 Gender + β4 Age + β5 Education + β6 DrinkC + β7 BuyC + β8 Conscious

Whereas T = taste, FTL = fair trade label, Pr = price, Alt = altruism, DrinkC = frequency of drinking coffee, BuyC = frequency of buying fair trade products, Conscious = consciousness of buying fair trade products. Note that equation four is not testing any specific hypotheses, but is rather for possible side effect and perhaps possible future research.

Results

In this section, this research will describe and analyze the data from the online survey by looking at all the regression models.

Regression Results

Before discussing the results of the regression models, this research first test the degree of multicollinearity between an independent variable with the other variables. A common test for this is looking at the ‘Variance Inflation

Factor’ (VIF). Multicollinearity means that two or more predictor variables are highly correlated with each other. In that case the regression model can still indicate how well the entire bundle of independent variables predicts the dependent variable, but it may not give a valid result about any individual independent variable. According to Hair et al. (2006) is a VIF score of ten or higher

regarded as perfect multicollinearity and a level of 4 is seen as a moderate level of multicollinearity. From table 3 results that the predictor variables used in this research are

Table 3:

Variance Inflation Factor (VIF)

Variable VIF

Fair Trade Label 1.059

Price 1.128

Altruism 1.071

Gender 1.378

Age 1.757

Education 1.135

(25)

24

absent of perfect multicollinearity. Hence, this research concludes that multicollinearity does not exist in this research.

The next sections will discuss all the hypotheses for all three aspects of perceived quality (dependent variable) separately as well as for the combined factor perceived quality.

Perceived Quality – Taste

First of all, this research will look at the results for the aspect taste of perceived quality and will go through all the hypotheses. The regression results for taste are shown in table 4 on the next page.

Of all the three perceived quality measures is taste probably the hardest to answer for participants. This because the quality of taste is hard to extract from a picture in an online survey. This is also reflected in the regression results. To start with, none of the models for ‘taste’ are significant and also none of the hypotheses are near significance. Due to the fact that even the models as a whole are not significant, this research will not go further in depth about this aspect of quality.

Hence, consumers will not rate a product with a fair trade label to be better on taste than when the product has no label (H1). This holds for both altruistic as non-altruistic

consumers (H4). Also, a price discount will not lead to a lower perception of taste (H2), and

there is no effect of a fair trade label on this relationship (H3). For the latter, there is also no

(26)

25 Table 4:

Regression Results (standardized coefficients) for Perceived Quality Aspect Taste

Hypothesis (Effect) Model 1 Model 2 Model 3 Model 4

Main Variables Beta Beta Beta Beta

Fair Trade Label 1 (+) .117 .085 .096 .091

Price 2 (+) -.017 -.049 -.028 -.047 Control Variables Gender .091 Age -.236 Education Level .064 Frequency of Drinking Coffee .165 Frequency of Buying Fair Trade Products .142 Consciousness of Buying Fair Trade Products -.069 Moderators Altruism -.124 Interaction Effect Fair Trade Label

* Price 3 (−) .055 .056 Altruism * Fair Trade Label 4 (+) .048 Altruism * Fair Trade Label * Price 5 (+) .192 R2 _.014 _.015 _.046 _.111 (Adjusted R2₎ _(-.003) _(-.011) _(-.005) _(.041) F-value .815 .577 .896 1.601d

(27)

26 Perceived Quality – Price-Quality Ratio

While the aspect taste did not have a regression model that was significant, for price-quality ratio this is different. Table 5 shows the regression results for the perceived quality aspect

price-quality ratio.

First of all, when looking at the F-statistics, the predictors of price-quality ratio lead to a good fit of the models, namely, model 1 with F = 11.686, p < .01, model 2 with F = 7.870, p < .01, model 3 with F = 4.467, p < .01, and model 4 with F = 4.024, p < .01.

Model 1 explains 15.4% (adjusted r2_{) of the variance in price-quality ratio. H}

1 expects

a positive relationship between a fair trade label and perceived quality. This hypothesis is not supported in this model. That means that whether a product has a fair trade label or not does not influence the score on price-quality ratio. This does not imply that a fair trade label has no effect at all on perceived quality, just that the score on price-quality ratio is not different for a product with or without a fair trade label.

In H2 this research expects that a price discount will result in a lower score on

perceived quality. This hypothesis is supported for pricequality ratio with p < .01 (β = -.407). This means that the higher the price (= regular price), the lower the score on quality. As a result, a discount (lower price) will have a positive effect on the rating on price-quality. Since price is closely linked to price-quality ratio, this is a rather obvious result. In all the models of price-quality ratio is price significant at a level of p < .01.

The second model tests for H3, which expects that a fair trade label will

counter-balance the effect of a price discount on the perceived quality of a product. Like mentioned before, the model is significant and a price discount still has a significant effect on price-quality ratio, but the moderating effect of a fair trade label on the relationship between price discount and price-quality rating does not exist (p = .546, β = .091).

In the third model, this research tests all the hypotheses of the conceptual model at once. Also in this model, only price discount has a significant effect. That means that if we measure perceived quality by price-quality ratio, then altruism is not of importance. Next to that, altruism has also no significant moderating effect on, firstly, the relationship between a fair trade label and price-quality ratio (H4), and secondly no significant moderating effect on

the moderator fair trade label (H5).

In other words, none of the hypotheses are supported by the models whereby perceived quality is measured in terms of price-quality ratio. Although in H2, price discount is

(28)

27

for quality ratio the relationship is expected to be negative. This because, for price-quality ratio specifically, the higher the price, the lower the value per monetary unit.

Table 5:

Regression Results (standardized coefficients) for Perceived Quality Aspect Price-Quality Ratio Hypothesis (Effect) Model 1 Model 2 Model 3 Model 4

Main Variables Beta Beta Beta Beta

Fair Trade Label 1 (+) .054 .002 -.003 .031

Price 2 (+) -.407a _-.461a _-.453a _-.416a Control Variables Gender -.072 Age -.159 Education Level -.122 Frequency of Drinking Coffee -.035 Frequency of Buying Fair Trade Products .283b Consciousness of Buying Fair Trade Products .138 Moderators Altruism -.047 Interaction Effect Fair Trade Label

* Price 3 (−) .091 .111 Altruism * Fair Trade Label 4 (+) .133 Altruism * Fair Trade Label * Price 5 (+) .058 R2 _.169 _.172 _.194 _.238 (Adjusted R2₎ _(.154) _(.150) _(.151) _(.179) F-value 11.686a _7.870a _4.467a _4.024a

(29)

28 Perceived Quality – Overall Quality

Even though the perceived quality measures taste and price-quality ratio did not find support for the hypotheses, maybe for the measure overall quality this research can find supporting evidence.

The first model in the regression results (table 6) show only the direct effects of a fair trade label and a price discount on perceived quality. In this section we measure perceived quality as a measure of overall quality. This first model explains 8.0% of the total variance in overall quality. Although this is a rather small percentage, the model itself is significant (p < .01, F = 6.117). When looking at H1 and H2, only H1 is supported (p < .01, β = .291). That

means that if a product has a fair trade label on it, people rate the product higher on overall quality. H2 is not supported (p = .235, β = -.106). Even though the effect is not significant,

the relationship in the model is negative, while this research expected a positive relationship. That means that while this research expected that a price discount would result in a lower score on overall quality, there is a tendency of a reverse effect. A higher price has the tendency to lead to a lower score on overall quality. This is an interesting point for future research.

Model 2 tests for H3 which expects that a fair trade label will counter-balance the

effect of a price discount on the perceived quality of a product. This moderating effect is expected to be negative, since a fair trade label will weaken the effect of high price, high rating. Like mentioned before, the high price, high rating expectation does not hold. There is a tendency that the relationship is the other way around. That might also be the reason why, although not significant, the moderating effect of a fair trade label on the relationship of price and perceived quality has a positive beta (p = .454, β = .118). This because a fair trade label was believed to have a negative moderating effect and the positive beta of .118 shows a direction that strengthens the previously found tendency of a negative relationship of price and perceived quality. Because the effect is not significant, H3 is not supported for

overall quality.

When looking at the other hypotheses, H4, the moderating effect of altruism on the

relationship between a fair trade label and perceived quality, is also not supported (p = .757, β = -.058). It seems that altruistic characteristics do not play a role in the effectiveness of fair trade. Additionally, the three way interaction effect in H5 is also not significant (p = .105,

(30)

29

On the next page is the table which shows the regression results for overall quality (table 6).

Table 6:

Regression Results (standardized coefficients) for Perceived Quality Aspect ‘Overall Quality’

Main Variables Beta Beta Beta Beta

Fair Trade Label 1 (+) .291a _.223c _.233c _.262a

Price 2 (+) -.106 -.175 -.168 -.130 Control Variables Gender .074 Age -.261b Education Level .026 Frequency of Drinking Coffee .140 Frequency of Buying Fair Trade Products .234b Consciousness of Buying Fair Trade Products -.068 Moderators Altruism -.040 Interaction Effect

Fair Trade Label * Price 3 (−) .118 .132 Altruism * Fair Trade Label 4 (+) -.058 Altruism * Fair Trade Label * Price 5 (+) .258d R2 _.096 _.101 _.138 _.228 (Adjusted R2₎ _(.080) _(.077) _(.091) _(.168) F-value 6.117a _4.250a _2.951b _3.792a

(31)

30 Factor Perceived Quality

In the above analysis, this research discussed the three different quality measures separately. The results show that for taste, none of the hypotheses were significant, whereas for price-quality ratio, only price is a significant predictor. Furthermore, for overall quality a fair trade label has a significant influence on the rating of this quality measure. To make the analysis complete, this research also combines the three quality measures, and see whether this will come to the same results.

Like in the other regression results tables, the first model only tests for H1 and H2. In

this model, both a fair trade label (p = .040, β = .187) and price (p = .036, β = -.191) have a significant influence on the rating of factor perceived quality. For a fair trade label in the dataset, a zero means no fair trade label and 1 means that it has a fair trade label. Since the beta for fair trade label on factor perceived quality is positive and significant, the following conclusion can be made: when a product has a fair trade label on it, the higher the perception of quality will be. H1 is therefore supported.

For price in the dataset, a zero means a discount price and 1 means a regular price. Therefore, the negative, significant beta implies that the higher the price, the lower the perception of quality. In other words, a discount price will result in a higher perception of quality. Although H2 is therefore not supported, the reversed effect is significant. In the

discussion chapter, possible causes of this result will be further analyzed.

Furthermore, the standardized beta’s for both fair trade label and price are in model 1 almost the same, which shows that they both are equally important in explaining the variance in the dependent variable.

When analyzing the other models for factor perceived quality, it looks quite the same as the other tables where every quality aspect is analyzed separately. For H3 there is no

significant effect, which implies that a fair trade label does not moderate the relationship of price on perceived quality. This also holds for H4 and H5, which tests for the effects of

altruism. Altruism does not moderate the relationship of a fair trade label on perceived quality, and has also no moderating effect on the moderator fair trade label on the relationship between price and perceived quality. Thus, altruism is not important concerning the effects of a fair trade label.

(32)

31

trade habits. The latter consisted of the variables frequency of drinking coffee, frequency of buying fair trade products and consciousness of buying fair trade products.

When looking in table 7 at model 4, there are some control variables which have a significant direct effect on the rating of the perceived quality. In fact, age has a significant standardized beta of -.259 (p = .031). This implies that the higher the age, the lower the rating of perceived quality. When looking at a possible interaction effect of age with a fair trade label on perceived quality, this effect is not significant (p = .407). This means that there is only a direct, negative effect of age on the rating of (perceived) quality.

Secondly, the frequency of buying fair trade products is also a significant predictor of the rating of perceived quality (p = .028, β = .252). This means that the more frequent someone buys fair trade products, the higher the perception of quality will be. For this result also holds that it is a direct effect on perceived quality. It could be that, because fair trade products are linked with something positive, a person who has a lot of experience with these positive products will transfer this feeling to general products as well. Since this conclusion is based on subjective expectations, this is a topic for future research.

Below table 7, there is a table which summarizes the hypotheses for this research related to factor perceived quality (table 8). This table shows that only H1 is supported.

Although H2 is not supported, the reverse effect occurred which was significant. The other

(33)

32 Table 7:

Regression Results (standardized coefficients) for Factor Perceived Quality

Main Variables Beta Beta Beta Beta

Fair Trade Label 1 (+) .187b _.128 _.135 _.157c

Price 2 (+) -.191b _-.251c _-.238c _-.215b Control Variables Gender .043 Age -.259b Education Level -.005 Frequency of Drinking Coffee .113 Frequency of Buying Fair Trade Products .252b Consciousness of Buying Fair Trade Products -.009 Moderators Altruism -.082 Interaction Effect

Fair Trade Label * Price 3 (−) .102 .116 Altruism * Fair Trade Label 4 (+) .041 Altruism * Fair Trade Label * Price 5 (+) .205 R2 _.072 _.075 _.113 _.181 (Adjusted R2₎ _(.055) _(.051) _(.065) _(.117) F-value 4.434b _3.080b _2.348b _2.839a

(34)

33

Discussion

Summary of Results and Comparison with Theory

This research focused mainly on the effects of a fair trade label on the perception of the quality of a product. In that, the expectation was that a fair trade label would increase the perception of quality in the mind of the consumer. Besides that, this research also looked at the influence of price, and more specifically a price discount. From previous research the expectation was that a higher price would increase the perception or expectation of quality by consumers for the product at hand. Consequently, a discount price, as used in this research, was expected to lower the perception or expectations surrounding quality of a product.

In addition, this research also looked at the possible moderating effect of altruism. Since fair trade is about compassion and caring (Luthans and Youssef 2007), and altruism is about behaving with the concern of the welfare of others, there seems to be an obvious link

Table 8:

Overview Hypotheses According ‘Factor Perceived Quality’

Hypothesis Result

H1: Consumers will rate a product with a fair trade label higher on perceived

quality than when the product has no label.

Supported

H2: A price discount will lead to a lower perceived quality of the product. Reversed effect

H3: A fair trade label will counter-balance the effect of a price discount on the

perceived quality of a product.

Not Supported

H4: For more altruistic consumers, the effect of a fair trade label on perceived

quality of a product is stronger.

Not Supported

H5: For altruistic consumers, the positive effect of a fair trade label is

stronger than the negative effect of a price discount on perceived quality.

(35)

34

between them. That means that beforehand this research expected that more altruistic people would be more influenced by a fair trade label and would therefore strengthen the relationship between a fair trade label and perceived quality.

Furthermore, as mentioned before, a fair trade label is expected to have a positive effect on the rating of perceived quality and a price discount is expected to have a negative effect on the rating of perceived quality. In light of this, a fair trade label was believed to have a counter-balancing effect on the negative effect of a price discount. For altruistic people, the effect of a fair trade label was expected to have a bigger impact than the negative effect of a price discount.

All these expectations were tested in regression models. First of all, this research had several aspects of quality tested. These aspects were taste, price-quality ratio, and overall quality. All the hypotheses were tested on the three quality aspects separately as well as on the underlying quality construct, called factor perceived quality, were the three aspects are combined.

Firstly, all the hypotheses were tested for taste. None of the models or hypotheses were significant for taste. This could be due to the fact that taste is probably hard to rate for a participant only from a picture. Despite this, this research can conclude that the perception of taste is not different whether the product has a fair trade label or not. Another conclusion from the regression results for taste is that a price discount does not have an effect on the rating for the perception of taste. Additionally, the other hypotheses, relating to altruism, were also not significant and are therefore not good predictors of the perceived score on

taste.

When looking at other studies which also researched the relationship between a fair trade label and taste, the results are different. This difference may be due to the fact that this research did an online survey instead of a field study or laboratory experiment. Compared to the study of Lotz et al. (2013), where participants needed to taste coffee or chocolate in a field study, participants rated the product to taste better when it had a fair trade label. Also, in the article by Lotz et al. (2013) taste was not expected to be different by participants before they actually taste the product. However, their taste evaluations contradicted their beliefs. Therefore, this research has reason to believe that taste might have shown the expected results when participants actually had to taste the product.

Secondly, the same hypotheses were tested for price-quality ratio. In contrast to the

(36)

35

the price, the lower the price-quality ratio. This is obvious, because when the value of the product stays the same and the price goes up, the value per monetary unit decreases (Gavious and Lowengart 2011). The other hypotheses were not significant, meaning that

altruism does not play a role. Whether a person is altruistic or not does not affect the score on the perceived quality aspect price-quality ratio.

Thirdly, overall quality is tested as the dependent variable perceived quality. In this case, a fair trade label does have a significant effect on the rating of overall quality. This means that when a product has a fair trade label on it, the consumer expects a higher overall quality of the product. For price, there is no significant effect, but the sign is negative, which is not what was expected.

After analyzing the different aspects of quality, the combination of the three in the form of an underlying construct, called factor perceived quality, is tested. From the results, this research can conclude that a fair trade label as well as price is a significant predictor of perceived quality. In other words, a fair trade label on a product will enhance the expectations of the product surrounding quality in the mind of the consumer. The perception of quality is higher with a fair trade label than without a fair trade label.

Similar results were found by Lotz et al. (2013). In the three studies of that research they found that a fair trade label had a positive significant effect on taste. The results are restricted to taste, as are the results of the article by Poelman et al. (2008). Like mentioned before, participants in those studies only evaluated taste to be better when a fair trade label was on it after they actually tasted the product. Therefore, this research adds the effect of a fair trade label to different quality measures (e.g. price-quality ratio, overall quality) to current knowledge. It seems that before tasting the product, taste is not expected to be different but the overall quality of the product is. Namely, this research found that a fair trade label enhances the perception of overall quality.

Furthermore, while the research of Lotz et al. (2013) only used undergraduates from a German university with a mean age of 23 years, this current research is represented by more layers of society. Therefore, this research shows that the effects of a fair trade label are also generalizable to other groups of society.

(37)

36

Whereas previous literature mentions a positive relationship between price and quality (Combris 1997; Grewal et al. (1998); Dodson et al. (1978)), in this research the reverse effect is present. At first sight this seems surprising, but when looking more closely at the determinants of this result, there is a possible explanation why this differs from previous literature. Namely, one of the three quality measures is price-quality ratio. In this measure, price was strongly negatively related, while other items did not show significant effects. As a result, the total factor perceived quality, including price-quality ratio, is also negatively related to price. Although this research believes that this is the reason for the negative relationship of price and perceived quality, the results of the quality measure overall quality also showed this sign in a weak, non-significant manner. Therefore, this can still be an interesting field of future research.

Besides the regular hypotheses, this research also tested a model with control variables. These control variables consisted of general social-demographic variables as gender, age and education level, and of variables concerning coffee c.q. fair trade habits. The latter consisted of the variables frequency of drinking coffee, frequency of buying fair trade products and consciousness of buying fair trade products. From these control variables, only age and frequency of buying fair trade product had a significant, direct effect on the rating of quality.

For age the direction of the effect was negative, meaning that the higher the age the lower the score for quality in general. Younger people expect a product, before they buy it, to have a higher quality than older people expect. Age has in this research a mean of 36,14 and a standard deviation of 12,8. The youngest participant was 20 years old and the oldest 63 years. That means that the sample consisted of participant who had a relatively large difference in age. It could be that younger people are, in general, more influenced by the looks of a product when making an expectation about the quality of a product. Older people could be more restrained, and might therefore have lower expectations before having experiences with a product. However, since age has a negative effect on perceived quality, it would be interesting to see whether the same result would be found when older people than 63 were included.

(38)

37

In comparison to theory, this research confirms the effect of a fair trade label on the perception of quality. Whereas the effect of price is described differently in literature, this research found a negative effect of price on the rating of quality. Furthermore, this research analyzed the role of altruism on this topic. This contributes to current knowledge, because the recent article by Lotz et al. (2013) thought that the positive effect of a fair trade label is more than a halo-effect based on automatic cognitive heuristics, in fact, they believe that it is also based on emotional processes. The same article mentions therefore that future research should concentrate on linking specific personality traits to the effects of fair trade.

According to this research, the personality trait of altruism is closely linked to fair trade (Poelman et al. (2008); Luthans and Youssef (2007); Ferran and Grunert (2007)). The results show that altruism has no significant effect on the relationship of a fair trade label and perceived quality. Therefore, the positive effect of a fair trade label is not different for people with an altruistic personality. It could be that altruistic people do not expect a fair trade product to be better than regular products and that the buying of these products is merely a form of pro-social behavior. That would explain the fact that an altruistic person does not rate a product with a fair trade label to be higher on quality than a non-altruistic person.

Practical Implications

The above stated results can be important for selling fair trade products. First of all, a fair trade label results in a higher perception of quality. Therefore, it is important, when designing packaging for a product that the fair trade label is easily recognizable and placed in a manner that the label is prominently present. In that way, the consumer can easily see that the product at hand is made according fair trade production.

The second result which can be important for fair trade organizations is the price. This research has found that a price discount is linked to higher perceptions of quality than a regular price. Even though there is an influence of the quality measure price-quality ratio on the negative sign of price, the aspect overall quality shows, although not significant, the same negative sign. This could mean that displaying the price as a discount will result in higher sales of fair trade products. This implication needs to be further analyzed since it is not clear to what extent this result is due to the negative relationship between price and price-quality ratio.

The influence of a fair trade label on the perception of quality

The influence of a fair trade label on

the perception of quality

Master Thesis Marketing

Johan J. Walda

The influence of a fair trade label on the perception of quality

Management Summary

Preface

Table of Contents

Introduction

Theory and Hypotheses

Research design

Results

Discussion