品牌, 原產國, 與知覺風險對消費者選擇行為的影響

全文

(1)國立臺灣師範大學管理學院管理研究所碩士論文 Graduate Institute of Management College of Management National Taiwan Normal University Master Thesis. 品牌, 原產國, 與知覺風險對消費者選擇行為的影響 The Role of Brand, COO, and Perceived Risk in Consumer Choice Behavior. 林佳芮 Carina Linortner. 指導教授：沈永正博士 Advisor：Yung-Cheng Shen Ph.D.. 中華民國 104 年 7 月 July, 2015.

(2) Abstract. Country of Origin has been a frequently discussed topic in marketing literature. While some researchers are convinced of country of origin being salient, others believe that it does not matter anymore where the product comes from in our modern and globalized world. This thesis, as one of only a few, takes into account consumers’ personal traits, namely loyalty and need for authenticity and the situational variable of a purchase’s perceived risk. It is shown that risk and gender directly influence the relative importance of brand vs. country of origin when making a choice between two products. Females are more likely to choose a product according to the better country than men. It is also more likely to be more influenced by country than by brand when the product bears risk. Effects on choice of two-way interactions between loyalty and risk, need for authenticity and gender, risk and gender, as well as of a need for authenticity, risk and gender three-way interaction have been detected.. Keywords: Brand, COO, Perceived Risk, Choice.

(3) 中文摘要. 來源國的研究是行銷領域的熱門研究題目之一。許多學者已證實了來源國效應存在於消費者行為中；但有些學者認為，在這個全球化的世界中，產品來源國並不會影響消費者的選擇。作為目前為數不多的研究之一，本研究將消費者個人特質，即品牌忠誠度、對於產品道地程度的需求以及情境變數——知覺風險度做為探討之重點。研究結果證實，當進行產品選擇時，性別和知覺風險會直接影響消費者對品牌或來源國的重視程度。具體來說，（1）女性比男性受到產品產地的影響相較於品牌之影響會更大。（2）當產品知覺風險越高時，產品產地的影響相較於品牌之影響會更大。品牌忠誠度與風險，對於產品道地程度的需求與性別，知覺風險與性別，這三組變數之間分別有交互作用。另外，對於道地程度的需求、知覺風險和性別這三個變數有交互作用。. 關鍵字: 品牌, 原產國, 知覺風險, 選擇.

(4) Table of Contents. 1. Chapter 1 - Introduction 1.1 Research Motivation 2. Chapter 2 – Literature Review 2.1 Consumer Inference 2.1.1. Diagnosticity and Accessibility. 3 6 8 9 9. 2.2 Brand Attitude. 11. 2.3 Country of Origin. 11. 2.4 Involvement and Loyalty. 16. 2.5 Need for Authenticity. 18. 2.6 Risk. 19. 2.7 Gender. 20. 3. Chapter 3 - Method. 21. 3.1 Development of Hypotheses. 21. 3.2 Experimental Design. 27. 3.3 Measurement. 29. 3.3.1 Loyalty Measurement. 30. 3.3.2 Need for Authenticity Measurement. 30. 3.3.3 Manipulation Check: Risk Measurement. 31. 3.4 Procedure 3.4.1 Pretest. 32 32. 3.4.1.1 Participants. 33. 3.4.1.2 Data Collection. 33. 3.4.1.3 Results. 35. 3.4.2 Main Questionnaire. 38. 3.4.2.1 Participants. 38. 3.4.2.2 Data Collection. 39. 3.4.2.3 Data Processing. 41. 4. Chapter 4 – Results 4.1 Reliability Testing. 41 41. 4.1.1 Cronbach’s Alpha: Loyalty. 41. 4.1.2 Cronbach’s Alpha: NFA. 42. 4.1.3 Cronbach’s Alpha: Risk. 42. 4.2 Manipulation Check Risk. 42.

(5) 4.3 Effect of Demographics. 44. 4.4 Hypothesis Testing. 44. 4.4.1 H1. 50. 4.4.2 H2. 50. 4.4.3 H3. 51. 4.4.4 H4. 51. 4.4.5 H5. 52. 4.4.6 H6a and H6b. 52. 4.4.7 H7a and H7b. 53. 4.4.8 H8. 53. 4.4.9 H9. 54. 5. Chapter 5 – Discussion and Conclusion. 55. 5.1 Summary of Findings. 55. 5.2 Academic and Managerial Implications. 56. 5.3 Limitations and Future Directions. 57. References. 60. Appendix. 67. 2.

(6) Chapter 1. Introduction. While some decades ago it was not yet necessary to think about the origin of a product when purchasing it, just because of the simple reason that everything was produced locally, maybe even at factories located in one’s own neighborhood, it has become increasingly important to consumers to know where their products are produced. Nowadays it does not really matter which product category we are speaking about, nearly everything is produced globally, transported throughout various countries until the products reach their final destination. Although it were companies which sold rather simple products like clothing which had been pioneers of global production in the past, there already is no limitation to which products can be produced abroad anymore. Starting from different kinds of foods and fruit provided in our supermarkets, up to high-tech products like mobile phones or laptops, we cannot be sure about the origin of any of these goods without taking a look at their country of origin label. Gaining knowledge about where one’s purchased products were manufactured can have different implications to different consumers. Some of us may prefer locally produced products, others may specially look out for imports because of their curiosity for foreign or exotic things and others may simply not care at all about country of origin. Regardless of consumers’ personal preferences it has become a common procedure, or rather has it become required by law, to inform people about a product’s origin. A company that wants to export its goods go the U.S. for example needs to make sure that its goods are marked with the country of origin. This marking must be legible and permanent enough for the ultimate purchaser to be made aware of the goods origin. The ultimate purchaser is defined as the person who will last purchase or receive the article in the condition which it was imported. Although there are exceptions to this 3.

(7) requirement, like goods that are incapable of being marked (i.e. fruit), also these products have to be labeled in a certain way. In the case of fruits it is common to mark the outer container. (U.S. Customs and Boarder Protection 2015). Globalization, a keyword that has become increasingly popular with plenty of authors of different research areas publishing articles about this topic and its effects in recent years, is not only happening in academic research or on manager level, but affects the everyday life of consumers. Globalization can be further categorized into political, economic, cultural and technical globalization, with either one of them affecting our everyday life. May it be a foreign minister visiting our country and therefore having roads blocked, that we usually take to get to work, or the internet, which connects people from all over the world, provides unlimited information and possibilities to network with people located thousands of kilometers away, both are effects of globalization that have changed our lives. Speaking about economy, the opening of markets and global trade has brought prosperity to many countries and people who know how to utilize their positions on this international economic platform. Country of origin literature has developed different views regarding the topic of globalization. It has been a hot topic of discussion, whether consumers have become indifferent to country of origin, because they are used to their products being produced far away already, or whether they take notice of the origin label even more now, because there are so many possibilities where their product could be from, so that one has to take an even closer look. While Johannson et. al (1985) state that country of origin of a product is not an important consideration, others disagree and find a significant influence of origin on product evaluation. (Nagashima 1970) Tse and Gorn (1993) find that even in the context of globalization the construct of country of origin did not lose on importance and suggest to managers to follow a strategy of developing new products made in 4.

(8) favorable (vs. unfavorable) countries.. On the one hand, it is possible that consumers are so globalized, that it is not a big deal for them to buy a product from Russia, because the perceived distance to Russia has become smaller through modern ways of communication and transportation, on the other hand there may be people deliberately avoiding products from certain countries, because they have seen negative information about these countries online or on television. Stereotypes have existed at all times, but the environment of information overload that we live in encourages the creation of stereotypes even more. People create stereotypes in order to cope with the huge amount of information we need to deal with every day, because it is easier for our brain to categorize certain information into groups. Just as there are stereotypes associated with people, also countries can be stereotyped. Stereotypes of countries are an important topic in country of origin literature and need to be considered additionally to considerations of a country’s economic situation or cultural background. Typical stereotypes enrooted in consumers’ minds include German products to be technically advanced or Chinese products to be imitations. Similar to stereotypes or imaginations about various countries, consumers are also trained to have certain images of brands in mind. In todays globalized world there are several well established brands that are well known all over the globe. Through intense marketing these brands were successful to let millions of people have the same image in their head when thinking of them. It does not matter whether you ask an American or a Taiwanese person, as a result of continuous marketing activities, either one will recognize the logo of Coca Cola or know what a LV purse looks like. Some people are strongly receptive to these marketing strategies, others less. Whereas there are brand loyal consumers who like to purchase the same brand repeatedly, there are others who care less about brand and choose a different brand in different situations. 5.

(9) People make multiple purchase decisions several times a day. May this purchase be made when buying lunch at a restaurant, may it be spending money for parking or just having a coffee in the afternoon. All these small purchases have become part of our everyday life and nobody spends too much time thinking about them, let out spending effort on information search. Usually these everyday expenses do not involve any risk, if they did, consumers would probably not make them this randomly and imprudently. When risk is high, consumers will spend significant effort on information search, as they fear the consequences of a wrong decision. Hence, the process of preparing to purchase a coffee differs significantly from what is going on when someone is planning to buy a house. Information search does not only depend on risk involved but also on consumer’s involvement itself, meaning how relevant a product purchase is to the consumer. Relevance may vary with situation or product category, therefore different product categories will be investigated in this thesis. Given a situation in which information is provided to the consumer about a product, it, among other factors, depends on the product’s relevance to the person, whether the consumer considers the given information for his or her purchase decision or chooses to neglect it.. 1.1. Research Motivation. Country of origin literature historically follows an urge to make COO effects more generalizable. Common practice is to evaluate countries on scales from unfavorable to favorable and to project this favorable or unfavorable image onto the product. Thakor and Kohli investigated the influence of country of origin on brand equity. Their work displays a fundamental difference between country of origin and brand origin, with COO seeing a country as a major exogenous variable that leads to analysis being 6.

(10) conducted at product, rather than brand level. (Thakor, Kohli 1996) Also Nagashima (1970, 1977) followed the scheme of letting consumers rate countries with the help of attributes like reliableness, workmanship or price level and Johansson and Nebenzahl tried to find a strategy, that helps managers determine which country of origin supports their brand best when planning multinational production. (Johansson, Nebenzahl 1986) Although authors like Lantz and Loeb (1996) or Zhang (1997) have taken steps to expand COO research to the individual level, there is still a gap to fill regarding varying COO effects when dealing with different personalities. Lantz and Loeb acknowledge that national identity, although being a salient aspect of the self, varies between individuals and hence influences COO effects. Zhang conducted research on individual differences in information processing, stating that factors operating at the individual’s psychological level provide better understanding of how consumers process country of origin information. According to Zhang the salience of COO information is affected by the individual consumer’s level of need for cognition. Consumers who are intrinsically interested in analyzing and processing information are less likely to be influenced by COO information, but will rather base their decision on the relevance and strength of product attribute arguments. Consumers, who enjoy the outcome rather than the process of thinking, will tend to base their decisions on more apparent characteristics, such as the favorableness of COO.. It varies from individual to individual, how important the fact is that a certain product is made in Germany and even if two individuals agree, that a German product is a good one there can still be several different reasons for such a mindset. It may be possible for someone to believe that the product is favorable because of Germany’s technical advance or maybe they perceive a “Made in Germany” label as a status symbol. Another reason might be, similar to a situation where a dress becomes very valuable just because 7.

(11) a famous actor or singer has worn it, a feeling of transferred essence, so that people feel that it is special to hold a product in hands, which came out of Germany. Although there are a few established brands who are able to use their brand equity to weaken or eliminate negative effects of unfavorable country of origins, the influence of a strong brand varies from person to person. Investigating a topic that intends to better understand or predict consumer preferences or choices, it is important to consider consumer needs that may either be attributed to the individual itself or to the situation in which a purchase is made. Therefore I will take a closer look at two constructs, namely loyalty and need for authenticity, which I believe to be personality features that influence COO effects. On the situational level, this thesis distinguishes between two different conditions. The first one considers consumers’ mindsets when purchasing low risk products. The second one investigates the decision process when dealing with products that bear risk. Research, which links consumer’s need for authenticity with strength of COO effect and loyalty with a brand’s ability to weaken the influence of an unfavorable COO is unprecedented in country of origin literature and will therefore increase our understanding of the limited generalizability of COO effects.. Chapter 2. Literature Review. In this chapter I will present an overview of diverse literature that I consulted or took into consideration when writing this thesis. The first part provides general information about consumer inference, as this thesis investigates processes of decision making, so that it is of help to get basic understanding of inference making and why certain information is accessible or diagnostic, while other is neglected. The second part 8.

(12) introduces measurement for brand attitude and work in the area of country of origin. Finding out about the relative importance of brand and country of origin in product choice of consumers will be related to need for authenticity, loyalty, risk and gender in this thesis. Finally, these independent variables, need for authenticity and loyalty, as well as risk, are introduced and hypotheses developed.. 2.1 Consumer Inference. Consumer inference describes the behavior of consumers, who form inferences beyond the information given about a product. Inferences can be inductive, that is inferences from specific instances to general principles, or deductive, that is from general principles to specific instances. Further categorization is possible into stimulus-based vs. memory-based and the judgement can include single as well as multiple products. (Kardes et al. 2004). 2.1.1 Diagnosticity and Accessibility. Diagnositcity and accessibility will influence the input that is considered in making a decision. It is essential to understand which factors influence the accessibility or retrievability of information, as this affects decision making. It needs to be differentiated between availability and accessibility, whereas information is available after it has been fully comprehended and encoded into long-term memory. However, only a very small part of available information is accessible at any given time. Certain 9.

(13) pieces of information can be accessible in one situation, but not in another because people are only capable of retrieving a small fraction of all available information. (Lynch, Srull 1982) Accessibility is defined by Lynch as the ease with which an input in memory is brought to mind. Diagnosticity is task specific and describes in how far consumers believe that one input alone can accomplish their decision goal. The likelihood of an input to be used for decision making is a positive function of the accessibility of the input, a positive function of the perceived diagnosticity of the input for the target decision and an inverse function of accessibility and diagnosticity for competing inputs. When consumers have different goals in mind, different items will be more diagnostic. When motivation is low, Lynch suggests that inputs will be used independently from their diagnosticity and that only high motivated consumers will spend effort to discount non-diagnostic information. (Lynch 2005) Diagnosticity is further affected by whether consumers engage in integrative or differentiative processing and identity consistent information is more relevant than identity inconsistent information. (Khare, Zhang 2009) Regarding the competition of brands, Keller writes that an important aspect of brands is their overall valence, calculated in terms of consumers’ overall ad and brand evaluations. Brands, who are equal in quality, will be strongly associated with each other in memory and simultaneously identified as high- or low quality brands. (Keller 1991) External information is used as retrieval cue, which is stored in the memory trace together with the to-be-remembered information. (Craik. 1979) Memory accessibility is a function dependent on the compatibility. between trace and cue information. (Tulving. 1974,. 1979). Keller mentions that. also factors like product category itself can serve as retrieval cue.. 10.

(14) 2.2 Brand Attitude. Knowing that a product one is thinking about to purchase is part of a certain brand signals the consumer information about expertise, trustworthiness, quality, risk and lets the consumer save information cost. Especially when consumer uncertainty is high or information is costly to obtain, brand credibility is an important factor to choice. (Tülin, Swait 2004) Consumers’ brand attitudes towards various brands along several different product categories had to be obtained for this thesis. Helpful suggestions were taken from Burton et al. (1998), who developed a scale to measure consumers’ attitude towards private label brands. Burton et al.’s scale includes a reverse coded question saying “In general, private label brands are poor quality products.”, which was changed to “This brand stands for quality” for the purpose of this thesis. Also did I make use of the question “Buying private label brands makes me feel good.”. which I modified into “I would feel good buying this brand’s product.”.. 2.3 Country of Origin. There is a huge amount of literature available in the area of country of origin (COO) research. Some authors even see COO as a multilevel construct (Brookshire 2012), made up of the individual components country of design (COD), country of parts (COP), and country of manufacturing (COM) (Chao, 2001; Essoussi & Merunka, 2007; Insch & McBride, 1998; Iyer & Kalita, 1997; Veale & Quester, 2009) On the actual marketplace however, consumers will only rarely see a detailed labeling like this. Commonly the country in which manufacturing took place is considered as 11.

(15) COO and displayed on the product. In most cases consumers are not informed about the source of raw materials or manufacturing sites that are used before the final step in a production process, and a single-country COO display may not be ideal to represent true COO. (Brookshire 2012). In this thesis I use the term “country of origin” referring to the country of manufacturing mentioned on a product’s label. It is further assumed, that the country of origin on the product’s label is the only manufacturing country the product was in contact with. The term country of origin needs to be distinguished for country of brand origin. Country of manufacturing gives information about where the product is produced, and, for simplicity, is the only country mentioned on the product’s “made-in”-label. Thakor (1996) defines the brand origin as “the place, region or country to which the brand is perceived to belong by its target consumers.” This place can differ from where the products are manufactured, or from where consumers believe that a brand’s products are manufactured. The term of brand origin is strongly connected to consumer perceptions. Therefore a brand’s country of origin is not necessarily the actual origin, but rather what is believed by consumer’s where the brand is from. A difference of actual and perceived brand origin can occur because of consumers’ ignorance, lack of information or because a brand wants to cover up an unfavorable origin. (Thakor 1996). Jashim et al. intend to explain country of brand origin and country of manufacturing image in their publication from 2013. Two regression models were tested in an attempt to identify factors that influence the significance of country of brand origin and country of manufacturing in consumer evaluation. They found that determinants that significantly influence country of brand include technological superiority, price level, reliability, performance and durability. The significance of country of manufacturing in 12.

(16) consumer evaluation is further influenced by price level, value for money and aesthetical design. (Jashim et al. 2013). Because globalization not only creates stronger competition for companies, but also more options to cut down cost through producing in countries, where raw materials or labor are available for a lower price, it becomes more and more seldom, that a certain product is manufactured and sold in one single country. Simultaneously more information about the various countries, their economic, political or social situation is available to consumers now than it was a decade ago. In these times it is very easy for consumers to get an idea about potential differences in product features, like product quality, when a product is produced in country A compared to another product that is made in country B. Additionally to their understanding of available country information comes the individual’s purely affective behavior, which is influenced by country stereotype. (Johansson, Nebenzahl 1986) Nagashima was one of the first authors, who did in depth research on consumer attitude in combination with country of origin stereotype. In 1970, Nagashima published results of his research about consumer attitudes towards products from different countries, for which he had conducted surveys among American and Japanese Businessmen, asking them about their attitudes towards American, Japanese, British, German, Italian and French products. Nagashima defines the “made-in” image as the picture, the reputation, the stereotype that businessmen and consumers attach to products of a specific country. Nagashima had products of different countries rated on five dimensions: Price & Value, Service & Engineering, Advertising & Reputation, Design & Style and Consumers’ Profile. (Nagashima 1977) His results showed that “Made in USA” was associated with automobiles, food, electronic products, computers, machinery and airplanes by Japanese businessmen. On 13.

(17) the other hand did American participants of the survey associate the “Made in Japan” label with cameras, electronic products, transistors, watches and ship building. Among all foreign labels that appeared in the survey, “Made in Germany” was found to be appreciated the most. (Nagashima 1970) From Nagashima’s work one can see that certain countries implicate certain product categories and one might infer, that a product category might also implicate a country, just as consumers might be provoked to think of France when evaluating a luxury perfume. This perceived match of a country and a product category further influences the consumer’s reaction, when weighing up the influence of a made-in label.. While some authors in the past are of the opinion that country of origin research loses relevance in the environment of nowadays’ globalization (Usunier 2012), Samiee states that there are also certain because of. issues concerning COO research, that only emerge just. globalization. Because of international production all around the globe the. sources of raw materials and components are blurred and products of international brands can be either sourced from their originating countries, locally manufactured or imported from third countries, with each of these option signifying a unique proposition for consumers. With respect to results of various researchers that reveal an existing preference among consumers towards products of a brand sourced from its brand origin, over products manufactured in a third country, Samiee supposes that this effect only occurs because of faulty brand marketing, that fails to educate consumers, or poor training of a brand’s global distribution channel. (Samiee 2010). Also Smith and Su refer to mixed results in country of origin and globalization literature. They conclude that this is due to different levels in the mind-sets among consumers. Smith and Su introduce the notion of “multi-layered consciousness”, which explains 14.

(18) the phenomenon of consumers being highly globalized at the economic level, while still being localized on the psychological level. Consumers therefore might perceive themselves as globalized, although still being prejudiced against foreign products. In collectivist cultures in-group and out-group biases may also strongly effect COO evaluations. Their work focuses on the dynamic nature of the antecedents to COO image, which may embrace notions like multiple stages of globalization and the dynamics between a nation’s culture and its globalization stage. (Smith, Su 2008) Country of origin is therefore seen to be subject to several culturally derived antecedents. These antecedents can be further categorized into endogenous vs. exogenous and cognitive antecedents that describe an individual’s information processing.. COO evaluation moderators include both product-based, as well as. individual consumer factors. (Pharr 2005) Country of origin was found to have a significant impact on purchase intentions when the product was a luxury product designated to be consumed in public. (Piron 2000). Finally also brand characteristics function as a moderator of COO effects, as brand image, for example, is known to act as a “protective shield” against negative COO evaluations. (Jo et al. 2003). While COO research underwent a shift from single cue studies to more complex investigations in the early period from 1965 to 1982, the increased interest spent on a reconceptualization of COO in terms of brand origin, product-country image and product-place image is a research trend particularly assigned to the timeframe 1993 to 2004 by Dinnie. It is also important to notice, that there has been a change over time and that nowadays COO effect is not exclusively related to tangible products anymore, but is becoming increasingly relevant in the service sector. (Dinnie 2003). Tse and Gorn find COO still being salient in the era of global brands. In a 2 by 2 design 15.

(19) they combined a positive vs. negative country of origin with an internationally known vs. unknown brand name. Evaluations were obtained before and after individuals tried the product. Tse and Gorn were not able to retrace evidence that a well-known global brand overrides the COO effect, which is a notion that runs like a common thread through post-industrial COO literature. Further they found that COO effect is salient, although reduced, even after experiencing a product. (Gorn, Tse 1993). 2.4 Involvement and Loyalty. Knowing that the amount of effort one is willing to spend influences the occurrence of assimilation or contrast effects, which again influence consumer choice, it is obligatory to consider other factors that might affect a person’s willingness or ability to spend effort. People with high involvement, for example, may have sufficient expertise and knowledge about a product category or it may be extremely important to them that they make the “right choice”, so that it is suggested that high involvement customers are willing to spend more cognitive effort. Antecedents of involvement can be categorized into three groups. These are the characteristics of a person, physical characteristics and varying situational factors. Characteristics of a person includes a person’s value system combined with personal experience. Physical characteristics refers to differences in type of product category or media channel used for advertising. Varying situational factors describe one’s personal state of need. If someone is planning to buy a car in near future, his or her involvement when watching a car ad will be higher than the involvement of a person, who is not planning to buy the product at that point of time. Involvement describes whether personal relevance exists. When a certain purchase consumption situation is personally relevant or important and when the value of the 16.

(20) product to the consumer is an essential element, then that purchase decision will be made very carefully. When a purchase decision is important for consumers, they are willing to expend effort to obtain information and reduce uncertainty. (Zaichkowsky 1986). Brand loyalty is explained by Tucker as being a biased response to some combinations of characteristics, not all of which are critical stimuli. Therefore one may also suspect loyalty to be a factor that needs to be considered in this thesis. Tucker chooses a quite simple approach to loyalty and states that brand loyalty is biased choice behavior with respect to branded merchandise. The degree of loyalty can be easily calculated via the frequency one choses a certain brand over another. (Tucker 1964) Literature provides many different approaches on how to operationally define brand loyalty, also different “sequence” definition, as that from Tucker, are available. Because of this big amount of definitions it is difficult and not commendable to compare and accumulate findings of different authors. Jakoby and Kyner believe that a single unidimensional measure is insufficient for measuring a complex multidimensional phenomenon like brand loyalty and they sadly detect that, although lots of literature has been written about brand loyalty, there a conceptual definition is still missing. Plain frequency cannot be used to speak of brand loyalty as there must be made a difference whether a consumer always buys brand A because it is the cheapest, whether another one buys brand A in every purchase situation because he or she simply prefers it, or whether a person keeps buying brand A just because this brand is the only available brand in the area. Brand loyalty therefore has to be differentiated from repeat purchase behavior. For this reason the authors developed a definition that is expressed by six necessary conditions. Brand loyalty therefore is defined as the biased (nonrandom), behavioral response (purchase), expressed over time, by some decision-making unit, 17.

(21) with respect to one or more alternative brands out of a set of such brands and is a function of psychological (decision making, evaluative) processes. (Jakoby, Kyner 1973). 2.5 Need for Authenticity. It is well established in COO literature that differences in manufacturing location can affect consumer preferences through lay inferences about production quality. These COO research findings can be paired with two other literature streams, namely authenticity search, which found that consumers prioritize connections with the source, and contagion research, which found that beliefs in essence transfer can play a role in the decision making progress. Dhar and Newman propose that beliefs in contagion influence authenticity perceptions of everyday and branded products and that it lets products from a company’s original factory be preferred over identical products made elsewhere. Belief in contagion is hence defined as the notion that objects may acquire a special aura or “essence” from their past. According to the authors consumers can differ in their sensitivity to contagion, with highly sensitive individuals preferring products made in the original place more strongly. (Dhar, Newman 2014). Although a few researchers see authenticity as goal dependent, Arnould and Price believe that authenticity is used in order to achieve self-authentication by the consumer co-creating product value and having a feeling that the consumption experience itself is part of ones self-authentication, or through a second mean, which they refer to as authoritative performance. Authoritative performance describes consumers using products as a display for their social units, like family, profession or ethnic group, or 18.

(22) for other important aspects of their lifes. Generally speaking, Arnould and Price find that through authenticity the consumer is purposefully linking the product to stories of the self. (Arnould, Price 2000) According to Beverland authenticity is a socially constructed interpretation of the essence of what it observed, and not directly linked to the actual properties of the object. (Beverland 2006, Beverland et al. 2008) Lewis and Bridger indicate in their research that need for authenticity is person dependent, rather than goal dependent, as they state that consumers with a need for authenticity have different consumption values and criteria than consumers without the same need. (Lewis, Bridger 2000) Another clue, that it is appropriate to view the authenticity variable in this thesis as person dependent comes from McNamara, who states that the attention given to authenticity is directly related to the consumers seeking happiness and personal meaning. (McNamara 1997). 2.6 Risk. Prior to this thesis risk has already been used as a moderator in literature, for instance in the work of Gürhan-Canli and Batra, who found risk to moderate the influence of two corporate image associations, innovation and trustworthiness, on product evaluations with the associations’ influence being stronger when consumers perceive high risk. Consumers are more likely to be concerned about the degree to which the product will perform as expected as well as seek and use a greater amount of and more various sources of information for research when perceived risk is high. (Gürhan-Canli, Batra 2004). Risk is defined in terms of the consumer’s perceptions of the uncertainty and adverse 19.

(23) consequences of buying a product (or service). Perceived risk is therefore being thought about in terms of the magnitude of consequences and the probabilities of these consequences to occur. Dowlong and Staelin’s (1994) findings are consistent with what has been supposed in most literature, namely that perceived risk affects information search behavior with more search activity being conducted in high risk categories.. Jacoby and Kaplan distinguish between five kinds of risk. These are financial, performance, physical, psychological and social risk, whereas psychological risk refers to the fear that a certain product may not fit well with one’s self-image and social risk meaning the risk of a product negatively influencing the way other people think of you. (Jacoby, Kaplan 1972) Roselius further introduced a sixth kind of risk. Risk of time loss exists when products fail and the consumer needs to waste time, convenience and effort in order to get it adjusted, repaired or replaced. (Roselius 1971). 2.7 Gender. Literature has found women to be more risk averse than men. As risk is expected to directly influence the relative importance of brand versus COO in choice, women and men are supposed to differ in their choices, as they behave differently at the same level of perceived risk. Hartog et al. (2002) were able to support the general belief that women are more risk averse and showed further that this had nothing to do with women having no income of their own. While the study of Hartog et al. was about a lottery situation, also Agnew et al. (2008) came to the same result, which is that women are more risk averse than men, analyzing the two genders’ behavior when purchasing annuities. The same results could further be found, regardless whether the sample group 20.

(24) consisted of non-student adults, as it was the case in Agnes et al.’s study, or of 15- to 16 – year old students, which were used as a sample by Borghans et al. (2009). Dwyer et al. (2001) analyzed the likelihood of risk-taking in fund investment decisions and find an impact of gender on risk taking, although this impact can be partly explained by knowledge disparities between the two genders.. Chapter 3. Method. 3.1 Development of Hypotheses. Consumers who are loyal to a certain brand, will make biased decisions when having the choice between “their” brand and one or more others. Because of their commitment to a brand, they do not only select that certain brand, but also “select out” others. (Jakoby, Kyner 1973) I suppose that the context of being produced in a favorable country of origin is not given attention to in this process of selecting out. Therefore high loyalty and low need for authenticity consumers will choose the better brand, not depending on the country of origin the brand is paired with.. H1: A consumer’s level of loyalty has a significant influence on the relative importance of brand vs. country of origin in product choice, with more loyal consumers tending to put more importance on brand.. H2: Product choice of high loyalty and low need for authenticity consumers is more influenced by brand than by country of origin. 21.

(25) In today’s time of globalization it is common for companies to manufacture their products in other places or countries different from the company’s origin or where they are headquartered. However, literature has found that the physical connection to a company’s or brand’s origin can be appealing to consumers, because consumers perceive products manufactured at the origin to embody a certain “core” or “essence” of the brand. The fact that consumers prefer products coming from a company’s or brand’s origin may also be influenced by contagion, that is a belief in transferred essence, that makes people perceive products from the original place as more authentic and valuable. This effect of preference varies with the level of sensitivity to contagion. (Dhar, Newman 2014) Authenticity is sought for in order to find the meaning of ones lives and is further connected to one’s personal goals. (Emmons 2005) This desire for authenticity is explained to be a response to standardization and homogenization in the marketplace by Thompson et al. (Thompson et al. 2006) Arnould and Price find that authenticity is part of cultural display and that it can be representative of a social unit. Consumers are known to link objects to stories of the self. (Arnould and Price 2000) People also match objects with their idea of how they should be, with belief systems and stereotypes. (Grayson, Martinec 2004; Bruner 1994; Rose, Wood 2005) Inferring from available literature, which states that objects are matched to a certain context in the framework of authenticity, it can be proposed that high need for authenticity and low loyalty consumers will concentrate more on the country than on brand, when making their decisions. When a certain product is made in Germany, whereas Germany is rated highly in a consumer’s value system or most people believing in a stereotype of good quality, technically advanced German products, this will have a great effect on the choice on someone with high need for authenticity. 22.

(26) H3: A consumer’s level of need for authenticity has a significant influence on the relative importance of brand vs. country of origin in product choice, with consumers having a stronger need for authenticity tending to put more importance on country.. H4: Product choice of low loyalty and high need for authenticity consumers is more influenced by country of origin than by brand.. Until now all hypotheses were proposed because loyalty can be connected to brand and need for authenticity can be connected to country of origin. However, it is not yet clarified what happens when consumers’ level of loyalty and level of need for authenticity are the same. According to the prior mentioned theory high loyalty and high need for authenticity consumers should be equally influenced by brand and country. However it is assumed that there exist further variables, which explain why sometimes it is the brand that is the decisive factor and why it is the country another time. Brand is known to be a signal for quality. A brand provides somewhat standardized products all over the world and one can be quite sure that one is getting the same or at least similar quality when buying a Zara T-shirt in Spain or in the US. Quality may either be perceived as high because of the brand name, or also because of a brands positioning. (Tülin, Swait 2004) Being positioned in the high end sector with consumers knowing that this brand’s products are high in price and them also knowing with what kinds of luxurious competitors a certain brand is competing with, gives them reassurance of the product’s quality to be favorable as well. However with consumers being more informed about the global manufacturing situation, it is supposed that when risk is high, consumers will think of production factors that influence the product they purchase. For globally operating brands it is increasingly hard to supervise and control production 23.

(27) sites in other countries, especially when cultural or legal barriers are high. More and more consumers are aware of the fact, that none of Nike’s managers has ever stood in a Nike factory in Bangladesh and personally caught a glimpse of how Nike shoes were produced there. This may not be a big deal when purchasing shoes, but it definitely becomes important to consumers, when the products they buy bear risk, may it be financial, personal or social, and they know that the brand has only limited ability to execute quality control at their production sites. It may not only be limited control, but quality could also differ because of legal matters, like certain countries’ laws that require companies producing there to use local raw material or create joint ventures with local firms in order to support the country’s economy. Nagashima (1970) found that perceived reliability of products is connected to their manufacturing country, e.g. German products being rated more reliable then American products. Consumers also have ideas about the level of technical advancement of different countries in their mind, which again influences how a product is rated. With all this knowledge given, I believe that consumers may ignore it when purchasing a product does not bear risk, but when risk is high, which also increases involvement, consumers will want to access more information in their search process and therefore country of origin will become more important for consumers’ decisions. The relationship of involvement and risk has been talked about in literature extensively. When risk is perceived situational involvement will be heightened. Also is risk of mispurchase high when the product’s price is high or when the product is a durable one, which consumers are not able to replace with a better one in short time. Involvement is known to strongly influence consumers’ decision processes and information search. (Laurent, Kapferer 1985). 24.

(28) H5: Level of perceived risk has a significant influence on the relative importance of brand vs country of origin in product choice, with the country clue becoming more important when risk is high.. H6a: Product choice of high loyalty and high need for authenticity consumers is more influenced by brand when risk is low. H6b: Product choice of high loyalty and high need for authenticity consumers is more influenced by country when risk is high.. Reasoning for H7a and H7b is similar to the theory behind the development of H6a and H6b, as like in the prior condition the level of loyalty and need for authenticity do not differ and therefore the importance of brand or country of origin cannot be predicted by this difference. I propose that low loyalty and low need for authenticity consumers will prefer the better brand when risk is low, just as high loyalty and high need for authenticity consumers do. When risk gets higher, there will be a shift to country of origin.. H7a: Product choice of low loyalty and low need for authenticity consumers is more influenced by brand when risk is low. H7b: Product choice of low loyalty and low need for authenticity consumers is more influenced by country when risk is high.. A possible relationship between loyalty and need for authenticity with risk needs to be investigated. Loyalty can be described as avoidance of trying new products (Schaninger 1976), which 25.

(29) infers that high loyalty consumers are very likely to perceive it as risky to get in contact with products or product features, that they do not know or are not used to, e.g. when they are told that a product they believe to be manufactured in France is actually manufactured in Russia. Consumers that are loyal to a brand are also stronger involved in a purchase decision than consumers who do not really care about which object they choose in the end. When relevance of a purchase decision is higher, consumers will be more sensitive to risk. Cunningham (1967) found in his research of low- medium- and high risk perceivers that there was a strong positive relationship between perceived risk and perceived brand commitment. Consumers with high need for authenticity develop this need in order to decrease the possibility of accidentally purchasing a fake product. This need for authenticity can therefore be seen as being a fear of the risk of purchasing a fake product or imitation. High need for authenticity participants of Liao and Ma’s research stated that authentic products are of finer quality, while there is no quality assurance when buying imitations. Therefore they were willing to spend more money for the authentic product. (Liao, Ma 2009) The quest for authenticity comes out of the consumer’s need for a feeling of assuredness. (Rose, Wood 2005) In other words the need for authenticity is developed in order to be secured and to avoid risk. Hence, I suppose that high loyalty and high need for authenticity consumers are more sensitive to risk than low loyalty and low need for authenticity consumers. If this assumption proves right, then one may infer that high loyalty and high need for authenticity consumers’ shift from basing their choice on brand to basing it on country happens faster, or at a lower threshold of risk, than it does for low loyalty and low need for authenticity consumers. As this threshold cannot be calculated within the limits of this thesis, only the hypothesis that high loyalty and high need for authenticity consumers are more sensitive to risk (versus low loyalty 26.

(30) and low need for authenticity consumers) will be tested.. H8: High loyalty and high need for authenticity consumers are more sensitive to risk than low loyalty and low need for authenticity consumers.. Prior literature has found that women were more risk averse than man. (Barsky et al. 1997) Borghans et al. are able to explain the difference in risk aversion of both genders partly by physiological traits, with people who are more agreeable, more neurotic and who have less ambition being more risk averse. (Borghans et al. 2009) As risk is assumed to be related to the likelihood of choosing the better country, H9 is developed.. H9: Women are more likely to choose the better country of origin, rather than the better brand, than men.. 3.2 Experimental Design. The objective of this thesis is to find out, why in certain situations COO effect is overridden by brand effect, but is still salient at other times. I suppose that the relative importance of brand or country of origin is correlated with a consumer’s tendency to be brand loyal and with his or her need for authenticity. High loyalty and low need for authenticity consumers are likely to put more importance onto brand in their decision process High need for authenticity and low loyalty consumers will be more focused on the country of origin cue (versus brand cue). When loyalty and need for authenticity are of the same level, either both high or both low, the relative importance a customer 27.

(31) attaches to brand vs. COO is related to perceived risk, with country of origin only then surpassing brand, when risk is high. Furthermore does gender influence the decision between country and brand.. Image 1: Expected Choices of Different Loyalty and NFA levels. Image 2: Framework. 28.

(32) Independent Variables: Loyalty and need for authenticity serve as independent variables and are split into low and high conditions by the median. Alongside loyalty and need for authenticity risk influences whether a consumer focuses on brand vs. country of origin cue when making a choice. Risk is manipulated via product category. As there are four product categories observed, there will be four risk conditions. The lowest risk condition is given when purchasing sports clothing as clothing relatively cheap in price and can be replaced easily in case of mispurchase. The highest risk condition is investigating choice in the area of automobiles. Automobiles are expensive, cannot be replaced easily and are perceived as status symbols, so that a mispurchase may negatively influence the way others think about the buyer. Furthermore is gender is used as an independent variable.. Dependent Variables: This thesis’ dependent variable s is a continuous variable, describing whether it is brand or country or origin that has a greater influence in product choice. This variable spans from 0 to 1, calculated from the answers to three multiple questions for each product category. A score under 0.5 displays that the participant chose the better brand more often than the better country. The better country was chosen more often if the score is over 0.5.. 3.3 Measurement. All following measurements are conducted using a 5-point Likert scale, ranging from “strongly agree” (非常同意) to “strongly disagree” (非常不同意). The lower the rating, the higher a consumer’s level of loyalty and need for authenticity and the higher is a 29.

(33) product’s risk. The questionnaire will be distributed in Chinese language, therefore scales taken out of prior literature had to be translated from English.. 3.3.1 Loyalty Measurement: (Amended from Arnould, Price 1999). 1. After having bought a certain brand, I usually feel a commitment to continue purchasing products of the same brand afterwards. 買過某個品牌的產品以後，通常我會對繼續購買該品牌產品有一種承諾感。. 2. I feel loyal to certain brands. 我對某些品牌很忠誠。. 3. When purchasing a product, I usually keep buying the same brand. 買東西的時候我通常會持續選擇同一個品牌。. 4. I usually continue to buy the same brands, even when their prices increase somewhat. 就算我習慣的品牌漲價, 我還是會繼續購買它。. 3.3.2 Need for Authenticity Measurement: (Amended from Newman, Dhar 2014). 1. A product manufactured in the place where the brand is originated is more valuable to me than the same product produced in another country. 30.

(34) 對我來說, 一個品牌, 其在品牌原產國生產的產品比該品牌在他國製造的相同產品更有價值。. 2. A product produced in the original factory is more authentic than if it was produced elsewhere. 一個在原廠製造的產品比在其他地方製造的該品牌產品更可信。. 3. I would rather buy a product that was produced in its brand’s country of origin than a product produced elsewhere, because it is of better quality. 我更願意買品牌原產國生產的產品，而不是該品牌在他國製造的相同產品，因為前者品質更好。. 4. A product that is produced in its brand’s country of origin is more worth than an identical product produced in another country. 一個品牌，其在品牌原產國生產的產品比該品牌在他國製造的相同產品更有價值。. 3.3.3 Manipulation Check: Risk measurement (amended from Jacoby, Kaplan 1972). A manipulation check is obligatory to check whether the two high risk products, car and luxury watch, and the two low risk products, sports clothing and electric kettle, are actually perceived to be of high or low risk respectively. Jacoby and Kaplan cataloged different varieties of risk and developed questions to measure each kind. Questions are asked as follows, with the empty space to be filled with the product category in the 31.

(35) actual questionnaire.. 1. If I make a wrong decision when buying …., this will negatively affect the way other people think of me. 如果我在購買……時做了錯誤地決定，別人會對我有不好的看法。. 2. Buying a … needs to be thought through carefully because this purchase is very risky. 購買 …時要謹慎考慮, 因為購買…的風險很高。. 3. When buying … I am very concerned that the product does not work properly or that I notice flaws after the purchase. 購買….時, 我很擔心購買後才發現產品有瑕疵或它不好用。. 4. Making a wrong decision when buying … will have severe negative consequences. 如果購買…時選錯的話，會有嚴重的負面結果。. 3.4 Procedure. 3.4.1 Pretest. As the main questionnaire will include favorable brand/unfavorable country and unfavorable brand/favorable country combinations to find out which one consumers will choose, brands and countries had to be evaluated in a pretest. The pretest was exclusively conducted via online questionnaires. 32.

(36) 3.4.1.1 Participants. For country evaluation I collected 31 questionnaires, 15 being filled out by male participants and 16 by females. All of the participants were Taiwanese. As people from different countries are expected to have different views on the countries of origin included in the questionnaire the nationality of participants had to be controlled. 19 people were between 21 and 25 years, five people between 26 and 30 years and seven people between 31 and 35 years, with the major part of participants being students.. Brands were evaluated by a total of 35 people, of whom 15 were male and 20 were female. All of the participants were Taiwanese. 22 people were between 21 and 25 years, six people between 26 and 30 years and seven people between 31 and 35 years. Again, a major part of participants were students, with 29 of them being graduate students.. 3.4.1.2 Data collection. A total of 17 countries were evaluated by being rated on four questions. Answers to these questions were given on a 5-point Likert scale, ranging from “strongly agree” (非常同意) to “strongly disagree” (非常不同意), so that a lower score represents a better country evaluation.. 1. Products made in this country are of good quality. 這個國家生產的產品品質優良.. 33.

(37) 2. I like products produced in this country. 我喜歡在這個國家生產的產品.. 3. Products made in this country are of better quality than products made in most other countries. 這個國家生產的產品比大多數其他國家產品品質要更好．. 4. Knowing that a product was made in this country will let me prefer it over others. 得知一個產品是這個國家製造的會讓我更有可能選擇該產品（而非選擇其他國家製造的產品. 48 brands were evaluated by asking five questions. Answers to these questions were given on a 5-point Likert scale, ranging from “strongly agree” (非常同意) to “strongly disagree” (非常不同意), so that a lower score represents a better brand evaluation. It was further asked whether the participant knows this brand. If it was stated that he or she does not know this brand, this person’s evaluation regarding the brand was not included in the overall rating.. 1. This brand is very valuable. 這個品牌很有價值.. 2. I like this brand. 我喜歡這個品牌.. 34.

(38) 3. This brand stands for quality. 這個品牌代表品質.. 4. I am willing to buy this brand's products. (Suppose you have the money) 我願意購買這個品牌的產品. (假設你/妳的錢足夠). 5. I would feel good buying this brand's product. 我相信購買這個品牌會讓我感覺良好.. 3.4.1.3 Results. Table 1 shows the results for country evaluations with the average 5 point Likert rating on the four asked questions, their sum and the Z score for each country. The Z score is needed later, as without it, it would not be possible to combine the date for countries and brands. The average evaluation for all countries is 11.066 with a standard deviation of 3.67. The best evaluated countries are Japan and Germany. China and Bangladesh are the two countries given the lowest ratings. Cronbach’s alpha was calculated and found to be 0.998.. 35.

(39) Table 1: Results for Country Evaluations. Table 2 shows the evaluations for 48 brands, of which only a small part will be used for the main questionnaire. Ratings on the five questions are shown as well as a sum and Z score. The two brands rated best are Apple and BMW. The two brands rated lowest are 566 and Motorola. Cronbach’s alpha was calculated and found to be 0.984.. 36.

(40) Table 2: Results for Brand Evaluations. 37.

(41) This pretest was conducted first to get a general understanding about the image of different countries and brands in the minds of Taiwanese consumers, and second to find suitable brand/country pairs that can be used in the main questionnaire. The main questionnaire asks participants to choose between two pairs of brand and country combinations. These are favorable brand/unfavorable country and unfavorable brand/favorable country pairings, whereas the Z distances between the favorable and unfavorable items need to be controlled and held constant. i.e.: when one wants to compare Apple (Z score -1.959) and HTC (Z score 0.412), which show a Z score distance of 2.371, then they need to be paired with two countries displaying the same Z distance. In this case Germany (Z score -1.433) and India (Z score 0.945) would be a good choice, as there lies a Z score distance of 2.378 between them. The resulting question asked in the questionnaire would then be: “Would you rather buy an Apple mobile phone produced in India, or a HTC mobile phone produced in Germany?”. 3.4.2 Main Questionnaire. The main questionnaire is included in the appendix.. 3.4.2.1 Participants. The main questionnaire was conducted online. A total of Taiwanese 221 people filled out the questionnaires, of whom 151 were students.. Five questionnaires turned out to. be non-valid, so that in the end 108 high risk questionnaires, asking about cars and 38.

(42) luxury watches, and 108 low risk questionnaires, asking about sports clothing and electric kettles, were collected. After analyzing their answers about brand loyalty and need for authenticity, 101 people were found to fit into the low loyalty and low need for authenticity group. 41 of those male, 60 female. The high loyalty and high need for authenticity group included 44 participants, 26 male and 18 female. Low loyalty and high need for authenticity included 44 people, 16 male and 28 female. Only 27 participants were found to be in the high loyalty and low need for authenticity group. 9 of those were male, 18 female. As a major part of participants were students, 115 of all participants stated that their monthly income was lower than 10,000 Taiwan Dollars, 56 participants rated their monthly income to be between 10,001 and 30,000 Taiwan Dollars.. 3.4.2.2 Data collection. The main questionnaire will be created in two versions, one being about products expected to be rather high risk, cars and luxury watches, and one being about low risk products, sports clothing and electric kettles. The first part will ask participants to make a choice between a favorable brand/ unfavorable country and an unfavorable brand/favorable country pair, whereas they should not decide for one of the selectable items when one or both brands are unknown to them.. i.e. Suppose you are planning to buy a new car, which one would you choose? 假設您打算購買一台新車, 您會選擇購買哪一個? (假設您錢足夠). 39.

(43) A. Audi manufactured in Thailand 於泰國製造的奧迪(Audi)汽車 B. Volvo manufactured in the U.S. 於美國製造的沃爾沃(Volvo)汽車 C.. I don’t know one or both of these brands 我不知道這兩個或之中一個品牌. Two product categories are featured per version with three of the upper introduced multiple choice questions asked for each product category. The next part includes measurement of consumer loyalty and need for authenticity and manipulation check for level of perceived risk. The third part directly asks participants to rate information cues or product features on their importance in decision making. 100 points have to be distributed to five items, which may influence a person’s decision, according to their importance. One question is asked per product category, whereby the five items which are given points stay the same regardless of product.. i.e. When buying a car I care most about: (you can give a total of 100 points) -Brand. -COO. -Price. -Color. -Service Attitude. 購買汽車時我最在意: (請給分, 最多可用 100 分) -品牌. -生產國家. -價錢. -顏色. -服務態度. The last part of the main questionnaire collects demographic data of the surveys participants.. 40.

(44) 3.4.2.3 Data Processing. Data acquired in the first part of the questionnaire, which are multiple choice questions, will be processed by one sample T-tests and Anova. For Anova the independent variables, loyalty and need for authenticity are made use of as categorical variables, split by their median. Risk and gender are also tested for their effect on choice. The dependent, continuous variable is whether the better brand or the better COO is chosen. The aspired result in this process is that consumers’ features concerning loyalty and need for authenticity, gender and risk condition are able to predict whether a strong brand overrides COO effect in this person’s decision making or whether country of origin has a strong influence on the consumer’s choice. Participants will be grouped into low vs. high loyalty and low vs. low need for authenticity conditions and T-tests will be used to investigate whether there exists any trend in different participants’ choices.. Chapter 4. Results. 4.1 Reliability Testing. 4.1.1 Cronbach’s Alpha: Loyalty. Four questions were asked to observe loyalty. However Cronbach’s Alpha was found to be at a low level of only 0.739. Therefore one of the questions will be removed. After 41.

(45) removing “I usually continue to buy the same brands, even when their prices increase somewhat”, Cronbach’s Alpha improves to 0.752.. 4.1.2 Cronbach’s Alpha: NFA. Cronbach’s Alpha for need for authenticity lies at 0.827. No improvement would occur, if one of the questions were removed. The proposed measurement is acceptable.. 4.1.3 Cronbach’s Alpha: Risk. The measurement for perceived risk shows a quite low reliability of 0.666. A small improvement is possible by removing one of the four questions asked in the questionnaire. Removing “If I make a wrong decision when buying …., this will negatively affect the way other people think of me” will bring reliability up to 0.682.. 4.2 Manipulation Check Risk. The first of the four questions used for risk measurement was taken out. Now, calculated out of three questions, the average perception of risk when making a car purchase lies at 5.018, which makes cars the product category with the most risk in this thesis. This is an average over all participants, regardless of loyalty and need for authenticity level. The risk score is calculated by taking the sum of three questions asking about risk, whereby “1” stands for “Strongly agree” and “5” for “Strongly disagree”, so that a 42.

(46) smaller score actually means higher risk. The risk perception of luxury watches is 6.459, stating that purchasing a luxury watch is seen less risky than buying a car. However the risk index of luxury watches is higher than one of the products originally being put in the low risk category. Risk for sports clothing lies at 7.973, which makes it the least risky product out of the four product categories used in this thesis. The average risk perception of electric kettles is 6.336. Therefore electric kettles may not be a suitable product for being used in the low risk condition, as it is perceived more risky as luxury watches and the distance between both is quite small. Electric kettles may further be received as even more risky, than what is observed in this thesis, as none of the questions in the questionnaire asked about risk of getting hurt or health concerns, which people might think about when dealing with electric products. As the products turned out to be not suitable to be split into two, low and high risk, groups, risk will be viewed per participant and product categories will be taken care of separately.. Table 3: Perceived Risk per Product Category. Product. Perceived Risk. Original Risk Condition. Car. 5.018. High Risk. Electric Kettle. 6.336. Low Risk. Luxury Watch. 6.459. High Risk. Sports Clothing. 7.973. Low Risk. 43.

(47) 4.3 Effect of Demographics. In order to check whether there are any differences in choice of different demographical groups, a possible influence of status was tested. The main point was finding out, whether there were any differences between the choices of participants, who were students, and participants who are working in a job. Jobs vs. Study was chosen over the age clue, as it is supposed that having a job influences ones choices and thinking stronger than age. For example may an 18-year old worker make more “mature” decisions than a 24-year old student who hasn’t made any experiences outside of school in her/his life. Each participant had to answer six multiple choice questions stating different situations in which the participant had to either choose the better brand or the better country. If the better brand was chosen, this was marked with “0”. The better country was marked with “1”. When the average result that a participant reaches in the multiple questions part lies over 0.5, this shows that this person chose the better country more often than the better brand. If the average of a person would be 0, this would mean that he or she chose the better brand in all six questions asked. An independent sample t-test was conducted and no significant difference of choice was found between participants that were still students (S) and those who had started their careers already (W) (t(219) = 0.981, p = 0.328).. 4.4 Hypotheses Testing. An Anova analysis is conducted with loyalty, need for authenticity and risk, as well as with gender, which also affects the relative importance of brand vs country in consumer 44.

(48) choice.. Table 4: Anova. df. F. Sig.. Loyalty. 1. 1.034. .310. NFA. 1. .356. .551. Risk. 3. 11.190. .000. Gender. 1. 6.828. .009. Loyalty * NFA. 1. .298. .586. Loyalty * Risk. 3. 2.275. .079. Loyalty * Gender. 1. .323. .570. NFA * Risk. 3. .163. .921. NFA * Gender. 1. 2.613. .107. Risk * Gender. 3. 2.261. .081. Loyalty * NFA * Risk. 3. .806. .491. Loyalty * NFA * Gender. 1. .544. .461. Loyalty * Risk * Gender. 3. 1.767. .153. NFA * Risk * Gender. 3. 2.618. .051. Loyalty * NFA * Risk * Gender. 3. .663. .575. Risk is found to have a significant effect on choice, whereby higher risk makes it more likely for people to choose the better country of origin rather than the better brand. Also the effect of gender is significant, with female consumers being more likely to choose the better country of origin.. A closer look can be taken at the two way interaction between loyalty and risk. Risk is split into four categories, which equals the product categories. 1 stands for the least risky product sports clothing (Risk: 7.973), 2 for luxury watch (Risk: 6.459), 3 for 45.

(49) electric kettles (Risk: 6.336) and 4 for cars, the product category with highest perceived risk (Risk: 5.018). At the lowest level of risk, choice of both low and high loyalty participants lies under 0.5, and therefore tends towards brand rather than country. With the other three product categories, speaking with increased risk, it becomes more likely for the better country to be chosen. The image below shows that the effect of risk is stronger on high loyalty participants, with high loyalty people being more likely to choose brand when risk is low (Average choice of sports clothing = 0.358, Average choice of cars = 0.682) compared to low loyalty participants (Average choice of sports clothing = 0.469, Average choice of cars = 0.571) and more likely to go for the better country when risk becomes higher (versus low loyalty participants).. Image 3: Loyalty * Risk. 46.

(50) An interaction between need for authenticity and gender was detected. There is only a small difference in choice between low and high need for authenticity males (Choice of high NFA = 0.534, choice of low NFA = 0.516). However, choice of low vs. high need for authenticity females differs greatly (Choice of high NFA = 0675, choice of low NFA = 0.589), with high need for authenticity females being much more likely to choose the product manufactured in the better country.. Image 4: NFA * Gender. Female participants were in three out of for product categories or risk conditions more likely to choose the better country than their male counterparts. The greatest difference between males and females can be seen when risk is at its lowest (Choice female = 0.513, choice male = 0.307), whereby men are very likely to choose the better brand 47.

(51) and women are rather indifferent.. Image 5: Risk * Gender. A three-way interaction between need for authenticity, risk and gender was found. Two images below show choices of low and high need for authenticity participants of each gender. When having to choose sports clothing, it is obvious that males were more influenced by the better brand then females. Females who had to choose cars were more likely to choose a weaker brand manufactured in a favorable country, with high need for authenticity females displaying this trend even stronger. In the male group however, only high need for authenticity participants tended to choose the better country, while low need for authenticity males chose the better brand more often, although risk was 48.

(52) high.. Image 6: Risk * NFA * Gender. 49.

(53) Image 7: Risk * NFA * Gender. 4.4.1 H1 The hypothesis that a consumer’s level of loyalty has a significant influence on the relative importance of brand vs. country of origin in product choice, with more loyal consumers tending to put more importance on brand could not be supported (F(1.410) = 1.034, p = 0.310).. 4.4.2 H2 H2 assumes that product choice of high loyalty and low need for authenticity consumers is more influenced by brand than by country of origin. The average of choice of high loyalty and low need for authenticity consumers over 50.

(54) all four products is 0.5698. Although there seems to be a trend towards country, it is not significant (t(26) = 1.312, p = 0.201). This may be due to the relatively low amount of people that fit in the group of high loyalty and low need for authenticity. Only 27 people fit in this condition, of whom 9 were male.. 4.4.3 H3 H3 supposes that a consumer’s level of need for authenticity has a significant influence on the relative importance of brand vs. country of origin in product choice, with consumers having a stronger need for authenticity tending to put more importance on country. The hypothesis is not supported (F(1.410) = 0.356, p = 0.551). 4.4.4 H4 H4, which states that product choice of low loyalty and high need for authenticity consumers is more influenced by country of origin than by brand, is supported. Making no difference between product categories, the average choice of all low loyalty, high need for authenticity participants is 0.619. Being over 0.5, this shows that low loyalty and high need for authenticity people tend to choose the better country. A one sample t-test shows that this result is significantly different from neutral 0.5 (t(43) = 2.662, *p < 0.05). Out of all participants, 44 people were of low loyalty and high need for authenticity background. However, as need for authenticity was not found to have an effect on choice in the Anova table, this hypothesis’ result being significant may be partly due to gender, with 28 participants being female and 16 being male in the low loyalty and high need for authenticity group.. 51.