The Effect of Perceived Country of Origin on Brand
Attitude, Brand Trust and Purchase Intentions
Master Thesis
Student:
Mariya Petrova
Student Number:
10993150
Date of submission:
29.01.2016
Supervisors:
Drs. ing. A.C.J. Meulemans
Drs. F. Slisser
Program:
MSc. in Business Administration – Marketing Track
University of Amsterdam
2
Statement of originality
This document is written by Mariya Petrova who declares to take full
responsibility for the contents of this document.
I declare that the text and the work presented in this document are original and
that no sources other than those mentioned in the text and its references have
been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision
of completion of the work, not for the contents.
3
Acknowledgements
I would like to thank Drs. Meulemans for his time, guidance, expert advice and feedback throughout the process of writing this thesis. Furthermore, I would like to express my gratitude to all of the participants in my survey whose responses provided valuable insights and helped me conduct this research. Lastly, I am grateful to my friends and family for their immense support and involvement.
4
Abstract
Recent research suggests that the country of origin effect is present regardless of consumers’ knowledge of the actual country of brand origin. Thus the concept of perceived country of origin emerged. The purpose of this thesis is to explore the effect of perceived country of origin on brand attitude, brand trust and purchase intentions. An online survey was conducted to answer the question to what extent the perception of a favorable country of origin, regardless of the actual country of brand origin, results in a positive effect on consumers’ attitude, trust and purchase intentions towards the brand. The questionnaire asked respondents to express their opinion on three brands of ice cream: Perry’s Ice Cream, Häagen-Dazs and Baskin-Robbins. A within-subjects design was used. All of these brands originate from the United States of America; however they employ different strategies concerning the usage of country of origin information. A repeated measures ANOVA was performed to compare the means of brand attitude, brand trust and purchase intentions across the three conditions of the perceived country of origin. The analysis showed that respondents had significantly higher attitude, trust and purchase intentions in the condition of positive image of the perceived country of origin. These results suggest that associating a brand with a favorable country, no matter the real country of origin, can lead to a higher attitude and trust towards that brand, as well as an increased likelihood to buy it. Regional differences were also found indicating that the effect of perceived country of origin varies among people from different geographical locations.
Keywords: country of origin, perceived country of origin, brand evaluation, brand attitude,
5 Table of Contents Acknowledgements ... 3 Abstract ... 4 1. Introduction ... 7 2. Literature review ... 10
2.1 Defining Country of Origin ... 10
2.2 Country of Origin as a Brand Building Tool ... 11
2.3 Country of Origin and Product Categories ... 12
2.4 Country of Origin Effect across Cultures ... 13
2.5 Developed and Developing Countries ... 15
2.6 Criticism on Country of Origin ... 17
2.7 Perceived Country of Origin ... 18
2.8 Brand Attitude ... 19
2.9 Brand Trust ... 20
2.10 Purchase Intentions ... 21
2.11 Conceptual Model and Hypotheses ... 22
3. Research Methodology ... 24
3.1 Research Design ... 24
3.2 The Pretest ... 26
3.3 The Main Questionnaire ... 28
3.4 Measurement of the Dependent Variables ... 29
4. Results ... 31
4.1 Sample Description ... 31
4.2 Perceived Country of Origin ... 32
4.3 Scale Reliabilities ... 34
4.4 Descriptive Statistics and Correlations ... 36
4.5 Hypotheses Testing: Repeated Measures ANOVA ... 36
4.6 Hypotheses Testing: Mixed ANOVA ... 41
5. Discussion, limitations and recommendations for future research ... 46
6. Conclusion ... 50
References ... 53
Appendix 1: Pretest ... 57
Appendix 2: Main Questionnaire ... 61
6
List of Figures
Figure 1: Conceptual Model ... 23
List of Tables Table 1: Brands chosen, actual and perceived COO and favorability of the perception ... 28
Table 2: Respondents’ Nationality and Region ... 32
Table 3: Perceived Country of Origin ... 34
Table 4: Reliability Statistics ... 35
Table 5: Means, Standard Deviations, Correlations ... 36
Table 6: Brand Attitude Pairwise Comparisons ... 38
Table 7: Brand Trust Pairwise Comparisons ... 39
7
1. Introduction
The growth of international trade made a wide variety of foreign products available worldwide. Marketers are struggling to differentiate their products and build a long-lasting brand image. In a globalized world country of origin (COO) may be a way of differentiating a brand and a source of competitive advantage if consumers hold favorable associations towards the country. Moreover, the importance of COO is increasing due to the growing popularity of online purchases. Most Internet advertising messages rely on extrinsic cues such as price, brand name, and country of origin because intrinsic cues (physical characteristics) are difficult to assess online without actually seeing the product (Insch & McBride, 2004). COO is an important extrinsic cue, therefore its importance in consumer product evaluations is growing (Klein et al., 1998, in Insch and McBride, 2004).
A wide body of research has investigated the effect of country of origin and country image on consumers' perceptions, product preferences and quality evaluations (Al-Sulaiti & Baker, 1998). Most researchers agree that COO positively influences these variables; however recent studies cast some doubt on these findings and claim that previous research has inflated the importance of COO (Samiee, Shimp, & Sharma, 2005). Samiee et al. (2005) argue that people do not have accurate perceptions of the country of brand origin and do not actively seek this information, therefore COO cannot be an important factor for their brand attitude.
Magnusson, Westjohn, & Zdravkovic (2011) refute this argument by exploring perceived COO as opposed to actual country of origin and find out that perceived country of origin matters regardless of whether the perception is accurate. Marketers seem to be aware of this and use it in their advantage by employing a foreign branding strategy and deliberately associating their brands with countries characterized with a positive image even though they are not the actual country of origin (Magnusson et al., 2011). Some examples include the
8
American ice cream Frusen Glädjé that uses Swedish words in the brand name to imply Scandinavian origin, the Ginsu knives that uses a Japanese sounding brand name despite its American origin, and the French dairy brand Président that uses the Greek alphabet and images from Greece in the packaging of its feta cheese. It is interesting to further explore the effect of this perceived country of origin on consumers’ perception of the brand.
The research on country of origin effect has been going on for decades; however the notion of perceived country of origin is relatively new and the research on the topic is limited. The purpose of this thesis is to explore the effect of perceived country of origin on brand attitude, brand trust and purchase intentions, and thus fill in the research gap between perceived and actual country of origin effect. I propose that if consumers associate a brand with a favorable country, this will lead to more positive brand attitude, higher brand trust and purchase intentions than in the case of associating the brand with an unfavorable country or with no country whatsoever.
This leads to the following research question:
To what extend does the association of a brand with a favorable country, regardless of
the actual country of brand origin, result in a positive effect on consumers’ perception of
the brand?
In order to answer this question a qualitative research method was used and an online survey was distributed among a student sample. The survey included three ice cream brands each representing different condition of the perceived country of origin: positive country image, negative country image and neutral condition (no country associations). Consequently the differential effect of perceived country image was analyzed across the three conditions.
This study contributes to the marketing literature by expanding our understanding of perceived COO as opposed to actual COO. Previous research by Magnusson, Westjohn, &
9
Zdravkovic (2011) already suggested that consumers’ perception of brand origin significantly affects brand attitude. The present research complements this by also investigating the effect of perceived brand origin on brand trust and purchase intentions. Moreover, this study also has a managerial application by offering practitioner a better understanding of the foreign branding strategy. The findings of the present research can be applied in the marketing practice as a way to create competitive advantage. Making a brand seem like it originates from a foreign country is a way to distinguish the brand and make it unique and easily recognizable in local markets. However, the country image should be carefully considered. The perceived country of origin only contributes to the brand image if it is a country with favorable associations. This research also emphasizes marketing practitioners’ need to consider the difference in the perceptions of people from different regions. The different culture and level of economic development across countries/regions might cause the effect of perceived country of origin to be stronger in some regions and weaker in others.
This thesis is structured as follows: the next chapter outlines the developments in the research on the country of origin effect, its criticism and the emergence of the concept of perceived country of origin. Chapter 3 discusses the research methodology, including results from a pretest and how they helped design the main questionnaire. Chapter 4 presents the statistical analysis of the data collected through this questionnaire and includes hypotheses testing. Consequently chapter 5 discusses the results and outlines some of the limitations of the research. The thesis ends with a brief conclusion in chapter 6.
10
2. Literature review
2.1 Defining Country of Origin
Country image and country of origin effect have long been of research interest. Many authors have attempted to combine the vast literature on the topic and create a review article that summarizes the findings on the topic and provides valuable insights. Al-Sulaiti & Baker (1998) provide a literature review of the research on the country of origin effect since the middle of the 1960s until 1997. Through the years different definitions of “country of origin” have emerged. Some authors use the term country of origin to refer to “the country of manufacture or assembly” of a product while others use the term “made in –“(Al-Sulaiti & Baker, 1998). Country of origin effect occurs when consumers prefer a certain product only because it is produced in a particular country. This bias is caused by the overall image of the country. Country image is defined as “consumers’ general perceptions of quality for products made in a given country” (Bilkey and Nes, 1982 in Han, 1989, p.222). The image of the country can affect consumers’ perceptions of the products manufactured in that country and influence product preference, quality evaluation, and buying decisions for products that are associated with the country - either designed, or manufactured, or assembled in that country (Han, 1989).
The first research to explore the significance of the country of origin information in consumers’ purchase intentions was conducted by Schooler in 1965. He proposed that consumers evaluate identical products differently and the basis for this difference is the national origin of the products (Schooler, 1965). He confirmed this hypothesis by conducting a research in Central America, thus proving for the first time that country of origin effect exists. The underlying explanation for the different product evaluation is based on people’s different evaluations of national segments (business, labor, government, and people) and travel experience (Schooler, 1965).
11
Nowadays brands are increasingly international and manufactured in a number of countries which leads to a change of the definition of country of origin. Country of origin is described as the country of brand origin where the corporate headquarters of the company are situated regardless of the location of actual manufacturing (Balabanis & Diamantopoulos, 2011). There are two different views on the role of country image in consumers’ product evaluations (Han, 1989). The view of country image as a “halo” suggests that consumers use country image as a sign for product quality when they encounter an unknown foreign brand (Han, 1989). The idea of country image as a halo is relevant only when consumers are not familiar with products from a certain country. However, if consumers are familiar with products of the country, country image is regarded as a summary construct in the same way as brand image is a summary construct (Han, 1989). The summary construct model suggests that consumers summarize information about a country’s products, then form product beliefs and based on them consumers form their perceptions of the country (Han, 1989). Other researchers propose that both country image and product beliefs simultaneously influence attitudes (Knight & Calantone, 2000).
2.2 Country of Origin as a Brand Building Tool
In the competitive environment nowadays, marketers use brand building tools in order to differentiate their brand. These include primary brand elements such as brand name, logo, slogan but also secondary sources of brand knowledge (Keller, 2005). Secondary sources of brand knowledge require associating the brand to an independent entity that has associations on its own – for example a person (when using a celebrity endorser) or a place (country/region of origin) - in order to borrow favorable associations (Keller, 2005). Those secondary sources can be people, places or other brands that have their own associations but are linked to the brand in a way that positive associations are transferred. The transfer of
12
meaning, feeling and associations from the country of brand origin to the brand involves risk because if the image of the country dramatically deteriorates (for example in case of bad policy decisions – starting a war, mistreating foreigners), it will have a negative effect on the brand itself. According to Keller (2005) the brand will benefit from these associations only if they are strong, favorable and unique.
2.3 Country of Origin and Product Categories
The country of origin effect is usually specific for product categories. For example, consumers may prefer automobiles from Germany but crystal from Ireland and shoes from Italy (Roth & Romeo, 1992). A lot of research investigates how country image varies across product categories. Roth and Romeo (1992) explore the fit (also referred to as “match”) between countries and product categories. They suggest that country image has a positive effect on product evaluations when there is a favorable country match. A product-country match occurs when a product-country is strong in an area that is an important feature for a product category, whereas a mismatch occurs when a country’s strengths are not important for the product category (Roth & Romeo, 1992). In the case of product country mismatch country image does not influence consumers’ perceptions of the product. For example, Italy is perceived as strong in the design area so there will be a positive match between Italy and fashion-oriented products. On the other hand, a mismatch will occur in the case of Japanese beer. Japan has an overall positive country image; however, the country’s strengths are related to innovation and technological precision that are not important for beer.
There are different variables that moderate the effect of country of origin on consumers’ perceptions such as consumers’ nationality, level of education, country familiarity, age, product involvement. Verlegh, Steenkamp, & Meulenberg (2005) propose that the impact of product-country image depends on consumers’ level of involvement with the ad. Under the
13
condition of low involvement consumers will use peripheral cues such as country of origin. They conclude that country of origin has a greater impact when consumers’ involvement is low (Verlegh et al., 2005). Their study supports the idea that country of origin plays an essential role in consumer product evaluations and supports the view of country of origin as a “cognitive shortcut”.
2.4 Country of Origin Effect across Cultures
Country of origin effect has an important cross-national component because perceptions of a country may be different depending on the consumers’ nationalities (Roth & Romeo, 1992). Insch and McBride (2004) decompose COO into country of product design, country of assembly and country of parts manufacture and test its effect on consumer perceptions using samples from two different countries – Mexico and USA. Their research discovers that customer perception of the importance of country of origin significantly varies between these two countries. For example, country of design has a stronger impact on US consumers compared to Mexican consumers in the sports shoes product category.
Country of origin effect also differs across countries because of cultural differences. Cultural differences have been vastly research during the years because they have important implications for management and marketing. Probably the most widely used and universally accepted research on national culture and cultural difference is the one by Geert Hofstede. Hofstede defines culture as “the collective programming of the mind which distinguishes the members of one category of people from another” (Hofstede, 1994, p. 1). He identifies five dimensions of culture: power distance, individualism vs. collectivism, masculinity vs. femininity, uncertainty avoidance, and long term vs. short term orientation and explores the implications of these in the family, in society and at the workplace (Hofstede, 1994). These dimensions of culture have also been used to explain the role of country of origin in
14
customers’ perceptions of a product ((Gürhan-Canli & Maheswaran, 2000) and (Insch & McBride, 2004)). For example, Gurhan-Canli & Maheswaran (2000) explore cultural variations in country of origin effect and find differences in product evaluations based on whether the society is individualistic or collectivist. There are significant variations in the evaluations of home country products and foreign products. Individualists evaluate a home country product more favorably only if it is superior to a foreign product; however collectivists evaluate a home country product more favorably, no matter whether the product is superior or inferior to a foreign product (Gürhan-Canli & Maheswaran, 2000). Another example is Insch & McBride’s (2004) study on differences in COO evaluations among American and Mexican consumers which suggests that the underlying reason for these differences is the difference in the power distance dimension between these two countries. Insch & McBride (2004) point out that Mexico and the United States have the opposite scores on three of Hofstede’s dimensions - individualism, power distance and uncertainty avoidance. Hofstede suggested that there is a negative correlation between power distance and education and between power distance and technological sophistication which might explain the variations in country of origin effect (Insch & McBride, 2004).
People from different cultures have different norms and values but also hold different stereotypes about the surrounding world. National stereotypes have been researched as an underlying explanation for the country of origin effect (Chattalas, Kramer, & Takada, 2008). Chattalas et. al (2008) use Stereotype Content Model to explore the relationship between national stereotypes and country of origin effect. The two dimensions of national stereotypes (perceived warmth and perceived competence) play a role in consumers’ product evaluation. Therefore, the contents of national stereotypes are viewed as antecedents of COO effect. Chattalas et. al (2008) find out that there exists a relationship between national stereotypes and attitude towards products and this relationship is moderated by a several product and
15
consumer characteristics such as cultural orientation, expertise and involvement. The most desirable position for a country is the quadrant characterized with high warmth and high competence (Chattalas et al., 2008).
People’s culture has an influence on their consumer behavior in general which might eventually impact their country of origin perceptions. For example, culture has been found to affect prices and consumers’ shopping behavior (Ackerman & Tellis, 2001). In a cross-cultural longitudinal study of American and Chinese consumers in California Ackerman & Tellis (2001) find out dramatic differences in the shopping practices of the two cultures. Even though the researched supermarkets are all situated in California, there is a huge difference in prices for identical products between Chinese and American supermarkets, indicating that culture can in fact affect prices. Moreover, people from these two countries exhibit different shopping behavior. Chinese customers are more price-sensitive and engage in a more extensive inspection of the products prior to purchase, for example by smelling the product. They view shopping as a family experience and a type of entertainment, whereas for Americans it is simply a task to perform. When shopping Americans value saving time, unlike the Chinese who focus on saving money, even when people from both cultures have the same socio-economic status (Ackerman & Tellis, 2001).
In sum, country of origin effect differs from country to country, therefore branding strategies based on origin need to be customized for each particular country.
2.5 Developed and Developing Countries
Some studies focus on variations of COO effect based on differences between developed and developing countries rather than culturally different countries (Batra, Ramaswamy, Alden, Steenkamp, & Ramachander, 2014). Consumers in developing countries show a higher preference towards foreign brands because of perceived product quality and social status.
16
Foreign goods are often associated with foreign lifestyle (high standard of living) and high quality. Moreover, Batra et al. (2014) find out that preference towards imported products is higher for people who appreciate Western lifestyle and for products that serve as a social signal. Sharma (2011) uses two constructs – materialism and value consciousness to highlight the differences between consumers’ perceptions of the country of origin in developed and developing countries. Materialism and value consciousness have significant effect on product preferences and purchase intentions only on consumers in emerging markets (Sharma, 2011). Consumes from emerging markets prefer products from developed countries because of perceived high quality and status enhancement; however, these consumers tend to be more value conscious and price sensitive (Sharma, 2011). Therefore, price might be a more influential extrinsic cue than country of origin. Local products might be preferred in emerging markets because they are associated with lower prices.
The differences in country of origin effect in developed and developing countries might be explained by the differences in the national culture of these countries. Cultural dimensions were previously discussed as influencing COO perceptions; however researchers argue that cultural dimensions are also correlated with the level of development of a country. Cox, Friedman & Tribunella (2011) explore the relationship between cultural dimensions and gross domestic product (GDP) and find out that power distance is negatively correlated with weighted gross domestic product per capita. However, individualism is positively correlated with weighted GDP per capita which indicates that with an increased economic development of a country, the society of that country is expected to become more individualistic (Cox et al., 2011).
17
2.6 Criticism on Country of Origin
Despite the numerous studies that prove the effect of COO, the field has been highly criticized. Recent research proclaims COO as irrelevant, suffering from overspecialization and problems with generalizability; however other authors provide strong arguments that COO is still an important cue (Josiassen & Harzing, 2008). The literature on country of origin is based on the assumption that consumers possess accurate knowledge of brands’ country of origin; however many researchers cast a doubt on this assumption (Balabanis & Diamantopoulos, 2008).
One of the reasons for criticism of COO is people’s lack of knowledge about brand origin (Samiee et al., 2005). Samiee et al (2005) develop a measure for brand origin recognition accuracy (BORA) and determine four factors that affect the BORA score: socioeconomic characteristics, international experience, demographics and ethnocentric tendencies. Their study of brand origin recognition accuracy of 84 brands finds out that respondents are capable of correctly recognizing the brand origin of only 22.3% of foreign brands and 49% of US (local) brands (Samiee et al., 2005). Consequently, this research concludes that consumers find COO information unimportant and unworthy of memorizing. Another study finds out that consumers have the ability to correctly identify the country of origin of different brands in only 22 % of the cases (Balabanis & Diamantopoulos, 2008). After discovering consumers’ inability to correctly classify brands’ origin, a research was conducted to explore the gains and losses of this misclassification (Balabanis & Diamantopoulos, 2011). In that research Balabanis & Diamantopoulos (2011) explore the consequences of brand origin misclassification in terms of brand image evaluation and purchase intentions. They suggest there are two types of misclassification – adverse and favorable - depending on whether the perceived country has a worse or a better image than the image of the actual country of origin. They propose that “weak brands have more to gain from favorable COO misclassification,
18
whereas strong brands have more to lose from adverse COO misclassification” (Balabanis & Diamantopoulos, 2011). However, their findings suggest that even favorable misclassification does not have a positive effect on brand image perceptions or purchase intentions – a conclusion that cast serious doubt on the effectiveness of foreign branding strategies.
However, Josiassen & Harzing (2008) point out the importance of the concept of foreign branding and by this refute the idea of COO becoming irrelevant. Consumers may easily confuse a brand as coming from a country other than its actual origin and this gives marketers the opportunity to associate their brand with any country, thus choosing one with a favorable image. Leclerc, Schmitt, & Dubé (1994) define foreign branding as “the strategy of spelling or pronouncing a brand name in a foreign language” which influences the way customers perceive and evaluate a product. For example, the American bottled water company Klarbrunn uses a German sounding brand name to elicit positive associations with the German Alps (Leclerc et al., 1994).
Another argument suggesting the lack of significance of country of origin cue is the increasing number of multinational companies with global production. Products of big international corporations are often produced in different countries and assembled in yet another country which makes COO as a term vague and even meaningless, therefore considered as unimportant cue for forming product preference and purchasing decisions (Pharr, 2005).
2.7 Perceived Country of Origin
Despite the lack of accuracy in country of origin perceptions COO effect matters (Magnusson et al., 2011). However, the attention is shifted from actual country of origin to perceived country of origin. The relevant country of origin nowadays is the country of association (Josiassen & Harzing, 2008).
19
Magnusson et al. (2011) provide evidence that perceived brand origin influences brand attitude regardless of the accuracy of the perception. Realizing this, marketers take deliberate actions to differentiate their brand by associating it to a country with positive image even if it is not the country of origin. However, the effectiveness of this type of branding strategy is not researched in depth. Moreover, it is unclear how the moderators that affect the actual country of origin will influence the relationship between perceived country of origin and brand attitude.
For the purposes of this research we define perceived country of origin as the country that customers associate with a brand. This can be the same country as the country of origin or a different country. Moreover, consumers’ perception for this country might be positive, neutral or negative depending on whether they hold favorable associations with this particular country. Perceived country of origin is the independent variable in this research. The next three sections provide definitions of the three dependent variables: brand attitude, brand trust and purchase intentions.
2.8 Brand Attitude
Keller (1993) defines customer-based brand equity as “the differential effect of brand knowledge on customer response to the marketing of the brand” (p. 8). Keller (2001) identifies six brand building blocks - salience, performance, imagery, judgments, feelings and resonance, that are organized in four levels of a pyramid. These four levels need to be reached by consumers so that the brand can build customer-based brand equity. The four levels relate to brand identity, brand meaning, brand response and customer-brand relationship (Keller, 2001). At the third level of this pyramid brand attitude is formed. It consists of brand judgments (relating to the more functional aspects of the brand) and brand feelings (relating to the emotional response to the brand).
20
Brand image and brand attitude should be differentiated. Brand image is a collection of brand related associations (Keller, 1993). Keller (1993) explores three types of associations - attributes, benefits and attitudes. Therefore, brand attitude is just one aspect of brand image. Brand attitude relates to the overall evaluation of the brand and is viewed as a function of the attributes and benefits that are associated with the brand (Keller, 1993). According to Keller (2001) achieving positive attitude towards the brand is a critical step towards achieving brand resonance which is the ultimate goal of any brand building program. If consumers hold a positive brand attitude, they can move to the next level of the brand pyramid that is characterized with brand loyalty, attachment, community and engagement (Keller, 2001). Positive brand attitude might be created by leveraging positive associations from the country of brand origin, which is an important source of secondary brand knowledge (Keller, 2005).
2.9 Brand Trust
Brand trust, also referred to as brand trustworthiness and brand credibility, is defined as: “the confident expectations of the brand’s reliability and intentions” (Delgado-Ballester & Luis Munuera-Alemán, 2005, p. 188).
Brand trust is based on consumers’ belief that the brand is consistent, honest and responsible (Delgado-Ballester & Luis Munuera-Alemán, 2005). To trust a brand means to believe or expect that the brand will deliver a positive outcome. Delgado-Ballester and Luis Munuera-Alemán (2005) investigated two distinct dimensions of brand trust – reliability and intentions. Reliability refers to the brand’s ability to satisfy customer wants and needs while intentions relates to the good rather than bad intentions of the brand; brand trust also involves the brand’s behavior - the extent to which a brand keep its promises to customers (Delgado-Ballester & Luis Munuera-Alemán, 2005). In a previous research Delgado-(Delgado-Ballester and Luis Munuera-Alemán (2001) explored the relationship among brand trust, satisfaction and loyalty.
21
They conceptualized brand trust as “the feeling of security held by the consumer that the brand will meet his/her consumption expectations” (Delgado-Ballester & Luis Munuera-Alemán, 2001, p. 1242). The findings of their research indicate that there is a direct effect of overall satisfaction on brand trust and this relationship is moderated by customer involvement, moreover brand trust is positively related to customer commitment. These results suggest that overall satisfaction is an antecedent of brand trust whereas brand loyalty is a consequence of brand trust.
Trust is a form of an interpersonal relationship; therefore the idea of brand trust implies that the relationship between a consumer and a brand is similar to personal relationships (Delgado-Ballester & Luis Munuera-Alemán, 2001). This is in line with the idea of Brand as Intentional Agents Framework (BIAF) that explores people’s perceptions of a brand using two dimensions – intentions and ability (Kervyn, Fiske, & Malone, 2012). These dimensions are comparable to the reliability and intensions dimensions used by Delgado-Ballester and Luis Munuera-Alemán (2005).
2.10 Purchase Intentions
Purchase intention is the most concrete attribute compared to brand attitude and brand trust. During the last decades researchers have explored the various effects of country of origin on product perceptions. Diamantopoulos, Schlegelmilch, & Palihawadana (2011) found out that country of origin image impacts purchase intentions indirectly and that indirect relationship is mediated by brand image. Different factors influence purchase intentions, for instance culture has an effect on price consciousness and price sensitivity and these might also have an effect on purchase intentions (Ackerman & Tellis, 2001). Moreover, people from developing countries are more likely to be price conscious due to low income. Culture might also affect
22
purchase intentions indirectly by influencing the values of a society. People from a materialistic culture will be more likely to engage in impulsive purchases.
2.11 Conceptual Model and Hypotheses
We have already defined the terms country of origin and perceived country of origin and provided evidence from previous research that the country of brand origin influences brand attitude, brand trust and purchase intentions. The aim of the present study is to investigate how the association of a brand with a country, without it being the country of origin, influences these three variables. We propose that the effect will be similar to the effect of actual country of origin but only when the perception is related to a country with a favorable image. The favorable country image will make the brand more likeable and trustworthy, thus resulting in a more positive brand attitude, higher trust and purchase intentions for the brand. Therefore, the following hypotheses will be tested:
Hypothesis 1: Favorable image of the perceived country of brand origin influences positively
consumers’ attitude towards the brand.
Hypothesis 2: Favorable image of the perceived country of brand origin influences positively
consumers’ trust towards the brand.
Hypothesis 3: Favorable image of the perceived country of brand origin influences positively
consumers’ purchase intentions towards the brand.
We investigated these propositions by comparing the condition of a favorable country image with the situations when the perception is negative and neutral. Moreover, we expect that consumers’ nationality, in other words their own country of origin, might influence this effect making it stronger for some consumers due to cultural differences and differences among consumers from developed vs. developing countries.
23
However, to make this research more viable we tested this proposition based on respondents region of origin, not nationality. More information about the four regional groups of countries is included in chapter 4. This leads to the last hypothesis:
Hypothesis 4: Consumers’ region of origin impacts their perception of the country of brand
origin in a way that the effect of perceived country of origin on brand attitude (4a), brand trust (4b) and purchase intentions (4c) will be different across regions.
Figure 1 shows a graphic presentation of the model.
H4a b c H1
H2
H3
Figure 1: Conceptual Model Perceived Country of Brand Origin Brand Attitude Brand Trust Purchase Intentions Respondents’ Region of Origin
24
3. Research Methodology
In this chapter the research method used for exploring the effect of perceived country of origin on brand attitude, brand trust and purchase intentions will be presented. The chapter includes a description of the research design, the pretest, the main questionnaire and lastly a description of the dependent variables and the scales used for their measurement.
3.1 Research Design
In order to explore the effect of perceived country of origin, a quantitative research was carried out using an online self-administered survey. The survey asked respondents to evaluate three ice cream brands. All survey participants answered questions about all three brands thus a within-subjects design was used.
The chosen brands are Perry’s Ice cream, Häagen-Dazs and Baskin-Robbins. All of these brands are originating from the United States of America; however they employ different branding strategies regarding the usage of country of origin information. Perry’s Ice Cream emphasizes it American origin by offering an “All American” ice cream. Häagen-Dazs uses a foreign branding strategy by choosing a German/Scandinavian sounding brand name. Baskin-Robbins does not include any country cues in its packaging and has a relatively neural brand name, thus we expect it will not cause any country associations in the mind of the respondents. In a way this brand serves as a control item for the research.
The respondents expressed their opinion about the three ice cream brands which provided insight about their brand attitude, brand trust and purchase intentions. The United States of America were chosen as a country of interest after a pretest showed a relatively negative quality perception towards food products coming from the US. The pretest indicated however a rather positive attitude towards German and Danish food products. Therefore, in this
25
research Perry’s ice cream represents a brand with unfavorable perceived country of origin, Häagen-Dazs represents a brand with favorable perceived country of origin and Baskin-Robbins stands for a brand with neutral perceived country of origin in terms of favorability.
Ice cream was selected as a product category in this research because of four reasons. First of all, it is a small daily purchase so consumers are expected to have a low involvement in this product category. The elaboration likelihood model suggests that under low involvement peripheral cues such as country of origin might be important for the purchase decision (Cacioppo & Petty, 1983, p.13). Secondly, ice cream was chosen because of consumers’ familiarity with this product category. Respondents can easily imagine the scenario of seeing the ice cream brands in their local supermarket and answer the questions regardless of their previous knowledge of the brand. Moreover, ice cream consumers are often variety seeking which means that strong customer loyalty to other brands will not interfere with respondents’ answer to the survey questions. When consumers are variety seeking there is also a higher likelihood that they will consider purchasing an unfamiliar brand. Thirdly, ice cream was chosen due to the availability of ice cream brands that employ the foreign branding strategy (e.g. Häagen-Dazs, Frusen Glädjé) and this made the research more feasible. Lastly, the findings in the ice cream product category are expected to be easily transferable to other food products. Therefore, the results of this research will contribute not only theoretically but might as well be actively applied in the marketing practice.
Existing brands were used in order to increase the similarity of the research to real life situations; however if respondents had previous experience with the brands this might lead to bias. In other words, respondents might be evaluating the brand based on their experience rather than the information provided in the survey.
26
The questionnaire explicitly instructed the respondents not to look up additional information about the brands while answering the survey questions in order to ensure that they make their answer choice based solely on the given information in the survey. The survey included a picture of the product where the brand name can be seen as well as a descriptor (e.g. in the case of Perry’s Ice Cream - “All American”). Thus respondents could make inferences about the country of brand origin.
All ice cream types included in the questionnaire were selected to have the same flavor (vanilla) to avoid biased answers as a result of personal preferences.
3.2 The Pretest
Before conducting the main survey, a pretest was carried out. The purpose of the pretest was to determine the favorability of country image for six different countries. The countries included in the pretest were: USA, China, Germany, Denmark, South Korea, and Greece. These were selected to represent countries with different level of development that were hypothesized to have different country image. Moreover, these countries are well-known therefore I expect that the survey respondents will have strong associations and hold an opinion about the products manufactured there. The pretest consisted of 5 statements for each country, 30 statements in total. The aim of these statements was to determine the favorability of each country in terms of country image in general, attitude towards products produced in these countries, quality perception of food products originating from each of these countries, as well as purchase intentions. An example statement is: “I believe products manufactured in country A are better than similar products produced in other countries”. Each statement was rated on a five-point Likert scale where 1 stands for Strongly Disagree and 5 stands for Strongly Agree. The survey ended with a few demographic questions asking about respondents’ age, gender and nationality. The pretest questionnaire is included in Appendix 1.
27
The pretest was distributed among university students, ages between 20 and 26. In total the survey received 15 responses. The subjects participating in the pretest have similar demographic characteristics with the respondents of the main survey. Most importantly, the participants in this pretest had a diverse national background including Dutch, Bulgarian, Polish, Vietnamese, Indian, Ecuadorian, and Chinese. None of the countries included in the survey was represented by a large number of respondents coming from this country, therefore eliminating the chances of bias due to ethnocentricity.
All of the statements used to evaluate the countries were positive thus a higher mean score indicates a more positive country image. The mean scores were: 3.4 for the United States, 2.68 for China, 4.05 for Germany, 3.78 for Denmark, 3.12 for Greece and 3.4 for South Korea. However, there were big differences between the overall attitude towards the countries and the attitude towards food products originating from these countries. While the total mean score of the USA was relatively high, the country had the lowest score on food products quality perception among all six countries (mean = 2.77). Since the focus of the main questionnaire is on a food item (ice cream), the USA will be considered as having an unfavorable country image. The findings of the pretest suggest that Germany and Denmark have the most favorable country image (both in general and in the food product category). Consequently, the USA was chosen as an actual country of origin for this research and the three chosen brands of ice cream had different perceived countries of origin with different favorability of the country image. Table 1 provides an overview of the brands chosen and their actual and perceived countries of origin.
28
Table 1: Brands chosen, actual and perceived COO and favorability of the perception
Ice cream brand Actual country of origin
Perceived country of origin
Country image of the perceived country of
origin
Perry’s USA USA Negative
Häagen-Dazs USA Denmark/Germany Positive
Baskin-Robbins USA Not country specific Neutral
3.3 The Main Questionnaire
The main questionnaire consisted of four parts. In the first part respondents were asked to evaluate the three brands of ice cream using established scales to measure brand attitude and brand trust. Moreover they were asked to express whether they are considering buying this brand or would like to receive more information about it. The only information included for each brand was a picture of the product. On the image the brand name was clearly visible, as well as the “All American” sub-name for Perry’s Ice Cream. Brand attitude was measured using a seven-point semantic differential scale while brand trust and purchase intentions were measured with a seven-point Likert scale where 1 stands for strongly disagree and 7 stands for strongly agree.
The second part of the survey asked the respondents to look at the three pictures of ice cream brands once again and indicate whether they were familiar with each brand prior to taking the survey. Respondents also had to write the name of the country they believe each brand comes from.
The next part of the survey asked respondents about the importance of the country of origin of products they buy. They had to evaluate how important the COO information is when purchasing durable goods and food products separately. Consequently, they indicated their
29
frequency of looking for COO information. These questions were only included so that we have a more accurate description of the sample.
The last part of the survey asked for respondents’ demographic information including age, gender and nationality.
The instructions made it clear that respondents should answer each question based only on the information included in the survey without looking up additional information. However, we have to acknowledge that some of the respondents had previous experience with these brands so it is likely that they used this experience in expressing their opinion. All respondents answered question about all three brands. The questionnaire is presented in Appendix 2.
3.4 Measurement of the Dependent Variables
The dependent variables in this study are brand attitude, brand trust, and purchase intention. Brand attitude and brand trust were measured using existing validated scales, whereas purchase intention was measured using a single item (likelihood to buy the product, evaluated on a seven-point Likert scale). Using a single item for measuring purchase intentions is recommended in previous studies because this variable is a concrete attribute and measuring it with only one item does not harm the validity of the research (Diamantopoulos et al., 2011).
Brand attitude was measured using a seven-point sematic differential scale adopted from the research by Li, Daugherty & Biocca (2002) and previously published in Bruner (1998) (Li, Daugherty, & Biocca, 2002). The scale includes six items: bad/good, unappealing/appealing, unpleasant/pleasant, unattractive/ attractive, boring/interesting, and dislike/like. The scale reliability is reported by Li et al (2002) as .91.
Brand trust was measured using three, seven-point Likert-type items (e.g. This brand is reliable.). The scale is reported by Bruner in Marketing Scales Handbook (2013, p. 373) and
30
is originating from Sheinin, Varki, and Ashley (2011). The Cronbach's Alphas reported are .88 (Study 1) and .89 (Study 2). The first scale item (This brand is dependable) is unclear and confusing when asked about an ice cream brand, therefore it was replaced with the statement: “This brand can deliver what it promises (e.g. good taste, premium ice cream)”. The questionnaire also included other possible statements to be an alternative of the first one. However, the above-mentioned statement showed a good correlation with the other two scale items and resulted in a good scale reliability, so it was the one used in computing the brand trust scale mean in the following chapter.
31
4. Results
In this chapter the data collected with the questionnaire will be statistically analyzed in order to explore the effect of perceived country of origin on brand attitude, brand trust and purchase intention. The chapter includes sample description, scale reliabilities, correlations and hypotheses testing.
4.1 Sample Description
In total the survey received 109 responses with a 13% dropout rate. After eliminating unfilled questionnaires or partial responses, we were left with 95 usable surveys.
They survey was distributed online mainly to students from the University of Amsterdam (UvA). UvA is an international university with over 3,000 international students coming from about 100 countries. Since it is important to ensure respondents with various national and cultural backgrounds, UvA is a suitable environment to carry out this research. The research used self-selection sampling technique. Respondents were invited to participate in the survey through various online communication channels that are currently popular among university students. The self-selection sampling technique means that respondents identify themselves as sample members (Saunders & Lewis, 2012, p. 140). The researcher invites people to participate in the survey by placing an advertisement announcing the research in a relevant media (in this case Facebook). This sampling technique is increasingly popular due to the variety of electronic media (blogs, social media, and discussion boards) however it has the disadvantage that people who select themselves to be part of the sample might have strong opinions about the researched topic or be particularly interested in it (Saunders & Lewis, 2012, p. 140). This means that they might not be representative of the population.
The majority of respondents were female (67%) and the average age was 25.4 years with median age 24. This was expected since most respondents were university students.
32
Most importantly, the sample included people with diverse cultural background. Respondents were nationals of 24 different countries including Austria, Bulgaria, Canada, China, Germany, the Netherlands, and the United States. The majority the respondents were from the Netherlands (40% indicated Dutch nationality), followed by 19% Bulgarians. Respondents were grouped into four regions based on their nationality: 8.8% are from Asia, 11% from North America, 29.7% from Central and Eastern Europe, and 50.5% from Western Europe. One respondent from Ecuador was not categorized in any of these regions as it was not statistically reasonable to make a group with only one respondent. Unfortunately, some respondents chose not to answer the question about their nationality
Table 2 shows the respondents’ distribution among regions and indicates which countries were included in each region. For a full overview of the respondents per country, please refer to Appendix 3.
Table 2: Respondents’ Nationality and Region
Region Number of Respondents
Percentage of Respondents
Countries Included in the Region
Asia 8 8.8% China, Indonesia, India, Iran,
Vietnam
North America 27 11% Canada, Mexico, the USA Central and
Eastern Europe
10 29.7% Bulgaria, Estonia, Hungary, Poland, Romania, Russia, Slovakia, Ukraine
Western Europe
46 50.5% Austria, Germany, Finland, France, the United Kingdom, Italy, Netherlands
4.2 Perceived Country of Origin
Respondents were asked to indicate whether they were familiar with each of the three ice cream brands prior to taking the survey. Only 4% of the survey participants were familiar with Perry’s ice cream, however 84% correctly identified the country of origin. A few people who did not indicate the correct country of brand origin believed that the brand comes from
33
the UK or Italy. The brand emphasizes its American origin and therefore even though the subjects of the questionnaire were mostly unfamiliar with the brand, their perception of the country of origin matches with the actual country of brand origin.
In the case of Häagen-Dazs, 59% of the respondents indicated they were familiar with the brand prior to taking the survey. Even though the majority of the respondents had previous experience with the brand, only 22% of the respondents correctly indicated the USA as a country of origin of Häagen-Dazs. However, 27% believed that the brand is from Germany while 13% thought it is Danish. Surprisingly 16% thought it is Dutch, perhaps misled by the fact that this is research made at the University of Amsterdam. The rest 23% mentioned other countries among which Austria, Sweden, and Switzerland. The difference in perceptions of country of origin is consistent with what we expected. Even though people are more familiar with this brand they are more confused about its origin as a result of the foreign branding strategy employed by Häagen-Dazs.
Baskin-Robbins was a familiar brand to 25% of the survey respondents. 57% of them indicated the correct country of brand origin - the USA, while 25% believed the brand is originating from the UK. The rest 18% had various ideas where the brand might be from such as Sweden, Finland, Iceland, and Spain. This brand served as a control condition in this survey. Unlike the other two brands, Baskin-Robbins neither emphasizes its American origin, nor uses foreign branding strategy. However, most survey participants still managed to identify the actual country of origin. Due to the English-sounding brand name respondents’ perceived countries of origin were mainly English speaking countries – the USA and the UK. Table 3 summarizes the information on respondents’ familiarity with the three ice cream brands and the accuracy of the COO perception.
34
Table 3: Perceived Country of Origin
Ice Cream Brand Percentage of Respondents Familiar with the Brand
Percentage of Respondents who Correctly Identified the COO
Perry’s Ice Cream 4% 84%
Häagen-Dazs 59% 22%
Baskin-Robbins 25% 57%
Regarding the importance of country of product origin, 45% of respondents consider country of origin information to be very important when buying durable good. Slightly less (44%) consider country of origin to be very important when buying food products. Only 9% of the survey participants think that COO is extremely important when purchasing durable goods, whereas 13% consider it extremely important in the case of food products. At the same time 31% of the respondents indicated that they rarely search for country of origin information of products they buy, while 33% look for this information sometimes. Only 7% of the research participants search for COO information every time.
4.3 Scale Reliabilities
Brand attitude and brand trust were measured using established scales previously used by marketing researchers. Before proceeding with analyzing the data, the scales reliabilities have to be evaluated. Cronbach’s alpha coefficient was used to check the scale reliability. Cronbach’s alpha is a measure of internal consistency and varies from 0 to 1. Value higher than 0.7 indicate a high internal consistency, thus a reliable scale.
Brand attitude was measured using a semantic differential scale including 6 items. The brand attitude scale has a good reliability with Cronbach’s alpha of 0.94 for Perry’s Ice Cream, 0.96 for Häagen-Dazs and 0.97 for Baskin-Robbins. The corrected item-total correlation indicates that all items have good correlation with the total score of the scales. The deletion of the fifth item (boring-interesting) in the Perry’s brand attitude measure will increase the Cronbach’s
35
alpha to 0.944 but this difference is less than 0.1 so the increase is negligible. Therefore the scale has a good degree of reliability without any items being deleted. As for Häagen-Dazs and Baskin-Robbins brand attitude measures, the deletion of none of the times will lead to an increase in the Cronbach’s alpha.
Brand trust was measure using a seven-point Likert scale including three items. The scale has a Cronbach's alpha of 0.85 for Perry’s Ice Cream, 0.94 for Häagen-Dazs and 0.92 for Baskin-Robbins. All Cronbach's alpha values are above the 0.7 threshold which means good degree of reliabilities for all scales. All items have corrected item-total correlation above 0.30 showing a good correlation with the total score of the scales. If the first item in the Perry’s brand trust measure is deleted, the alpha will increase from 0.847 to 0.855. If the same item is deleted in the Häagen-Dazs and Baskin-Robbins brand trust measures, the Cronbach's alphas would rise from 0.937 and 0.922 to 0.953 and 0.944 respectively. However, these differences are less than 0.1 therefore the deletion is not recommended or necessary. Consequently, we can conclude there is a good reliability of the questionnaire design. Table 4 summarizes the Cronbach's alpha values for the brand attitude scale and the brand trust scale for Perry’s Ice Cream, Häagen-Dazs and Baskin-Robbins.
Table 4: Reliability Statistics
Brand Attitude Scale
Cronbach's Alpha
Brand Trust Scale Cronbach's Alpha
Perry’s 0.941 0.847
Häagen-Dazs 0.960 0.937
36
4.4 Descriptive Statistics and Correlations
After the preliminary analysis of the scale reliabilities, the means of the scales were computed. Table 5 shows the means, standard deviations and correlations for the main variables of interest. It is clear that Häagen-Dazs has the highest mean brand attitude, brand trust and purchase intention. Häagen-Dazs is followed by Baskin-Robins in the brand attitude and brand trust measures, however not in respondents’ purchase intentions. Respondents showed slightly higher intentions to buy Perry’s Ice Cream compared to Baskin-Robbins.
Table 5: Means, Standard Deviations, Correlations
Age is not significantly correlated with any of the dependent variables. Gender is significantly correlated only with Häagen-Dazs brand attitude (r = 0.331), brand trust (r = 0.278) and purchase intentions (r = 0.3). The brand attitude, brand trust and purchase intentions for each of the three ice cream brands are significantly positively correlated with each other for each individual brand. This was anticipated because a more positive brand attitude will go along with higher brand trust and purchase intentions.
4.5 Hypotheses Testing: Repeated Measures ANOVA
In this section the first three hypotheses were tested using repeated measures ANOVA. Hypothesis 1 stated that the perception of a favorable country of brand origin has a positive
37
effect on customers’ attitude towards the brand. We tested this hypothesis by examining the mean brand attitude across three conditions: favorable perceived country of origin, unfavorable perceived country of origin and neutral perceived country of origin (or no country associations). Using repeated measures ANOVA the null hypothesis of no statistically significant differences between the means was explored. If the results show a significantly higher brand attitude in the favorable country image condition but no difference in the other two conditions, the hypothesis would be supported.
Brand Attitude
A repeated measure ANOVA is an appropriate test for this research because it explores the mean difference for within-subjects design. All participants in the survey answered questions about the three conditions of perceived country of origin; therefore we repeatedly measured the same subjects and want to analyze the mean difference in their brand attitude in the three scenarios. The dependent variable was brand attitude and the independent variable was country image (of the perceived country of origin) consisting of three categorical groups: negative image of the perceived country of origin (in the case of Perry’s Ice Cream), positive image of the perceived country of origin (Häagen-Dazs) and neutral image (Baskin-Robbins).
In order to perform repeated measures ANOVA, the data should meet the sphericity assumption meaning that the variances of the differences between all combinations of related groups are equal (Laerd Statistics, 2013). One way to test this assumption is by using Mauchly's test of sphericity in SPSS Statistics (Mauchly, 1940). The Mauchly’s test was not significant (p = .086) therefore the sphericity assumption is fulfilled. The mean brand attitude respondents indicated was 4.365 in the first condition (negative perceived country image), 5.514 in the second condition (positive image of the perceived COO) and 4.379 in the third condition (neutral). The repeated measures ANOVA showed that these means are
38
significantly different, F(2, 186) = 24.16, p<.0005. The pairwise comparisons with Bonferroni correction revealed that subjects had significantly higher brand attitude towards the positive perceived COO condition compared to the negative country image (mean difference = 1.149,
p < .0005) and compared to the neutral condition (mean difference = 1.135, p < .0005).
However, the brand attitude in the neutral and the negative conditions did not differ significantly. In other words, there is a statistically significant difference between the mean brand attitude towards the positive and the negative perceived COO and between the positive and the neutral perceived COO suggesting that respondents preferred the ice cream brand coming from a country with a favorable image. The results of the repeated measures ANOVA support hypothesis 1. Table 6 shows the pairwise comparisons of the three conditions.
Table 6: Brand Attitude Pairwise Comparisons
Country Image Mean
Difference Std. Error Sig.
95% Confidence Interval for Difference Lower Bound Upper Bound
Negative x Positive -1.149* .195 .000 -1.624 -.674
Negative x Neutral -.014 .168 1.000 -.423 .395
Positive x Neutral 1.135* .205 .000 .636 1.634
Brand Trust
Hypothesis 2 proposed that customers have a higher brand trust towards brands they perceive as origination from a country with a favorable image. Repeated measures ANOVA was used once again to test whether there is a significant difference in the brand trust means across the three conditions of country image. The mean brand trust was 5.47 in the positive country image condition, 4.33 in the negative and 4.44 in the neutral. The Mauchly’s sphericity test was not significant (p = .651) therefore the sphericity assumption is met (Mauchly, 1940). The repeated measures ANOVA revealed a significant difference in the means of brand trust
39
across the three ice cream brands that represent the three conditions, F(2, 186) = 28.29, p < .0005. The post hoc test indicated there was a significant difference between the positive and negative condition (mean difference=1.144, p < .0005), as well as between the positive and neutral (mean difference = 1.034, p < .0005). There was no statistically significant difference between the brand trust in the negative and neutral condition. These results are similar to the results obtained for brand attitude. The analysis showed that respondents had a significantly higher brand trust towards the brand which they perceived as originating from a country with positive image. These results are consistent with the proposition of Hypothesis 2. The pairwise comparisons are presented in table 7.
Table 7: Brand Trust Pairwise Comparisons
Country Image Mean
Difference Std. Error Sig.
95% Confidence Interval for Difference Lower Bound Upper Bound
Negative x Positive -1.144* .173 .000 -1.565 -.722
Negative x Neutral -.110 .160 1.000 -.499 .279
Positive x Neutral 1.034* .170 .000 .618 1.449
Purchase Intentions
Hypothesis 3 suggests that perceived COO has an effect on purchase intentions in a way that favorable image of the perceived COO leads to a higher likelihood of purchasing the product. In order to test this hypothesis the same procedure was performed. Mauchly's test of sphericity was not significant (p = .801) thus the sphericity assumption is met once again. The mean values for purchase intentions were 5.39 for Häagen-Dazs, 4.26 for Perry’s and 4.19 for Baskin-Robbins. The repeated measures ANOVA showed that the difference in the mean purchase intentions is significant F(2, 186) = 17.49, p < .0005. Similarly to the results for brand attitude and brand trust, the pairwise comparisons indicated that there was a significant mean difference between respondents purchase intentions in the positive and negative country image condition (mean difference =1.138, p < .0005), as well as in the positive and neutral
