Thesis summary Business administration: Boycotting behavior in general product - A case of research in Vietnam

Chia sẻ: _ _ | Ngày: | Loại File: PDF | Số trang:25

Thêm vào BST

Báo xấu

25
lượt xem 1
download

Download Vui lòng tải xuống để xem tài liệu đầy đủ

This study explores the nature of the boycott and its effects on the Vietnamese boycott behavior of China product. The thesis compares the similarities and differences of the components discovered with previous studies. At the same time, this study proposes a theoretical model and verification of the factors that influence the boycotting behavior.

Chủ đề:

Bình luận(0) Đăng nhập để gửi bình luận!

Lưu

Nội dung Text: Thesis summary Business administration: Boycotting behavior in general product - A case of research in Vietnam

MINISTRY OF EDUCATION AND TRAINING UNIVERSITY OF ECONOMICS HCMC Cao Quoc Viet BOYCOTTING BEHAVIOR IN GENERAL PRODUCT: A CASE OF RESEARCH IN VIETNAM Major: BUSINESS ADMINISTRATION Code: 9340101 THESIS SUMMARY Ho Chi Minh City – 2018
The thesis was completed in University of Economics Ho Chi Minh City Tutor: Asso. Prof. Ph.D. Vo Thi Quy Opponent 1: ................................................... …… Opponent 2: ........................................................... Opponent 3: ........................................................... The thesis will be protected in front of the Council at: At ………. hour ……… …… ……………….. 2018 The thesis can be found at: …………….……………………………………
1 CHAPTER 1. OVERALL INTRODUCTION 1.1. Research Motivations A comprehensive understanding of consumer behavior in international business plays an important role not only for multinational companies (MNCs) but also for domestic enterprises. Amine (2008) recommends that three prerequisites for a successful manager in each market they invade. First, the manager must have a detailed knowledge of the current and historical relations of that country with bilateral trading partners to draw conclusions about consumer behavior. Secondly, managers must constantly observe changes in consumer perceptions and their behaviors in the target market. Third, the manager must plan a regular and long-term market research to monitor the continuous changes in perceptions, emotions, and consumer behaviors. When a country triggers events in another country, consumers in the affected country view the events from different perspectives, and they will have different emotions; affecting their consumption behavior (Harmeling et al. 2015). International disputes, cultural, political and economic conflicts are events that one country causes to another. This negatively affects consumer sentiment through negative emotional states such as anger, frustration, sadness, anxiety or confusion. These emotions cause the corresponding negative consumption behavior of the product originating in the country causing the conflict. The boycott is one of the negative consumer behaviors that companies do not want to encounter. Therefore, understanding the boycotting behavior of consumers and the factors that influence the boycotting behavior is a matter that needs to be studied seriously and thoroughly. For many years, the boycott of goods originated in a certain country of consumers has been a concerned problem not only in Vietnam but also spread throughout the world. In Vietnam, the boycott of Chinese goods excels and attracts public opinion because of poor quality goods, toxic substances. Especially, the HD 981 oilrig was placed illegally by China on the continental shelf of Vietnam in April 2014. This action has led to the urgent, anger of a large number of Vietnamese people. They have boycotted the Chinese product to show their patriotism. Consequently, the boycott behavior of consumers is a phenomenon that has already existed in the market. The study of the nature of the boycott and the factors that influence the boycotting behavior need to be clarified. Findings may be significant evidence in behavioral science. The research will have many valuable implications for managerial practice. 1.2. Research context This thesis uses negative events to infer the relationship between consumer perceptions, emotions, and boycotts of goods originating from the country of events. China is the chosen country because of the negative historical events that have taken place throughout China's thousands of years of aggression and domination. The Chinese government's economic events (e.g., trade deficit, currency devaluation, etc.) are also negative events, which in turn lead to boycott China products. For China products, the boycott of consumers in Vietnam excels and attracts public opinion and is reflected through mass media because the quality of the product is poor, contains many toxins from vegetables, roots, and fruits for food, clothing, equipment, and machinery. Since July 2012, since the establishment of Tam Sa in the Paracel archipelago of Vietnam. Especially, the HD 981 rig event set up in the Vietnamese territorial waters in April 2014 by the Chinese government. This action has led to the urgent, anger of a large number of Vietnamese people lead to call for boycotting China products to show their patriotism. Consequently, the boycott behavior of consumers is a phenomenon that has already existed in the market. The study of what the nature of the boycott is, what components are included, what factors influence the boycotting behavior should be clarified as the results of the study may be significant evidence in the behavior field. This thesis brings some important implications for management practices. 1.3. Problem Statement Previous researchers used the concept of consumer ethnocentrism to refer to an independent variable influencing the boycotting behavior and explaining this relationship based on social identity theory. The term
2 ethnocentrism has been studied for a long time, eg (Sumner, 1906; Shimp and Sharma, 1987). Shimp and Sharma (1987) are the authors who build the CETSCALE scale for measuring the ethnocentric of consumers. Since 1987, much of the replicated, revised and expanded research that has been interested in researchers because of the controversial and conflicting results. Klein et al. (1998) proposed the concept of consumer animosity and testified about the relationship between the construct and the boycott behavior. Klein et al. (1998) for the first time published the evidence explaining the boycott of Japan product of Chinese consumers. Klein's research is based on the refusal to use Japanese goods because the people of Nanjing are still angry at the aggression and massacre of hundreds of thousands of Japanese civilians during the 1937 war. Similar to consumer ethnocentrism, consumer animosity plays an important role in the explaining for the boycotting behavior. However, the findings are inconsistent with the research. At the time of the study, no research in Vietnam was published that provided scientific evidence to clarify these relationships. Some previous studies in the world have focused on the boycott of China product. Repeated research by Witkowki (2000) dismisses previous research by Klein et al. (1998), as there is no evidence to support the relationship between economic animosity and boycotting behavior. However, Huang et al. (2010a) showed that economic animosity has a positive effect on the boycott of China product in Taiwan. Lee and Lee (2013) found evidence to support the consumer animosity and the boycott of Japanese consumers toward China goods. However, repeated studies six months later did not confirm this finding. More recently, the study by Garcia- de-Frutos and Manuel Ortega-Egea (2015) in the Spanish has resulted in a similar relationship between the Spanish consumer animosities towards the boycotting behavior of China product. Thus, repeated and extended studies showed the important role played by the relationship between the animosity factors and the boycotting behavior. There is, however, a lack of consistency in the results of research between different markets for the factors. 1.5. Research Objectives This study explores the nature of the boycott and its effects on the Vietnamese boycott behavior of China product. The thesis compares the similarities and differences of the components discovered with previous studies. At the same time, this study proposes a theoretical model and verification of the factors that influence the boycotting behavior. Specific research objectives are as follows: The first objective: to build and develop the components of the boycotting behavior. The second objective: to investigate the impact of patriotism on the animosity beliefs. Determine the indirect impact of patriotism on the product quality judgment through consumer ethnocentrism; to examine the indirect impact of consumer ethnocentrism and the boycotting behavior through product quality judgment. The third objective: to build, develop and measure the elements of the animosity beliefs; examine the indirect impact of these factors on the boycotting behavior through negative emotions. Explore the impact of animosity beliefs on consumer apathy; to explore and test the impact of patriotism on the boycotting behavior through consumer apathy, and discover the moderating role of individualism/collectivism and the factors (gender, education level, income, age) on the relationship encourage the boycotting behavior. 1.6. Research subjects, research scope, and research methods The research subject of the thesis is the boycotting behavior of consumers for goods originating from China. China product is any commodity produced in China. Research scope: The dissertation has focused on individual consumers living in Ho Chi Minh City. The study was conducted from 2014 to 2016 with three main phases. Stage one (early 2014 - end of 2015): literature review, qualitative research. Phase two (early 2016): expert research. Phase three (second half of 2016): quantitative research. To increase the generality of the theory, this study did not focus on a particular commodity. All phrases such as products or merchandise used in the research and in the main survey were general products.
3 The mixed method was designed in two main phases. The qualitative phase used the method of netnography, combining grounded theory to explore the categories. The result of this phase was to propose a new, exploratory research model in comparison with the previously studied categories. The quantitative research phase included expert research, preliminary quantitative research, and formal quantitative research. Expert research helped to evaluate and select appropriate construct and items in the proposed research model. After this period, new components and adjusted scales were included in the pilot test for quantitative research at the next stage. After that, the thesis evaluated the conditions related to the verification of the measurement model, the multivariate analysis method using the structural equation modeling (SEM) technique for the purpose of the verification the proposed research model. To test the moderating role, the multi-group analysis method was used to compare the level of impact of the model on two groups of consumers based on the level of individualism/collectivism, male/female, low/high income, education level. 1.7. Value, new contribution in theory and practical implications The prior research considered boycotting behavior is unidimensional construct with a representative component of the willingness to boycott. This study explored and demonstrated that boycotting behavior was a multidimensional construct consisting of two components (i.e, willingness to boycott and boycott calls). The exploration of the components of the boycotting behavior and proven reliability, as well as validity, contribute to the measurement theory. In addition, the relationship of consumer animosity independently or adversely affects product quality judgment originating from the offended country also contributed meaningfully to the theory. This study supported the new approach of Harmeling et al. (2015) based on cognitive appraisals of emotions theory. The thesis separated the concept of consumer animosity into two components, including cognitive appraisal and negative emotions. Different negative emotions that result from various cognitive evaluations will lead to different behavioral responses. The thesis explored an emotional component of the original research model. This emotional component contributes to the relationship between animosity beliefs and product quality judgment. Furthermore, the process of cognitive appraisals in the Harmeling research model was based solely on the political and war conflict between the two countries. This study added the content of economic animosity beliefs and relationship animosity beliefs. Qualitative results showed that patriotism and apathy are two new components that contribute to the expansion of the boycotting behavior model. This result was verified by quantitative research that contributed to the extension of the theoretical model of the boycott. In practice, exploring the components involved in boycotting is very important and meaningful not only to multinational corporations (MNCs), but to local companies. Studying the motives behind the boycott will help managers orient their business strategies and marketing their products and services. Managers of MNCs need to anticipate the risks from the markets in which they operate. The basic risks in international business may come from the macro environment related to politics, culture, and economics, which risks altering consumer behavior. In the context of globalization, MNCs have multiple manufacturing divisions in different countries, identifying the level of consumer boycotting for the country, triggering the event will help MNCs set the risk response strategy for boycotts. Animosity beliefs, negative emotions are two important factors that explain the product quality judgment originating in the country that caused the event. If a country produces goods negatively rated by the consumer, this will indirectly cause damage to those companies. Thus, the negative emotions of consumers can harm the brand of a business. However, the reasons for these negative emotional reactions do not come from the businesses themselves. Comprehensive understanding of the level of consumer behavior in a country in terms of cognitive appraisals, emotions and their behaviors is very significant for MNCs and domestic companies. Because they can anticipate the dangers, that lead to boycotting the goods and services provided in a target market. Examining the nature of patriotism and the relationship between patriotism and other factors in the pattern of boycotting behavior helps managers to plan business strategies (i.e., marketing, and branding) based
4 on the consumer patriotism. In the context of the integration into the world economy of our country, tax barriers will gradually be removed and technical barriers will gradually shift to a unified common standard between nations. Therefore, the success of domestic enterprises will depend greatly on the support of patriotic consumers. Understanding the influence of patriotism on consumer behavior will have significant implications for managers and policymakers. CHAPTER 2. LITERATURE REVIEW 2.1 Consumer Boycotting Behavior Boycotting behavior is the term proposed by Kozinets and Handelman (1998), derived from the term consumer boycott proposed by Friedman (1985), and operationalized under the construct of willingness to boycott (Abosag and Farah, 2014; Abdul-Talib et al., 2016). The boycotting behavior is a collective act (Friedman, 1985; Garrett, 1987; Yuksel and Mryteza, 2009; Shebil et al., 2011). It may also be a personal act of an individual in society (Kozinets and Handelman, 1998). Boycotting behavior is an action to call on others not to buy the product of the boycotted object (Friedman, 1985). It is the act of refusing, abandoning, or abstaining from consumption of goods is discredited (John and Klein, 2003, Shebil et al., 2011, Yuksel and Mryteza 2009). Boycotting behavior is the rejection of marketing activities (Garret 1987) while marketing is just one chain in the value chain of a business. Conclusions from the concepts of previous authors show that there is no definitive and consistent definition of consumer boycotts. A comprehensive and integrated proposed boycotting behavior definition for this thesis as follows: “Consumer boycotting behavior is personal or collective intended acts for one or more groups of people to give up or call on consumers to give up; as opposed to consuming products of one or more organizations that are alleged to have caused adverse or wrongdoing to them. " Thus, the task of this study has been to develop a proposed boycott behavior for Vietnamese consumer. 2.2 Foundation Theories Theoretical summation shows five major theories. Theories used include (1) Cognitive Appraisal of Emotions Theory; (2) Attribution Theory; (3) Cognitive Dissonance Theory; (4) Social Identity Theory and (5) Bystander Effect and Diffusion of Responsibility Theory. A set of five theoretical backgrounds on the formation of the theoretical framework used in this thesis. The theoretical framework used for the thesis is illustrated in Figure 2.1 as follows: Social Identity Theory Consumers’ opinions/views Negative Product Boycott Cognitive Emotions Judgment Behavior Negative Events Cognitive Appraisal of Emotions Theory Attribution Theory Cognitive Dissonance Theory Bystander Effect and Diffusion of Responsibility Theory
1 Figure 2.1: Theoretical Framework for the Study of the Boycott Behavior. 2.2.1. Cognitive Appraisal of Emotions Theory The cognitive appraisal of emotions is also called Lazarus’s theory, a theory of cognitive psychology. The theory of Lazarus presents the relationship between cognition and emotion. Lazarus (1982) states that cognition and emotion are interconnected spontaneously. A cognitive appraisal is a way that people explain when they fall into a certain situation at a given moment and this appraisal determines their emotional response. Lazarus (1982) concludes that cognitive appraisal is a necessary and sufficient condition for emotions. Harmeling et al. (2015) based on the cognitive appraisal of emotions theory to explain the relationship between constructs in their proposed research model. The Harmeling team used the context of the Senkaku / Diaoyu Islands dispute between Japan and China as compared to the political conflicts between Russia and the United States. The authors suggested that this might be the appropriate event to test and expand the research model related to the concept of consumer animosity previously proposed by Klein et al. (1998). When the two countries have the conflict (ex., in the past/present), whether people will display their animosity and they lead them to (1) negative product quality judgment of offending country or not. Moreover (2) does this animosity lead to the refusing to buy goods of the consumer from that country? Harmeling et al. (2015) argued that previous researchers have combined the cognitive and emotional components in the same construct as consumer animosity. This aggregation led to inconsistent results as researchers responded to the above questions. The Harmeling and partners recognized that in the context of international disputes between the two countries, the cognitive appraisal of a person in a country of negative events (ex., economics, politics, and war) would lead to their negative emotional reactions. The deduction is in line with Lazarus’s theory. These different emotional responses will lead to different behaviors. 2.2.2. Attribution Theory Ang et al. (2004); Leong et al. (2008); Mrad et al. (2014) grounded on Weiner's attribution theory to deduct the relationships involving boycotting behavior. The Ang and Leong groups were based on the attribution to categorize events. Ang et al. (2004) classified animosity into four groups named stable animosity, situational animosity, personal animosity and national animosity. Where the stability animosity refers to negative emotions stemming from historical contexts such as economic or military interstate, situational animosity refers to the negative emotions related to a particular situation that has just occurred. Mrad et al. (2014) deducted their model in light of attribution theory, but they have an approach done under a different epistemology. When consumers in a country judge an offended country that causes harm to them, consumers will show their anger. They will refuse to buy goods from this country because they think their actions are helping the country. In addition, when consumers experience an event that can be controlled by an organization (a company/a country), their anger will increase. At this moment, anger plays a mediating role in the relationship between their animosity and their boycotting behavior. 2.2.3. Cognitive Dissonance Theory Shoham et al. (2006); Harmeling et al. (2015) applied cognitive dissonance theory to explain the relationship between consumer animosity, product quality judgment originating in the offended country and boycotting behavior. Festinger's (1957) theory suggests that if every individual has discrepant attitudes, or inconsistent behavior and attitudes, each individual experiences an uneasy roof, an unpleasant situation. This discomfort is called cognitive dissonance. Conflict motivates people to change to a reconcilable state which it has no discord. To achieve this psychological equilibrium, individuals must change their attitudes or their behaviors. Because the buying process is a decision-making process, a consumer decides to buy a product or service; he or she will experience a cognitive dissonance. Directly related to boycotting behavior, Shoham et al. (2006) found that consumers who felt animosity to the negative situation of a country (i.e., conflict) would have a high willingness to boycott toward product from this country. On the other hand, the higher their perception of the quality of the goods of the conflicting country the more they will be willing to buy. This causes inconsistency in an individual’s perception and he or she has to return to a harmonious state. To get to this state, they may judge negatively the product quality of
2 the conflict country. This means that the higher the animosity of the consumer, the lower the quality of the product. Results of Shoham et al. (2006) confirmed the relationship, however, some later studies have been inconsistent (e.g., Maher and Mady, 2010; Lee and Lee, 2013). For that reason, Harmeling et al. (2015) had a different deduction in the comparison with the Shoham group when applying cognitive dissonance theory. Harmeling et al. (2015) have suggested that only consumer's fear and anxiety negatively affect the product quality judgments. Because high emotional anxiety people are those who do not cope with the event as angry people, they will regress in front of the situation and focus their attention on avoiding risks and uncertainties. These consumers are deliberate, careful thinking about the threats. Thus, the fear of avoiding products from the conflicting country, and similar to Shoham’s argument, they will evaluate the decreasing quality product of the country causing anxiety for them. The Harmeling and colleagues’ findings were confirmed; they proposed that similar basic emotions such as shame, sadness should be tested in another context. 2.2.4. Social Identity Theory The Social Identity Theory proposed by Tajfel and Turner (1979) derives from studies related to social categorization, intergroup relations, and prejudice. The central philosophy of social identity is that people feel a desire and inclination to build themselves a positive identity that can be demonstrated by their identification with different groups followed by many ways (Tajfel, 1981). Individuals in a society classify themselves into groups which identify themselves through something of their own identity and distinguish themselves from outgroups. People share the same views, behaviors, and people within their group. They will move outside the group in a collaborative, conflicting, stereotyped, or a heterogeneous way through a process that Tajfel (1982) called social comparison. Concerning the field of consumer behavior, many scholars have borrowed social identity and rhetoric concepts to explain consumer behavior in different countries. Shimp and Sharma (1987, p. 280) noted that “From the perspective of ethnocentric consumers, purchasing imported products is wrong because, in their minds, it hurts the domestic economy, causes loss of jobs, and is plainly unpatriotic; products from other countries (i.e., outgroups) are objects of contempt to highly ethnocentric consumers.” Herche (1994, p6) pointed out, “ethnocentric consumers will tend to reject people, symbols, and values that are culturally dissimilar, while intra-cultural objects will become recipients of pride and attachment.” Lantz et al. (1996) argued that a person might make reasonable judgments for the consumption of domestic goods because either that would bring collective strength to the national economy or it could be morality. Social identity can also explain the relationship between the concepts of nationalism and patriotism to consumer ethnocentrism. Because the high nationalist groups are more competitive, aggressive, aggressive and biased towards other nations and peoples. Patriotism, meanwhile, refers to the patriotic attitude of a person who is willing to sacrifice his country, attaching personal interests to national interests (Druckman, 1994). Patriots will find themselves responsible for protecting the economy and helping domestic producers (Balabanis et al., 2001). 2.2.5. Bystander Effect and Diffusion of Responsibility Theory Darley and Latané (1968), Latané and Darley (1968, 1969) are the first authors to explain the phenomena of indifference and dispersion of human responsibility. Latané and Darley (1969) conducted a number of psychological experiments and concluded that humans were only willing to help others in emergencies if five conditions were met. First, they pay attention to the situation. Second, they infer it urgent. Third, they make assumptions about responsibility. Fourth, they know how to assist and eventually, they know they can make the decision to take help. When an emergency occurs, the more people witness, the more difficult they are to support. As many people witnessed, fewer people decided to help the victims. The reason for this is the diffusion of responsibility. Observers of a person’s accident situation will argue that there will be other people helping them so that they stand to watch or walk away. In the case of only one observer, they will ignore it if they think the responsibility to help is not relevant to them (Darley and Latané, 1968). This creates a bystander effect or bystander apathy. Since the theory of indifference and diffusion of responsibility has been established, many researchers have used it to account for a wide range of other issues in society. In terms of consumer behavior, Chan (1999,
3 2000) explored consumer attitudes toward the environment and found that these groups were less likely to buy environmentally friendly products than those who care about their environment. Directly linked to the boycott, Klein et al. (2004) used the theory to explain the boycott behavior of European consumers. As a European company closed down and relocated its factory, many consumer protesters boycotted the company because of the loss of jobs for workers. According to Klein et al. (2004), those who opposed the boycott might argue that others have boycotted, that their participation was unnecessary (i.e., diffusion of responsibility). Their participation in the boycott also causes more serious job loss. In this thesis, the bystander effect and diffusion of responsibility theory has been used to explain the relationship between consumer apathy and their indifference to national events. They do not care about politics; they do not care about the negative events that happen every day. Therefore, they will consume any goods according to their needs; their boycott behavior may be reduced. 2.3. The relationship between consumer ethnocentrism and boycotting behavior Klein et al (1998) found that Chinese consumer ethnocentrism toward Japan and the boycotting behavior Japanese products are related positively. This result was accepted by researchers (Ang et al., 2004; Russell and Russell, 2006; Nakos and Hajidimitriou, 2007; Ishii, 2009; Rose et al., 2009; Huang et al. Cui et al., 2012; Wang et al., 2013). However, some other results did not provide statistically significant evidence to accept this relationship (Ettenson and Klein, 2005; Maher et al., 2010; Russell et al., 2011; Ma et al., 2012). 2.4. The relationship between consumer animosity and boycotting behavior The inconsistency in the definitions of consumer animosity is research gaps, which have been identified. Leong et al. (2008) based on the classification of Ang et al. (2004) argues that consumer animosity may be a construct of two components: situational animosity and stable animosity. Hoffmann et al. (2011) commented on the model of Klein et al. (1998) is too limited because of its generalization in a multinational context. The authors argued that the general animosity is not only derived from the conflicts of war, politics, and economics, but also it comes from the perceived threat, antithetical political attitudes, and negative personal experiences. These factors may affect directly the general animosity and indirectly to the boycotting behavior. Little and Singh (2015) accepted the definition of Klein et al. (1998), but the authors suggested that consumer animosity of a country should not be confined to another country in the context of a war, economic or military conflict. It should be widened for any events. Most previous authors have considered consumer animosity under a concept that exists under a variety of different constituents, including war animosity, military animosity, economic animosity, ethnic animosity. The events that create that animosity are war, military conflict, and ethnicity. Findings showed that the positive relationship between consumer animosity and boycotting behavior toward the products of a hostile country. However, there have still been many studies that did not give statistical evidence to accept this effect (e.g., Nijssen and Douglas, 2004; Nakos and Hajidimitriou, 2007; Leong et al., 2008; Chan et al. Maher et al., 2010; Hoffmann et al., 2011; Ma et al., 2012; Mrad et al., 2014). The inconsistency in the results suggests that there is a need to re-evaluate the relationship from either another perspective on the method or measurement. Thus, Harmeling et al. (2015) based on the cognitive appraisal of emotions theory (Lazarus and Folkman, 1985; Roseman, 1996) to propose the new concept of consumer animosity approached from the perspective of animosity beliefs. Harmeling et al. (2015, p. 4-5) stated that: “In the animosity context, the cognitive appraisal refers to the consumer’s beliefs about the extent of damage and/or potential future threat attributable to the offending country. We refer to such cognitive appraisals as animosity beliefs.” CHAPTER 3. RESEARCH METHODOLOGY – QUALITATIVE RESEACH AND FINDINGS 3.1. Research Process The research process for the thesis was designed through two stages of research. Specifically, in the first phase, the qualitative research was done with the combination of two commonly used methods in the
4 marketing and behavioral field (Netnography and Grounded Theory). The results of the first phase were the proposed research model with emerging components. In the second phase, new components were developed to scale and refine the content validity through expert research. A preliminary quantitative study was done for assessing the reliability and validity of the constructs. The formal quantitative study was conducted at the next step to evaluate the measurement model and testing the proposed research model. 3.2. Qualitative research and findings Applying previous measurement scales to a new research context without adjustments will result in the inappropriate findings (Riefler and Diamondman, 2007). Thus, the qualitative approach in the first phase of the study has two main objectives: (1) assessing fit, adjusting scales and models; (2) exploring new components (Debus, 1988). Whereas the quantitative approach is used for verifying the proposed model. For qualitative research, this study applied the Netnography method proposed by (Kozinets, 2002, 2010, 2015) to explore consumer cognitive appraisals, their emotions, and their behaviors. The thesis has analyzed the content of 240 comments on the topic “boycott Chinese goods are negative patriotic” on dantri.com.vn. This e- newspaper was chosen because it was the website that opened the comments section under each article so readers could read and showed their views and attitudes. To confirm the trustworthiness of the qualitative research, the thesis diversified the approach by combining content analysis from group discussions with two groups of undergraduate students in the third year; a graduate student group, and in-depth interview with five non-students consumers. Qualitative samples were collected in a convenient manner based on their willingness to participate in interviews with invited people. The sample size was based on the level of information saturation after the coding of keywords, grouping the content into the categories of the interview recording according to Yin, (2010). The content analysis of 240 comments showed that 14 components have emerged in which the content was consistent with the construct of the willingness to boycott -WTB. In addition, the new composition has been entitled “boycott call – Bcall”. In the comparison with the previous definitions of boycotting behavior, the boycott call is the term that has been mentioned in Friedman (1985, 1991). However, the term refers to the general definition of the concept of boycotts and Friedman's definition refers to the behavior of a group of people rather than individual behavior. Qualitative results have provided evidence that calling for a boycott is a newly emerging component with a rich and varied content reflecting the description of the individual boycotting behavior. They call, suggest, urge others and organizations to boycott. Thus, if the act of boycotting is personal behavior (Kozinets & Handelman, 1998), the act of calling for a boycott is an individual act and the behavior has distinct characteristics with the willingness to boycott or participate in the boycott. With this critical argument, the author of the thesis proposed boycott call could have been a contributing component to the boycotting behavior. It exists independently of willingness to boycott. The ingredients could have a negative impact on boycotting behavior (i.e., boycott call and willingness to boycott) including product quality judgment, apathy, counterarguments. Factors could have the positive impact on boycotting behavior including perceived risk; three negative emotions (i.e., anger - AE, disgust - DG, worry - WE). The indirect positive factors influencing the boycotting behavior include current relationship animosity beliefs, historical relationship animosity, economic animosity beliefs, patriotism, and consumer ethnocentrism. As shown in the 14 components above, new components compared to previous studies were identified as boycott call, apathy, disgust, current relationship animosity beliefs, and historical relationship animosity. In order to avoid the complexity of the measurement, the composition of the current relationship animosity beliefs, historical relationship animosity was proposed as relationship animosity beliefs - RA. The proposed formal research model is composed of patriotism - PA, relationship animosity beliefs - RA, economic animosity beliefs - EA, negative emotions (i.e., anger, disgust, worry), product quality judgment - PJ, consumer ethnocentrism - CET, boycotting behavior - BB (including willingness to boycott - WTB and boycott call - BCall). As the result, the proposed model of official study has been adjusted after the qualitative research phase as follows:
5 CHAPTER 4: QUALITATIVE RESEARCH 4.1. Expert Research and Findings The expert research was conducted to assess the suitability of proposed research models, construct and scales. The procedure was based on the guidelines of Hardesty and Bearden (2004) for recording the results of experts’ assessment of the proposed definition in terms of generalization, ambiguity, value, and the content validity of each construct. To ensure objectivity, the previous constructs of the boycotts were encoded as A, B, and C and compared with the definition D in this study. In the next step, the statements of each construct related to the willingness to boycott, participate in the boycott, boycott call labeled by experts. Experts were unaware of the items of constructs; they only knew the list of statements and then based on the content of the definition to infer the suitable items in accordance with given label. This means that the expert will judge the conceptual content from the given statements attached to the label of the construct based on its definition. Twenty experts are lecturers and research majors in management, marketing who have experience teaching, and research were invited. Fourteen experts agreed to participate in the study. Findings Based on 75% of the consensus of experts, the items of the construct willingness to boycott to be assimilated to its label. Experts have coined the additional content of the new component (i.e., boycott call). Thus, it is proved that the components of the boycott call satisfied the content validity, face validity, and to be eligible for further evaluation at next steps. Qualitative research findings have also shown that the keywords expressing feelings of disgust emotion have been mentioned in studies of cognitive theories of emotion (Horberg et al., 2009; Smith and Ellsworth, 1985). The results confirm that boycott call (BCall) is a new component developed by the thesis on exploratory research and expert research. The new component consists of four main themes that call for others to boycott Chinese-made goods, urging businesses not to buy Chinese goods or call for a boycott of Chinese goods. The measurement of the relationship animosity beliefs was developed on the previous scale of Harmeling et al. (2015); six additional items come from exploratory and expert research. Other scales are reused from prior studies. 4.2. Pilot Test and Findings A pilot study was conducted on 238 responses for preliminary evaluation of the measurement scale. Scales were evaluated for reliability and validity through the Cronbach Alpha and the exploratory factor analysis - EFA. Results showed that the scales are quite suitable and qualify for the formal testing in the next stages. Thus, after eliminating two items of the IVD scale, the total number of items for the formal quantitative study was 52 observed variables.
6 4.3. Formal Quantitative Research The official quantitative study was conducted on a sample of 1138 respondents, collected through a direct survey. The main survey was approached with the non-probability method. James and Sonner (2001) proposed that part-time students should be used as a viable replacement for consumers. Peterson and Merunka (2014) found that the use of students to test theories in marketing and consumer behavior was an indispensable and unconstrained trend. Based on this, the thesis approached the part-time class to conduct the survey and then used the snowball sampling method. Each student was asked to take 5 more questionnaires from their friends, their colleagues, and their relatives in order to diversify respondents. The snowball sampling method has been used by many authors around the world (e.g., Maher and Mady, 2010; Cleveland et al., 2014). Scales were re-evaluated for reliability, validity, and final modeling through structural equation modeling technique - SEM. CHAPTER 5. FORMAL QUANTITATIVE RESEARCH FINDINGS 5.1. Descriptive statistics After two months of collecting samples, 1,500 questionnaires were distributed and 1336 questionnaires were collected (the response rate was about 89 percent). After eliminating inadequate information sheets (> 10 percent) and abnormal responses, the number of official surveys used for the study was 1138. Approximately 34 percent of respondents were males, while 66 percent of respondents were females. About 60 percent of respondents got an intermediate or lower level of education, 40 percent of respondents had the undergraduate level or above. The age group of 20 to 29 accounted for 77 percent of the sample. Nearly 30 percent of surveyed people were aged 30 or over. About 50 percent of respondent’s income ranged from 5 to 10 million (VND), 36 percent each belonged to an income group of below VND 5 million and nearly 15 percent of respondents had an income of VND 10 million or more. 5.2. Scale Reliability and Scale Validity 5.2.1. Scale Reliability 12 scales with 52 observational variables were evaluated for reliability. The results showed that all variables had the Cronbach Alpha index, ranging from 0.72 to 0.92. The corrected correlation coefficient item - total of all items were greater than 0.3. In short, all the observational variables were reliable so that they could be analyzed in the next steps. 5.2.2 Scales Validity For validity evaluating via EFA, the observation variables had a much lower factor coefficient than 0.4, which was not satisfactory for the exploratory factor analysis, such as RA4, RA8, and PJ4. Item EA4 and WTB5 were less than 0.4, but in terms of content validity, two variables have an important role in its contents that have been retained for CFA analysis. If the CFA asserts that these two variables were unsatisfactory, the EA4 and WTB5 would be eliminated after completion of the CFA evaluation in the saturated model. Cronbach Alpha results were evaluated again for scales, including RA (item 4 and 8), PJ (PJ4) and IVD (IVD3). The results showed that the Cronbach Alpha indexes met reliability requirements (ranged from 0.72 to 0.92). In summary, the EFA results were satisfactory for subsequent analyzes. 5.2.2.1 Testing of Multivariate Normality Before testing the CFA, observational variables were evaluated for skewness and kurtosis to test the normality assumption. The analysis showed that the variables had skewness and kurtosis index in the suitable range. Although the variables were approximately standardized distributions (in practice, it is so rare to find the standardized normal distribution), they were eligible to choose the maximum likelihood (ML) estimation method. 5.2.2.2. Confirmatory Factor Analysis – Multidimension Measurement Model The CFA results showed that two items (WTB2 and WTB5) had standardized weights of less than 0.5, which were excluded from the scale. The model fit indexes such as GFI, TLI, CFI were good. The remaining groups were evaluated for CFA. The third running result eliminated variables, which had the standardized
7 loadings of less than 0.5 (including RA1, RA3, RA7, and EA4). Other indexes such as TLI, CFI, and RMSEA have shown a fairly good fit with the data. 5.2.2.3. Saturated CFA Model The saturated CFA model was evaluated with all interrelated freedom variables, it also means that willingness to boycott (WTB) and boycott call (BCall) were tested in the model. The standardized CFA results showed that the saturated model had good indices (TLI, CFI > 0.93 and RMSEA < 0.4). Thus, the unidimensionality of the scales had no potential threat in the model. However, except for the patriotism (PA) scale because of the error correlation between the PA1 and PA2. 5.2.2.4. Examination of Composite Reliability (CR), Cronbach Alpha and Total Variance Explained (AVE) Composite Reliability and Cronbach Alpha results showed that measurement scales had composite reliability and Cronbach Alpha were ranged from 0.72 to 0.92. For total variance explained, the scales exceed the threshold of 50%, except for the WTB and CET scale that was less than 50% (0.48 and 0.45). Thus, it can be concluded that the measurement scales in the model could not get the reliability and validity perfectly. 5.2.2.5. Examination of Convergent Validity Criteria for convergent validity were all standardized indicators loadings are greater than 0.5 and statistically significant (Gerbing and Anderson, 1988). This is true for all scales, and this can support the convergent validity. 5.2.2.6. Examination of Discriminant Validity The discriminant validity was evaluated by the difference of correlation coefficient test in the comparison with 1 (Steenkamp and van Trijp, 1991). Results showed that the correlation between coefficients and 1 (p-value < 0.05). It supported the discriminant validity of the constructs. 5.2.2.7. Examination of Common Method Biases The common method bias can occur in survey studies due to the same provided number for each item of respondents (Podsakoff et al., 2003). In practice, this study followed the precautionary process used by earlier authors, for example, using the reversed questions, changing the position of the dependent variables on the questionnaire whilst other variable groups in the model are arranged in random order. Sampling is also done in the diversification of respondents. In addition, the usual way is to use statistical indicators to test for method biases. This study examined the Harman single factor test in CFA. The results showed that the one- factor model had low indices (χ2 = 15355.660, df = 702, TLI = 0.351, CFI = 0.385, RMSEA = 0.135). They were much lower than the saturated CFA model (TLI, CFI> 0.93 and RMSEA
8 Figure 5.4. The Result of Structural Model Assessment - Source: author’s data 5.4. Reversed Structural Model Assessment The structural research model based on the cognitive appraisal of emotions theory with the cognitive evaluation process occurs before the emotions when each individual confronts an event. Therefore, Harmeling et al. (2015) test the reversed model (the emotional factor is assumed to occur prior to the cognitive assessment process). Thus, the dissertation tested the reversed saturated model to compare the results. Findings showed
9 that the TLI and CFI indices of the reversed model were lower than the saturated theoretical models. In addition, the study followed Akaike's Information Criterion (AIC), which is considered to be good when the comparison of one model with rival models. The lower value of AIC shows a good fit. In the thesis, the reversed model has AIC = 3130.340 > AIC theoretical model (2895.223). The second reversal model (emotional factors and cognitive appraisal included in the isolation component) had AIC = 5763.053 > AIC theoretical model (2895.223). The lower the AIC, the better the model is chosen, so there is evidence that the proposed theoretical model is more suitable than the two inverted models. 5.5. Proposition Testing Results The 25 proposed hypothesis test results showed that there were 20 accepted propositions (P1, P2, P3, P4, P5, P6, P7, P8, P9, P10, P11, P12, P13, P14, P15, P17, P19, P20, P23, P25) and 5 unacceptable propositions (P16, P18, P21, P22, P24). Specifically, the P1 proposition (λ = 0.231, p
10 This suggests that emotions such as anger and disgust may explain the change of product quality evaluation instead of worry. There is evidence that when the author of the thesis runs a theoretical model by removing the two variables (anger, disgust) from the model, the relationship between the worry and the product quality judgment has the weight of λ = -0.139 (p
11 χ2 was statistically significant (p
12 evidence of a negative impact on the apathy of the relationship animosity beliefs and economic animosity beliefs. Apathy has been demonstrated validity through independent existence in the measurement model. This construct was also negatively related with patriotism. However, there was no significant statistical evidence to accept the negative relationship between apathy and the boycotting behavior. The final goal of the thesis is to examine the moderating role of individualism/collectivism and the factors (gender, education level, income groups, and age groups) on relationships. The moderating role of gender, education level, income groups, and age groups were confirmed. However, this study did not find evidence for the moderating role of the individualism/collectivism group. The mixed method (qualitative and quantitative combination) was used for this thesis. Qualitative research was conducted in two phases by two main approaches: netnography and grounded theory. The author collected 240 comments, conducted group discussions with 3 groups of students (15 people), in-depth interviews with 5 consumers to confirm the explored categories. The findings of the study were a set of components that were similar and different from the existing constructs. These components have been the basis for the development of measurement scales for quantitative research. Prior to the quantitative study was conducted, the expert research has been done immediately after the exploratory study to evaluate the ambiguity, duplication, content validity and the face validity of the scales. The results of the expert research were the selected scales to carry out the survey for quantitative research. Quantitative research was conducted through two stages of pilot testing with a sample size of 238 and formal study with a sample size of 1138 respondents. The purpose of the quantitative research was to confirm the reliability, validity of measurement scales and verification of the relationship among constructs in the proposed research model. The proposition test results showed that out of the 25 propositions, 20 propositions were accepted and 5 propositions were not accepted. For proposition about the moderating role of groups, this study found no difference between relationships when comparing individualism and collectivism group. However, the remaining groups (i.e., gender, income, education level, age) changed at least one of the relationships between the independent and dependent variables in the proposed research model. 6.2. Contributions on the applied of methods The methodology section is one of the carefully and closely synthesized sections of the thesis. The author has a combination of traditional and modern methods done in qualitative research to explore the research model. Expert research helps adjust the measurement scales to match the research context. The methods are published discrete through various works in the world, this study has synthesized, presented them in detail, and systematic. The author has also provided the scientific evidence and practical application in the study. Thus, this approach is a valuable contribution to researchers in Vietnam of both academic and applied behavior. Over the last twenty years, Kozinets' netnography has proven its worth in the context of the development of internet and technology. Netnography is useful when researching topics that are sensitive and need to exploit information from each individual (Langer and Beckman, 2005). Netnography provides a wealth of information that is both useful for academic research (e.g., exploratory research) and valuable for applied research (e.g., market research) (Kozinets, 2015). Since its inception in 1998, netnography has contributed to the creation of hundreds of highly publicized scientific publications using this method and cited over 1700 (Bartl et al., 2016). Netnography is also present in many areas such as travel, management, retailing, and marketing. In particular, in the context of the social networking is growing, as consumer behavior tends to gradually shift their shopping habits from offline to cyberspace and social networking have an important role in behavioral research (Belk et al., 2012). Vietnam is the seventh country in the world in terms of the number of social network users (Facebook), Vietnam has nearly 35 million social network users in 2016 and this number is expected to increase to nearly 46 million users in 2021 (Statista.com 2017). This is an indication that consumers are increasingly tending to use social networking. They used this channel to reflect their opinions, attitudes, perceptions, feelings, and behaviors. During the study, the author found that many subjects related to behavioral research outside of marketing could be exploited through this channel in other areas such as entrepreneurship and human resources. In practice, netnography can be used as a substitute for other traditional qualitative methods used by market research firms. For example, before launching an advertising campaign featuring video clips about the product, marketers can use social networks to collect consumer attitudes and emotions toward the video clips.
13 Consumer feedback is a reliable reference channel for decision makers (Belk et al., 2012). Moreover, with the development of the app on smartphones, Kozinets (2015) also recommends researchers gathering information, attitudes, views, and behaviors right on the apps. The Vietnamese market has experienced a number of incidents related to the products and services of companies that caused waves of social protests and public comments. For example, the case of Xphone - a Vietnamese phone brand established in 2015 (Bkav.com.vn, 2015). In the launch phase, this brand has created strong excitement, as this is a 100% phone produced in Vietnam. However, after that, the brand has created mixed comments with the ads shock and uncontrolled quality products (Cafebiz.com, 2017). If the marketer takes advantage of the social network and analyzes the content of the netizens, the coping strategies will work. 6.3. Discussion and new contribution to the measurement model In the field of consumer behavior, theories should be tested in many markets in a variety of approaches and methods. The main reason is that a research cannot cover and reflect all existing phenomena in a market. Preliminary studies have considered boycotting behavior are unidimensional construct and measure it through the construct of willingness to boycott. This thesis demonstrates that boycotting behavior is a multidimensional construct. In addition to the willingness to boycott, the component added a boycott call with four newly developed items from qualitative research and expert research. The newly added construct demonstrates the value (uni-dimension, discriminative value, convergent value, nomological network value). The scales were further adjusted to suit the market and the context of the study, including relationship animosity beliefs, apathy, and patriotism. Scales that are reused include the willingness to boycott, product quality judgment, negative emotions (ex., anger, and disgust), economic animosity beliefs, consumer ethnocentrism and individualism/collectivism. Adjustable scales have proven reliability. Findings significantly contribute not only to behavioral research but also to those who practice in the international market. 6.4. Discussion and new contribution to the structural model In addition to these measurement scales, the study explores several significant relationships that contribute to the new theoretical research model of boycotting behavior. First, the relationship between patriotism and factors such as relationship animosity beliefs, economic animosity beliefs, angry, disgust, and anxiety. The relationship between patriotism and consumer ethnocentrism, the impact of consumer ethnocentrism on boycotting behavior is also confirmed. The author of the dissertation has divided the concept of animosity into two components: cognitive appraisals and emotions. The impact of patriotism on the relationship animosity beliefs, economic animosity beliefs, and negative emotions. The findings are new theoretical contributions. Second, this thesis also found patriotism that influenced consumer ethnocentrism and boycotting behavior. These are new contributions to the consumer behavior field, which has been of interest to many academic and applied researchers for the last thirty years. Third, the influence of relationship animosity beliefs and economic animosity beliefs about negative emotions are statistically significant. The findings reflect a logical deduction from the cognitive appraisal of emotion theory. Thus, results are also a new theoretical contribution; as compared with the previously published work by Harmeling et al. (2015). The authors do not measure the construct of economic animosity beliefs in relation to the nomological network of negative emotions and boycotting behavior. Fourth, the significant negative impact of emotions (i.e., anger, disgust) to product quality judgment comes from the country that caused the event. Previous research by Harmeling et al. (2015) demonstrates in negative emotions that only worry emotions have the opposite effect on product quality judgment. The authors also suggest that disgust may also affect product quality evaluation. The thesis proved this relationship through quantitative research. Fifth, angry and disgust has a statistically significant effect on the boycotting behavior. Thus, patriotism, relationship animosity beliefs, economic animosity beliefs indirectly influence the boycotting behavior through negative emotions (i.e., angry, disgust). The affirmation of relationships is very important and has meaningful managerial implications for theory and practice.
14 Sixth, the higher the quality of the imported products from the country, the more likely they are to have a boycott. This demonstrates the importance of evaluating the quality of the product for consumer behavior expressed by their assessment of the manufacturing technology, the workmanship, the durability and reliability of the product. The tested relationship again confirms the previously published results on the role of product quality evaluation in the boycott model and has many practical management implications. Seventh, findings by comparing the models show that the proposed theoretical research model has better-fit indicators. This proves that the approach of the thesis is appropriate when considering the relationship in the model towards the cognitive appraisals appears before the negative emotions. Finally, the results of the moderator analysis with individualism/collectivism show that there is no difference in relation to these groups. However, the results of multiple-group analysis of sex, education, income, and age indicate differences in some of the relationships in the boycotting behavior model. This result shows the unique and specific role of sex, education, income, and age in the analysis of consumer behavior. 6.5. Managerial and Policy Implications A boycott is a form of consumer behavior that companies do not want to happen to their business. However, boycotting behavior is very common in practice. For that reason, the boycotts research is eager to provide an explanation, a theoretical perspective based on scientific evidence in many markets. It is helpful for managers to anticipate the boycott. The managers also have the appropriate coping strategies. Tian and Pasadeos (2012) proposed that, once multinationals identify the degree of consumer boycotting in the target market relative to a particular country, the localization strategy should be chosen in comparison with the globalization strategy. In international business, a multinational company chooses the localization strategy if the manager judges that the level of local demand adaptation is high and the pressure on cost reduction is low. In contrast, a global standardization strategy is chosen when the manager judges the company is dealing with high-cost pressures (due to many competitors appearing) and assessing the demand- side pressure localities are down (Hill, 2012, Peng, 2014). In order to implement a localization strategy, an enterprise must transfer its capital into a market, and then either do greenfield investment or mergers and acquisitions (M & A) in the host country. In order to implement a global standardization strategy, businesses must select a country that meets the criteria of low salary, a policy of attracting foreign direct investment, the presence of suppliers or related industries, sufficient resources to maximize production advantage by economies of scale, learning effects and experience curve (Hill, 2012, Peng, 2014). Tian and Pasadeos (2012) explained that the emotions and behavior of consumers towards a country are extremely important and the standardization strategy will not be appropriate if the chosen country has ever caused or causing events for other countries such as war, economic and political conflicts. These actions cause a negative reaction from consumers. Localization strategies will be more appropriate as it contributes to reducing their negative emotions. This means that if a multinational company is implementing a global standardization strategy in an X country. If consumers in other countries evaluating negative emotions and boycotting towards X, then the company should consider moving from the global standardized strategy (from X) into a localization strategy. For example, a company can do the foreign direct investment or make a strategic alliance with a certain company in the Y market. Since the mid-1990s, China has emerged as a country that has attracted foreign investment, with the cost advantage of labor, increasing investment in infrastructure, diversified supporting industries and big home demand (Liu et al., 2005). As a result, China has become the world's largest factory attracting thousands of companies to implement their global standardization strategy (Peng, 2012). For multinationals with factories located in China, scattering their plants to other production sites is a strategy that should be considered because the wave of the boycott of Chinese goods is occurring all over the world (Bongiorni, 2007; Rohrer, 2008; Akdeniz Ar and Kara, 2014). In addition, over the years, China has always been a country with unusual behavior towards its neighbors (e.g., Philippines, Indonesia, Japan, Vietnam), which have caused negative appraisal and emotions from consumers. Thus, a firm should locate its various manufacturing activities where it can avoid the lower product quality judgment and deal with consumer boycotts. Managers should consider this hedging strategy. From a broader perspective involving the relationship between animosity beliefs, negative emotions of consumers and their boycott in target markets, managers of MNCs should think towards a strategic alliance to produce hybrid products between. That is, the product is manufactured in one country