社會認同對團購網站使用者再購意願的干擾作用 - 政大學術集成

全文

(1)國立政治大學企業管理研究所(MBA 學位學程) 碩士學位論文. 政治大. 立社會認同對團購網站使用者再購意願的干擾作用 ‧. ‧ 國. 學. Nat. io. sit. y. The Moderating Effect of Social Identity on n. er. Users’ Repurchase Intention in the Context of al iv n Ch i U e n g c hBuying Online Group. 指導教授：洪叔民博士研究生：陳貞妮. 中華民國一百零四年七月.

(2) Acknowledgment I would like to express my sincere gratitude to National Chengchi University for accepting me as a student and providing me with the resources to fulfill my dreams. I would also like to thank the Master Business Administration program for giving me the opportunities to further the development of my knowledge and for providing me with scholarships to assist my learning in NCCU and in several countries abroad. Furthermore, I would like to express my appreciation to my advisor Prof. Shwu-Min Horng for his continuous support during my MBA study, and for his time,. 政治大 and the completion of this thesis. I could not imagine having a better advisor and 立. coaching, patience, and immense knowledge. His guidance helped me in my research. ‧ 國. 學. mentor for my MBA study.. Personally, I would like to thank my father who supported me through all my. ‧. years of education. I would also like to thank him for always being there to support. Nat. sit. y. me when I encounter challenges in my life.. n. al. er. io. Last, I would like to dedicate this thesis to all those who have assisted me along. i Un. v. the way and those who are interested in this area of research.. Ch. engchi. Jennifer Chan 陳貞妮 July 6, 2015.

(3) 摘要近年來，使用線上團購網站來購物的消費者逐日增長。隨著提供線上購物服務的電子商務業者增多，業者之間的競爭狀況也日漸激烈。因此，團購網站業者必須了解影響消費者再購意願的因素，並藉此建立除了價格優惠之外的優勢來提升自己的競爭力。本研究使用計畫行為理論來建立研究架構，並分別探討消費者對於產品和團購網站的滿意度、信任度及再購意願之間的關聯性。此外，本研究也探討消費者在團購網站中的社會認同程度對於信任度及再購意願關聯性的干擾作用。. 政治大份有效問卷，再使用偏最小二乘回歸分析法分析並獲得結論。分析結果顯示，消立. 本研究以曾經使用過團購網站來購物的消費者為研究對象，並蒐集了 300. ‧ 國. 學. 費者對於產品及團購網站的滿意度、信任度及再購意願之間皆為正向關係。此外，消費者的社會認同程度對於信任度及再購意願之間的關聯性具有負向干擾作用，. ‧. 表示社會認同程度越低的消費者如果對於所購買的產品感到信任，他對於同項產. sit. y. Nat. 品的再次購買意願和再次使用同購物網站的意願的關聯性就會越高。最後，基於. al. n. 有所貢獻。. er. io. 以上研究分析結果，本研究將提出管理意涵及建議，並希望能對相關領域的研究. Ch. engchi. i Un. v. 關鍵詞：團購、社會認同、滿意度、信任度、再購意願、電子商務.

(4) Abstract Online group buying has become increasingly popular in recent years. As the number of e-commerce businesses escalates, so has the competition among these businesses. Companies need to understand what factors influence customers’ repurchase intention, and provide more than just discounted prices in order to compete in this challenging market. Built upon the theory of planned behavior (TPB) model, this research discusses how customer satisfaction could influence trust and how trust could influence. 政治大 examined separately, and. repurchase intention. Satisfaction, trust, and repurchase intention of group buying. 立. websites and products were. their relationships were. ‧ 國. 學. investigated as well. Furthermore, customers’ social identity of the group buying websites that they have used to make purchases have been measured, and the. ‧. moderating effects of social identity toward the relationship between trust and. Nat. sit. y. repurchase intention are discussed in this research as well.. n. al. er. io. 300 effective samples were collected, all of which are customers who have made. i Un. v. purchases on group buying websites before. PLS regression analysis reported that. Ch. engchi. there are positive relationships between customer satisfaction and trust, and also between trust and repurchase intention. Additionally, this research has found a negative moderating effect of social identity toward the relationship between “trust toward the product” and “repurchase intention of the same product traded any website”, and also between “trust toward the product” and “repurchase intention of any product traded in the website used”. Last, managerial implications of my findings are outlined and discussed. Keywords: Group Buying, Social Identity, Satisfaction, Trust, Repurchase Intention, E-Commerce.

(5) Table of Contents Chapter One: Introduction. 1. 1-1 Research Background and Motivation 1-2 Research Scope and Purpose 1-3 Research Process. 1 5 7. Chapter Two: Literature Review 2-1 Online Group Buying 2-2 Theory of Planned Behavior 2-3 Social Identity 2-4 Satisfaction 2-5 Trust 2-6 Repurchase Intention. 立. 8 8 9 12 13 14. 政治大. 16. Chapter Three: Research Framework. 19. ‧ 國. 學. 3-1 Research Framework 3-2 Hypotheses. ‧. n. al. Chapter Five: Conclusion. Ch. engchi. sit er. io. 4-3 Reliability and Validity Analysis 4-4 Partial Least Square Regression 4-5 Moderating Effects of Social Identity. y. Nat. Chapter Four: Analysis 4-1 Questionnaire Development and Data Collection 4-2 Descriptive Analysis. i Un. v. 5-1 Research Results 5-2 Managerial Implications 5-3 Research Limitations and Future Research Directions. 19 20 26 26 28 29 34 36 38 38 39 41. References. 43. Appendix - Questionnaire. 50.

(6) List of Figures Research Process Theory of Reasoned Action (TRA) Theory of Planned Behavior (TPB) Research Framework PLS Regression Result. 立. 7 10 12 20 35. 政治大. 學 ‧. ‧ 國 io. sit. y. Nat. n. al. er. Figure 1 Figure 2 Figure 3 Figure 4 Figure 5. Ch. engchi. i Un. v.

(7) List of Tables Description of the Common Group Buying Websites in Taiwan Descriptive Analysis of Questionnaire Respondents Factor Loadings and Cronbach α of Questionnaire Items Measurements of Research Constructs CR, AVE, and Discriminant Validity of Questionnaire Constructs Moderating Effect of Social Identity Summary of Research Results. 立. 政治大. 學 ‧. ‧ 國 io. sit. y. Nat. n. al. er. Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Table 7. Ch. engchi. i Un. v. 27 29 31 32 34 37 38.

(8) Chapter One: Introduction 1-1 Research Background and Motivation Worldwide B2C e-commerce sales have grown considerably in the past decade. It had amounted to $1.251 trillion USD in 2013, and has been forecasted to reach $2.357 trillion USD in 2017 (eMarketer, 2014). This growth came primarily from the increases of internet and mobile users, advancement of shipping and payment methods, and the desire to win international markets by major companies. It has also been reported that worldwide digital buyer penetration, the percentage of global. 治政 internet users who had purchased products online, had大 reached 41.3% in 2013, and is 立 expected to reach 47.3% in 2018 (Statista, 2015). ‧ 國. 學. According to Institute for Information Industry (III, 2015) of Taiwan, the. ‧. e-commerce annual sales was 461.9 trillion in 2010, and is expected to reach 1,006.9. sit. y. Nat. trillion NTD by the end of 2015, indicating an annual growth rate around 16.8%.. io. er. Furthermore, III estimated that B2C e-commerce sales in Taiwan would be 613.8. al. trillion NTD and C2C e-commerce sales would be 393.1 trillion NTD in 2015.. n. iv n C Ministry of Economic Affairsh (MIC, 2014) ofUTaiwan reported that the most engchi. common methods of online purchasing are online direct purchase (76.7%), online auction direct purchase (45.5%), and online group buying (32.5%). The top reasons for customers to purchase online are lower product price (68.9%), convenience (49.2%), and more discount promotions (43.3%). According to this report, the discounted price of the products is one of the main reasons contributing to the growth of online purchases. As customers’ acceptance toward this new method of purchasing raises, so has the prevalence of online group buying websites, an e-commerce model that aims to provide discounted price for customers. For example, Groupon, one of the largest group buying websites in the world, has earned $3,191.69 million USD as their 1.

(9) revenue worldwide in 2014, which is ten times the amount they earned in 2010 (Statista, 2015), indicating the success and potential growth of online group buying websites. Group buying is a method of obtaining discounted prices from the selling parties by forming a group of customers who are willing to purchase the same item together. By gathering a group of customers who wish to purchase the same item, the volume of purchase can be increased, and therefore, raises the bargaining power of the customers (Sharif-Paghaleh, 2009). Online group buying websites provide internet platforms that overcome distance barriers of the customers, and enable them to. 治政 congregate online to raise their bargaining power 大 and obtain discounted prices. 立 Several possible motivation for buyers to purchase through group buying websites are: ‧ 國. 學. (1) uncertain if the product would be worthwhile of the price, it is safer to try it when. ‧. it is on discount, (2) because of the discounted price, the customer is willing to pay for. sit. y. Nat. something that they don’t usually purchase, and (3) taking advantage of the. io. al. to substitute other brands or similar products.. er. discounted prices, the customer would take the opportunity to stock up the product or. n. iv n C As for the selling parties, using buying websites enables them to reach h egroup ngchi U. more customers and build customer awareness in a shorter period of time and at a lower cost. Examples of motivation for sellers to participate in group buying websites are: (1) allows the seller to congregate large amount of customers quickly, (2) cutting down on marketing expenses allows the cost to be devoted to potential customers, allowing the customers to purchase the products at a lower price, (3) allows the seller to test the potential markets of a new product. With the reasons above, we can conclude that online group buying is a win-win business model that benefits both the sellers and the buyers. Most of the group buying websites do not sell products of their own. Their 2.

(10) source of profit are mostly from the result of how they match the sellers and buyers together. Therefore, how they spit the profit with the sellers would be critical to an online group buying business model. The percentage of discount from the original prices of the products sold on group buying websites varies greatly, but usually, the discounted selling price would be 50% of original price. On average, group buying websites take 30% ~ 40% of the discounted selling price of the products as their commission (MIC, 2011). For example, an item with an original price of $2000 would be sold as around $1000 on a group buying website. Among that $1000, around $300 ~ $400 would be the commission earned by the group buying website, and revenue. 治政 that the seller receive would be around $600 ~ $700. 大 Therefore, this creates the 立 mutual goal of the group buying websites and the sellers to reach as many customers ‧ 國. 學. as possible.. ‧. Online group buying business was initiated from the idea of group buying in. sit. y. Nat. which a person, the initiator, invites his or her friends to purchase a product together. io. er. in order to obtain a discounted price. This method of transaction were later evolved. al. into online group buying businesses that allows the group buying websites to take role. n. iv n C of “initiator”, and save buyers the trouble other customers. In Taiwan, h e nofgcongregating chi U online group buying business began around 2008 when 17Life was merged by PayEasy, an online B2C shopping website, which quickly took over more than half of the online group buying market back then. In 2010, Groupon, an online group buying business from the US, merged Atlaspost, a social networking website in Taiwan, and entered the Taiwan online group buying market. In the same year, Gomaji, an online group buying business initiated in Taiwan, entered the competition, and expanded its shopping platform from PC to mobile applications in 2013. In 2011, other competitors such as Yahoo and PChome began to join the online group buying market in Taiwan quickly, causing the market to hold up to 50 players at the same time a few years later. 3.

(11) Due to the fierce competition within the online group buying market in Taiwan, those 50 players were later trimmed down to 20 players that could survive in the competition. Through all these years of evolution, Groupon, Gomaji, and 17Life, the players that entered the market early, are the dominant players in the market, and they are currently taking over 80% of the online group buying business market share now in Taiwan (MIC, 2011). There are several group buying websites in Taiwan with different features. Some group buying websites sell a wide variety of products, while others sell a more specific type of products, such as providing only food or baby related products.. 治政 Another approach to classify the websites is by understanding 大 their original purpose. 立 Some of the websites were established as a group buying website, while some others ‧ 國. 學. were originally social platforms, such as BabyHome, that provide group buying. ‧. function due to its ability to congregate a large group of members.. sit. y. Nat. When people participate in a group or an organization, they would build a sense. io. er. of connection with that group or organization. Making purchases with group buying. al. websites that does not have a social networking platform would mean that the buyers. n. iv n C usually don’t know each other even they share a common goal of obtaining h ethough ngchi U discounted price. Participating in group buying websites with social networking platforms would allow the buyers to interact with each other while they are trying to bargain for a lower price. These two different types of online group buying model may be one of the features that could lead to different levels of social identity of the customers toward the group buying websites that they are using. Therefore, one of the purpose of this research is to investigate if social identity of the customers within the group buying websites that they are using would have a moderating effect on their repurchase intention. With the convenience and discounted prices provided by group buying websites, 4.

(12) this method of online purchasing has become more popular in the past decade. However, as more group buying websites were established, the competition among them became fierce as well. Although price is still the main incentive that attracts customers to shop via group buying websites, these websites still need to provide more than discounted prices if they want to stay in the competition. Most researches conducted before focused on the economic value provided by the group buying business model (Chen et al., 2009). However, few approached this topic from the customers’ point of view, or discussed the reason of purchasing from a particular group buying website instead of another. Therefore, this research would be. 治政 discussing how satisfaction, trust, and social identity 大 toward a group buying website 立 could influence customers’ intention of using the same website to purchase products ‧ 國. 學. again, hoping to contribute to future development of e-commerce business models.. ‧. 1-2 Research Scope and Purpose. y. Nat. io. sit. This research would focus on customers in Taiwan who have used group buying. n. al. er. websites as my research targets. Furthermore, this research would be targeting 11. Ch. i Un. most commonly used group buying websites in Taiwan.. engchi. v. The following are the research purposes that this research would like to achieve: 1. Investigate if customers’ satisfaction of the product they have purchased from a group buying website would influence their trust toward that particular product and the website that they have used. 2. Investigate if customers’ satisfaction of the group buying website that they have used to purchase a product would influence their trust toward the particular product that they have purchased and the website that they have used. 3. Investigate if customers’ trust of the product they have purchased from a group buying website would influence their repurchase intention of that particular product 5.

(13) and their intention of using the website that they have used to make purchases again. 4. Investigate if customers’ trust of the group buying website that they have used to purchase a product would influence their intention of using the same website to make purchases again. 5. Investigate if customers’ social identity of the group buying website that they have used would have a moderating effect on the relationship between customers’ trust and repurchase intention toward the product they have purchased and the website they have used.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 6. i Un. v.

(14) 1-3 Research Process. Define Research Objective and Scope. Literature Review. Establish Research Structure and Hypotheses. 學. ‧ 國. 立. 政治大. Establish Research Method. ‧ y. Questionnaire Design. n. al. er. io. sit. Nat. Measurements Development and. Ch. engchi. i Un. Data Collection and Analysis. Research Results and Discussion. Figure 1. Research Process. 7. v.

(15) Chapter Two: Literature Review In this chapter, I would be going through the relevant literature done by other researchers. First, I would be discussing literature related to online group buying. Then I would discuss the theories that were used to formulate my research model, which are the theory of reasoned action model (TRA) and theory of planned model (TPB). Last, I would discuss the literature related to social identity, satisfaction, trust, and repurchase intention.. 政治大 The idea of group buying model is to raise the bargaining power of the customers 立. 2-1 Online Group Buying. ‧ 國. 學. by gathering a certain amount buyers. They could then obtain a lower price for the products that they are interested in. For the selling parties, the group buying model. ‧. enables them to gather a large group of buyers at a lower cost and at a shorter period. Nat. sit. y. of time. Therefore, the group buying is a win-win model that is beneficial for both the. n. al. er. io. buyers and sellers (Kauffman et al., 2010).. i Un. v. The group buying model is a dynamic pricing mechanism where the buyers and. Ch. engchi. sellers negotiate the final transaction price for the products or services. The number of existing orders has a positive effect on the amount of new orders placed. As the price falls, the number of orders would raise as well (Kauffman & Wang, 2001). The main benefit of participating in online group buying is to reduce the price of the products. Therefore, the participants of online group buying are extremely price sensitive. At the same time, these price conscious consumers also want to obtain these low price benefits with a minimum searching cost, so convenience is another important factor that influences the participation of online group buying (Hsu & Yang, 2012). Cheng et al. (2012) discussed customer purchase intention of group buying. 8.

(16) websites integrating perceived risk and social influence on group buying websites. They have found that perceived critical mass has the strongest effect on intention to use group buying websites. In addition, their findings shows that perceived usefulness and a sense of virtual community have significant effect on intention to use group buying websites, and also, both perceived ease of use and website quality would influence perceived usefulness. Last, perceived risk has negative effect on intention to use group buying websites. There are other factors that may hinder the success of an online group buying business, such as the challenges from the marketplace, competitors, customers, and. 治政 suppliers (Kauffman & Wang, 2001). The selling parties 大 of online group buying 立 business depend on the volume of orders to realize cost savings in order to pass lower ‧ 國. 學. prices on to consumers. Also, the number of existing orders would influence the speed. sit. y. Nat. crucial for the success of an online group buying business.. ‧. and amount of new orders. Therefore, developing a critical mass in the marketplace is. io. er. The group buying websites are targeting price sensitive customers, therefore they. al. may be attracting customers who cares more about the price of the products, but do. n. iv n C not necessarily have high loyalty toward The group buying websites are h e ntheg websites. chi U not just facing competition from other group buying websites, but from all sorts of competitors providing products at low prices. For example, bricks-and-mortar retail chains could compete by saving their cost through purchasing in large bulks and selling them in wholesale prices. If the customers find another site or venue to purchase the same product at a lower price, the group buying website may lose the customers to competition easily.. 2-2 Theory of Planned Behavior Theory of planned behavior (TPB) is extended from theory of reasoned action 9.

(17) (TRA). It is a model often used to predict human behavior that was proposed by Ajzen and Fishbein (1980). They believed that behavioral intention is the key factor leading to the actual behavior. Therefore, the purpose of the TRA model is to discuss the key factors influencing behavioral intentions. It proposed that behavioral intention depends on the person’s attitude about the behavior and subject norms. Attitude would be the sum of beliefs about a particular behavior weighted by the person’s evaluations of these beliefs. Subjective norms is the opinions of others in one’s social environment on its behavioral intentions, which means that the person would be influenced by the beliefs of others weighted by the importance that the person. 治政 attributed to each of their opinions. The TRA model assumes 大 that people have the free 立 will to control their own behavior, and have the ability to make rational decisions for ‧ 國. 學. themselves. The TRA model is expressed in the following figure:. ‧ y. sit. Attitude. n. al. er. io. Outcome Evaluation. Nat. Behavioral Beliefs. Normative Beliefs. Ch. engchi. iv n U Behavioral Intention. Behavior. Subjective Norm. Motivation to Comply. Figure 2. Theory of Reasoned Action (TRA). (Source: Ajzen, I. & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. Englewood Cliffs, NJ: Prentice-Hall.) The TRA model assumes that all behavior are voluntary, however, later the 10.

(18) researchers have found the behaviors under most circumstances are not entirely free from the influence of its environment and situation. Therefore, in 1991, Ajzen modified his TRA model and proposed the TPB model, which could make a better prediction of behavioral intention by adding the volitional control into account. The TPB model suggested that if a person is influenced by external causes, the predictability of behavioral intention by the TRA model would be reduced. The TPB model not only takes one’s attitude and subjective norm into account, but also its perceived behavioral control.. The TPB model predicts deliberate and planned. behavior. The TPB model suggested that behavioral beliefs, normative beliefs, and control beliefs could influence. 立. 治政 one’s attitude, subjective 大. norm, and perceived. behavioral control, respectively, and they in turn could influence the person’s. ‧ 國. 學. behavioral intention.. ‧. Perceived behavioral control indicates the individual’s perception of the. sit. y. Nat. difficulty to carry out a certain behavior. This usually also reflects the individual’s. io. er. past experiences and its predictions of the difficulties and challenges that he or she. al. will encounter. The more resources and opportunities the individual believes it has,. n. iv n C believes encounter, the h ehenwould gchi U. and the less obstacles he. higher his perceived. behavioral control would be and the more likely that this would lead to behavioral intention and the actual behavior. The TPB model is expressed in the following graph:. 11.

(19) Behavioral Beliefs. Attitude. Normative Beliefs. Subjective Norm. Control Beliefs. Perceived Behavioral Control. Figure 3. Behavioral Intention. Behavior. 政治大. Theory of Planned Behavior (TPB). 立. ‧ 國. 學. (Source: Ajzen, I. (1991). Theory of Planned Behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211.) However, after Ajzen’s proposal of the TPB model and the concept of perceived. ‧. behavioral control in 1991, other scholars challenged this proposal indicating that. y. Nat. io. sit. Ajzen’s concept of perceived behavioral control was incomplete. In 2006, Pavlou and. n. al. er. Fygenson added self-efficacy and controllability to the TPB model to further explain. Ch. i Un. v. the concept of perceived behavioral control. If this model is implicated on. engchi. e-commerce, it can be used to discuss whether customers have the access to obtain sufficient product information and if they enough resources and ability to purchase the product online.. 2-3 Social Identity Social identity is one’s self-concepts by their connections with a social group or organization. Social identity theory explained intergroup behavior, and how one perceives himself within a group, such as belongingness and social status. This theory also states that social behavior varies along a continuum between interpersonal. 12.

(20) behavior and intergroup behavior. Complete interpersonal behaviors are behaviors determined solely by the individual characteristics and interpersonal relationships between two or more people. However, complete intergroup behaviors are behaviors determined solely by the social category memberships of two or more people. The social identity theory also stated that the extreme ends on the interpersonal and intergroup behavior continuum are unlikely to exist in real social situations. Therefore, behaviors are expected to be somewhere between the two extremes (Tajfel & Turner, 1979). Another statement of the social identity theory is that individuals are intrinsically. 治政 motivated to obtain positive distinctiveness or self-concept. 大 Also, it has been found 立 that there are conditions where individuals would endorse resource distributions that ‧ 國. 學. would benefit the positive distinctiveness of an ingroup in contrast to an outgroup. ‧. even if doing so would sacrifice its personal self-interest (Tajfel & Turner, 1979).. Nat. io. sit. y. 2-4 Satisfaction. n. al. er. One of the earliest concepts of consumer satisfaction was proposed by Cardozo. Ch. i Un. v. (1965). He stated that satisfaction would lead to repurchase behavior of the customer,. engchi. and would also lead to other purchasing behaviors. His research also showed that customers’ satisfaction toward a product would be influenced by the amount of effort to acquire the product, such as time and money devoted. Kotler (1997) made further investigation based upon Cardozo’s research and proposed that customers’ satisfaction not only would lead to repurchase behavior, but also lead to discussion with others about the product, and negligence toward competing brands and similar products. Oliver (1980) used the expectation disconfirmation theory to investigate customer satisfaction and stated that customer satisfaction is a function of expectation and expectancy disconfirmation. Customers would compare perceived performance of 13.

(21) the product or service with their expectations before their purchase, and the difference between the performance of the product or service and their expectations would be their level of expectation disconfirmation. Customer satisfaction is lower when the product or service does not match their expectations, and it is found that this would influence customers’ repurchase behavior. Other than customer satisfaction toward products, satisfaction with purchase channel has been researched as well. Devaraj et al. (2002) discussed customer satisfaction toward e-commerce channel, and proposed that satisfaction with e-commerce channel can be predicted by usefulness, ease of use, time saving, price. 治政 savings, and assurance. Further, their study also 大 found 立 determinant of channel preference.. that satisfaction is a. ‧ 國. 學. Müller (1991) proposed that customer satisfaction would be the key success. ‧. factor of businesses because customer satisfaction has direct influence toward loyalty,. sit. y. Nat. reputation, brand image, customer complaints, and market share. Since these factors. io. al. n. satisfaction is crucial toward the success of a business.. 2-5 Trust. Ch. engchi. er. strongly influence revenue and the competitiveness of the organization, customer. i Un. v. Trust is an attitude believing in someone or something. Trust can be attributed to relationships between people and social groups. It can also be an attitude toward products, services, or technology. Gundlach and Murphy (1993) believed that trust plays an important role when we are making decisions and would have a strong influence upon our behavior. For example, customers would evaluate if the suppliers are trustworthy before they make a purchase. Madhok (2006) defined trust as a mutual interaction between two parties, and they would fulfill their duties expected by their partners. Garbarino and Johnson 14.

(22) (1999) defined trust as how transaction partners believe each other with confidence and willingly rely on each other. Chaudhuri and Holbrook (2001) stated that trust is an attitude between the customers and suppliers caused by their previous interaction. Trust is also one of the key components of whether they could build a long lasting relationship. Only under the circumstances of mutual trust could the relationship be built upon long term benefits instead of short-sighted goals. This would eventually lead to strengthened competitiveness and lowers mutual transaction cost (Zeithaml et al., 1996). Although the definition of trust differs among the researchers, we could conclude that trust is a positive attitude indicating confidence, belief, and willingness to rely on something or someone.. 立. 政治大. Mayer et al. (1995) discussed the relationship between the trustor and trustee.. ‧ 國. 學. They explained that other than factors influencing levels of trust of the trustor,. ‧. qualities of the trustee could influence the levels of trust that it will receive from the. io. er. to be trusted are ability, benevolence, and integrity.. sit. y. Nat. trustor as well. According to their research, the qualities that the trustee possess for it. al. 1. Ability: Ability is indicating how someone or something has the competencies,. n. iv n C skills, and characteristics to make an within a specific domain. The domain h einfluence ngchi U of the ability is specific because the trustee may be competent and gaining trust from. others in an area, but may have little knowledge or ability to excel in another area. Therefore, trust is domain specific. 2. Benevolence: Benevolence indicates how the trustee is believed to want to benefit the trustor, and that the action is not caused by an egocentric profit motive. It is a special bonding between the trustee and trustor that makes the trustee to be willing to benefit the trustor even when there will not be any return or incentives for the trustee. 3. Integrity: Integrity indicates the trustor’s perception that the trustee is following a set of principles or values that the trustor finds acceptable. Both the alignment of 15.

(23) principles and the ability to follow to those principles are important for the trustee to be considered to have integrity, and therefore, trusted. Trust is especially important when it comes to online purchase and transactions done online because of the higher level of uncertainty faced by the customers (Heijden et al., 2003, & Lim et al., 2004). For example, customers could not know whether the products they have ordered online would be exactly as they were described and advertised on the website. Also, when customers are making payments online, they face the risk of property loss and identity theft. Therefore, the online service providers would need to build trust with the customers in order to battle. 治政 product uncertainty, service uncertainty, and information 大 asymmetry faced by the 立 customers. ‧ 國. 學. Lee and Turban (2001) have identify three key success factors of online shopping. ‧. models, and they are (1) customers’ trust toward the supplier, (2) customers’ trust. sit. y. Nat. toward the products provided by the supplier, and (3) customers’ trust toward the. io. al. n. when it comes to online service business model.. Ch. 2-6 Repurchase Intention. engchi. er. digital transaction system. Therefore, we could see that trust is an important factor. i Un. v. According to Peppers and Rogers (1998), it takes businesses 6~9 times of cost to obtain new customers than to generate repurchase behaviors from the existing customers. Therefore, other than winning new customers, retaining previous customers is crucial for the success of businesses as well. Cronin and Taylor (1992) defined repurchase intention as the willingness to purchase a product again after the customer has used to product. Kotler (1997) further defined repurchase intention as the willingness to repurchase or recommend others to purchase the same product or service after being satisfied with the service provided. 16.

(24) Hellier et al. (2003) proposed that repurchase intention is influenced directly and indirectly by seven important factors, which are service quality, equity, value, customer satisfaction, past loyalty, expected switching cost, and brand preference. His research found that the main factors that have the strongest direct influence on repurchase intention are brand preference and customer satisfaction with the other factors preceding them. The study of Tsiros and Mittal (2000) showed that satisfaction directly influences repurchase intentions. Also, without satisfaction, customers would have a higher rate of complaints and purchase regret. Therefore, we could conclude that repurchase. 治政 intention is significant to the success of a business. 大立 Research regarding the relationship between customer complaints and repurchase ‧ 國. 學. intention has also been investigated. Halstead and Page (1992) have found that,. ‧. interestingly, complainers do not have higher repurchase intentions than. sit. y. Nat. non-complainers. However, for those complainers who were dissatisfied with the. io. al. er. product, their satisfaction with complaint resolution did lead to significantly higher. n. repurchase intentions. Therefore, post-purchase services are very critical if companies. Ch wish to retention their customers.. engchi. i Un. v. Zeithaml et al. (1996) proposed that customer repurchase intention could be decomposed into three constructs, which are loyalty to company, negative propensity to switch, and willingness to pay more. Their descriptions are as follows: 1. Loyalty: Loyalty is the customer’s attitude toward the product or service. A customer with high loyalty would say positive things about the product, recommend others to use the product, and would also consider the product as their first choice when they have relevant needs. 2. Negative Propensity to Switch: Switch intention is the intention of the customer to use less of the product or service and switch to competitors’ services instead. A 17.

(25) customer with low switch intention would not be lured to use competitors’ services easily. 3. Pay More: Pay more is the willingness of the customer to continue the use of the product or service even when they are charged more than before and also when it is charged higher than similar products or services provided by the competitors.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 18. i Un. v.

(26) Chapter Three: Research Framework Based on the literature review above, this chapter would explain my research model, research factors, evaluation method, questionnaire design, and measures.. 3-1 Research Framework As the theory of planned behavior (TPB) model that I have discussed in section 2-2, the basic structure of Ajzen’s TPB model is: belief  attitude  intention  behavior Hellier et at., (2003) concluded from the TPB model and several other researches. 治政 that satisfaction and attitude are the major antecedents 大 of customer repurchase 立 intention, and attitude is viewed as a post-purchase construct. Therefore, he proposed ‧ 國. 學. the following model:. ‧. satisfaction  attitude  repurchase intention. sit. y. Nat. In his model attitude is the customers’ learned disposition of a service, company,. io. negative.. er. or brand as a result of past evaluative experiences, and it could be positive, neutral, or. al. n. iv n C In this research, I am buildinghmy research framework e n g c h i U upon Azjen’s TPB model. and Hellier’s model. I am using satisfaction as the belief, trust as the attitude, and repurchase intention as the intention of discussion. My proposed conceptual model would be: satisfaction  trust  repurchase intention Using the TPB model and the literature discussed in the previous chapters, I have formed several hypotheses and a research model to be tested. In this model, I will be discussing how customers’ satisfaction toward the products and websites would lead to trust toward the products and websites. Also, I will be discussing how customers’ trust toward the products and websites would lead to repurchase intention of the same 19.

(27) products and using the same websites to purchase any products again. Last, I will discuss how customers’ social identity of the group buying websites that they are involved in could influence the degree of relationship between trust and repurchase intention. In my model, the “product” of discussion means the particular product with the same brand that the customer has bought the last time he or she made a purchase by a group buying website. The “website” of discussion means the particular group buying website that the customer has used to make a purchase the last time he or she bought something by a group buying website. Figure 4 illustrates the research framework that I am proposing.. 立. 政治大. ‧ 國. y. sit. H3a. n. er. io. al. Ch. H2b. Satisfaction toward the Website. H5c. ‧. Trust toward the Product. H1a. H1b. 學 H5a. Nat. Satisfaction toward the Product. H5b. Social Identity. H2a. Figure 4. i Un H3b. engchi Trust toward the Website. v. H4. Repurchase Intention of the Same Product Traded in any Website. Repurchase Intention of Using the Same Website to Purchase any Product. Research Framework. 3-2 Hypotheses According to the research model above, I will be illustrating the supporting literature of the ten hypotheses that will be discussed in this research.. 20.

(28) 3-2-1 Satisfaction toward Products Ranaweera and Prabhu (2003) discussed how satisfaction and trust have strong positive effects on customer retention. Their research also showed that switching barriers have a moderating effect on the relationship between satisfaction and customer retention. Service providers may be able to retain some dissatisfied customers if those customers were facing high switching barriers. However, customers participating in online group buying face low switching barriers. Therefore, the relationship between satisfaction and customer retention of online group buying business models remains strong.. 政治大 H1a: Satisfaction toward the product will positively influence trust toward the 立. ‧ 國. 學. product. Many researches have supported that satisfaction have a positive influence on. ‧. trust. For example, Tax et al. (1998) have found that satisfaction with a service has a. Nat. sit. y. direct impact on trust and commitment. Therefore, I hypothesize that satisfaction. n. al. er. io. toward the product will positively influence trust toward the product.. Ch. i Un. v. H1b: Satisfaction toward the product will positively influence trust toward the website. engchi. Oliver (1980) has proposed that satisfaction would influence attitude change and purchase intention. Therefore, I would like to investigate if satisfaction toward the product would influence trust toward the website.. 3-2-2 Satisfaction toward Websites Shankar, Smith, and Rangaswamy (2003) discussed how levels of customer satisfaction could be related to loyalty of online and offline services. They discovered that customer satisfaction for a service delivered online is the same as when it is delivered offline. Furthermore, customers’ loyalty toward to the service provider is 21.

(29) higher when it is delivered online than offline. This shows the importance of gaining customer satisfaction for the e-commerce service providers. H2a: Satisfaction toward the website will positively influence trust toward the website Abdul-Muhmin (2010) has found that attitude is positively influenced by overall satisfaction and negatively influenced by experience with online purchase problems. Devaraj at el. (2002) have proposed that customers’ satisfaction with an e-commerce channel is a determinant of channel preference. Research has also shown that. 政治大 (Kassim & Abdullah, 2010). Therefore, I hypothesize that satisfaction toward the 立 customer satisfaction is also positively related to trust on e-commerce settings. ‧ 國. 學. website will positively influence trust toward the website.. H2b: Satisfaction toward the website will positively influence trust toward the. ‧. product. y. Nat. io. sit. When products are sold online mediated by the internet, the products’ physical. n. al. er. information are reduced, and this creates information asymmetries between the buyers. Ch. i Un. v. and sellers. Wells at el. (2011) used the signaling theory as their framework to. engchi. understand how extrinsic cues can be used by sellers to convey product quality information to consumers and reduce information asymmetries. They have found that website quality would influence consumers’ perception of product quality, which would influence online purchase intentions. Therefore, I hypothesize that satisfaction toward the website will positively influence trust toward the product.. 3-2-3 Trust toward Products and Websites Singh and Sirdeshmukh (2000) proposed that trusted service providers are those that can be relied on and would deliver what they have promised. Their research also showed how trust could influence the relationship between consumers and service 22.

(30) providers. Morgan and Hunt (1994) stated in their research that trust and commitment are important factors to maintain the relationship between customers and service providers, and they would also influence customers’ willingness to associate with the same providers again. When customers are using the services provided online, or purchasing products online, they are concerned with receiving products that are different as described on the website and do not match up with their expectations. Also, customers are concerned with the transaction process conducted online due to security issues. There are evidence that online trust is built through order fulfilment, privacy, responsiveness. 治政 and contact (Chiu et al., 2009). Therefore, it is important 大 for online service providers 立 to build trust and maintain long term relationships with the customers in order to be ‧. ‧ 國. 學. successful.. H3a: Trust toward the product will positively influence repurchase intention for the. Nat. sit. y. product traded in any website. n. al. er. io. There are plenty of studies discussing the relationship between trust and. i Un. v. purchase intention. Laroche et al. (1996) examined the relationships among brand. Ch. engchi. familiarity, brand attitudes, and purchase intention. Their results showed that consumers’ familiarity with a brand influences their confidence and attitude toward a specific brand, which in turn would affect their purchase intention for that particular brand. Also, Ha et al. (2010) have found that trust directly affected repurchase intention of the consumer. Therefore, I hypothesize that trust toward the product will positively influence the repurchase intention for the same product. H3b: Trust toward the product will positively influence repurchase intention for any product traded in this website Wells et al. (2011) have found that consumers’ perceptions of product quality 23.

(31) would affect online purchase intentions. Therefore, I hypothesize that trust of the product will positively influence the repurchase intention for any product traded on that group buying website. H4: Trust toward the website will positively influence the repurchase intention for any product traded in this website Research has shown that the success of a relationship is based on mutually profitable relationship between the buyer and supplier (Selnes, 1998). Because of the concern associating with security, safety, and credibility of e-commerce services, trust. 政治大 It has also been found that attitude toward online purchasing has a positive effect 立. of online services is particularly important because it reduces sacrifices for the buyer.. ‧ 國. 學. on repurchase intentions (Abdul-Muhmin, 2010). Research has shown that satisfaction and trust on the website are determinants of website loyalty (Horppu et al.,. ‧. 2008). Other research has shown that online relationship quality and perceived. Nat. sit. y. website usability impact customer repurchase intention. And online relationship. n. al. er. io. quality is positively influenced by perceived vendor expertise in order fulfillment,. i Un. v. perceived vendor reputation, and perceived website usability. Also, the research has. Ch. engchi. shown that distrust in vendor behavior would negatively influence online relationship quality. Therefore, in this research, I hypothesize that trust toward the website will positively influence the repurchase intention of any product traded on the group buying website.. 3-2-4 The Moderating Effect of Social Identity H5a: Social identity will have a negative moderating effect on the relationships between trust toward the product and the repurchase intention of the product traded in any website. 24.

(32) H5b: Social identity will have a negative moderating effect on the relationships between trust toward the product and the repurchase intention for any product traded in the particular website used H5c: Social identity will have a negative moderating effect on the relationships between trust toward the website and the repurchase intention for any product traded in the particular website used Venkatesh et al. (2012) investigated the moderating effect of experience on hedonic motivation and behavioral intention and they have found that the moderating. 政治大 al. (2012) investigated individual-level tie strength and group-level identification as 立 effect would be stronger when the person has less experience. Furthermore, Wang et. ‧ 國. 學. antecedents of product attitude and purchase decisions. They have found that participants who have stronger tie strength and identification with peer group have a. ‧. positive effect on peer communication, which has a positive effect on product attitude. Nat. sit. y. and purchase intention. Therefore, I speculate that customers with higher levels of. n. al. er. io. social identity made their purchase on a group buying website due to social pressure. i Un. v. or their desire to conform with the group instead of trusting the product or website. Ch. engchi. themselves. On the other hand, customers with lower levels of social identity made their purchase because they trust the product and website, and therefore, are more likely to purchase the product again and to use the website again to make purchases. For these reasons, I hypothesize that social identity would have a negative moderating effect between trust and repurchase intention.. 25.

(33) Chapter Four: Analysis 4-1 Questionnaire Development and Data Collection To develop the questionnaire used in this research, three steps were taken. First, I have chosen the appropriate factors and developed the questions according to researches done by others. There are 7 constructs in total, including (1) satisfaction toward the product, (2) satisfaction toward the website, (3) trust toward the product, (4) trust toward the website, (5) repurchase intention of the product traded in any website, (6) repurchase intention of any product traded in the particular website used,. 治政 and (7) social identity of the customers toward the website. 大 Taking the questionnaires 立 used by other studies, their questions were modified to fit the purpose of my research ‧ 國. 學. and to form the first draft of my questionnaire.. ‧. Second, depth interviews were conducted with 3 group buying website users in. sit. y. Nat. order to check if my research model and hypotheses reflect the reality, and also to. io. al. er. modify the questions whose descriptions are potentially misleading or ambiguous.. n. The three interviewees are a female aged 20~30, a female aged 40~50, and a male aged 40~50.. Ch. engchi. i Un. v. Next, two pretests were conducted. The first pretest was conducted by passing out printed questionnaire to 40 undergraduate students, while the second pretest was conducted by iSurvey Eastern Online Company, Ltd. (http://www.isurvey.com.tw/), a well-known marketing and consumer behavior research company in Taiwan. 300 samples were collected in the second pretest, and the questionnaire was modified according to the results of the pretests, which led to the final version of my questionnaire. The questionnaire was designed in Traditional Chinese since the respondents are Taiwanese. 11 most commonly used group buying websites in Taiwan are listed at the 26.

(34) beginning of the questionnaire for the respondents to select the website that they are most acquainted with. The questionnaire used a 7-point Likert scale, asking the respondents to choose their answers according to how much they agree to the descriptions. The data were collected by iSurvey Eastern Online Company, Ltd. by their online platform and the respondents were selected randomly. The time length of data collection was a week, and 300 effective samples were collected, all of which are customers that have participated in online group buying before. The group buying websites examined in this research are described in Table 1.. 治政 These websites were selected according to a research 大 done by the Ministry of 立 Economic Affairs, R.O.C. (2011), which listed the top 10 group buying websites in ‧ 國. 學. Taiwan according to number of customers, revenue, and market share. However,. ‧. among those 10 group buying websites, 5 websites were closed down by the time this. sit. y. Nat. research was conducted, therefore they were taken out of my final questionnaire.. io. er. During my pretest, I have found 6 more group buying websites that are often used by. al. customers in Taiwan, and they were added into my final questionnaire.. n. iv n C Description of the Common Buying Websites in Taiwan h eGroup ngchi U. Table 1. Group Buying Websites Groupon (酷朋). GOMAJI (夠麻吉). Website Features The largest group buying website worldwide. Provides new discounts on different products every day. Products are categorized into food and restaurants, beauty and fashion, sports and entertainment, courses and services. Provides discounts on food, restaurants, beauty services, entertainment, hotels…etc. Also provides a smartphone application with group buying functions for customers to follow updated discounts.. 17Life. Acquired by PayEasy in 2008. Provides a large variety of products. Allies with Familymart and provides a smartphone application with group buying functions. 27.

(35) Yahoo!奇摩超級商城. Provides discounts for clothing, food and restaurants, digital products, and entertainment. Smaller in scale compared with other group buying websites. 123 團購網. Provides a large variety of products. Other than categorizing products by their functions, its products are also categorized by speed of delivery.. PTT BuyTogether Board PTT is a free and open online bulletin board in Taiwan (PPT 合購版) with around 1.5 billion members. PTT BuyTogether board is a platform for customers to congregate and initiate group buying activities by themselves. Does not provide delivery service. All of the services need to be initiated and arranged by the members. TaoBao (淘寶網). A Chinese online purchasing website of Alibaba Group that also provides services for customers in Taiwan. Provides mostly discounts for clothes and restaurants, and also provides easier access to Chinese products for Taiwan customers.. 立. ‧ 國. 學. ‧. y. Mostly provides discounts for clothes, food, fish related products, digital products, and home appliances.. sit. io. 瘋狂賣客. Nat. 生活市集. An online purchasing website with group buying function. Most of the products sold by the group buying function are food.. Mostly provides discounts for digital products, home appliances, apparels, travel and hotels.. er. PChome. 政治大. al. n. iv n C A socialhbulletin platformUof babies and children related engchi topics and attracts parents as their main source of. BabyHome 好物. members. Provides a group buying board for parents to purchase babies and children related products.. 4-2 Descriptive Analysis According to the descriptive analysis of the respondents shown in Table 2, we could see that the majority of the respondents are over the age of 33 (72.7%). The gender of the respondents was very balanced, and the marital status of the respondents is quite balanced as well, with 55.7% married and 44.3% being single. Most of the respondents have more than 24 months of experience with online group buying 28.

(36) (60.3%). We could also see that most respondents use their desktop for online group buying (60.3%). Last, 58.4% of the respondents have less than 20,001 NTD of average disposable income per month. The descriptive analysis of the respondents is provided in the following chart: Descriptive Analysis of Questionnaire Respondents. Variables Gender. Age (Years). Frequency. Percentage (%). Male. 152. 50.7. Female. 148. 49.3. 18 and under. 4. 1.3. 19~25. 28. 9.3. 26~32. 50. 16.7. 167. Single. 133. 44.3. 6 months and less. 27. 9.0. 7 to 12 months. 29. 9.7. 13 to 18 months. 35. y. 9.3. 19 to 24 months. 28. more than 24 months. 181. desktop. 181. 11.7 60.3. a l or similar devices 80i v notebook n C U tablets h e n 14 i h gc. 60.3. smartphones. 25. 8.3. 10000 and under. 86. 28.7. 10,001~20,000. 89. 29.7. 20,001~30,000. 48. 16.0. 30,001~40,000. 23. 7.7. 40,001 and above. 54. 18.0. n. Average Disposable Income per Month (NTD). 55.7. er. io Most Frequently Used Device for Online Group Buying. 46.7. Married. Nat. Experience on Online Group Buying. 26.0. ‧. ‧ 國. 治 78 33~39 政大140 40 and above 立學. Marital Status. Category. sit. Table 2. 26.7 4.7. 4-3 Reliability and Validity Analysis 4-3-1 Reliability Analysis This research was conducted in 7-point Likert scale. The respondents could select only one answer for each question, and they were asked to respond by selecting 29.

(37) “1” if they strongly disagree with the statement and by selecting “7” if they strongly agree. Cronbach ɑ would be used as a method of verifying the reliability of the questionnaire. In general, Cronbach ɑ over 0.7 would be considered as a good reliability rating (Guilford & Fruchter, 1978). The Cronbach ɑ of all of my constructs are well above 0.7, and they are provided in Table 3 to prove that the constructs are consistent and reliable. The chart also includes the standardized loadings of the items in the questionnaire, which represent the correlation between each item and their corresponding constructs. The higher the standardized loadings are, the stronger the. 治政 relationship between the items and their corresponding 大constructs. The standardized 立 loadings of the items of this questionnaire range from 0.64 to 0.88, which proves them ‧ 國. 學. acceptable. The items of my questionnaire are provided in Table 4 and the complete. ‧. questionnaire is provided in the appendix section at the very back of this research for. sit. y. Nat. readers’ reference. The factor loadings, item average, item standard deviation are also. io. al. n. according to the item order in the questionnaire:. Ch. engchi. 30. er. provided in the Table 3 along with Cronbach ɑ and their listed order are arranged. i Un. v.

(38) Factor Loadings and Cronbach α of Questionnaire Items Factor Item Standard Cronbach ɑ Constructs Item Average Loadings Deviation. Trust toward the Website. 1.02. 0.75. 4.99. 1.06. 0.70. 5.19. 1.08. 0.73. 5.17. 1.02. 0.72. 5.11. 1.09. 0.68. 5.12. 1.13. 0.68. 5.10. 1.17. 0.68. 5.07. 1.11. 0.82. 5.21. 1.12. 0.83. 5.20. 1.15. 5.22. 1.13. 0.68. 5.18. 1.10. 0.64. 5.16. 1.16. 0.66. 5.16. 1.16. 0.70. 5.52. 1.14. 0.72. 5.53. 1.14. y. Satisfaction toward the Product. 5.10. 0.67. 5.48. 1.22. 5.64. 1.12. 0.88. 4.84. 1.49. 0.82. 5.06. 1.44. 0.84. 5.00. 1.32. 0.83. 4.98. 1.24. 0.87. 4.81. 1.31. 0.87. 4.61. 1.37. 0.77 0.77. al. 0.88 0.87. 政5.23 治大1.17 5.26 1.12. 0.966. 0.971. ‧. Social Identity within the Website. 0.70. n. Repurchase Intention of the Same Product Traded in any Website. io. to Purchase any Product. Nat. Repurchase Intention of Using the Same Website. 0.977. 學. Trust toward the Product. 立. ‧ 國. 0.68. 0.955. 0.978. er. Satisfaction toward the Website. 0.76. sit. Table 3. iv n C h 4.98 e n g c h i U1.39 4.87 1.40. 0.977. 0.963. 4-3-2 Validity Analysis This section would be discussing the validity of the questionnaire by checking its content validity and construct validity. Content validity refers to the extent to which 31.

(39) the measures in the questionnaire represent all aspects of their corresponding construct. In general, content validity can be justified if the items were built upon other researches. The items of my questionnaire were modified after depth interviews with three people, and were also modified from other researches. References that I have used to develop this questionnaire are provided in the chart below: Table 4 Constructs. Measurements of Research Constructs Questions. References. 1. 我對於在此網站購買產品時的決定 Cheng, Yang, & Chen. 感到非常滿意 (2011) Satisfaction toward the Website. 政治大. 2. 我相信在此網站購買產品是一項明 Choi, Sohn, & Lee. (2010) 智的決定 Martin, Camarero, & San 3. 我在此網站消費的經驗是愉快的 Jose. (2011). 立. ‧ 國. 學. 4. 整體而言，我滿意在此網站的消費經 Wen, Prybutok, (2011) 驗. & Xu.. 5. 對於擁有 P 產品我感到滿意. ‧. 6. 我對於購買 P 產品這項決定感到滿 Ha, Janda, & Muthaly. 意 (2010) 7. 我相信購買 P 產品是一項明智的決 Hume, & Mort. (2008) 定 Kauffman, Lai, Ho. (2010). 8. 整體而言，我滿意 P 產品. n. al. 9. 此網站會實踐它的承諾與保證. Ch. engchi. 10. 此網站會完成它所保證的服務. er. io. sit. y. Nat. Satisfaction toward the Product. i Un. 11. 此網站是可靠的 Trust toward the 12. 整體而言，此網站是值得信任的 Website. vCheng,. Yang, & Chen.. (2011) Green, & Pearson. (2011) Heijden, Verhagen, & Creemers. (2003) Martin, Camarero, & San Jose. (2011). 13. 我購買的 P 產品有發揮其當初保證的用途與效用 14. 我購買的 P 產品所提供的功能符合 Trust toward the Ha, Janda, & Muthaly. 它所表明的功用 Product (2010) 15. 我相信 P 產品會發揮其當初保證的用途與效用 16. 整體而言，P 產品是值得信任的. 32.

(40) Same Product Traded in any Website. Ha, Janda, & Muthaly. 23. 我有很高的機率會再次購買 P 產品 (2010). 政治大再次購買 P 產品立 25. 我認為我是此網站團購社群中的一 24. 下次我有類似的需求時，我會考慮. 份子. 學. 26. 我在此網站團購社群中是被接納的 Dholakia, Bagozzi, 27. 我在此網站團購社群中是有價值的 Pearo. (2004). ‧. Social Identity within the Website. 22. 我希望能夠再次購買 P 產品. sit. Nat. 28. 我對於此網路團購社群有很高的歸屬感. y. Repurchase Intention of the. 18. 我願意再次使用此網站來購買商品 Choi, Sohn, & Lee. (2010) 19. 我有很高的機率會繼續使用此網站 Heijden, Verhagen, & Creemers. (2003) 來購買商品 20. 我會考慮再次使用此網站來購買商 Hung, Cheng, & Chen. (2012) 品 Wen, Prybutok, & Xu. (2011) 21. 我預期未來會有再次購買 P 產品的意願. ‧ 國. Repurchase Intention of Using the Same Website to Purchase any Product. 17. 我預期未來會有再次使用此網站購 Chiu, Chang, Cheng, & 買商品的意願 Fang. (2009). n. al. er. io. Construct validity can be further divided into convergent validity and. i Un. v. discriminant validity. Convergent validity refers to the degree which two or more. Ch. engchi. measures of a construct that theoretically should be related are indeed related. Discriminant validity tests whether the measures that are supposed to be unrelated are indeed unrelated. This research would be using Composite Reliability (CR) and Average Variance Extracted (AVE) to test the convergent validity of the questionnaire. Composite reliability assesses the internal consistency of a measure, and higher CR would indicate greater convergence among the measures. In general, it is recommended that the CR should be greater than 0.7 (Bagozzi & Yi, 1988). Average Variance Extracted is a stricter and more conservative measure of convergent validity, and a higher AVE 33. &.

(41) would indicate greater convergent validity. In general, AVE above 0.5 is usually treated as an indication of convergent validity (Hair et al., 2010). CR and AVE of this research are shown in Table 5. We could see that they are all well above 0.7 and 0.5, which indicate the convergent validity of this research. Discriminant validity can be tested by the square root of AVE. The square roots of AVE are compared with the correlations between the construct and other constructs. The square roots of AVE of this research are shown as the diagonal numbers in Table 5. All of them are greater than the numbers in the corresponding rows and columns, which demonstrates adequate discriminant validity for this research.. 6. 7. 學. ‧ 國. Table 5. 政治大 CR, AVE, and Discriminant Validity of Questionnaire Constructs 立 CR AVE 1 2 3 4 5 0.97. 2. Product Trust. 0.98 0.92. 0.83 0.96. 3. Repurchase Intention - Product. 0.98 0.94. 0.65 0.68 0.97. 4. Repurchase Intention - Website. 0.98 0.94. 0.73 0.78 0.60 0.97. 5. Website Satisfaction. 0.97 0.88. 0.83 0.73 0.60 0.76 0.94. 6. Website Trust. 0.98 0.91. 0.66 0.74 0.52 0.78 0.70 0.95. io. sit. y. ‧. 0.98 0.93. Nat. 1. Product Satisfaction. n. al. er. 7. Social Identity 0.97 0.90 0.53 0.61 0.54 0.65 0.57 0.57 0.95 CR: Composite Reliability AVE: Average Variance Extracted Diagonal Numbers: Root squares of AVE Numbers under the Diagonal: Correlation of the vertical and horizontal constructs. Ch. engchi. i Un. v. 4-4 Partial Least Square Regression In this section, I would be discussing the Partial Least Square (PLS) regression of my research model. PLS regression is a technique that reduces the predictors and dependents to principal components and performs a least squares regression on these components instead of on the original data itself. In PLS regression, the standardized coefficients reflect the importance of each predictor in the model and correspond to the standardized dependent variables and standardized independent variables. The. 34.

(42) t-value indicates the relationship between the predictors and dependent variables. The higher the t-value is, the stronger the relationship. The PLS regression in this research was performed by SmartPLS 3.0, and the results are shown in Figure 5. The numbers in parentheses are the t-values and the numbers above them are the standardized coefficients. All of the causal relationships of the hypotheses in my model have positive t-values, and most of them have confidence levels higher than 99% indicating significant positive relationships. However, the causal relationship between “satisfaction toward the website” and “trust toward the product” has a confidence level of 90%, but it still indicates a significant positive relationship.. 學. ‧ 國. 立. Social Identity. n. (R2=0.696). Ch. 0.25*** (3.23). engchi. sit. io. al. Trust toward the Product. 0.70*** (16.60). er. 0.74*** (11.80). y. ‧. Nat. Satisfaction toward the Product. 政治大. i Un. v. Repurchase Intention of the Same Product Traded in any Website (R2=0.460). 0.42*** (6.29). 0.124* (1.90). Satisfaction toward the Website. 0.55*** (7.66). Trust toward the Website (R2=0.513). 0.48*** (8.14). Repurchase Intention of Using the Same Website to Purchase any Product (R2=0.670). ***: α=0.01, **: α=0.05, *: α=0.1, numbers in parentheses are t-value Figure 5. PLS Regression Result 35.

(43) 4-5 Moderating Effects of Social Identity In the section, I would be discussing the moderating effects of the customers’ social identity of the group buying websites that they have been participating in on the relationship between (1) trust toward the product and the repurchase intention of the product traded in any website, (2) trust toward the product and the repurchase intention of any product traded in the particular website that they have used, and (3) trust toward the website and the repurchase intention of any product traded in the particular website that they have used.. 政治大 with measured social identity equal to the median value, which is 5, were removed. 立. To generate the high and low groups of social identity, 40 out of 300 samples. ‧ 國. 學. This left 134 samples with measures smaller than 5 for the “low group” and 125 samples with measures larger than 5 for the “high group” (Hair et al., 2010).. ‧. According to the numbers shown in Table 6, the differences of the t-values. sit. y. Nat. between the low and high groups are 2.550, 1.345, and -0.369 for “Product Trust –. n. al. er. io. Product Repurchase”, “Product Trust – Website Repurchase”, and “Website Trust –. i Un. v. Website Repurchase”, respectively. The moderating effects of “Product Trust –. Ch. engchi. Product Repurchase” are significant at 99% confidence level. Customers with low social identity (β = 0.701) are more likely to purchase the product again if they trust the product compared to those with high social identity (β = 0.445). Also, the moderating effects of “Product Trust – Website Repurchase” are significant at 90% confidence level. Customers with low social identity (β = 0.426) are more likely to use that website to purchase products again if they trust the product that they have purchased compare to those with high social identity (β = 0.257). However, the moderating effects of “Website Trust – Website Repurchase” are not significant.. 36.

(44) Table 6. Moderating Effect of Social Identity Moderator Coefficient for Hypothesis (Social repurchase Identity) intention (t-value) Low 0.701*** (13.374) Trust (Product) Repurchase (Product) High 0.445*** (5.029) Trust (Product) Repurchase (Website) Trust (Website) Repurchase (Website). Low. 0.426*** (4.131). High. 0.257*** (3.722). Low. 0.441*** (4.943). High. 0.487*** (5.681). Coefficient difference (L-H) (t-value). P. 0.256*** (2.550). .006. 0.169* (1.345). .090. -0.046 (-0.369). .644. ***：α=0.01；**：α=0.05；*：α=0.1. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 37. i Un. v.

(45) Chapter Five: Conclusion The main purpose of this research is to investigate if social identity would influence the repurchase intention of customers. Also, it investigates if the satisfaction toward the product and website is related to trust toward them, and if trust toward the product and website is related to repurchase intention. After collecting and analyzing 300 samples, I will be discussing the results, potential contributions, and limitations of this research in this chapter.. 5-1 Research Results. 治政 This research focuses on the customers who have 大 used online group buying 立 websites as a mean of purchasing. It examines the relationship between satisfaction, ‧ 國. 學. trust, and repurchase intention of the customers, and investigates if social identity. ‧. would have a moderating effect on the relationship between trust and repurchase. io. Hypotheses. y. n. al. sit. Summary of Research Results. er. Table 7. Nat. intention. The results of my hypotheses are summarized in the following table:. i Un. v. Results. H1a. Product Satisfaction  Product Trust. H1b. Product Satisfaction  Website Trust. H2a. Website Satisfaction  Website Trust. Supported. H2b. Website Satisfaction  Product Trust. Supported. H3a. Product Trust  Repurchase Intention of Product. Supported. H3b. Product Trust  Repurchase Intention of Website. Supported. H4. Website Trust  Repurchase Intention of Website. Supported. H5a. Moderating Effect of Social Identity on H3a. Supported. H5b. Moderating Effect of Social Identity on H3b. Supported. H5c. Moderating Effect of Social Identity on H4. Not Supported. Ch. engchi. 38. Supported Supported.

(46) The majority of my hypotheses are supported. We could see that all of the hypothesized casual effects between satisfaction, trust, and repurchase intention are supported, which aligns with other studies done before. As for the moderating effects, hypotheses H5a and H5b are significant as well. This means that customers with lower social identity would have stronger relationship between “trust toward product” and “repurchase intention of the product traded in any website”. They would also have a stronger relationship between “trust toward the website” and “repurchase intention using the same website”. However, hypothesis H5c of my research is not supported, suggesting that social. 治政 identity is not a moderator between “trust toward website” 大 and “repurchase intention 立 using the same website”. ‧ 國. 學. 5-2 Managerial Implications. ‧. This study provides the following three recommendations for the managers. y. Nat. er. io. sit. running an online group buying business.. 5-2-1 Importance of theaWebsite. n. iv l C n According to the results of this hresearch, i U see that customer satisfaction e n g cweh could. has a positive influence toward product trust and website trust, and this in turn would lead to repurchase intention of the product and using that particular website to make future purchases again. Therefore, it is important for the group buying websites to provide a stable system, convenient web design, fast transaction speed, and a wide array of products. I would also recommend taking satisfaction surveys regarding the website as a form of managerial reference for failure prevention and future improvements. Other than satisfaction of the website, customers’ trust toward the website is another important factor influencing customers’ intention of using the website to 39.

(47) make purchases again. One of the factors that reduces customers’ willingness to make purchases online is the uncertainty and security concerns involved. Therefore, group buying websites should ensure that they provide updated and adequate amount of product information to reduce information asymmetry between the sellers and buyers. By increasing the product information provided, and by ensuring that those information are accurate, the website would be earning customers’ trust, which would lead to repurchase intention of the customers. Also, the group buying websites should ensure that the transaction process is safe, and the personal information provided by the customers is secure. By doing so, the website would be earning customers’ trust,. 治政 and this would lead to repurchase intention as well. 大立 ‧ 國. 學. 5-2-2 Importance of the Products. Although all of the casual effects hypothesized in this research are significant,. ‧. the relationships between “satisfaction toward product”, “trust toward product”, and. sit. y. Nat. “repurchase intention of the product” are greater than the relationships between their. n. al. er. io. website counterparts. This indicates that although the websites are important, the. i Un. v. products sold on the websites have a stronger impact on customers’ repurchase. Ch. engchi. intention than the impact of the websites. Therefore, managers of group buying businesses should not neglect the importance of the products. Managers should ensure that the products sold on their website are of good quality and would deliver the functions and quality that live up to the customers’ expectations. As this research has shown, customers’ satisfaction and trust of the products would have a positive influence toward customers’ intention of making purchases with that website again, therefore products sold on the websites should be carefully chosen.. 5-2-3 Social Identity on Repurchase Intention I have hypothesized that customers’ social identity within the group buying 40.