遊戲設計策略對衍生商品購買意願之影響 - 政大學術集成

全文

(1)國立政治大學資訊管理學系碩士學位論文指導教授：尚孝純博士. 治政遊戲設計策略對衍生商品購買意願之影響大立 ‧ 國. 學. EFFECT OF GAME DESIGN STRATEGY ON PURCHASE INTENTION OF. ‧. EXTENSION PRODUCTS. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. 研究生：林冠宇中華民國 102 年 7 月.

(2) 中文摘要現今應用程式軟體（APP）不斷推陳出新，吸引不少企業與廠商投入應用程式設計的行列，原因來自於使用者高度熱衷使用這類型的程式，而此高使用度也可能正面影響使用者對贊助品牌的認同度。由於近年來資訊科技與通訊發展迅速，因而帶動應用程式設計業者設計更多動態功能與互動性功能的應用程式以期留住顧客，鼓勵他們重複性購買該應用程式之衍生商品，以提升顧客忠誠度。但隨著有越來越多的社群與手機遊戲應用程式漸漸. 政治大. 在現今生活中形成一股重要潮流後，企業應瞭解添加遊戲元素至應用程式中將能夠帶給顧. 立. 客哪些價值感受，並選取最重要的元素作為主要訴求。本研究中整理不同的遊戲設計元素. ‧ 國. 學. 及不同人口統計特徵與消費行為，用以探討不同類型顧客的消費購買意向，在檢閱過遊戲設計的相關文獻後，本研究歸納出七種遊戲設計元素，並以個案研究方式驗證這些元素. ‧. 後，進而提出研究假設，試圖找出遊戲設計對不同類型顧客的影響。接著，為驗證本研究. y. Nat. er. io. sit. 提出的假設，我們調查五百名受試者對遊戲元素的感受，希望本研究成果能提供應用程式設計者參考，設計出更貼近使用者需求的應用程式，幫助他們從衍生產品中獲得更高的收. n. al. 益。. Ch. engchi. 關鍵字衍生產品、遊戲設計元素、購買意向. 1. i n U. v.

(3) ACKNOWLEDGEMENTS I would like to express my gratitude to all those who helped me during the writing of this thesis. I gratefully acknowledge the help of my Professor, Ms. Shari S.C. Shang, who has offered me valuable suggestions in the academic studies. In the preparation of the thesis, she has spent much time reading through each draft and provided me with inspiring advice. Without her patient instruction, insightful criticism and expert guidance, the completion of this thesis would not have been possible. I also appreciate the help from Professors, Ms. Ya Ling Wu, Mr. Eldon Li, and Mr. Minder Chen, who gave me valuable comments on the research framework and have instructed and helped me a lot with my research, analysis, and surveys.. 政治大. I would also like to extend my gratitude to my friends, including Jen Jiang, Ting-Shiuan Wu, Jen-Jen Jiang, Yi-Chen Yeh. Without their generous support and help, my study life would be. 立. much tougher. Last but not least, I would like to express my sincere appreciation to my family,. ‧ 國. 學. especially my parents. Their continuous and unselfish love and support motivate me to accomplish my study.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. 2. i n U. v.

(4) ABSTRACT Applications software (“apps”) have generated substantial interest among marketers, primarily because of their high level of user engagement and the positive impact this presumably has on a customer’s attitude toward the sponsoring brand. Due to the advancement of information and telecommunication technologies, more dynamic and interactive applications have been developed to retain customers and encourage repeat purchasing for extension products and further enhance customer loyalty. As the increasing application of social and mobile games plays an important trend in today’s culture, enterprises need to understand the value of adding frequent game design elements into every customer encounter. This study consolidates different. 政治大. game design elements and demographics to explore the purchase intentions of different types of customers. We have reviewed the game-design literature and identified seven motivational. 立. elements. Using a case analysis, we verified these elements and proposed hypotheses on game. ‧ 國. 學. design for different types of customers. Then the hypotheses was tested by collecting responses from 500 participants regarding their perceptions of the elements of the game activity. We hope that the findings can provide app designer useful references in generating more revenue. ‧. from extension products.. y. Nat. Keywords. n. er. io. al. sit. Extension product, game design elements, purchase intention. Ch. engchi. 3. i n U. v.

(5) TABLE OF CONTENTS CHAPTER 1 INTRODUCTION ............................................................................................... 6 CHAPTER 2 LITERATURE REVIEW ................................................................................... 9 2.1 Means-End Chain Model ...................................................................................................... 9 2.2 The Design of Game ............................................................................................................. 9 2.3 The Design of Game and Consumer-Purchase Intention for Extension Products ...............11 2.4 The Design of Game for Different Types of Consumers .................................................... 15 2.5 Demographics ..................................................................................................................... 20 CHAPTER 3 METHODOLOGY ............................................................................................ 22. 政治大 3.2 Prototype Design and Data 立Collection................................................................................ 22 3.1 Research Framework .......................................................................................................... 22. ‧ 國. 學. CHAPTER 4 RESEARCH RESULTS .................................................................................... 25 4.1 Survey Administration ........................................................................................................ 25. ‧. 4.2 Data Analysis ...................................................................................................................... 26 4.2.1 Measurement Model .................................................................................................... 26. sit. y. Nat. 4.2.2 Structural Model ......................................................................................................... 28. io. er. 4.3 Summary Results ............................................................................................................... 29 CHAPTER 5 DISCUSSIONS AND LIMITATIONS ............................................................. 31. n. al. Ch. i n U. v. 5.1 Summary ............................................................................................................................. 31. engchi. 5.2 Implications for game design strategy ................................................................................ 31 5.3 Limitations .......................................................................................................................... 33 5.4 Conclusion .......................................................................................................................... 33 REFERENCES ............................................................................................................................ 35 APPENDIX. QUESTIONNAIRE .............................................................................................. 40. 4.

(6) LIST OF TABLES Table 1. Description of the Studied Game Programs...............................................................15. Table 2. Analysis of Game Design of Selected Studies.............................................................19. Table 3. Constructs Definition and Sources of Measurement...................................................23. Table 4. Demographic Information of Respondents (N = 369) ...............................................25. Table 5. Descriptive Statistics for the Constructs.....................................................................27. Table 6. PLS Confirmatory Factor Analysis and Cross-Loadings............................................27. Table 7. Path Coefficient of Hypotheses....................................................................................30. 政治大. LIST OF FIGURES. 立. Constructs Definition and Sources of Measurement....................................................22. Figure 2. Interaction effect model : consumer type as a moderator...........................................28. ‧. ‧ 國. 學. Figure 1. n. er. io. sit. y. Nat. al. Ch. engchi. 5. i n U. v.

(7) CHAPTER 1 INTRODUCTION Application software, also known as an application or an “app“, is computer software designed to help the user to perform singular or multiple related specific tasks. Marketers and Designers have developed increased interest in creating branded apps, conceptually defined as software downloadable to a mobile device which prominently displays a brand identity, often via the name of the app and the appearance of a brand logo or icon, throughout the user experience. One reason for the popularity of branded apps as a marketing device is their high level of user engagement and the positive impact this presumably has on attitudes toward the sponsoring brand (Hutton and Rodnick, 2009). And, in contrast to other forms of advertising, branded apps. 政治大. are welcomed as “useful”, which suggests that they may be one of the most powerful forms of advertising yet developed.. 立. To date, however, no research has tested impact of branded apps on consumers. Previous. ‧ 國. 學. research on the effectiveness of game-derived elements has largely concentrated on the effects of user engagement, especially within the game industry and the game studies community. However,. ‧. various game design mechanics also have been used by companies for many years to encourage repeat purchasing and enhanced customer loyalty. For instance, airlines, hotel chains, and retail. Nat. sit. y. stores have created various loyalty programs for developing customers. These elements have. io. er. shown to be successful for encouraging user participation and maintaining user contribution (Hunter, 2011). The increased application of social and mobile games plays an important trend in. n. al. i n U. v. culture and technology, from our iPhones to our hybrid cars and primes consumers to be. Ch. engchi. instinctive players. Enterprises should understand the value of adding game design elements into customer encounters. For instance, Foursquare offers a loyalty program as a service layer of reward and employs game design elements such as points, badges, levels, and leader boards. Starbucks has been testing Foursquare by encouraging their customers to “check in.” With check in, Starbucks customers can receive a “Barista Badge” if they check in five times. Those who receive this badge will be eligible for rewards, similar to a frequent buyer card or other traditional loyalty programs. The common business goal of game design strategy is to pull together and engage a group of customers with a common passion or interest and bring their intention to buy or stay loyal to the brand (Bunchball Inc., 2010). Although diversified elements can be applied dynamically in designing interactions between companies and customers, the ultimate goal is acceptance by the target customers.. 6.

(8) Extension products often constitute physical products as well as the associated accessories or services (Gasós and Thoben, 2003). The combined package is supposed to make the purchase of product “attractive” to the customer. Thus, depending on the type and core competencies required to supply the associated services, there may be several business partners collaborating very closely towards the common goal of making the sale of the package attractive. Most importantly, companies widely employ strategies to target customers because of the belief that these establishments can build strong brand positioning, develop awareness and enhance customer loyalty by lessening new product risk for consumers. Successful strategies positively influence the parent brand and other extension products (Swaminathan, 2003). When managed well, extension products are great source of revenue, because they reinforce brand meaning and. 政治大 The objective of this study is to understand the relationship between game design elements 立. quality, thereby helping to build brand equity (Keller and Sood, 2003).. and intention to buy extension products and to encourage further analyses into how different. ‧ 國. 學. types of consumers may react to different design elements. The research question is focused on how different game design elements can affect consumers’ intention to buy the extension. ‧. products and how customers of different attributes may moderate the effect on the relationship between game design elements and purchase intention for extension products. To answer these. y. Nat. sit. questions, we first reviewed the literature on game design and classified various game-design. al. er. io. concepts into the taxonomy of game-design elements. Then we tested these elements on 20 game. n. applications on the Web between 2010 and 2013 and identified customer profiles of different. Ch. i n U. v. kinds of game activities. Based on the target customer profile of the 20 tested cases, the study. engchi. proposed a relationship between game design elements and consumer intention to buy extension products and further proposed the moderating effect of different consumer attributes on the relationship between game design elements and consumer intention to buy extension products. The study examines the impact of game design on consumer intention to buy promoted extension products or services and further analyzes the effects of consumer differences on the game-design impact. First, we review the literature on game design and identify motivational elements that have been incorporated in the design of game. We then verify the elements of 20 programs. The case analysis helped form seven propositions on the impact of game design on consumer-purchase intentions. Further findings from the case analysis helped form the eighth proposition that differences between consumers can affect the impact of game design on. 7.

(9) consumer-purchase intentions. A proposition about the differences of consumer demographics is also suggested. We developed a research model with which to test different types of online consumers. As explained in the theory of planned behavior and verified in the cases studied, both personal and social factors can influence a consumer’s intention towards purchasing. However, it can be further inferred that instead of intrinsic and social factors influencing game design of consumers’ intentions directly and passes through consumer attitudes before impacting on intentions (Andrews and Kandel, 1979; Cheng Ang, Lim, and Tambyah, 2001).. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 8. i n U. v.

(10) CHAPTER 2 LITERATURE REVIEW 2.1 Means-End Chain Model Means-end theory, originally developed for consumer marketing by Gutman (1982), has been used in several studies of consumer-purchasing behavior (Baker, Thompson, and Engelken, 2002; Costa, Dekker, and Jongen, 2004; Futopoulos, Krystallis, and Ness, 2003; Grunert and Grunert, 1995; Walker and Olson, 1991). It is important to consider a consumer’s motivation for buying products and the means-end chain theory is a way to understand consumers’ cognitive networks related to consumption behavior. This theory explains how consumers’ knowledge about a product attributes (A) are correlated with consumers’ perceived positive consequences (C),. 政治大 abstract (e.g., taste) product characteristics. Consequences are any result (functional or 立 psychosocial) the product is perceived to deliver to the consumer. Values are intangible, higher. which may in turn be also related to personal values (V). Attributes are concrete (e.g., color) or. ‧ 國. 學. order outcomes or end-states of the consumer (Feunekes and den Hoed, 2001). For Peter and Olson (1999), values can be instrumental values which are seen as objectives and terminal. ‧. values as needs, that represent the broadest and most personal outcomes that people try to achieve in their lives. Means-end chain theory can be applied to game design strategy for many. y. Nat. sit. reasons. For example, it allows for measurement of consumer preferences and attitudes toward. er. io. specific products under different attractiveness levels (Gutman, 1982). In addition, it provides a possibility of explicitly linking consumers’ needs (consumer type) and product characteristics. n. al. Ch. i n U. v. (game-design elements), and reveals consumers’ goals/motivation (consumer-purchase intention. engchi. for extension products) in purchasing a product. 2.2 The Design of Game. The use of game design and elements has been increasingly applied to many promotional programs by enterprises—from pragmatic issues of human-computer interaction to aspects of emotion and joy in user experience. Game design elements are defined as features that motivate people’s attitudes toward online activities. In earlier studies, education experts and psychologists came to realize the importance of different types of motivation. Malone (1981) game-design elements for instructional purposes that are intrinsically motivational (challenging, fantasy, curiosity, control, cooperation, competition, and recognition) and as evidence that a person is intrinsically motived to engage an. 9.

(11) activity for its own sake rather than for any outside reward or external punishment. In the past, there were a few theorists that have argued for intrinsic motivation with respect to competence and challenge (Harter, 1978; White, 1959), whereas others have explained it with respect to optimal levels of arousal or stimulation (Hunt, 1961; Piaget, 1952), and still others focused on concepts of control and self-determination (Condry, 1977; DeCharms, 1972). However, psychometric characteristics of the questionnaire have not yet been analyzed. In other studies on perspectives of massively multiplayer online role-playing games (MMORPGs) design, there has been a lot of research that has explored the critical design components and important design features and derived experiences of MMORPG users. Lo and Wen (2010) found that users’ sense of importance of design features includes both the game-. 政治大 services and virtual item shops). The hierarchical system-structure model is a simple and 立. content system (e.g., sound and light effects) as well as a value-added system (e.g., customer. comprehensive framework that encompasses many design factors. Lin and Lin (2011) obtained. ‧ 國. 學. similar results applying an “Attributes-Consequences-Values’’ model for MMORPG players. That study found that role playing, interface design, multiplayer gaming, independent play,. ‧. popularity and virtual pets were the order of game attributes users took into consideration when playing MMORPGs. In addition, in the conceptual models of the components of game or game. y. Nat. sit. experience, according to Hunicke, LeBlanc, and Zubek (2004), a mechanics-dynamics-aesthetics. al. er. io. (MDA) framework is a formal approach to understanding games, one which attempts to bridge. n. the gap between game design and development, game criticism, and technical game research.. Ch. i n U. v. This framework depicts the relationship between the designer, who designs the mechanics or. engchi. rules of the game, and the player. These rules are embedded at playtime and forming dynamics, which are the run-time behavior of the mechanics. Aesthetics are the emotional responses the player experiences as a result of the dynamics. Although the MDA framework has provided a useful conceptual model for designing and analyzing gameplay, it has not yet addressed the existence of other aspects, including storytelling, user experience, or other influences of external conditions on the design. Overall, in understanding the various concepts of game design, the gamer’s motivation for game design seems to be an important aspect of understanding the nature of game design strategy. From a cognitive psychology perspective, Demetrovics et al. (2011) measured 129 motivational dimensions and represented 7 key motivators from 56 items (i.e., statements). These motivators. 10.

(12) were derived from a combination of at least two of the following features : (1) coping-escape (i.e., the person wants to escape reality and deal with real problem (stress, aggression, and anxiety in the virtual world); (2) fantasy (i.e., the person likes to be immersed in the virtual world to explore things); (3) skill development (i.e., the person plays games in order to improve his/her coordination, concentration, or other essential skills); (4) omnipotence-power (i.e., the person has the power to do all things); (5) recreation (i.e., the person can enjoy relaxing to engage in games or activities); (6) competition (i.e., the person likes to compete with and defeat others in order to acquire a sense of achievement); and (7) social (i.e., the person likes meeting, chatting, and playing with other players).. 治政 The consumer’s beliefs, attitudes, and behavior factors construct 大 consumer-purchase intention, 立predicting consumer-purchasing behavior (Fishbein and Ajzen, and these are important factors in 2.3 The Design of Game and Consumer-Purchase Intention for Extension Products. ‧ 國. 學. 1975). There is much research that focuses on situations where continual purchases by a buyer are used to predict customer behavior, such as Anderson and Mittal (2000), who discussed. ‧. satisfaction and performance and Dick and Basu (1994), who researched loyalty (especially pledges as well as trust using loyalty establishment and satisfaction) to make predictions about. Nat. sit. y. customer-purchasing intentions. These studies provide more understanding about the predictors of purchase intentions and have been used in satisfying diverse consumer needs.. io. n. al. er. Although consumer-purchase intention is perhaps the best predictor of actual behavior, it has. iv n C U Supporting h e n(Morwitz, purchases g c h i 2001).. long been recognized that the answer to stated consumer-purchase intention questions are not perfectly correlated with actual. consumers’ actual. purchase behavior may lead to more satisfied consumers and increase purchase intention. On the basis of analysis across the different service industries, Zeithaml, Berry, and Parasuraman (1996) stated that customers’ actual purchase behaviors and attitudinal intentions are mainly associated with a service provider’s ability to get its consumer to (1) remain loyal to them, (2) recommend them to relatives or other consumers, and (3) pay more under increased pricing. In order to trigger customer-purchase intentions, manufacturers/retailers have to explore the impact of shopping orientations. The tactic of game design nowadays is one of the emerging strategies in which to entice customer attention and involvement in an environment linked with the enterprise. Through carefully designed activities that target customer-intrinsic motivation, users could either become customers or customers could purchase more products or services.. 11.

(13) Below we discuss the impact of motivation elements of game-design strategy that have a positive influence on consumers’ intention to purchase. Lazarus (1993) found that the coping motivator stems from stress-management literature. Under stressful or threatening situations, the coping motivator presents a personal style in decision making or problem solving (Lazarus, 1993). Thus, coping should have direct behavioral implications. Some studies suggest that coping with stress and uncertainty directly affect people’s choice of behavior or decisions (Creyer and Kozup, 2003; Luce, Payne, and Bettmam, 2000). If consumers tend to adopt avoidance-coping strategies, it is expected that they will have a negative attitude toward adopting a new technology and will be less likely to purchase a new product. Meanwhile, consumers using confrontation-coping strategies are more likely to develop. 政治大 that coping has a direct effect on the attitude toward adoption and purchase intention. Thus we 立. a more positive attitude and a stronger intention toward adopting a new technology. It is expected. can form the following the hypothesis :. ‧ 國. 學. H1. The design of coping has a positive influence on consumers’ purchase intention. ‧. The escape motivator comes from escape theory, which was formulated as a hybrid of self-. y. Nat. sit. awareness theory and action-identification theory (Baumeister, 1990). According to escape. al. er. io. theory, playing a game is often the result of consumers needing a way to escape from, or mask,. n. their own self-awareness and the reality of their lives. Many consumers recognized there were. Ch. i n U. v. unhappy with the circumstances of their daily lives and turned to gaming as a way of forgetting. engchi. about their problems and escaping as a means of avoidance. Consequently, their unwillingness and inability to manage the details of their lives caused their situation to become more difficult and their need to escape becomes greater. In turn, this behavior resulted in more purchasing (Faber and Vohs, 2003). Thus we can form the following the hypothesis :. H2. The design of escape has a positive influence on consumers’ purchase intention. The fantasy motivator is related to flow theory. Csinkszentmihalyi (1997) characterized flow as involving machine interactivity, enjoyment, loss of self-consciousness, and self-reinforcing. Flow has been linked with other types of consumer activities where the individual becomes so. 12.

(14) engrossed with the activity as to create a pleasurable experience (Mannell, Zuzanek, and Larson, 1988). Hoffman and Novak (1996) argued that an online shopping environment could bring about a state of flow, which in turn leads to more browsing and, ultimately, purchasing. Meanwhile, Smith and Sivakumar (2004) propose that flow facilitates online behaviors such as browsing, shopping, and repeat purchases. Thus, this belief is that consumers who are deeply engrossed in a product that is visually appealing, entertaining, informative, or perceived to be useful may experience a state of flow. Thus we can form the following the hypothesis :. H3. The design of fantasy has a positive influence on consumers’ purchase intention. 政治大 1977; Ghani and Deshpande 1994; Hoffman and Novak 1996; Novak, Hoffman, and Yung, 2000). 立. One of the most important antecedents of flow is an individual’s level of skill (Csikszentmihalyi,. The skill-development motivator was measured as perceived by the user and not through. ‧ 國. 學. observation or a standard efficacy and defined as an individual judgment of one’s capability to use a computer (Compeau and Higgins, 1995). Computer self-efficacy has been found to affect. ‧. computer use, often through its effect on the emotional state of the user by reducing his computer anxiety (Marakas, Mun, and Johnson, 1998). Similarly, skills have been consistently found to be. y. Nat. sit. significant antecedents to flow. Therefore, we expect that as online consumers perceive their self-. al. er. io. skills to be higher, they will be more likely to have positive emotional and cognitive responses to. n. the company/brand they purchase. Thus we can form the following the hypothesis :. Ch. engchi. i n U. v. H4. The design of skill development has a positive influence on consumers’ purchase intention The recreation motivator’s major component is enjoyment. Prus and Dawson (1991) identified recreational shopping as embracing "...notions of shopping as interesting, enjoyable, entertaining and leisurely activity." In other words, recreation is characterized by the shopper-experience gratification from the shopping process per se, either in conjunction with or independent of the acquisition of goods and services (Bellenger and Korgaonkar, 1980; Guiry, Mägi, and Lutz, 2006). It is important for enterprises to note that the recreation motivator that is attendant to the shopping process is anticipated by the shopper and hence becomes a goal that is sought. 13.

(15) deliberately and may even be the impetus for a shopping trip. Thus we can form the following the hypothesis :. H5. The design of recreation has a positive influence on consumers’ purchase intention. As most games are goal directed, competitive situations or competition motivators are commonly present in game-play contexts (Frederick-Recascino, Schuster-Smith, and Frederick-Recascino, 2003). Early research revealed that competition in games plays a vital role in increasing intrinsic motivation. For example, Ryan, Rigby, and Przybylski (2006) tested how well self-determination theory (SDT) applies to game-playing motivation; results of them an experiment suggested that. 政治大 effect of competition upon intrinsic motivation may not work the same manner for all people 立. competence and autonomy were significantly associated with game enjoyment. However, the. (Locke, Latham, Smith, Wood, and Bandura, 1990). Harackiewicz and Sansone (1991) compared. ‧ 國. 學. individuals with high in motivations to individuals with low in motivations and predicted that individuals high in motivations should respond positively to competition. Consider together, the. ‧. company can be perceived by a consumer as positively competition in the same way an game is challenging or competing. The process of discovering information and using the competition. y. Nat. sit. motivator in game design can makes consumers feel challenging and ambitious with positive. n. al. er. io. effects on consumers’ purchase intention. Thus we can form the following the hypothesis :. Ch. i n U. v. H6. The design of competition has a positive influence on consumers’ purchase intention. engchi. The social motivator is related to social-identity theory. Originally from social identity theory (Ellemers, Kortekaas, and Ouwerkerk, 1999), affective identification exerts an influence on purchase intention and behavior (Chu and Li, 2012) because consumers often go beyond their personal identity to develop a social identity with the hope of articulating their sense of self (Brewer, 1991). Thus, they might identify with companies that provide a good service or product (Scott and Lane, 2000). Affective identification causes consumers to become psychologically attached to the company and care about the company (Bhattacharya and Sen, 2003), which positively motivates their purchase intention. Because consumers identify with a company rather than just its products or services, their purchase intention is likely to be immune to minor. 14.

(16) variations in product (or service) formulation and extend, ceteris paribus, to all the products and services provided by the company (e.g., Bhattacharya and Sen, 2003). Thus we can form the following the hypothesis : H7. The design of social activities has a positive influence on consumers’ purchase intention 2.4 The Design of Game for Different Types of Consumers In order to further understand the customer behavior of game playing, we use the taxonomy of game-design elements by Demetrovics et al. (2011) to verify the recently developed game programs by several enterprises between 2011 and 2013 (see Table 1 for a description of the studied game programs). These game programs were developed to promote brands, sell products. 治政 elements in the design of the game activities to target different 大 kinds of customers (see Table 2 for the analysis of game design 立 of selected studies).. or services, or enhance customer relationships. The authors applied various motivational. n. 2.Uniqlo Lucky Machine. 3.Uniqlo Heat Tech TShirt. 4.Decode Jay-Z and Bing 5.Nike Plus. Ch. engchi. 15. New Product Promotion. y V. sit. io. al. Customer loyalty. V. er. Uniqlo Lucky Line is for people eagerly awaiting the opening of a new UNIQLO Taipei store. Instead of queuing at the real store, the store site provided a chance to stand in line on the UNIQLO website via Facebook and Twitter. http://www.youtube.com/watch?v=MGCzGS2 6waw Uniqlo Lucky Machine is a pinball flash-based game offering players the chance to win prizes and online uniqlo products. http://www.uniqlo.com/luckymachine/tw/ Uniqlo Heat-Tech T-Shirt Games is giving away free Heat-tech T-shirts in Taiwan. A campaign ran from October through November 2012. Customers can play a game in real life or online to win a free piece of Heat-Tech T-shirt. http://www.uniqlo-events.com/heattech/ A multi-platform search experience and interactive game that helps fans decode Jay-Z’s life and lyrics. http://www.youtube.com/watch?v=XNic4wf8A Yg Nike+ Basketball is an interactive experience. Brand awareness. ‧. ‧ 國. Case Description & Source link. Nat. 1.Uniqlo Lucky Line. 學. Case Selected. Objectives. i n U. v. V. V. V. V. V. V. V. V. V. V. V.

(17) 7.TippEx: Hunter and Bear's 2012 Birthday Party (2012). V. 政治大V. V. 學 V. 12.Expedia Tag Me If You Can. n. 11.Yahoo Fantasy Basketball. Ch. engchi U. 16. sit er. io. 10.Recycle Bank. al. V. y. Nat. 9.Photoshop. ‧. 8.Girls Generation SHAKE. 立. V. ‧ 國. 6.Tipp-Ex: A Hunter Shoots a Bear (2010). that lets you track your in-game activity with Nike+ basketball shoes and later review, analyze, and share information using the Nike+ Basketball mobile app. http://nikeplus.nike.com/plus/ Tipp-Ex designed an exciting interactive video entitled, “A hunter Shoots a Bear (2010)” that used TippEx whip-out to erase a word from a sentence and allowed people to fill in the blank with words and come up with different endings. http://www.youtube.com/watch?v=4ba1BqJ4S 2M Tipp-Ex is ran “Hunter and Bear's 2012 Birthday Party,” a sequel to the 2010 Shoot The Bear Youtube campaign. Partying away at the Bear’s birthday bash in 2012, the Hunter and his furry friend are faced with an earthdestroying meteor in the sky above them. http://www.youtube.com/watch?v=eQtai7HMb uQ Girls' Generation SHAKE rewards the user with cards when a game is completed. These cards can be collected and include unreleased pictures of Girls' Generation, or they can be used in games like an item and users can earn higher scores. https://play.google.com/store/apps/details?id=c om.dooub.shake.ggshake Points are awarded for the completion of each mission, passing quizzes, reading the help articles, and winning virtual award badges, and social sharing, when available. http://success.adobe.com/microsites/levelup/in dex.html At Recycle Bank, you earn points for everyday “green” actions such as learning how to use less water or making greener purchases. Users can then users these points to get rewards. https://www.recyclebank.com/ Create or join a NBA league and manage your team with FREE live scoring, stats, scouting reports, news, and expert advice. http://basketball.fantasysports.yahoo.com/ “Tag Me If You Can” is an globally online interactive geo-game. Participants who have to sign up first and then tag the location based on the short video clues given by a guy within 10 meters http://www.expedia.com.au/p/tagme. v ni. V. V. V. V. V. V.

(18) Lexus is looking to connect with a younger audience by running mobile ad integration in Zynga’s popular Draw Something game. http://www.youtube.com/watch?v=4ZZ4lgEvy M8 Foursquare is a location-based social networking Website for mobile devices, such as smartphones. Users "check in" at venues using a mobile website, text messaging, or a 14.Foursquar device-specific application by selecting from a e list of venues the application locates nearby. Each check-in awards the user points and badges. https://foursquare.com/ Shopkick is a shopping app for mobile devices that offers customers rewards for walking into 15.Shopkick participating stores. It uses points called “kicks.” http://www.shopkick.com/ Catch “Buffer Monsters” and drop them off at retail locations using a 2D barcode to receive 16.Vodafone discounts and other rewards. http://www.vodafonebufferbusters.de/index_en.html#index In 2010, Mini Cooper combined a location17.Mini based sensor, AR, and social network sites to Cooper: develop an app on smartphone to challenge Getaway people for searching a virtual Mini Cooper in Stockholm the city of Stockholm. Bame (2010) http://www.youtube.com/watch?v=WMWu1h_ 6OfE Mini Cooper developed the first multiplayer 18.Mini 3D console-quality race game for Facebook, Cooper: 3D- based upon the MINI John Cooper Works. The Race Game game is on the recently re-launched Facebook (2011) page for Germany. http://www.miniusa.com/crm/racinggame.jsp Users can explore a list of all the top cheats 19.Harry and step-by-step walk-through guides until Potter they arrive at the right place. http://www.pottermore.com/ The game is available as a free download and serves to educate users of Microsoft Word, 20.Ribbon Excel, PowerPoint, and OneNote in Microsoft Hero 2 Office 2007 and 2010 how to use the ribbon interface. It is a sequel to Ribbon Hero. http://www.ribbonhero.com/ 13.Lexus Draw Something Game. V. V. 政治大V. V. V. V. V. V. V. n. engchi. sit. y. V. er. io. Ch. V. i n U. v. V. V. V. V. Table 1. Description of the Studied Game Programs. 17. V. ‧. Nat. al. 學. V. ‧ 國. 立. V. V.

(19) Based on the previous corresponding investigations, because gaming-related customers are mainly online customers, they have more shopping outlets than ever with available technology and less time to think about a purchase. Therefore, understanding impulsive buying habits and planned buying habits is important for retailers to utilize the available resources and increase sales. In early studies, the classification of a purchase as “planned” or “impulse” began with the Stern (1962) study in which the author provided the basic framework of impulse buying by categorizing a behavior as planned, unplanned, or impulse. Planned purchases involved time consumption and an information search combined with rational decision making, whereas unplanned buying referred to all shopping decisions made without any advance planning. However, impulse buying is distinguished from unplanned buying in terms of quick decision. 政治大 sudden, strong, and irresistible urge to buy. 立. making. In addition to being unplanned, an impulse purchase also involves experiencing a. The research on planned purchases remains scarce, and previous studies have mainly. ‧ 國. 學. concentrated on the anteceding factors of planned purchase. Planned-purchase buyers may have more extensive evaluative criteria and information needs than impulsive buyers. Cobb and Hoyer. ‧. (1986) found that in-store factors have a limited influence on planned purchasers whereas instore factors play a significant influence on impulse purchasers. Moreover, Bucklin and Lattin. y. Nat. sit. (1991) found that there is a significant relationship between consumer-purchase behavior and if. al. er. io. those consumers have purchase plans in advance.. n. Previous research of impulse purchase focuses mainly on characteristics and predicting. Ch. i n U. v. factors. In terms of researches on the characteristics of impulse purchase, previous studies have. engchi. found that product category has a large impact and that impulse purchase is extremely high in some product categories; for example, for fashion and jewelry products, up to 62% consumer purchases are impulse purchases whereas in other category such as pharmaceutical products, there is a low-impulse purchase ratio. In terms of anteceding factors of impulse purchase, researchers have undertaken extensive research. Stern (1962) concludes that these factors which determined impulse purchase such as low prices, the potential demand for the product, largescale distribution, self-service, large-scale advertising, in store display, short life cycle, small package size and easy storage, etc. Beatty and Elizabeth Ferrell (1998) found that impulse purchase comes from situational factors (abundant time or money) and individual factors (pleasure of shopping and the intention of impulse purchase) by building a structural-equation. 18.

(20) model. Recent studies have shown that impulse purchase was influenced by consumer sentiment (Beatty and Elizabeth Ferrell, 1998), personality (Donovan, Rossiter, Marcoolyn, and Nesdale, 1994), and demographic factors (Rook and Fisher, 1995). Skill development. V. V. V. V. 立. V. 政治 V大 V. V. V. V. Social. V. V. V. V. V. V. V. V. V. V. V. V V. V. V. ‧ V. io. V. al. V. V. V. V. V. y. V. sit. V. er. V. V. Competition. 學. V. Game design elements OmnipotenceRecreatFantasy Power ion. V. Nat. V. n. 1.Uniqlo Lucky Line 2.Uniqlo Lucky Machine 3.Uniqlo Heat tech T-Shirt 4.Decode JayZ and Bing 5.Nike plus 6.TippEx A hunter shoots a bear (2010) 7.TippEx Hunter and bear's 2012 birthday party (2012) 8.Girls generation shake 9.Photoshop 10.Recycle bank 11.Yahoo fantasy basketball 12.Expedia Tag me if you can 13.Lexus draw something game 14.Foursquare 15.Shopkick 16.Vodafone 17.Mini cooper Getaway Stockholm game (2010) 18.Mini. CopingEscape. ‧ 國. Case Selected. Ch. n e nVg c h i U. iv. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. V V V. V. V V V. V. V. V. V. V. V. V. V. V. V. V. V. V. V. 19.

(21) cooper 3DRace Game (2011) 19.Harry Potter 20.Ribbon Hero 2. V. V. V. V. V. V. Table 2. Analysis of Game Design of Selected Studies In categorizing the different types of consumers in attitude and preference for buying, Rook and Fisher (1995) demonstrated that the impulsive-buying trait was significantly related to impulsive-buying behavior, but they did not test its moderating effect. Thus, the current research attempts to add to the existing literature on impulsive buying and planned buying as well as the. 政治大 Combining findings from the game-program analyses and literature regarding different types 立 of online consumers, one can see that individuals attach different levels of importance onto. moderating role of buying impulsiveness and non-impulsiveness.. ‧ 國. 學. product attributes, consumption preferences, and the opinions of others. This results in significant differences in the experiences of game activities.. ‧. H8. The impact of different game design on consumer-purchase intention is affected by different. y. sit er. io. 2.5 Demographics. Nat. types of consumers.. al. n. iv n C h eThere factors also have to be taken into account. i U factors among online users : age, n gare c hdifferent In order to fully understand consumers’ attitudes to engage in online buying, demographic. gender, education, and income (Burke, 2002). Age differences indicate that younger people are. more interested in searching for non-specific product information, and comparing and evaluating alternatives with new technologies, such as a mobile phone (Wood, 2002). Elderly consumers have a greater preference for product-detail explanation, usage instructions, usage experience, and other product-specific functions. Gender is related to the purchase decision and has always been academically applied to many product categories (Dholakia, 1999; Hawfield and Lyons, 1998). Women have a more favorable opinion of product price, promotions, or coupons than men. Compared with female consumers, men preferred to shop with the assistance of various technologies. The next factor of interest was income. Consumers with higher incomes ($60,000/year, or above) had a greater willingness to purchase than those with lower incomes. In. 20.

(22) addition, higher-income consumers preferred credit cards over debit cards, while lower-income consumers showed otherwise. Educational background has a positive relationship with an individual’s level of Internet literacy (Li et al., 1999). For example, higher-educated consumers had a greater tendency to shop on the Internet compared with lower-educated consumer. This study assumed that education and income are significant factors that affect online-buying behavior. One reason may be that higher household incomes are often positively correlated with the possession of computers, access to the Internet, and higher education levels (Lohse, Bellman and Johnson, 2000).. H9. Different types of consumers have significant differences in demographic variables.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 21. i n U. v.

(23) CHAPTER 3 METHODOLOGY 3.1 Research Framework The paper describes a study to develop a better understanding of game-design elements that potentially influence a consumer's intention to buy extension products while participating in a game activity. It also combines game-design elements with demographics to establish a conceptual framework in which to explore purchase intentions of consumers of different types. Differences in demographics (such as age, income, etc.), and game-design elements (such as fantasy, coping, escape, etc.) are examined in Fig. 1. In addition, because Liebermann and Stashevsky (2009) have pointed out that the demographic variables on consumers may influence. 政治大. their purchase behavior, this study uses age and income as control variables in the theory of. 立. planned behavior model.. io. y. al. n. Social. sit. Competition. Nat. Recreation. Demographic (e.g. age, income, etc.). Purchase intention. er. Escape. ‧. Skill development. ‧ 國. Fantasy. 學. Coping. Ch. engchi. i n U. v. Consumer type (Impulsive Buyer v.s Planned Buyer). Figure 1. Constructs Definition and Sources of Measurement 3.2 Prototype Design and Data Collection To test the conceptual framework, data were collected with a survey instrument that uses multi-item scales to measure all research constructs (see Table 3 for construct definitions and measurement sources). The scale measuring the game-design elements was adapted from Demetrovics et al. (2011). Measures for the intention-to-buy extension products were adapted from Z (eithaml, Berry, and Parasuraman 1996). Items used to measure the consumer-type variables were adapted from Rook and Fisher (1995). The objective of this study was to. 22.

(24) empirically test the hypothesized relationship between the attributes of game-design elements and intention to buy the extension products. All survey items used a five-point interval scale (see Appendix for questionnaire). Item. Description. Sources Demetrovics, Z., Urbán, R., Nagygyörgy, K., Farkas, J., Zilahy, D., Mervó, B., Harmath, E. (2011). Game-design elements. Coping, escape, fantasy, skill development, recreation, competition, and social.. Intention to buy. Loyalty to company (loyal), propensity to switch (switch), willingness to pay more (pay more), external responses to a problem (external response), and internal responses to a problem (internal response).. Different types of consumers. Impulsive buyer vs. planned buyer. 立. 政治大. Zeithaml, Berry and Parasuraman (1996). Rook and Fisher (1995). ‧ 國. 學. Table 3. Constructs Definition and Sources of Measurement. ‧. To ensure the clarity and suitability of the questionnaire items, this study amended the questionnaire using a pretest and a pilot test. The pretest was administered to various faculty. sit. y. Nat. members as well as personnel in information systems, information engineering, and business administration to verify the face validity of the items. A total of 30 faculty members were. io. er. selected for the pretest process. The purpose of the pretest was to address any misunderstandings. al. n. iv n C U questionnaire was distributed to question items were reworded. Thereafter, h ethencorrected g c h ipretest. in the wording of the questions. In response to comments from the faculty members, several 60 people from the general public to conduct the pilot test. Cronbach’s α and factor analysis were used to verify the reliability and validity of the scales. To estimate reliability, statistical reliability tests were conducted for individual items. The Cronbach’s α of the deleted items was used to determine which items highly contributed to the reliability of the scales. The questionnaire items were considered to represent a measure of sufficient internal consistency if the total Cronbach’s α was greater than 0.6 (i.e., 60.0%–90.9%). From the factor loading attained from factor analysis, these question items exhibited a factor loading of greater than 0.5 (i.e., 51.9%–90.2%). In addition, the questionnaire was developed for study purposes, attained in accordance with a literature review, and resulted from repeated discussions and corrections. Thus, the questionnaire possessed content validity.. 23.

(25) The survey instruments were mainly used to measure the short-term and long-term future intentions to purchase extension products after participating in the game. A probability-sampling technique was developed. Formal questionnaires were distributed to the attendees of the lab. The survey was administered over a period of three days to visitors of the lab after they experienced the game activities. A total of 500 surveys were collected. A structural equation model (SEM) will be used to test each of the hypotheses with the endogenous variables. In addition, the collected survey data will analyzed using SmartPLS 2.0. First, a component-based estimation procedure with criteria of reliability, convergent validity, and discriminant validity will used to evaluate each questions. Second, a moderator analysis was used in this model, and t testes were used to examine the interaction-effect model including consumer type as a moderator.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 24. i n U. v.

(26) CHAPTER 4 RESEARCH RESULTS 4.1 Survey Administration The research model was tested with data from the customers of the Candy Crush Saga in 2012. This game is chosen because it is the most popular game on Facebook, with 45.6 million average monthly users. It is a variation of “match three” games such as Bejeweled. Each level has a different-shaped game board (the maximum game board size is 9 rows by 9 columns) filled with three-to-six different colored candies and sometimes obstacles. The basic move of the game consists of swapping two candies' positions to align sets of three of the same color, whereupon they disappear, causing candies above them to collapse into the space left behind, often resulting. 政治大 must be met before the player runs out of moves (or time on timed levels); in every case, the 立 player must earn enough points to get the first star (out of three) in the scoring system. In order. in chain reactions. Moves and alignments must be horizontal/vertical. Each levels’ objectives. ‧ 國. 學. to target the online consumers of the Candy Crush Saga, a Web-based survey was employed. A banned with a hyperlink to our survey was published via a number of bulletin board systems. ‧. (BBS), chat rooms, and virtual communities. In addition, individuals with online-game experience were cordially invited to support our survey. The respondents were instructed to. y. Nat. sit. answer all of the questions based on their game experience with Candy Crush Saga. The Web. al. er. io. survey yielded a total of 369 complete, valid responses for the data analysis (See Table 4 for lists. n. of demographic information about the respondents). Measure Gender. Age. Income. Items. Male Under 21 21-30 31-40 41-50 50-64 Over 65 Less than 10,000 10,001-20,000 20,001-35,000 35,001-50,000 Over 50,000. i vItems Freq. Measure n C256 U h e nGender Female i h c g 160 209 0 0 0 0 30 287 31 21 0. Freq. 113. Education. High school & Below College Graduate school. 50 258 61. Online game experience (in years). <1 1-2 3-5 5~. 22 42 95 210. Table 4. Demographic Information of Respondents (N = 369). 25.

(27) 4.2 Data Analysis A two-step approach, recommended by Anderson and Gerbing (1988), was adopted for the data analysis. The first step involves the analysis of the measurement model while the second step tests the structural relationships among the latent constructs. The aim of the two-step approach is to establish the reliability and validity of the measures before assessing the structural relationship of the model. SmartPLS 2.0 was used because it allows latent constructs to be modeled as formative or reflective indicators. SmartPLS 2.0 places minimal restrictions on the measurement scales, sample size, and residual distribution. 4.2.1 Measurement Model. 政治大 development, fantasy, social, recreation, and competition motivator should be regarded as 立 forming the game-design elements rather that the other way around. Second, each game-design. First, according to the conceptual definition of game-design elements, coping, escape, skill-. ‧ 國. 學. element is clearly unique, distinguishable, and not interchangeable. Third, each game-design elements is independent. They are not highly correlated. While this approach repeats the number. ‧. of manifest variables used, the model can be estimated using the standard PLS algorithm (Chin,. sit. Nat. approximately equal numbers of indicators for each construct.. y. Marcolin, and Newsted, 1996). The repeated-indicators approach can be used with. er. io. The adequacy of the measurement model was evaluated based on the criteria of reliability,. al. iv n C reliability values. Table 5 shows that all of the values areU h e n g c h i above 0.7, satisfying the commonly acceptable level. The convergent validity of the scales was assessed by two criteria (Fornell and n. convergent validity, and discriminant validity. Reliability was examined based on the composite. Larcker, 1981): (1) all indicator loadings should be significant and exceed 0.7 and (2) the average variance extracted (AVE) by each construct should exceed the variance due to the measurement error for that construct (i.e., AVE should exceed 0.50). Table 6 shows that all of the items exhibit a loading higher than 0.7 on their respective constructs, and Table 5 shows that all of the AVEs range from 0.6214 to 0.7591, thus satisfying both criteria for convergent validity. Discriminant validity was examined using the following two tests. First, the cross-factor loadings (see Table 6 for PLS confirmatory-factor analysis and cross-loadings) indicate good discriminant validity because the loading of each measurement item on its assigned latent variable is larger than its loading on any other construct (W. W. Chin, 1998). Second, the square. 26.

(28) root of the AVE from the construct is much larger than the correlation shared between the construct and other constructs in the model (Table 5) (Fornell and Larcker, 1981).. COM CP ES FA RE SK SO PUR. Items 3 3 3 3 3 3 3 3. AVE 0.6345 0.6214 0.6988 0.7413 0.705 0.6228 0.7149 0.7591. Composite Reliability R Square Cronbach's alpha 0.8384 0 0.7263 0.83 0 0.7128 0.8743 0 0.7857 0.8958 0 0.826 0.8763 0 0.8066 0.8315 0 0.6985 0.8826 0 0.8015 0.9041 0.520 0.8403. 政治大 ES FA RE SK. Table 5. Descriptive Statistics for the Constructs. n. Ch. engchi. 0.4726 0.5003 0.3816 0.4656 0.5398 0.418 0.431 0.4335 0.4552 0.5416 0.4306 0.4735 0.2676 0.5238 0.5617 0.7529 0.7553 0.8551 0.4358 0.4098 0.4879 -0.1934 -0.17 -0.2376. PUR. 0.6194 0.3975 0.4833 0.3998 0.415 0.3341 0.2884 0.3595 0.398 0.5621 0.3528 0.4105 0.2703 0.5008 0.4201 0.4483 0.32 0.4794 0.8503 0.8207 0.865 -0.202 -0.1445 -0.2418. -0.1269 -0.1201 -0.0578 -0.1094 -0.0536 -0.0654 -0.1167 -0.1441 -0.1113 -0.2802 -0.2406 -0.2479 -0.0429 -0.1509 -0.0878 -0.1529 -0.174 -0.2148 -0.1954 -0.1678 -0.211 0.8741 0.8149 0.9216. y. sit. ‧ 國. io. al. 0.4966 0.2845 0.3443 0.3556 0.3317 0.8179 0.8721 0.8165 0.5469 0.4879 0.6099 0.2402 0.3739 0.4109 0.3714 0.4794 0.3964 0.312 0.3219 0.4137 -0.1289 -0.1068 -0.1526. 0.4219 0.3402 0.3656 0.6213 0.616 0.537 0.303 0.3296 0.4213 0.3566 0.2629 0.3502 0.723 0.9486 0.832 0.4159 0.4318 0.487 0.4355 0.406 0.4304 -0.1216 -0.0771 -0.1385. SO. ‧. 0.3534 0.3299 0.8785 0.7019 0.7745 0.2552 0.3474 0.475 0.3649 0.2583 0.3365 0.4689 0.6711 0.6961 0.4708 0.3857 0.517 0.4071 0.3481 0.4441 -0.142 -0.0236 -0.1017. 0.3324 0.5532 0.3205 0.2812 0.3991 0.2578 0.4555 0.6117 0.5166 0.8653 0.8449 0.8725 0.1925 0.378 0.3194 0.4509 0.3429 0.5289 0.433 0.3764 0.4981 -0.2615 -0.2354 -0.282. 學. 0.8546 0.793 0.7378 0.3199 0.4601 0.3515 0.3399 0.3807 0.3995 0.5559 0.3774 0.3839 0.1921 0.4666 0.4247 0.3987 0.4047 0.544 0.5083 0.5702 0.5193 -0.1191 -0.0805 -0.1537. Nat. COM_1 COM_3 COM_4 CP_1 CP_3 CP_4 ES_2 ES_3 ES_4 FA_2 FA_3 FA_4 RE_1 RE_2 RE_3 SK_1 SK_3 SK_4 SO_1 SO_2 SO_3 PUR_1 PUR_2 PUR_3. 立 0.2651 0.3854 CP. er. COM. i n U. v. Table 6. PLS Confirmatory Factor Analysis and Cross-Loadings. 27.

(29) 4.2.2 Structural Model In formulating and testing for interaction effects using PLS, we need to follow a hierarchical process that is similar to that used in multiple regression, in which we examined the interactioneffect model including consumer type as a moderator. Fig. 2 indicates that part of the paths exhibit a P-value of less than 0.05, such as skill development, recreation, social, . Overall, the model accounts for 51.3% of the variance of purchase intention. In addition, the path coefficient between consumer type and purchase intention is 0.688. After analyzing the interaction-effect model, we examined the R-square in order to assess the interaction effect (W. Chin, Marcolin, and Newsted, 1996). Figure 2 shows that the inclusion of the interaction effect with a strong beta. 政治大. between -0.449 and 0.945 increases the R-square for consumer-purchase intention to 0.513. The interaction-effect model possesses significantly high-explanatory power. Therefore, consumer. 立. type has a significantly positive effect on the relationship between game-design elements and. ‧ 國. 學. consumer-purchase intention. * p < .05, ** p < .01, *** p < .001. 0.077. Coping. ‧. -0.395* 0.412***. 0.124. -0.207. Escape. Purchase intention. Competition. al. n. 0.349**. 0.501**. Ch -0.068. 0.310* -0.449*. Social. -0.378 0.355*. Demographic (e.g. age, income, etc.). er. io. 0.025. Recreation. y. (R square = 0.513). Nat. Skill development. sit. Fantasy. 0.945*. engchi. 0.688**. i n U. v. -0.135. 0.342. Consumer Type (Impulsive Buyer v.s Planned Buyer). Figure 2. Interaction effect model : consumer type as a moderator Further, we also examined the effect of the respondents answers to nine consumer-type questions about purchase intention. We conducted another analysis by remodeling consumer type with five indicators with a higher-loading factor. The path coefficient for the relationship between the consume type and purchase intention becomes 0.688, whereas the path coefficient for the original model with the consumer type having nine indicators is 0.766. The path coefficient difference is 0.087, which is insignificant (t=0.77) based on the formula proposed by. 28.

(30) (Johnson Jr, Johnson, and Buse, 1987). Accordingly, the nine consumer-type questions do not cause significant problems. 4.3 Summary Results The results indicate that part of game-design elements have a significant effect on purchase intention for extension products and consumer type exerts a positive moderating effect on the relationship between game-design elements and intention to buy extension products (see Table 7 for path coefficient of hypotheses). The path coefficient of interaction term of game-design elements and consumer type on purchase intention for extension products is significant, moreover, the path coefficient is positive, further a stronger moderating effect of consumer type. 政治大. will increase the influence of game-design elements on purchase intention for extension products, so that hypothesis H8 can be accepted.. 立. In addition, with regard to the direct antecedents of purchase intention for extension products,. ‧ 國. 學. the analysis yields against three positive (H4, H6 and H7) and a negative (H3) answers. The perception of game-design elements, indeed, seems to be a highly important predictor of. ‧. purchase intention in a skill-development : a path coefficient of 0.612 at a high degree of significant leaves no doubts. In contrast, the other three expected antecedents of purchase. Nat. sit. y. intention (H1, H2, H5), coping, competition, and fantasy, and escape, seems to have no influence. io. zero.. er. on purchase intention at all in the present analysis, with an insignificant path coefficient close to. al. n. iv n C h e n g ceducation across respondents’ h i U levels. Further data analysis indicates that respondents’ attitudes toward online buying show significant mean differences. and income. Specifically,. Burke (2002) showed that statistically significant mean differences exist between respondents with some college experience and postgraduate school experience. These results reinforce those from previous studies (Lohse, Bellman and Johnson, 2000), which suggest that education has been positively associated with individual’s level of Internet literacy (Li, Kuo, and Rusell, 1999). Those who use the Internet as a routine tool and have higher incomes with a higher intention to shop online. However, in the current study there were no statistically significant differences of mean scores by income levels and education levels in consumers’ product attitudes and purchase intentions, so that hypothesis H9 is rejected. It is possible to assume that the majority of respondents were university students in the current study, thus their household incomes do not reflect each respondents’ individual income level.. 29.

(31) Table 7. Path Coefficient of Hypotheses. Path. ßeta. t-statics. H1. Coping → Purchase Intention. 0.077. 0.468. H2. Escape → Purchase Intention. -0.207. 1.231. H3. Fantasy →Purchase Intention. -0.395*. 1.974*. 0.412***. 3.471***. H5. Recreation → Purchase Intention. 0.025. 0.274. H6. Competition → Purchase Intention. 0.349*. 2.121*. H4. Skill-Development → Purchase Intention. H7. Social → Purchase Intention H8. Consumer type as a moderator. 立. 2.069* 2.591**. -0.135. 0.876. ‧. ‧ 國. 學. H9. Demographic → Consumer type * p < .05. ** p < .01. *** p < .001.. 0.501* 治政 0.688**大. n. er. io. sit. y. Nat. al. Ch. engchi. 30. i n U. v.

(32) CHAPTER 5 DISCUSSIONS AND LIMITATIONS 5.1 Summary The purpose of this study is to thoroughly examine the complex relationships between gamedesign elements, consumer type, and purchase intention in the context of game design strategy. We proposed two research question : 1) how different game design elements can affect consumers’ intention to buy extension products ? and 2) how consumers of different attributes may moderate the effect on the relationship between game design elements and purchase intention for extension products ? For the first question, we intent to classify various gamedesign concepts into the taxonomy of game-design elements through literature review, then we. 政治大 customer profiles of different kinds of game activities. Based on the target customer profile of 立 the 20 tested cases, we proposed a relationship between game design and consumer intention to tested these elements on 20 applications on the Web between 2010 and 2013 and identified. ‧ 國. 學. buy extension products. Further study is planned to test the hypotheses by collecting responses from 500 participants regarding their perceptions of the elements of the game activities. The. ‧. design element of game can affect consumers’ intention to buy extension products : Skill Development, Competition, Social, and Fantasy. For the second question, we discuss the. y. Nat. sit. different types of consumers with different attitude and preference for buying, characteristics,. al. er. io. then propose proposed the moderating effect of different consumer attributes on the relationship. n. between game-design elements and consumer intention to buy extension products : Recreation,. Ch. i n U. v. Social, Fantasy, and Skill Development. These findings can facilitate the design of game by. engchi. asking such question : how different game design elements can affect consumers’ intention to buy extension products ? and how consumers of different attributes may moderate the effect on the relationship between game design elements and purchase intention for extension products ? We hope that the findings can provide app designer useful references in generating more revenue from extension products. 5.2 Implications for game design strategy Based on the findings of this study, we would like to share some points that are worthy of consideration. First, many studies consider purchase behavior as being guided by reasoning. Accordingly, much effort has been devoted to trying to explain purchase intention from the perspectives of perceived usefulness, perceived ease of use, perceived risk, trust, online purchase. 31.

(33) experience, etc. However, as game design strategy becomes more popular and more people shop online without much reasoning process, consumer type may play a role in consumer-purchase intention. This study holds important implications for consumer research and can be assessed from both a theoretical and a practical perspective. Most importantly, this study provides a model that examines inter-relationship between the factors important to consumer-purchase intention. In the past, several studies have attempted to explore different types of consumer, but there has been no study that applied an overall framework. Besides, no study has investigated the impact of game design on consumer intention and the moderating effect of different consumer attributes on the relationship between game-design elements and consumer intention to buy extension products.. 政治大. We therefore hope that the present study will contribute to the further development of the. 立. concept of the game design strategy.. Second, our findings show that the link between game-design elements and consumer-. ‧ 國. 學. purchase intention is rather complex. Adding consumer type as a moderator is a first step toward establishing a better understanding of this relationship. Future research should continue to. ‧. explore the complex relationship between consumer type, game-design elements, and consumerpurchase intention. For example, what would be the threshold level of game-design elements at. y. Nat. sit. which the influence of game-design elements on consumer-purchase intention will increase and. al. er. io. the role of consumer type will become influential ? It would also be interesting to know whether. n. in an extreme case, the effect of game-design elements will be nullified by the effect of consumer. Ch. i n U. v. type. In other words, at what level of game-design elements will the consumer type become the. engchi. main driver of consumer-purchase intention ?. Third, impulsive buying and planned buying play an important part of consumer behavior. More purchases result from impulse than from planning buying (Sfiligoj, 1996). However, compulsive or addictive online buying is a form of unregulated consumer behavior (LaRose and Easti, 2002). LaRose and Eastin (2002) found that addiction is a powerful predictor of onlineshopping activity. The role of addiction has been a popular topic in economic studies on consumer behavior (Martin, 1999). Accordingly, an interesting area for future research would be to investigate the relationship between consumer type and addiction to see if the latter has a stronger suppressive effect than consumer type on the relationship between game-design elements and consumer-purchase intention.. 32.

(34) Finally, although consumer type cannot be determined by simply looking at the frequency of past behavior (Verplanken and Orbell, 2003), shopping frequency is still a necessary condition for online-buying behavior to develop. Product types are correlated with online-shopping frequency. For example, an individual may purchase clothes every season but only purchase luxury goods once a year. Therefore, an interesting area for future research would be to examine the relationship between product type and online-buying behavior. 5.3 Limitations We note that our findings must be interpreted in light of the study’s limitations. First, the data were collected from a single game, which also has a reputation as the most popular game on. 治政 programs in Taiwan are also multi-purpose sites. 大 Uniqlo Lucky Machine 立 (http://www.uniqlo.com/luckymachine/tw/), for example, was known primarily as a famous. Facebook. However, Candy Crush Saga is not unique in this respect. Other well-known game. ‧ 國. 學. application in 2011. More important, most applications use very similar principles and policies. Nonetheless, the generality of the model and findings to other applications requires additional. ‧. research. Second, the results may have been impacted by self-selection bias. Our sample comprises only university students and post-graduate students as current consumers. Perhaps. Nat. sit. y. other consumers who had already participated in Candy Crush Saga might have different perceptions about the influence of the seven dimensions of coping, fantasy, escape, recreation,. io. er. competition, social, skill-development and consumption behavior, and so could have been. al. n. iv n C consumer-purchase intentions of currenthconsumers. i U the results can be generalized to e n g c hWhether. differently affected by them. Therefore, the results should be interpreted as only explaining the. nonparticipants or to disaffected participants will require additional research. Third, as the data are cross sectional and not longitudinal, the posited causal relationships could only be inferred rather than proven. Finally, although this study suggests that game-design elements could be applied in the context of applications, the link may not be as strong as that in the context of development of platform. 5.4 Conclusion In this paper, we have extended the body of knowledge with respect to understanding the impact of game design on consumer-purchase intentions and to encourage further analyses into how different types of consumers may react to different design elements. The major concerns of. 33.

(35) this research are the moderating effect of consumer type on the relationship between gamedesign elements and consumer-purchase intention and the future antecedents of consumerpurchase intention. As verified by our data, there is an interaction effect between game-design elements and consumer type in the prediction of consumer-purchase intention. Although online consumer-purchase intention is influenced by the rational analysis of game-design elements in this Candy Crush Saga game, consumer type is an even more important driving force. Further, the four antecedents of consumer-purchase intention are recreation, social, fantasy, and skill development ranked in descending importance. In conclusion, our study highlights the importance of the role of the consumer type and antecedents of consumer-purchase intention in the achievement of online-customer retention. More refined knowledge about the proposed. 政治大 attempts to cope successfully with the challenges posed by the perpetually versatile applications. 立. variables and relationships may ultimately benefit enterprises and consumers alike in their. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. 34. i n U. v.

(36) REFERENCES 1.. Anderson, E. W., & Mittal, V. (2000). Strengthening the satisfaction-profit chain. Journal of Service Research, 3(2), 107-120.. 2.. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological bulletin, 103(3), 411.. 3.. Andrews, P. and D. B. Kandel. (1979). Attitude and Behaviour: A Specification of the Contingent Consistency Hypothesis. American Sociological Review 44, 298–310.. 4.. Ang, S. H., Cheng, P. S., Lim, E. A. C. and Tambyah, S. K. 2001. Spot the difference: Consumer responses towards counterfeits. Journal of Consumer Marketing,18(3), 219 – 235.. 5.. Baker, S., Thompson, K., & Engelken, J. (2002). Mapping the values driving organic food choice. Germany vs. the UK. European Journal of Marketing, 38(8), 995–1012.. 6.. Beatty, S. E., & Elizabeth Ferrell, M. (1998). Impulse buying: modeling its precursors. Journal of Retailing, 74(2), 169-191.. 7.. Bellenger, D. N., & Korgaonkar, P. K. (1980). Profiling the recreational shopper. Journal of retailing, 56(3), 77-92.. 8.. Bhattacharya, C. B., & Sen, S. (2003). Consumer-company identification: a framework for understanding consumers' relationships with companies. Journal of marketing, 76-88.. 9.. Brewer, M. B. (1991). The social self: On being the same and different at the same time. Personality and Social Psychology Bulletin, 17(5), 475-482.. 立. 政治大. ‧. ‧ 國. 學. sit. y. Nat. 10. Bunchball Inc. (2010). Gamification 101: An Introduction to the Use of Game Dynamics to Influence Behavior. http://www.bunchball.com/gamification/ gamification101.pdf.. n. al. er. io. 11. Burke, R. R. (2002). Technology and the customer interface: what consumers want in the physical and virtual store. Journal of the Academy of Marketing Science, 30(4), 411-432.. i n U. v. 12. Chu, K.-K., & Li, C.-H. (2012). The study of the effects of identity-related judgment, affective identification and continuance commitment on WOM behavior. Quality & Quantity, 46(1), 221-236.. Ch. engchi. 13. Chin, W. W. (1998). The partial least squares approach for structural equation modeling. 14. Chin, W., Marcolin, B., & Newsted, P. (1996). A partial least squares latent variable modeling approach for measuring interaction effects: results from a Monte Carlo simulation study and voice mail emotion/adoption study. 15. Compeau, D. R., & Higgins, C. A. (1995). Computer self-efficacy: Development of a measure and initial test. MIS quarterly, 189-211. 16. Condry, J. (1977). Enemies of exploration: Self-initiated versus other-initiated learning. Journal of Personality and Social Psychology, 35(7), 459. 17. Costa, A., Dekker, M., & Jongen, W. (2004). An overview of means-end theory:Potential application in consumer-oriented food product design. Trends in Food Science & Technology, 15, 403–415.. 35.