行動裝置APP價值創新與擴散之研究─以LINE為例 - 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: 姜國輝博士. 立. 政治大. ‧ 國. 學. 行動裝置 APP 價值創新與擴散之研究. ‧ sit. y. Nat. ─以 LINE 為例. n. er. io. A Research into the Value Creation and Diffusion a l APPs on Cases of v LINE of Mobile i n Ch engchi U. 研究生：黃詩貽中華民國一○三年七月.

(2) Acknowledgements I would like to express my great appreciation to those who have supported me during these two years of study. I would not have accomplished it without them. I must express my sincere gratitude to my advisor, Dr. Johannes Chiang, for guiding me and inspiring me. I would not have found the topic I am interested in researching without the guidance of him. He offered me opportunities to co-work with doctoral seniors on theses and conference papers. From the experience I learnt to research and was inspired to have a deeper understanding of the topic.. 政治大 discuss papers with me and 立gave me advices. Every brainstorming with them made I would like to thank the doctoral seniors and my classmates who spent time to. ‧ 國. 學. my train of thought clearer. Among them, I especially want to thank TaYu Chen, Collin Suen and Joshra Lin. They were such a great support during the time I was. ‧. working on this study.. Nat. sit. y. I would also like to thank the committee for sharing their knowledge with me. n. al. er. io. and giving me advices on revision. Furthermore, their encouragements emboldened. i n U. v. me to pay more effort into the research. I would like to thank Dr. Chang-Sung Yu for. Ch. engchi. giving me ideas on value creation, Dr. Linda Fu for coaching me the writing of research method, and Dr. Yan-Ping Chi for showing consensus toward my idea which is such an encouragement for me. Moreover, I want to thank the department of MIS in NCCU for creating such a wonderful environment for study and research. I am grateful to study here. The faculty of the department had also assisted me in many ways. Last but not least, I would like to thank my parents, my grandparents and my siblings. They are always there for me to support me and to encourage me through all kind of difficulties I encountered. Their wishes are my strength to move on..

(3) 摘要自從 Apple 於 2007 年推出了第一隻 iPhone，智慧型裝置的時代也來臨。HTC 也從 2008 年開始陸續發表了數隻 Android 為系統的手機。iPhone 與各廠牌 Android 系統手機的競爭炒熱了智慧型手機的市場。於 2013 年，智慧型手機的全球銷售量終於超過了一般功能性手機 (Feature Phone)。智慧型裝置的流行也帶動了行動應用程式(Mobile Apps)的發展。智慧型手機系統的兩大龍頭 Google 與 Apple 皆於 2013 年宣布其應用程式商店的應用程式數量已經超過五千萬，而且下載量也超過一百萬次。隨著應用程式產業的蓬勃發展，也出現了各式各樣的營. 政治大本研究試圖了解影響使用者購買應用程式內商品的變數，並以有多樣種類商立. 利模式，其中應用內購買(In-App Purchasing)是最近比較流行的一種模式。. ‧ 國. 學. 品的 LINE 為研究案例。本研究的假設是建立於創新擴散理論 (Innovation Diffusion Theory)、深思可能性模式(Elaboration Likelihood Model)以及延伸的科. ‧. 技接受度模型(Extended-TAM)。. sit. y. Nat. 本研究以問卷的方式調查使用者行為並收集了 301 個有效樣本，並以. al. er. io. SmartPLS 分析資料。研究結果發現：一、社會網絡對於使用者購買應用程式內. v. n. 商品的態度沒有直接關系；二、社會網絡影響使用者的感知趣味性以及個人涉入. Ch. engchi. i n U. 程度；三、使用者的感知趣味性以及個人涉入程度對於使用者購買應用程式內商品的態度有直接影響。. 關鍵字：創新擴散理論(Innovation Diffusion Theory)、思考可能性模式(Elaboration Likelihood Model)、延伸的科技接受度模型(Extended-TAM)、個人涉入、應用程式內購買(In-App Purchases).

(4) Abstract Since Apple introduced the first iPhone in 2007, the market share of smart devices is raging. HTC introduced several Android-based smart phones in 2008. The competition between iPhone and Android phones has heated up the market of smartphone. By 2013, worldwide sales of smartphones surpassed sales of feature phone. The prevailing state of smart devices has also stimulated the development of mobile application (Apps) industry. In year 2013, both Apple and Google announced that the number of Apps published in their stores has reached over 1 million Apps and. 政治大 of Apps, in-App purchasing立 is becoming the favorable one.. the number of Apps downloaded has reached over 50 billion. Among all profit models. ‧ 國. 學. This study attempts to understand the causes that would influence users’ behavior when deciding to purchase products in Apps such as LINE. The model of the. ‧. study is based on Innovation Diffusion Theory, Elaboration Likelihood Model (ELM). Nat. sit. y. and the Extended Technology Acceptance Model (Extended-TAM).. n. al. er. io. In this study, data was collected from 301 valid respondents through both online. i n U. v. questionnaires and paper copies. SmartPLS was employed as data analysis tools.. Ch. engchi. Result revealed that: (1) Social network effects hardly have direct impact on the attitude toward in- App purchases; (2) Social network effects have impact on a user’s perceived playfulness and personal involvement; and (3) A user’s perceived playfulness and personal involvement have impact on the attitude toward in-App purchases.. Keywords: innovation diffusion theory, elaboration likelihood model (ELM), extended technology acceptance model (extended-TAM), Personal Involvement, in-App purchases.

(5) Content List of Figure ............................................................................................................... iii List of Table ................................................................................................................. iv Chapter I. Introduction .......................................................................................... - 1 1.1. 1.2.. Research Background ........................................................................... - 1 Research Motivation ............................................................................. - 3 -. 1.3.. Research Objective ............................................................................... - 5 -. 1.4. Research Process................................................................................... - 7 Chapter II. Literature Review ............................................................................... - 9 5.1. Diffusion of Innovation......................................................................... - 9 -. 5.2 5.3. Extended Technology Acceptance Model (TAM).............................. - 14 Elaborating Likelihood Model (ELM) ................................................ - 15 -. 5.4 5.5. 政治大 Involvement ........................................................................................ - 18 立 Value Creation Cycle (VCC) .............................................................. - 19 3.2.1. Social Network Effects and Personal Involvement.................. - 22 Perceived Playfulness and Attitude toward in-App Purchase .. - 23 -. 3.2.5 3.2.6. Personal Involvement and Attitude toward in-App Purchases. - 23 Attitude toward in-Apps purchasing and Purchase Intention... - 24 -. sit. y. 3.2.4. n. al. er. Social Network Effects and Attitude toward in- App Purchase - 23 -. io. 3.4. Social Network Effects and Perceived Playfulness.................. - 22 -. Nat. 3.2.2 3.2.3. 3.3. ‧ 國. Research Framework........................................................................... - 21 Hypothesis........................................................................................... - 22 -. ‧. 3.1 3.2. 學. Chapter III. Research Methodology ................................................................... - 21 -. Ch. engchi. i n U. v. Variables and Operational Definitions................................................ - 24 3.3.1 Social Network Effects............................................................. - 24 3.3.2. Perceived Playfulness............................................................... - 26 -. 3.3.3 3.3.4. Personal Involvement ............................................................... - 27 Attitude toward in-App Purchase ............................................. - 29 -. 3.3.5. Intention to perform in-App purchase ...................................... - 30 -. Research Design.................................................................................. - 31 3.4.1 Questionnaire Design ............................................................... - 31 3.4.2. Target Subjects ......................................................................... - 31 -. 3.4.3 Data Collection......................................................................... - 31 3.5 Data Analysis ...................................................................................... - 32 3.5.1 3.5.2 3.6. PLS-SEM ................................................................................. - 32 Model Verification ................................................................... - 33 -. Pretest.................................................................................................. - 34 i.

(6) 3.6.1. Demographic analysis for Pretest............................................. - 34 -. 3.6.2 Measurement Assessment for Pretest....................................... - 36 Chapter IV. Result ................................................................................................ - 38 4.1. Demographic analysis ......................................................................... - 39 -. 4.2. Confirmatory Factor Analysis............................................................. - 42 4.2.1 Reliability Test ......................................................................... - 43 4.2.2. 4.3 4.4. Validity Test ............................................................................. - 43 -. Structural Model Assessment ............................................................. - 45 Hypothesis Testing.............................................................................. - 47 -. Chapter V. Discussion .......................................................................................... - 48 5.1 Discussion ........................................................................................... - 48 5.1.1. Summary of Results ................................................................. - 48 -. 5.1.2 Contribution and Key Insights ................................................. - 49 5.2 Research Limitations........................................................................... - 50 5.4. 學. ‧ 國. 5.3. 政治大 Conclusion .......................................................................................... - 50 立 Future Research and Suggestions ....................................................... - 51 -. Reference................................................................................................................ - 52 Appendix I. Questionnaire of the Study ............................................................. - 57 -. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. ii. i n U. v.

(7) List of Figure Figure 1: App revenue growth .................................................................................. - 1 Figure 2: App Revenue in Key Countries ................................................................. - 5 Figure 3: LINE’s revenue by products...................................................................... - 6 Figure 4: An example of game purchase .................................................................. - 6 Figure 5: An Example of free sticker........................................................................ - 7 Figure 6: An example of paid sticker........................................................................ - 7 Figure 7: Research Process ....................................................................................... - 8 Figure 8: The Elements of Diffusion ........................................................................ - 9 Figure 9: Perceived Characteristics of Innovations ................................................ - 10 Figure 10: Innovation-Decision Process ................................................................. - 11 -. 政治大 Figure 13: The two routes to 立persuasion ................................................................. - 16 -. Figure 11: Variables determining the rate of adoption of innovations ................... - 11 Figure 12: Extended Technology Acceptance Model............................................. - 15 -. ‧ 國. 學. Figure 14: Comparison of two supply chain (Yu, 2012) ........................................ - 19 Figure 15: Research Framework ............................................................................. - 21 Figure 16: PLS-SEM Procedure ............................................................................. - 33 -. ‧. Figure 17: Path Analysis ......................................................................................... - 46 Figure 18: Model indicated by Research Result ..................................................... - 49 -. n. er. io. sit. y. Nat. al. Ch. engchi. iii. i n U. v.

(8) List of Table Table 1: Comparison of App store and Google Play ................................................ - 2 Table 2: Business Model for Apps............................................................................ - 3 Table 3: Previous Study Basing on Diffusion of Innovation .................................. - 13 Table 4: Construct Definition of Social Network Effects....................................... - 25 Table 5: Construct Definition of Perceived playfulness ......................................... - 26 Table 6: the original PII Scale................................................................................. - 28 Table 7: Construct Description of Personal Involvement ....................................... - 28 Table 8: Construct Description of Attitude toward in-App Purchase ..................... - 29 Table 9: Construct Description of Intention to perform in-App purchases ............ - 30 Table 10: Demographic analysis of pretest sample ................................................ - 34 -. 政治大 Table 13: Demographic Analysis............................................................................ - 39 立. Table 11: Pretest Overview..................................................................................... - 36 Table 12: Questionnaire Sections and Constructs .................................................. - 38 -. ‧ 國. 學. Table 14: Comparison between Genders on Average of each Construct ............... - 41 Table 15: Comparison between users on Average of each Construct .................... - 42 Table 16: Overview of CFA.................................................................................... - 42 -. ‧. Table 17: Matrix of Loadings ................................................................................. - 44 Table 18: Comparison between Root of AVE and Correlations............................. - 45 -. sit. y. Nat. Table 19: t-Value and Path Coefficient................................................................... - 46 -. io. n. al. er. Table 20: Summary of Hypotheses Test Result...................................................... - 47 -. Ch. engchi. iv. i n U. v.

(9) Chapter I. Introduction 1.1. Research Background Since Apple introduced iPhone in 2007 and HTC introduced several Android-based phones started from year 2008, smart devices have taken down the world steps by steps. In year 2013, worldwide sales of smartphones surpassed sales of feature phone devices for the first time, according to Gartner’s latest market estimates on February 13, 2014 – with 968 million smartphone device units sold to end users in 2013 out of a total of 1.8 billion mobiles sold.. 政治大. The adoption of iPhone, Android-based smartphone, iPad and other smart. 立. devices are booming the development and usage of mobile applications (To be. ‧ 國. 學. short, I will called it “Apps” in the following paragraphs). Furthermore, it is driving the App revenue. According to the report from App Annie, a mobile. ‧. analytics company, and IDC (International Data Cooperation), the massive. y. Nat. sit. growth in smart handsets have dramatically expanded opportunity to reach. n. al. er. io. consumers which also means the chances of gaining cash flows have also. i n U. v. increased. In other words, App revenue growth is driven by gains in device adoption. and. Ch. revenue. i e nperg c hdevice. Figure 1: App revenue growth (Source: App Annie and IDC) -1-. (See. Figure. 1)..

(10) Although there are many players in the market, Apple’s iOS and Google’s Android system are the biggest players. Each player in the market runs its own digital distribution platform for Apps. For example, Apple runs “App Store” and Google held “Google Play” (Formerly, the “Android Market”). Other than Apps, these markets also include music, magazines, books, movies and television programs. In year 2013, both Apple and Google announced that the number of Apps in their stores has reached over 1 million Apps published and over 50 billion Apps have been downloaded according to The Verge and PhoneArena,. 政治大. websites for technology news and media networks. Table 1 shows comparison of these two markets.. 立. 學. ‧ 國. Table 1: Comparison of App store and Google Play App Store. Google Play. ‧. Icon. Nat. Numbers of published. Over 1 millions. sit. er. n. al. y. Over 50 millions. io. Numbers of downloads. Ch. engchi. i n U. Market share. Quarterly Revenue. (Source: TechRpublic) -2-. v.

(11) Many App vendors provide free-trial to attract potential customers to try their products. Apps with this type of profit model are usually referred to as “Paid Apps” since the customer would need to make purchase if they want to use the App continuously after the trial. This type of business model has traditionally been used in PC software industry. Another business model is in-App purchasing which is a more innovative model nowadays. It allows users to acquire the App for free and then purchase bonus feature as they wish. These Apps are referred to as “Free Apps”, also called the “Freemium”. Purchases made in a freemium App. 政治大 Therefore, the model is potential in creating a more consistent revenue stream 立. are not a traditional one-time purchase. They are small but continuous purchases.. which makes it more appealing to venders and becomes the trend in App markets.. ‧ 國. 學. 1.2. Research Motivation. ‧. The APP market is thriving. There are plenty of business model for App.. sit. y. Nat. They are freemium, paid, paidmium, and in- App advertising. Table 2 shows a. io. er. brief introduction of these four models. Table 2: Business Model for Apps. n. al. Business Model. Ch. engchi. iv n U Top 3 Apps by iOS and. How does it work?. Google play Revenue in 2013* Puzzle & Dragons. Free. download. with. Freemium. Candy Crush Saga in-App purchases Clash of Clans Minecraft-Pocket Edition Paid download with no. Paid. Pages in-App purchases Whatsapp messenger** -3-.

(12) FIFA 13 Paid. download. with. Paidmium. Grindr Xtra in-App purchases Bloon TD 5 Contains ads (banner ads,. In-App Advertising. N/A video ads, etc). *Based on App Annie Intelligence estimates. **WhatsApp Messenger ranked as the #3 paid app based on its iOS revenue up until early August 2013, when it went from paid to freemium on iOS.. 政治大 As mentioned previously, freemium model is in vogue. It allows vendors to 立 (Source: App Annie and IDC, 2014). ‧ 國. 學. cast a wider net and monetize in the long-run through a potentially more consistent stream. Figure 2 shows App revenue is key countries such as Brazil,. ‧. Canada, France, Germany, India, Japan, Russia, South Korea, United Kingdom. sit. y. Nat. and United States. As the figure indicates, the revenue of paid and paidmium. io. al. er. model are shrinking, but the revenue of freemium model and in-App advertising. n. model is increasing. Clearly, free is working. The revenue of freemium model is especially notable since. v i n itsCgrowth between 2012 and 2013 is more than U h e nrate i h gc. 200%. It is obvious that freemium is king. While a fairly strong majority — 83 percent of the top thousand apps on each of the iOS app store and Google Play — monetize via freemium, an even bigger proportion of revenue generated by those top 2,000 apps arrives via freemium: 92 percent.. -4-.

(13) Figure 2: App Revenue in Key Countries. 政治大. (Source: App Annie & IDC, 2014). 立. 1.3. Research Objective. ‧ 國. 學. According to previous research, we can see that freemium is growing fast and generating notable revenue, therefore, is it worthy to learn why it is working.. ‧. The goal of this study is to realize the acceptance of freemium apps. In other. sit. y. Nat. words, the objective is to figure out what factors influence users’ decision toward. n. al. er. io. making purchases within.. i n U. v. Since each category of apps would include different variables, I would. Ch. engchi. focus on one App which is outstanding and considered innovative comparing to other Apps belonging in the same category. I choose “LINE” from the category of “communication” in Google Play and “Social Networking” in App Store. “LINE” is very popular in Taiwan. According to the statistics from App Annie, the App has been on the Top Chart for all time in both Google Play and App Store.. -5-.

(14) 政治大立(Source: Business Insider). Figure 3: LINE’s revenue by products. LINE shows the potential for chat apps as platforms, after generating $338. ‧ 國. 學. million USD in Revenue for 2013. Its income comes mostly from games and. ‧. stickers (see figure 3). In games, LINE sells virtual currency and power-up. sit. y. Nat. property and so on (see figure 4). Stickers are emotion icons that can be sent in. io. er. chat rooms. Some LINE stickers are free and others are not (see figure 5 and figure 6). The stickers are the reason why LINE can stick out among various. n. al. communication Apps.. Ch. engchi. i n U. v. Figure 4: An example of game purchase. -6-.

(15) Figure 5: An Example of. Figure 6: An example of. free sticker. paid sticker. In this study, I would take LINE as an example of freemium app and try to. 政治大. understand what factors influence users’ decision toward adopting Apps and. 立. making purchases in it.. ‧ 國. 學. 1.4. Research Process. Figure 7 illustrated the research process. The process started with. ‧. establishing research direction and scope. After collecting information and. Nat. sit. y. reviewing previous studies regarding the subject, I established the research. n. al. er. io. framework. According to the operation definitions, I designed questionnaire to. i n U. v. collect data and to testify my hypothesis. Finally, I came to conclusion and. Ch. suggestions for future research.. engchi. -7-.

(16) 立. 政治大. ‧ 國. 學 Figure 7: Research Process. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. -8-. i n U. v.

(17) Chapter II. Literature Review 5.1 Diffusion of Innovation Diffusion of innovation is a theory that seeks to explain how innovated ideas or technology spread in a social system. Therefore, in this study we use it to depict the adoption and diffusion of Apps. The diffusion of innovation is a process of socialization which an innovation is introduced through communication channels over time among the members in a social system (Rogers, 1983). This definition points out the four elements of. 政治大. diffusion (figure 8). Each of them is explain as following.. 立. Innovation. Diffusion. ‧. ‧ 國. 學. Social System. Channels. n. Time. Ch. engchi. er. io. sit. y. Nat. al. i n U. v. Figure 8: The Elements of Diffusion An innovation is an idea, practice, or object that is perceived as new by an individual or other unit of adoption. An innovation may be describes by five characteristics (figure 9) which are (1) relative advantage, (2) compatibility, (3) complexity, (4) trialability, and (5) observability. How an individual perceive an innovation affect the rate at which the innovation is diffused and adopted. The first two characteristics mentioned above are especially influential.. -9-.

(18) Relative Advantage. Observability. Compatibility. Innovation. T rialability. Complexity. 政治大 The essence of the diffusion process is the information exchange by which 立 Figure 9: Perceived Characteristics of Innovations. one individual communicates a new idea to one another. Communication. ‧ 國. 學. channels include mass media channels and interpersona l channels which involve. ‧. a face-to-face exchange between two or more individuals.. sit. y. Nat. Time is an important element in the diffusion process. The time dimension is. io. er. in the innovation-decision process (figure 10), by which an individual passes from first knowledge of an innovation through its adoption or rejection. The. al. n. v i n innovation-decision processCis the process through h e n g c h i U which an individual passes from first knowledge of an innovation to forming an attitude toward the. innovation, to a decision to adopt or reject, to implementation of the new idea, and to confirmation of this decision. A social system is defined as a set of units interrelated that are engaged to accomplished a common goal as a whole. The members or units of a social system may be individuals, informal groups, organizations, and/or subsystems. It is important to remember that diffusion occurs within a social system, because the social structure of the system affects the innovation's diffusion in several ways. - 10 -.

(19) Knowledge Persuasion. Decision Implementation Confirmation Figure 10: Innovation-Decision Process The concept of innovation-decision process or called the adoption process, mentioned previously is noteworthy therefore I will explain more about it. At the stage of knowledge, people grow awareness of the innovation. They find out the. 政治大. existence of the innovation and how it functions. At Persuasion stage, favorable. 立. or unfavorable attitude toward the innovation are formed. People begin to. ‧ 國. 學. generate personal opinion about the innovation. Next stage, decision, they would decide to adopt the innovation or not. Implementation stage occurs when an. ‧. individual put an innovation into use. At the last stage they would confirm the. y. Nat. n. al. er. io. sit. adoption which means they will use it continuously or abandon it.. Ch. engchi. i n U. v. Figure 11: Variables determining the rate of adoption of innovations (Source: Rogers, 1983) - 11 -.

(20) The rate of adoption (figure 11) is influence by variables that can be grouped into. five. categories:. perceived. attributes. of. innovations,. type. of. innovation-decision, communication channels, the nature of the social system, and the extent of change agents’ promotion efforts. The perceived attributes of innovations are especially important. Within the five attributes, relative advantage and compatibility matter most. Both of them have positive effects to the rate of adoption which means the higher the relative advantage and compatibility an innovation has, the higher the adoption rate. Relative advantage. 政治大 the one it is replacing. Compatibility is regarding to the compatibility of an 立 refers to how superior the innovation is (economically or socially), comparing to. innovation to the culture of the social system or the need of the people in the. ‧ 國. 學. system.. ‧. According to Rogers, different types of communication channel not only. y process. differently.. At. io. sit. innovation-decision. Knowledge. stage,. public. er. Nat. have different influence over types of adopter, but also affect each stage of the. communication channels like mass media are more important; however, at. al. n. v i n Ch Persuasion stage, local communication channels such as interpersonal channels engchi U. are more important than the others. Sometimes, the adoption of one’s kith and kin equals to the adoption of the person himself in some degree. In some cases, a person would have a positive attitude toward an innovation but not adopt it. This is known as Knowledge-Attitude-Practice (KAP) gap (Rogers, 2003) which indicates that potential adopters may consider adopting an innovation if they feel a need for it (Hassinger, 1959). A “need is a state of dissatisfaction or frustration that occurs when one’s desire outweighs one’s actualities” (Rogers, 2003). Although innovation diffusion model is a good predictor of social and technical change (Katz et al., 1963) and has been - 12 -.

(21) implemented in multiple fields, it has some limitations. For example, the technology adoption can be influenced by other factors besides the five perceived characteristics of innovations listed by Rogers (Moore & Benbasat, 1991). There are several previous empirical studies on diffusion of software or mobile technologies based on diffusion of innovation (See table 3). Table 3: Previous Study Basing on Diffusion of Innovation Literature. Type of App.. Zmud, 1982. Software. 立. Key Conclusion Compatibility and observability. of the innovation are important to 政治大 users.. 1990. communication were dominant in. software. channels. of. ‧. all phases of adoption decision making. PU,. external. motivation. and. io. er. Web-based. sit. y. Nat. Hsu, 2009. ‧ 國. Interpersonal. 學. Brancheau & Wetherbe, Spreadsheet. n. social recognition does not have a l Learning v i n Ch Communities i U influence on the e n g c hsignificant diffusion.. Bruce, 2013. Mobile. Three. of. technology. areas—tension. the. innovation–system. five for. DOI change,. fit,. and. support and advocacy—showed INGOs to be far along toward the adoption of mobile technology. - 13 -.

(22) 5.2 Extended Technology Acceptance Model (TAM) The Technology Acceptance Model (TAM) stands for a theory that models how users come to accept and use a technology. For years, TAM has been used by many researchers to investigate user acceptance of internet, e-commerce, mobile technology, and so on. The model suggests that when users are presented with a new technology, a number of factors influence their decision about how and when they will use it. TAM is one of the most influential extensions of Ajzen and Fishbein’s. 政治大 Davis in 1989. TAM replaces many of TRA’s attitude measures with two 立. Theory of Reasoned Action (TRA). TAM was originally developed by Fred. technology acceptance measures – “Perceived Ease of Use” and “Perceived. ‧ 國. 學. Usefulness” which are extrinsic motivators. In 2001, Moon and Kim extended. y. Nat. on the concept of Flow by Csikszentmihalyi.. ‧. TAM by including an intrinsic motivation factor, “Perceived Playfulness”, based. er. io. sit. Motivation theories have been used to understand an individual’s IT acceptance behavior. Extrinsic motivation refers to the perceived helpfulness. al. n. v i n toward the activity, such asCimproving job performance h e n g c h i U ; and intrinsic motivation refers to the performance of an activity for no apparent reason other than the process of performing it (Moon & Kim, 2001). Moon and Kim added this variable owing to the fact that technology is not used only for work nowadays. Take the example of World-Wide-Web usage, people use it not only for work, but also for shopping, playing games, communication, killing time, etc. So unlike traditional information technology, WWW is used both for work and pleasure, which is, in their words, perceived playfulness. In this research, I included this variable into the model since App is also a technology that could be used both for work and pleasure,. - 14 -.

(23) The extended model is shown in figure 12. The definitions of the two important features by Davis (1989) are as following: . Perceived Usefulness (PU): the degree to which a person believes that using a particular system would enhance his or her job performance.. . Perceived Ease-of-Use (PEU): the degree to which a person believes that using a particular system would be free from effort. Moon and Kim (2001) defined Perceived Playfulness as “The extent to. which the individual perceives that his or her attention is focused on the. 政治大 the interaction intrinsically enjoyable or interesting.” In their study, they found 立. interaction with the World Wide Web; is curious during the interaction; and finds. that Perceived Playfulness had a significant positive relationship with Attitude. ‧ 國. 學. toward Using.. ‧. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 12: Extended Technology Acceptance Model. 5.3 Elaborating Likelihood Model (ELM) The Elaboration Likelihood Model (ELM) is an example of a “dual process” approach to persuasion (Figure 13). It posits that attitude change may occur through one of two different processing routes: the central route or the peripheral route. The person’s motivation and ability to elaborate on a message determine the route taken, that is, the amount and nature of the thinking that a person does - 15 -.

(24) about a persuasive message is a vital determinant of the kind of persuasion that occurs (Petty & Cacioppo, 1981, 1986; Petty & Wegener 1999). The ELM is based on an assumption that people are bombarded with too many persuasive communications so that it is not possible for them to fully elaborate and carefully evaluate each of them. Therefore, they are likely to put more elaborative effort on some matters and less on the others. Considerable research has demonstrated that motivational and ability factors play a role in the model influencing the persuasion process. Motivational factors include personal. 政治大 Petty, 1981; Tetlock, 1983), and anticipated interaction regarding the 立. involvement (Petty, Cacioppo, & Goldman, 1981), accountability (Harkins & issue. (Chaiken, 1980). Ability factors include distraction (Petty, Wells &Brock, 1976),. ‧ 國. 學. message repetition (Cacioppo & Petty, 1989), as well as time pressure. ‧. (Kruglanski, Freunds & Obertynski, 1996).. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 13: The two routes to persuasion (Source: Petty and Cacioppo, 1986) - 16 -.

(25) The central route is taken when high elaboration is taken place. It involves paying careful attention to the relevant information in the message and generating new implications of the information. When persuasion follows the central route, the extent of attitude change depends on the valence of the thoughts, the amount of thoughts, and the confidence that people have in their thoughts. Attitudes formed this way are expected to be (1) relatively easy to call mind (accessible) (2) relatively persistent over time (3) relatively resist to challenge from competing message (4) relatively predictive of the person’s attitude-relevant judgments and. 政治大 Peripheral route is taken when people rely on simple cues and shortcuts, 立. behaviors (Petty, Haugtvedt, & Smith, 1995).. rather than engage in thoughtful consideration. This is referred to as following the. ‧ 國. 學. “heuristics” principles. There are different types of heuristics. One is the. ‧. credibility heuristics, in which message receiver is guided by the apparent. sit. y. Nat. expertise of the communicator. A second heuristics is the liking to the. io. er. communicator by the receiver. A third one is the consensus heuristics, in which the receiver is influenced by other’s reaction toward the message. The influences. al. n. v i n would C decrease, as elaboration h e n g c h i U increases. of such heuristics. (O’keefe, 2008).. According to ELM, attitudes formed or changed taking peripheral route is less accessible, less persistent, less resistant and less predictive of behavior than those taking central route (Petty et. al., 2005). However, central route and peripheral route represent the extreme level of elaboration. At intermediate levels of elaboration, the combinations of these two routes are possible. For example, a person’s attention might be drawn to a colorful advertisement, but then start thinking about the slogan and generate thought regarding the product description. In the example, peripheral cues take place but do not have much impact. - 17 -.

(26) 5.4 Involvement There are two aspects of involvement, one is “personal involvement” and another is “product involvement”. Involvement has received attention from researchers because of its importance in influencing customers’ cognitive and behavioral responses which makes it especially importance for marketing. Personal involvement and product involvement are two side of one coin. Personal involvement refers to the feelings of interest, enthusiasm, and excitement one would have about specific product categories (Bloch, 1986). Product involvement. 政治大 or drive that evoked by a product class (Bloch, 1981; Mittal & Lee, 1989; 立 is defined as an internal state variable that indicate the amount of arousal, interest. Dholakia, 2000).. ‧ 國. 學. People who are involved in certain products feel that these product. ‧. categories are relevant to their lives. Hobbies, collections or specialties such as. sit. y. Nat. car lovers are example of this type of customer behavior (Zaichkowsky, 1994).. io. er. To explain customer behaviors, Park and Young (1986) break down personal involvement into two view (1) cognitive involvement and (2) affective. al. n. v i n involvement. The first one C is more rational and based h e n g c h i U on the product’s functional performance. The second one is based on emotional need to express a self- image to the outside world. Measuring only the thinking without feelings when referring to involvement may lead to omission in capturing the relevance of the object to the individual (Vaughn, 1980; 1986). When a person is getting involved into purchasing event, the aspects of product also matter. There are two type of product involvement (1) Situational Involvement (2) Enduring involvement. Situational involvement reflects product involvement that occurs only in specific situations, such as a purchase. Enduring involvement represent an ongoing concern with a product that transcends - 18 -.

(27) situational influences (Houston & Rothschild, 1978; Laurent & Kapferer, 1985; Rothschild, 1979). For example, in the case of refrigerators, people don’t usually show interests in it, unless they plan to purchase one and this is when situational involvement is aroused. An example for enduring involvement is car lovers. In a high enduring involvement case, a product is interesting and occupies the consumer's thoughts without the stimulus of an immediate purchase.. 5.5 Value Creation Cycle (VCC) The theory of Value Creation Cycle (VCC) developed by Yu (2012) conveys. 政治大 since they are far from the customer so that they hardly can get any feedback to 立 that the traditional supply chain make it hard for independent designer to succeed. adjust their products to the market. Fortunately, internet has provided designer a. ‧ 國. 學. simple and far-reach way to get in touch with their customers. It offers designers. ‧. opportunities to establish a VCC that maximize customers’ feedback. Through. sit. y. Nat. ecommerce, designer can easily build up brand and online store to sell products.. io. er. Furthermore, they can get valuable feedback from customers. Feedback is very important for designers’ innovative product design. This also means that. n. al. Ch. marketing intelligence is vital to the designer’s. engchi. i n success. U. v. (1). (2) Figure 14: Comparison of two supply chain (Yu, 2012) - 19 -.

(28) Figure 14 depict the comparison between traditional supply chain and the one with internet. Figure 14(1) depicts that the independent designer upload their innovative products with designer’s branding on the internet, and then after promotions, the customer can participate and make responses. Later on, marketing intelligence (MKI) can be obtained to show the market performance of the products. On the other hand, Figure 14(2) depicts the traditional supply chain. However, owing to time limited for the study we did not have the chance to review the relationship between vendors and users. Therefore in the end, the. 政治大. research model of this study did not put emphasis on this literature.. 立. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. - 20 -. i n U. v.

(29) Chapter III. Research Methodology 3.1 Research Framework The purpose of this study is to achieve a better understanding of In-App purchasing behavior and to explore why people like to purchase stickers or game properties in Apps such as LINE. The research framework of this study is based on Innovation Diffusion Theory, and also Elaboration Likelihood Model (ELM) and extended Technology Acceptance Model (TAM). During the construction of the conceptual model of the study, I conducted a few interviews aiming to shed. 政治大 model is showed in figure 15. 立. some light on how people make in-App purchasing decision. The conceptual. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. Figure 15: Research Framework The conceptual model includes five constructs. According to the Innovation-Decision Process (figure 10) adopted from Innovation Diffusion Theory, the first stage of decision process is “knowledge” which is acquired from channels. The concept of information being acquired can be interpreted as “Network effects”. The second and third stage of decision is “Pe rsuasion” and - 21 -.

(30) “decision” which is the stage when user form attitude toward in- App purchasing. The next stage of decision process is “implementation” which is presented by “intention to purchase”.. 3.2 Hypothesis 3.2.1 Social Network Effects and Perceived Playfulness Social network effects are influences from a social system in which a person belongs. Social network effects are delivered through channels which are means that an individual acknowledge information of an idea or a product,. 政治大 According to Innovation Diffusion Theory by Rogers 立. in my case, information regarding LINE stickers or game properties. (1983), channels. include mass media and interpersonal reference.. ‧ 國. 學. The concept of perceived playfulness is adopted from extended. ‧. Technology Acceptance Model (Moon & Kim, 2001). It conveys that the. sit. y. Nat. subject discussed is fun and interesting to indulge in. In the case of the study,. io. er. LINE is a communication platform which also combines games and emotion stickers. These two features are meant to make communication more fun. I. n. al. thus propose:. Ch. engchi. i n U. v. H1: Social network effects are positively associated with perceived playfulness of the App 3.2.2 Social Network Effects and Personal Involvement After receiving information from channels, a person would form an impression of the product, LINE sticker or game property, and personal involvement is aroused. I thus propose: H2: Social network effects are positively associated with Personal Involvement.. - 22 -.

(31) 3.2.3 Perceived Playfulness and Attitude toward in-App Purchase Previous study shows that positive attitudes, such as pleasure and satisfaction, are highly correlated with the user’s perceived playfulness. Furthermore, the positive attitude is resulted from the user’s playfulness experience (Webster & Heian, 1990). Therefore, if a user is enjoying chatting through LINE or playing LINE game, it is more likely that the user form positive attitude toward making purchase within LINE. I thus propose: H3: Perceived Playfulness is positive associated with Attitude toward. 政治大 3.2.4 Social Network Effects and Attitude toward in-App Purchase 立 in-App Purchase. During the construction of the conceptual model of the study, I. ‧ 國. 學. conducted a few interview aiming to shed some light on how people make. ‧. in-App purchasing decision. In the interviews, I found that when making. sit. y. Nat. in-App purchasing decisions, people tend to be affected by network channels. io. er. such as Facebook fan pages, App stores or peers. I thus propose: H4: Social network effects are positively associated with attitude. n. al. Ch. toward in-Apps purchasing.. engchi. i n U. v. 3.2.5 Personal Involvement and Attitude toward in-App Purchases Although LINE user with positive attitude toward in-App purchasing shall be more likely to purchase LINE stickers and game properties, the occurrence of KAP gap is possible. Therefore, to enforce persuasion, motivational factors are required. According to ELM, personal involvement is an important variable which also include some illustrated attributes. I thus propose: H5: Personal involvement is positively associated with the attitude toward download and use. - 23 -.

(32) 3.2.6 Attitude toward in-Apps purchasing and Purchase Intention According to acceptance models mentioned in this study, extended TAM and Innovation Diffusion, attitude plays a role in influencing decision making. User with positive attitude toward in-App purchasing shall be likely to perform in-App purchasing. Therefore, I propose: H6: Attitude toward in-App purchasing is positively associated with intention to purchase.. 3.3 Variables and Operational Definitions. 政治大. There are five constructs in the conceptual model. They are Network. 立. Effects, Perceived Playfulness, Personal Involvement, Attitude toward in-App. ‧ 國. 學. Purchase and Intention to perform in-App purchase. I will explain each of them in the five sections below.. ‧. 3.3.1 Social Network Effects. Nat. sit. y. Social network effects are influences from a social system in which a. n. al. er. io. person belongs. Social network effects are delivered through channels which. i n U. v. are means that an individual acknowledge information regarding LINE. Ch. engchi. stickers or game properties. According to Innovation Diffusion Theory by Rogers (1983), channels include mass media and interpersonal reference. Mass media includes official websites, blogs, advertisement, ranking in App stores, and comments in App stores. Interpersonal reference refers to suggestion from friends and family. The operational definition and questions designed for the construct are showed in table 4.. - 24 -.

(33) Table 4: Construct Definition of Social Network Effects Construct. Social Network Effect. Operational. The extent to which users tend to be effected by. Definition. social network effects they are exposed.. Questions*. (1) My colleges and friends use LINE to chat or play LINE game frequently. (2) Elders I know use LINE to chat or play LINE game frequently.. 政治大 tend to take friends’ suggestion into. (3) When performing in-App purchasing, I. 立. ‧ 國. 學. consideration.. (4) When performing in-App purchasing, I. ‧. tend to refer to comments on the internet.. tend to refer to the item’s ranking in the. n. al. er. io. sit. y. Nat. (5) When performing in-App purchasing, I. store.. Ch. (6) When performing. engchi. iv n in-App U purchasing,. I. tend to refer to the item’s description on official website. (7) When performing in-App purchasing, the advertisement of the item makes it more tempting. *Use 7-point Likert Scale with “1” standing for “Strongly disagree”, “4” for “Neither agree nor disagree”, and “7” for “Strongly agree”. - 25 -.

(34) 3.3.2 Perceived Playfulness The concept of perceived playfulness is adopted from extended Technology Acceptance Model (Moon & Kim, 2001). It conveys that the subject discussed is fun and interesting to indulge in. In the case of the study, LINE is a communication platform which also combines games and emotion stickers. See table 5 for the operational definition and questions designed for the construct. Table 5: Construct Definition of Perceived playfulness Construct. 政治大立The extent to which users perceive Perceived playfulness. Operational. enjoying experience toward LINE stickers or LINE game properties**.. (1) When using LINE, I am not aware of time. ‧. Questions*. 學. ‧ 國. Definition. fun and. (2) When using LINE, I am not aware of. n. al. distraction around me.. er. io. sit. y. Nat. as it elapses.. v i n (3)CWhen LINE, IUoften forget other h e nusing gchi commitment.. (4) When using LINE, I often forget what I was doing (before using LINE). *** (5) I think using LINE is enjoyable. *** (6) I think using LINE is fun. *** Reference. Moon and Kim (2004). *Use 7-point Likert Scale with “1” standing for “Strongly disagree”, “4” for “Neither agree nor disagree”, and “7” for “Strongly agree” - 26 -.

(35) **In the full questionnaire, the participants were asked to choose a scenario when answering questions. *** These questions are developed by this research. 3.3.3 Personal Involvement Personal involvement refers to a person’s perceived relevance of the object based on inherent needs, values and interests (Zaichkowsky, 1985, p.342). People who are involved in certain products feel that these product categories are relevant to their lives. Hobbies, collections or specialties such. 政治大. as car lovers are example of this type of customer behavior (Zaichkowsky, 1994).. 立. Zaichkowsky (1985) designed a Personal Involvement Inventory Scale. ‧ 國. 學. with 20 questions to measure a person’s involvement with products. The. ‧. original PII scale is showed in table 6. Based on PII, I developed questions. sit. y. Nat. for this study. I deleted some questions from PII since they are not suitable. io. er. for in-App purchasing items. The questions included in the construct are showed in table 7.. n. al. Ch. engchi. - 27 -. i n U. v.

(36) Table 6: the original PII Scale. 立. 政治大. ‧ 國. 學. Table 7: Construct Description of Personal Involvement Personal involvement. The extent to which a user’s involvement with the. sit. y. Nat. Operational. product, LINE sticker or LINE game property*.. io. er. Definition. to my life. a l (1) I think Product A is relevant v i n C hI think Product A U (2) is important. engchi. n. Questions**. ‧. Construct. (3) I think Product A is valuable. (4) I think Product A is beneficial. (5) I think Product A is useful. (6) I think Product A is appealing. (7) I think purchasing Product A is essential. (8) I think purchasing Product A is exciting.. Reference. Zaichkowsky (1985) - 28 -.

(37) *Use 7-point Likert Scale with “1” standing for “Strongly disagree”, “4” for “Neither agree nor disagree”, and “7” for “Strongly agree” **In the full questionnaire, the participants were asked to choose a scenario when answering questions. *** These questions are developed by this research. 3.3.4 Attitude toward in-App Purchase An Attitude is a favor or disfavor expression toward a certain object or behavior. This construct is meant to measure user’s attitude toward. 政治大 construct is showed in table 8. 立. purchasing within LINE. The operational definition and questions for the. Table 8: Construct Description of Attitude toward in-App Purchase. ‧ 國. 學. Construct. Attitude toward in-App Purchase. The extent of attitude of a user toward purchasing. ‧. Operational. the product*, sticker or game property, in LINE.. sit. y. Nat. Definition. (1) I think purchasing product A in LINE is a. io. n. al. good idea.. er. Questions**. v i n C hI like to purchase product (2) e n g c h i U A in LINE.. (3) I have enough money to purchase Product A. Reference. Wu & Liu (2007) Huang & Hsieh (2011). *Use 7-point Likert Scale with “1” standing for “Strongly disagree”, “4” for “Neither agree nor disagree”, and “7” for “Strongly agree” **In the full questionnaires, the participants were asked to choose a scenario when answering questions. Therefore, “product A” could be LINE stickers or game properties depending on which scenario is chosen. - 29 -.

(38) 3.3.5 Intention to perform in-App purchase Intention is the action of intending which is performing in-App purchasing. In this construct I measured a user’s ability, willingness and in-App purchases experiences. The operation definition and questions of the construct are showed in table 9. Table 9: Construct Description of Intention to perform in-App purchases Construct. Intention to perform in-App purchase. Operational. The extent of Intention of a user toward. Definition. purchasing the product*, sticker or game property,. 立. (1) I am aware of the characteristic of product A.. ‧ 國. 學. Questions**. 政治大 in LINE.. (2) I am capable of performing in-App purchase. (3) I download LINE stickers or game frequently. io. er. (including freebies).. sit. y. Nat. to buy product A a l (4) I perform in-App purchases v i n C hfrequently. U engchi. n Reference. ‧. in order to buy product A.. Wu & Liu (2007) Huang & Hsieh (2011). *Use 7-point Likert Scale with “1” standing for “Strongly disagree”, “4” for “Neither agree nor disagree”, and “7” for “Strongly agree” **In the full questionnaires, the participants were asked to choose a scenario when answering questions. Therefore, “product A” could be LINE stickers or game properties depending on which scenario is chosen.. - 30 -.

(39) 3.4 Research Design 3.4.1 Questionnaire Design The questionnaire of the study is based on previous research and theory. Questions adapted from previous study were revised according to the research objective and operational definition. Questions developed in the research were based on relevant theory and definition. The questions adjusted and developed were discussed with my advisor and two doctoral seniors. Few of my classmates have also given me suggestion regarding the. 政治大. description of the questions after participated pretest. 3.4.2 Target Subjects. 立. The objective of the study is to understand purchase behavior in APPs. ‧ 國. 學. and taking LINE as an example. Therefore, the target subjects of the study. ‧. are LINE users. LINE game players are also included in the research. The. y. Nat. first question in the questionnaire is design with the purpose to filter. er. io. sit. participants. The first question is “Are you a LINE user?” 3.4.3 Data Collection. n. al. To affirm the. v i n C h and the validity reliability of the engchi U. questionnaire, the. collection of data included two stages. The first stage was pretest and the other was the formal sample collection. The questionnaires were hand out in two forms, paper copies and online version. The online questionnaire was powered by google drive. I posted the link of the questionnaire on groups and timelines of Facebook and LINE. The paper copies were handed out around campus.. - 31 -.

(40) 3.5 Data Analysis To validate the research framework, I conducted a survey to test the hypotheses. I analyzed the data collected using SmartPLS to apply variance-based structural equation model (PLS-SEM). 3.5.1 PLS-SEM Hair et al (2011) had suggest this method to be known as silver bullet since there are a lot of advantages comparing to covariance-based structural equation model (CB-SEM). For example, PLS-SEM works. 政治大 Afthanorhan (2013) also have proven that the confirmatory factor analysis 立. better at maximizing the explain variance of latent constructs.. (CFA) conducted by PLS-SEM is more reliable and valid. Based on the. ‧ 國. 學. result section, the value of factor loadings/outer loadings, and average. sit. y. Nat. using the same data provided.. ‧. variance extracted (AVE) in PLS-SEM is better than CB-SEM even when. io. er. The reason why I chose PLS-SEM for analyzing data was also because it makes analyzing abstract concept possible. Taking advantage of. n. al. PLS-SEM, I got. v i n toCanalyze vague andUcomplicated hengchi. concept such as. involvement. In PLS-SEM, abstract concepts are constructs which contain several variables to explain the idea. Figure 16 illustrates the procedure for executing PLS-SEM.. - 32 -.

(41) Defining Research Question and Research Design Literature Review Constructing Model and Choosing Measure Tool. Data Collection Model Verification Interpretation. 政治大 3.5.2 Model Verification 立. Figure 16: PLS-SEM Procedure. The model verification includes two parts: confirmatory factor. ‧ 國. 學. analysis (CFA) and hypothesis testing. CFA includes the assessments of. ‧. goodness-of- fit, composite reliability, and construct validity of the model. sit. y. Nat. proposed in the study. Constructive validity includes convergent validity. io. er. and discriminant validity.. To achieve convergence validity, the factor loading of each variables. al. n. v i n C h0.5 (Hair et al, 2010), should be larger than average variance extracted engchi U. (AVE) should be larger 0.5, composite reliability and Cronbach’s alpha should be at least larger than 0.6 (Hair et al, 2010). To achieve discriminant validity, the root if average variance extracted (AVE) should be larger than the correlations of latent variables. The goodness-of-fit could be calculated with the average of AVE and the average of R square.. - 33 -.

(42) 3.6 Pretest Cooper and Schindler (2008) suggested that the number of pretest sample shall be around 25 to 100 unit. I collected total 86 samples for pretest. Three samples were considered invalid due to that the participants were not LINE user or the answers given by the participants were making no sense. Therefore, there were 83 valid samples. 3.6.1 Demographic analysis for Pretest Among these 83 samples, there are 42 female partic ipants and 41 male. 政治大 participants of the pretest are 21 to 30 years old. The education level of 立. participants, so the gender of the sample is quite equal. Most of the. most of the participants of pretest is college and most of them are student.. ‧ 國. 學. The monthly income of most of the participant of pretest is below. ‧. 10,000NTD. The overall demographic data of the participants o f pretest are. sit. y. Nat. showed in table 10.. io. al. Numbers of. iv n Sample U. Percentage (%). Female. 42. 51%. Male. 41. 49%. < 20 Years Old. 24. 29%. 21-30 Years Old. 33. 40%. 31-40 Years Old. 17. 20%. 41-50 Years Old. 6. 7%. 51-60 Years Old. 3. 4%. Postgraduate. 24. 29%. n. Category. er. Table 10: Demographic analysis of pretest sample. Item. Ch. engchi. Gender. Age. Education. - 34 -.

(43) Level. Undergraduate. 31. 31%. Senior High School. 18. 22%. Junior High School. 10. 12%. 3. 4%. Governmental. 2. 2%. Education/ Research. 5. 5%. 3. 4%. Financial And Insurance. Public Relations. 政治大立Information 9 And Marketing. Occupation. ‧ 國. 11%. 學. Technology. Medical. 5. Clergy. 1. y. 1%. Art. 2. 2%. n. er. io. al. ‧. 3. Nat. Manufacturer. sit. Category. v ni. Service. 5. C h e n g c h i U 43 Student. 4% 6%. 6% 52%. Unemployed. 2. 2%. Below 10,000NTD. 38. 46%. 10,001-20,000NTD. 7. 8%. Monthly. 20,001-30,000NTD. 8. 10%. Income. 30,001-40,000NTD. 9. 11%. 40,001-50,000NTD. 11. 13%. Over 50,001NTD. 10. 12%. - 35 -.

(44) 3.6.2 Measurement Assessment for Pretest The overview of quality criteria of the pretest and the factor loading of each variable is shown in table 11. As shown in Table 12, all composite reliability and Cranbach’s Alpha are larger than 0.7. Also, most of the AVE is larger than 0.5, except for the AVE of the construct, Network Effects, is only 0.41 which indicate that the construct is not well explained by the variables. Looking at the factor loading, we can see the first and second question. 政治大 of factor loading should be larger than 0.5, NE1, NE2, and I4 are only 0.45, 立. of the construct might be the one causing problem. Furthermore, the values. 0.3 and 0.43. Therefore, I decide to delete these three questions from the. ‧ 國. 學. questionnaire. The goodness-of-fit of the model is 0.51 which is acceptable.. ‧. Therefore, reconstructing the model is not necessary. I will only adjust the. sit. Table 11: Pretest Overview. n. aItem l C h. er. io Constructs. y. Nat. variables.. Factor AVE. Loading engchi U. NE1. 0.45. NE2. 0.30. NE3. 0.71. NE4. 0.67. NE5. 0.72. NE6. 0.66. NE7. 0.73. Composite. Cronbach’s. Reliability. Alpha. 0.82. 0.75. v ni. Network Effects. - 36 -. 0.411.

(45) Perceived. PP1. 0.73. PP2. 0.82. PP3. 0.77 0.60. PP4. 0.77. PP5. 0.78. PP6. 0.78. PI1. 0.84. PI2. 0.85. ‧ 國. io. PI6. 0.87. PI7. 0.74. PI8. 0.86. A1. 0.93. al. n. toward. 0.88. Ch. A2. in-App Purchase. 0.95. 0.87. ‧. Nat. Attitude. 0.73. PI5. 學. Involvement. y. 立PI4. Personal. 政0.92 治大 0.85. sit. PI3. 0.87. er. Playfulness. 0.90. 0.95. engchi. A3. 0.65. I1. 0.86. I2. 0.81. I3. 0.89. I4. 0.43. I5. 0.62. i n U. 0.73. v. 0.89. 0.80. 0.90. 0.87. Intention toward in-App Purchase. - 37 -. 0.59.

(46) Chapter IV. Result The formal questionnaire was online during June 8 th to June 13th . Besides online questionnaire, paper copies were also handed out around campus during the time. Total 312 samples were collected, but 11 of them were considered invalid and discarded due to that the participants answering the questionnaire are not LINE users or the answers provided are meaningless. Therefore, there were 301 units of valid samples. In the formal questionnaire, questions were grouped into five sections. In each. 政治大 scenario. The five sections contained general questions about user, user’s experience, 立 section, there was a short description to explain and lead the participant into a. ‧ 國. 學. experience regarding in- App purchases, user’s perspective toward in-App purchases and user’s ability regarding purchase in LINE. The matching of the constructs and the. ‧. sections are showed in table 12.. io. al. User’s Experience Purchase Experience. User’s perspectives. No. of Questions. v ni. n. General Questions. Construct. er. Section. sit. y. Nat. Table 12: Questionnaire Sections and Constructs. C hPerceived Playfulness engchi U. 6 7. Social Network Effects. 5. Personal Involvement. 8. Attitude toward In-App 2 Purchasing Attitude toward In-App 1 Purchasing. User’s Ability Intention toward In-App 4 Purchasing - 38 -.

(47) Personal Information. 6. 4.1 Demographic analysis Among the 301 samples, 146 are female participants and 155 are male participants. Most of the participants are around 21 to 30 years old. Most of them are students and their education levels are at college or postgraduate degree. Most of them earn no more then 10,000NTD a month. 62.5% of them are Android devices users and 36.9% of them are iOS devices users. Table 13 shows the demographic analysis of the samples.. 政治大 Numbers of. Table 13: Demographic Analysis. 立Item. Category. Percentage (%). 146. Male. 155. 51.5%. < 20 Years Old. 24. y. 5.3%. 21-30 Years Old. 33. sit. ‧ 國. Female. ‧. 學. Sample. 77.7%. Nat. io. al. n. Age. i n U. 31-40 Years Old. Ch. er. Gender. 17. engchi 6. 41-50 Years Old. v. 48.5%. 13.0% 2.7%. 51-60 Years Old. 3. 1.3%. Postgraduate. 119. 39.5%. Education. Undergraduate. 161. 53.5%. Level. Senior High School. 8. 2.7%. Junior High School. 13. 4.3%. Student. 138. 45.8%. 32. 10.6%. Occupation Information Category Technology - 39 -.

(48) Service. 21. 7.0%. Manufacturer. 14. 4.7%. Financial/ Insurance. 13. 4.3%. Education/ Research. 10. 3.3%. Military/Police. 10. 3.3%. Architect. 9. 3.0%. Medical. 7. 2.3%. Governmental. 7. 2.3%. 政治 7大 Business 6 立. Public Relations/ 6 Marketing. 2.0%. ‧. Distribution/ Retail. 2.3%. 學. ‧ 國. Unemployed. 5. Nat. sit. 4. n. er. io. Travel Agent. a lArt Ch. 1.7%. y. Transportation/. 2.0%. i n U. 4. engchi 3. Legal Professions House Keeping. v. 1.3%. 1.3% 1.0%. 2. 0.7%. 2. 0.7%. Real Estate. 1. 0.3%. Below 10,000NTD. 111. 36.9%. Monthly. 10,001-20,000NTD. 38. 12.6%. Income. 20,001-30,000NTD. 39. 13.0%. 30,001-40,000NTD. 53. 17.6%. Entertainment/ Publishing. - 40 -.

(49) 40,001-50,000NTD. 27. 9.0%. Over 50,001NTD. 33. 11.0%. From the demographic analysis, I found some differences between genders. According to the analysis, female participants are more indulged in LINE since they tend to forget time and what they should be doing. Male participants tend to be effected by social network more than females. However, in the construct of network effects, females tend to be influenced by commercials or advertisements more than males. Regarding the intention to purchase in LINE,. 政治大 understanding on how to purchase in LINE. Table 14 shows the comparison 立 males have higher intention in purchasing than females and have a better. ‧ 國. 學. between genders.. 3.89. 3.92. al. n. Ch. 3.85. 4.05. 4.23. 3.56. 3.76. 3.99. 0.33. 0.09. engchi U. y. 3.60. -0.32. I. 3.97. er. PI. io. Difference. NE. ‧. Female. PP. Nat. Male. A. sit. Table 14: Comparison between Genders on Average of each Construct. v ni. 0.08. 0.24. *PP stands for Perceived Playfulness; NE for Network Effects; PI for Personal Involvement; A for Attitude toward in-App Purchasing; I for Intention toward in-App Purchasing. In addition to the difference between genders, I also compare iOS and Android users. Although these two types of user show no significant difference in terms of user’s experience, they are difference in terms of the attitude toward in-App purchases and the intention toward in-App purchases. IOS users show higher attitude and intention toward purchasing in LINE which means that iOS users think that purchasing in LINE is a good idea and they tend to purchase - 41 -.

(50) more than Android users do. Table 15 shows the comparison of iOS users and Android users. Table 15: Comparison between users on Average of each Construct PP. NE. PI. A. I. iOS. 3.70. 3.73. 3.88. 4.13. 4.27. Android. 3.80. 3.74. 3.78. 3.95. 4.03. Difference. -0.09. -0.01. 0.09. 0.18. 0.24. *PP stands for Perceived Playfulness; NE for Network Effects; PI for Personal. 政治大. Involvement; A for Attitude toward in-App Purchasing; I for Intention toward in-App Purchasing.. 立. ‧ 國. 學. 4.2 Confirmatory Factor Analysis. Before testing the hypothesis, I will evaluate the reliability and validity of. ‧. the model first. See table 16 for the overview of CFA.. y AVE. Reliability. n. al. Cronbach’sα. er. Composite. Item. io. Construct. sit. Nat. Table 16: Overview of CFA. Ch. e n0.84 gchi. iv n U 0.87. 0.54. PP. 6. NE. 5. 0.80. 0.86. 0.54. PI. 8. 0.93. 0.84. 0.67. A. 3. 0.95. 0.97. 0.91. I. 4. 0.75. 0.84. 0.57. *PP stands for Perceived Playfulness; NE for Social Network Effects; PI for Personal Involvement; A for Attitude toward in-App Purchasing; I for Intention toward in-App Purchasing.. - 42 -.

(51) 4.2.1 Reliability Test In general, if the value of Cronbach’s α is larger than 0.7, then value of composite reliability is larger than 0.6 and the value of AVE is larger than 0.5, we consider that the measurements are consistent and stable which means the questions are with well reliability (Bagozzi & Yi, 1988; Fornell & Larcker, 1981). As shown in table 16, all Cronbach’s α is larger than 0.7, all composite reliability is larger than 0.6 and the value of AVE is larger than 0.5, therefore, the constructs are reliable.. 政治大 Validity test include convergent validity and discriminant validity. 立. 4.2.2 Validity Test. Convergent validity refers to that the items in the same construct are related. ‧ 國. 學. to each other. Discriminant validity refers to that the items from different. ‧. constructs are in fact different than each other.. sit. y. Nat. To examine the convergent validity, we look at Cronbach’s α and. io. er. composite reliability. These two values have to be over 0.7. As shown in table 16, the standard is achieved. As for discriminant validity we have to. al. n. v i n C h with cross- loading, compare the factor loading and the root of AVE with engchi U the correlation between each construct. The factor loading of the construct. should be larger than cross- loadings. The root if AVE should be larger than the correlation between each constructs. See table 17 for the matrix of loadings and table 18 for the comparison of root of AVE and correlations. As shown in table 17, all factor loadings are larger than cross- loadings. In table 18, all roots of AVE are larger than the correlation with other constructs. Thus, the validity assessment is achieved.. - 43 -.

(52) A. I. NE. PI. PP. A1. 0.98. 0.65. 0.26. 0.61. 0.43. A2. 0.90. 0.66. 0.26. 0.58. 0.42. A3. 0.98. 0.65. 0.26. 0.70. 0.43. I1. 0.47. 0.77. 0.18. 0.51. 0.29. I2. 0.40. 0.76. 0.12. 0.36. 0.23. I3. 0.71. 0.86. 0.18. 0.67. 0.35. I4. 0.39. 0.41. 0.23. NE1. 0.23. 0.29. 0.30. NE2. 0.07. 0.01. 0.72. 0.15. 0.17. 0.16. 0.10. 0.79. 0.23. 0.26. 0.26. 0.22. 0.71. 0.29. ‧. 0.31. 0.20. 0.16. 0.75. 0.32. y. 0.41. 0.62. 0.45. 0.29. sit. Table 17: Matrix of Loadings. 0.43. 0.44. 0.31. n. al. PI2. 0.59. PI3. 0.73. PI4. 0.67. 0.55. PI5. 0.71. PI6. 0.77. er. io. PI1. Nat. NE5. ‧ 國. NE4. 學. NE3. 立. 0.70 治 0.09 政大 0.14 0.72. v ni. 0.79. 0.45. 0.88. 0.41. 0.32. 0.85. 0.38. 0.66. 0.25. 0.87. 0.36. 0.72. 0.62. 0.25. 0.84. 0.36. PI7. 0.63. 0.48. 0.39. 0.74. 0.34. PI8. 0.77. 0.58. 0.33. 0.78. 0.41. PP1. 0.17. 0.17. 0.28. 0.20. 0.71. PP2. 0.18. 0.17. 0.26. 0.25. 0.70. PP3. 0.12. 0.06. 0.30. 0.17. 0.66. C h 0.59 i U e n g c h 0.25. - 44 -.

(53) PP4. 0.20. 0.12. 0.33. 0.24. 0.69. PP5. 0.50. 0.43. 0.36. 0.51. 0.82. PP6. 0.53. 0.46. 0.31. 0.52. 0.80. *PP stands for Perceived Playfulness; NE for Social Network Effects; PI for Personal Involvement; A for Attitude toward in-App Purchasing; I for Intention toward in-App Purchasing. * The colored areas indicate the factor loadings of the construct. Table 18: Comparison between Root of AVE and Correlations A. 立. 政I 治NE 大. PI. I. 0.68. [0.75]. 0.27. 0.20. [0.74]. 0.84. 0.67. 0.37. [0.82]. 0.45. 0.37. 0.42. 0.48. y. Nat. PP. ‧. PI. [0.73]. sit. NE. ‧ 國. [0.95]. 學. A. PP. al. er. io. *PP stands for Perceived Playfulness; NE for Social Network Effects; PI for. v. n. Personal Involvement; A for Attitude toward in-App Purchasing; I for. Ch. engchi. Intention toward in-App Purchasing.. i n U. *The values in the brackets indicate root of AVE of the construct.. 4.3 Structural Model Assessment Structural model examined the relationships between different constructs, hence, from the analysis I tested if the significance of the hypotheses. The analysis was done by SmartPLS applying regression and bootstrapping. Figure 17 shows the path analysis output of the model.. - 45 -.

(54) Figure 17: Path Analysis. 政治大 The values of the coefficients 立 are between 1 and -1. If the coefficient is positive, Path analysis shows the relationships and intensity between each constructs.. ‧ 國. 學. it means that two construct are related positively. If the coefficient is negative, it indicates that two constructs have an inverse relationship. The t-value and the. Nat. sit. Table 19: t-Value and Path Coefficient. n. al. t-Value. Coefficient NE → PP. 0.420. NE → PI. 0.467. PP → A. Ch. e n9.15 gchi. er. io. Path. Construct. y. ‧. path coefficient are shown in table 19.. p-Value. i n U. v. Vitrified. <0.001. Significant. 7.58. <0.001. Significant. 0.079. 2.29. <0.05. Significant. NE → A. -0.064. 1.89. <0.1. Not Significant. PI → A. 0.824. 30.27. <0.001. Significant. A → I. 0.684. 21.44. <0.001. Significant. Since SmartPLS did not provide p-values, I estimated it using two-tail test. As shown in table 19, all relationships except for the relationship of “Social Network Effects toward Attitude” have a t-value larger than 1.96 which indicates - 46 -.

(55) a p-value smaller than 0.05. The relationship between network effects and attitude toward in-App purchases only has a p- value around 0.1 which is not significant.. 4.4 Hypothesis Testing According to the path analysis and t-value we conclude that all hypotheses except for H4 are significant. See table 21 for hypotheses testing summary. Table 20: Summary of Hypotheses Test Result Hypothesis. t-Value. 政治大 associated with perceived playfulness of the 9.15 立. p-value. Support. <0.001. Yes. H1: Social network effects are positively. ‧ 國. 學. App. H2: Social network effects are positively 7.58. <0.001. io. al. toward. in- App. y. positive. n. purchase. with attitude. is. 2.29. <0.05. Yes. 1.89. <0.1. No. 30.27. <0.001. Yes. 21.44. <0.001. Yes. er. associated. playfulness. sit. Perceived. Nat. H3:. ‧. associated with Personal Involvement.. Yes. Ch. engchi. H4: Social network effects are positively associated with attitude toward. in-Apps. i n U. v. purchasing H5: Personal involvement. is positively. associated with the attitude toward download and use. H6: Attitude toward in-App purchasing is positively. associated. with. intention. purchase. - 47 -. to.

(56) Chapter V. Discussion 5.1 Discussion The prevailing state of smart handsets also stimulates the development of App industry. Among all profit models of Apps, in- App purchasing is becoming the favorable one. This study strives to understand the causes that would influence users’ behavior when deciding to purchase in-App products. 5.1.1 Summary of Results The result of current study shows that social network effects have an. 政治大 network has effect on both perceived playfulness and personal involvement; 立. impact on users’ perceived playfulness and personal involvement. Social. ‧ 國. 學. however, it influences perceived playfulness more than it does on personal involvement.. ‧. We also found that personal involvement and perceived playfulness. sit. y. Nat. have an impact on attitude toward in- App purchases. Personal involvement is. io. al. er. the strongest predictor of attitude toward in-App purchase. Both perceived. n. playfulness and personal involvement plays a role in influencing the attitude,. Ch. engchi. but personal involvement plays a more. iv n important U. role than perceived. playfulness. Surprisingly, social network effects do not have a direct impact on attitude toward in-App purchases. The reason could be that people nowadays have higher ego consciousness and the information is more transparent than before as a consequence of the prevailing the internet. Also, attitude toward in-App purchases have an impact on intention to purchase in- App product. This hypothesis is supported just as it is in technology acceptance model (TAM).. - 48 -.