Tại sao người dùng lại sáng tạo nội dung - Ứng dụng của thuyết hành vi có kế hoạch

A questionnaire was developed for use in the data collection process with 12 questions directly related to our research. Measurements for attitudes, subjective norms and cognitive behavioral control are adjusted from previous studies. For each element designed with three questions, we used the 5-point Likert scale for the study. In the process of developing a questionnaire, we refer to studies that apply TPB in explaining other human behavior. The SPSS analysis process applied to the thesis applied a lot of formulas. Among them, the formula for determining the minimum sample size for research is reliable. The size of the sample applied in the study is based on the requirements of the Exploratory Factor Analysis (EFA) and the multivariate regression. Based on research by Hair, Anderson, Tatham and Black for reference on expected sample size, accordingly the minimum sample size is 5 times the total number of observed variables. This is a suitable sample size for the study using factor analysis (Comrey, 1973; Roger, 2006). n = 5 * m, note that m is the number of questions in the lesson. So the accepted minimum is 60. For multivariate regression analysis: With Tabachnick and Fidell fomular, the minimum sample size to be obtained is calculated by the formula n = 50 + 8 * m=74. A total of 78 questionnaires were distributed to respondents on Facebook. The data were then analyzed using SPSS version 16. Descriptive analysis, reliability analysis, factor analysis and regression analysis were then performed on the data. Demographic A total of 78 respondents from Facebook community, majority of the respondents were female (43 respondents or 55.1%), between the ages of 18 and 30 years of age ( 38 respondents or 48.7%), most of them are high school students or not who have just only finished high school or not graduated from high school yet (40 respondents or 51.3%) . The background of the respondents is presented in bellow table ( Table 1).

pdf7 trang | Chia sẻ: hachi492 | Ngày: 14/01/2022 | Lượt xem: 16 | Lượt tải: 0download
Bạn đang xem nội dung tài liệu Tại sao người dùng lại sáng tạo nội dung - Ứng dụng của thuyết hành vi có kế hoạch, để tải tài liệu về máy bạn click vào nút DOWNLOAD ở trên
Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 08 (2018) 65 TẠI SAO NGƢỜI DÙNG LẠI SÁNG TẠO NỘI DUNG - ỨNG DỤNG CỦA THUYẾT HÀNH VI CÓ KẾ HOẠCH Zhou Xiao Hong 1 , Bùi Thị Thúy2 Tóm tắt Lý thuyết về hành vi có kế hoạch (TPB) kể từ khi phát triển khoảng 30 năm trước đã được chứng minh là một cách tiếp cận mạnh mẽ để giải thích hành vi của con người. Nó đã được áp dụng thành công cho một loạt các hành vi. Theo lý thuyết, hành vi của người tiêu dùng là một chức năng của ý định thực hiện hành vi được đề cập; hành vi dựa trên thái độ, chuẩn mực chủ quan và kiểm soát hành vi đối với hành vi; và các yếu tố này được xác định, tương ứng, bởi thái độ đối cá nhân đối với hành vi, chuẩn mực và kiểm soát. Với sự phát triển của công nghệ, nội dung do người dùng tạo ra (UGC) được coi là một phần của truyền miệng điện tử, được tạo ra và chia sẻ giữa người tiêu dùng có tầm quan trọng lớn đối với các nhà tiếp thị. Nghiên cứu này giải thích lý do tại sao người dùng tham gia vào việc tạo nội dung. Lý thuyết về hành vi có kế hoạch (TPB) đã được sử dụng để giải thích hành vi này. Thông qua một số câu hỏi khảo sát đã được thực hiện vào tháng 10/2018, sử dụng SPSS 16 với 78 người đã được hỏi về thông tin liên quan đến nội dung được tạo và chia sẻ trên internet, kết quả kiểm tra cho thấy ý định tạo nội dung của người dùng được xác định theo thái độ cá nhân, chuẩn mực và kiểm soát. Từ khóa: Lý thuyết về hành vi có kế hoạch (TPB), nội dung do người dùng tạo ra (UGC), truyền miệng điện tử (eWOM). WHY USERS GENERATE CONTENT AN APPLICATION OF THE THEORY OF PLANNED BEHAVIOR Abstract The theory of planned behavior (TPB) since its apprerance about 30 years ago has been proved to be a powerful approach to explain human behavior. It has been successfully applied to a variety of behaviors. According to the theory, the consumer’s behavior is a function of intention to perform the behavior in question; the behavior is based on attitude, subjective norm, and perceived behavioral control; and these factors are determined, respectively, by behavioral, normative, and control beliefs. With the development of technology, the user-generated content (UGC) is considered as a part of electronic word of mouth created and shared between consumers, which has a major importance to marketers. This study explains why users are involved in creating content. The theory of planned behavior (TPB) has been used to explain this behavior. Through survey questionnaires in 10/2018, using SPSS 16 with 78 respondents who were asked about information related to the generated and shared content on the internet, the results showe that the user's intention to generate content is determined by personal attitude, Subject norm, Perceived behavioral control. Keywords: Theory of planned behavior (TPB), User-generated content (UGC), electronic Word of mouth (eWOM). 1. Introduction Vietnam is currently ranked 7th in the number of Facebook users with about 60 million users. Zalo currently has about 40 million monthly users. Mocha of Viettel has about 4.5 million users. According to the 2017 survey results of Pew Research Institute, Vietnamese people ranked 4th in the world in terms of reading news online. (Trong Dat, 2018) With the development of technology, people need to change the way they communicate. Electronic Word-of-mouth (WOM) has been recognized as one of the most influential resources of information transmission. With the Internet, even ordinary Web users can conveniently create and disseminate media content. The notion of User- Generated Content captures the user-as-producer feature and refers to content that is not generated or published by professionals on the Internet, unlike traditional media. Defined in terms of situations where consumers suggest products or services to other consumers on the Internet, Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 07 (2018) 66 eWOM is closely related to UGC. (Ye Wang, Shelly Rodgers, 2011) UGC is related to, but not identical with, electronic word-of-mouth(eWOM), which is defined as being “any positive or negativestatement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet”(Hennig-Thurau et al.2004, p. 39). The impact of user generated content is undeniable. Brand engagement also increases when users share content. According to a ComScore study, brand engagement increases by 28% when consumers are exposed to a mixture of branded and user-generated content (Comscore, 2012). UGC also works wonders no matter which generation of your target audience. When Bazaarvoice asked a pool of Millennials and Baby Boomers how much user-generated content played into their purchase decisions the received answers were: 84% of Millennials said that UGC had at least some influence; 70% of Baby Boomers said that UGC had at least some influence; 20%+ of each generation said that UGC had a lot of influence. (Bazaarvoice, 2012). Understanding the factors that influence UGC creation is important for modern marketing. However, researches in this area are limited. With the application of TPB and accreditation to find answers to this problem is the purpose of this study. 2. Background 2.1. Electronic Word-of-mouth Social media has impacted various facets of modern life and it has a profound influence on interpersonal communication. People need interaction to fulfill their social needs and social media has become a preferred medium for communication with the proliferation of digital and mobile technologies (Kalpathy, 2017). People have grown up with the Internet as an important part of their everyday life and interaction rituals. They suggest that the reason is coming from the decrease in the amount of time they spend interacting face-to-face (Brignall and Van Valey, 2005). The advances in information technology and the emergence of online social network sites have changed the way information is transmitted and have transcended the traditional limitations of word of mouth (Mohammad Reza, 2010). Electronic word-of-mouth (eWOM) communication refers to any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet (T. Hennig- Thurau,2004). The web has created both challenges and opportunities for electronic word- of-mouth (eWOM) communication (R. E. Goldsmith, 2006). eWOM allows consumers to not only obtain information related to goods and services from the few people they know but also from a vast, geographically dispersed group of people, who have experience with relevant products or services. 2.2. User-Generated Content We mention definitions and outline for our understanding of UGC, which is often referred to within the scope of Web 2.0 and social media. One of the most quoted definitions of UGC is provided by the Organization for Economic Cooperation and Development (OECD) (Vickery and Wunsch-Vincent 2007). OECD uses the term of user-created content (UCC), which is considered synonymous with UGC. According to Vickery and Wunsch-Vincent (2007), UGC has three central characteristics: (1) publication requirement, (2) creative effort, and (3) creation outside of professional routines and practices. Base on the features, we can see some kinds of UGC normal in real life: Video on Youtube (Review, Parody commercials, Introduction product, Tutorial), Picture and Video on Instagram, Post on Facebook, Twitters, Rating and comments on websites (like E-commercial Shopee, Lazada or main website of products). In the past, there have been too many successful marketing campaigns, we can mention that Old Spice- Video Responses, Coca-Cola: Share a coke, Starbucks- White Cup Contest, ect by focusing on UGC. (Delhi school of internet marketing, 2016). But some of UGC became disasters such as McDonalds with hashtag # Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 08 (2018) 67 McDStories; Kia Sorento with creating Kia- themed memes; Starbucks with hashtag#SpreadTheCheer; Walgreens with hashtag#IloveWalgreens (Duel, 2017). They are good evidence to show up the power of UGC. 2.3. Theory of planned behaviour The Theory of Planned Behavior (TPB) developed by Ajzen (1985) is an explanatory model that has been widely applied in diverse studies on behavioral intention (Lee, Cerreto & Lee, 2010 ). TPB stipulates that voluntary human behavior is preceded by intention to engage in such behavior (Shirly & Todd, 2001). Then it postulates that behavioral intention in turn is determined by three major determinants – attitude towards behavior (AB), subjective norm (SN) and perceived behavioral control (PBC). Fig 1. Model depicting the theory of planned behavior (Ajzen, 1991) PBC judgments are determined by beliefs pertaining to the extent to which one has access to resources or opportunities necessary to carry out the behavior effectively, subjected to the perceived power of each factor to enable or prevent the behavior (Ajzen, 1991). 3. Conceptual framework and hypotheses In order to develop our research framework, we begin by examining the relationships between each element and UGC that appear in the literature. Based on the TPB, intention signifies the motivational components of behavior. It represents the conscious effort that a person is willing to invest in a behavior. Human action is guided by three kinds of readily accessible beliefs: behavioral beliefs are those about the likely consequences of the behavior, normative beliefs are those about the normative expectations and actions of important referents, and control beliefs are those about the presence of factors that may facilitate or impede performance of the behavior (East, 2000). In their respective aggregates, behavioral beliefs bring on a favorable or unfavorable attitude (ATT) toward the behavior; normative beliefs give rise to subjective norms (SN) or perceived social pressure (which also contribute to the forming of attitudes), and control beliefs result in perceived behavioral control (PBC). Fig 2. Proposed research model Attitude Subject norm Perceived Behaviour control Intention Behaviour Attitude to UGC Subject norm Perceived Behaviour control Intention to create UGC Create UGC Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 07 (2018) 68 So, base on TPB model, we suggest below hypotheses: H1: Consumer UGC attitude affects content creation on the network. H2: Social influence has effect on the content creation behavior on the network. H3: Behavior control affects the content creation behavior on the network 4. Methodology There is no official TPB questionnaire, however, base on the original research (Ajzen, 1991) and instrument of "Constructing a theory of planned behavior questionnaire" and sample questionnaire are provided by Ajzen (2013) we developed the questionnaire for this research. We focus on a behavior creat video upload to facebook then interaction between them and their facebook friend. To measure for “Behavior”, we create 3 questions about frequency of the action record, create Video, Edit Video and Share Video about a product. We use Likert 5 point scales with 1= rarely and 5=Usually for measurement behavior. The "Attitude" scale consists of 3 items that reflect the implementation of actions towards action, the respondents feel interesting, value, happy when share information, we use Likert 5 point scales with this variation. “Subject norm” shows the reaction toward the action by other people from the community, they might friends, another member from society, so we develop 3 questions your friend always creates and share contents on the Internet, your friend react positively whenever you share contents on the Internet, your friend appreciated your contents, we use Likert 5 point scales with this variation with 5 is highest point for positive reaction. “Control” mention the ability of respondents when we tend to do the behavior. So in this case, we develop 3 questions to describe the ability of the user, do they meet any difficulty when they want to do the action which is showing up through they photograph skills, supporting equipment, Editing skills. And we use Likert 5 point scales with this variation with 5 was for master skills and 1 for novice skills. A questionnaire was developed for use in the data collection process with 12 questions directly related to our research. Measurements for attitudes, subjective norms and cognitive behavioral control are adjusted from previous studies. For each element designed with three questions, we used the 5-point Likert scale for the study. In the process of developing a questionnaire, we refer to studies that apply TPB in explaining other human behavior. The SPSS analysis process applied to the thesis applied a lot of formulas. Among them, the formula for determining the minimum sample size for research is reliable. The size of the sample applied in the study is based on the requirements of the Exploratory Factor Analysis (EFA) and the multivariate regression. Based on research by Hair, Anderson, Tatham and Black for reference on expected sample size, accordingly the minimum sample size is 5 times the total number of observed variables. This is a suitable sample size for the study using factor analysis (Comrey, 1973; Roger, 2006). n = 5 * m, note that m is the number of questions in the lesson. So the accepted minimum is 60. For multivariate regression analysis: With Tabachnick and Fidell fomular, the minimum sample size to be obtained is calculated by the formula n = 50 + 8 * m=74. A total of 78 questionnaires were distributed to respondents on Facebook. The data were then analyzed using SPSS version 16. Descriptive analysis, reliability analysis, factor analysis and regression analysis were then performed on the data. Demographic A total of 78 respondents from Facebook community, majority of the respondents were female (43 respondents or 55.1%), between the ages of 18 and 30 years of age ( 38 respondents or 48.7%), most of them are high school students or not who have just only finished high school or not graduated from high school yet (40 respondents or 51.3%) . The background of the respondents is presented in bellow table ( Table 1). 5. Result 5.1 Reliability statistic Cronbach’s Alpha Cronbach‟s Alpha of them from 0.6 to 0.9 (Behavior- 0.845; Attitudes - 0.821; Subjective Norm – 0.836; Control - 0.780) and Corrected item- Total correlation >0.4 so and Cronbach‟s Alpha if item Deleted isn‟t bigger than the Total Cronbach Alpha so don‟t need to delete any question (Table 1). Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 08 (2018) 69 Table 1: Reliability statistic Cronbach’s Alpha Item-Total Statistics Scale Mean if Item Deleted Scale Variance if Item Deleted Corrected Item- Total Correlation Cronbach's Alpha if Item Deleted Cronbach's Alpha BH1 6,87 4,11 0,74 0,76 0,845 BH2 6,88 4,26 0,65 0,84 BH3 6,94 3,49 0,75 0,74 ATT1 7,59 3,78 0,68 0,75 0,821 ATT2 7,37 4,37 0,61 0,82 ATT3 7,42 3,78 0,75 0,68 SN1 7,21 3,26 0,8 0,68 0,836 SN2 7,09 3,33 0,64 0,83 SN3 7,09 3,36 0,66 0,81 TBC1 7,53 2,49 0,65 0,66 0,78 TBC2 7,5 2,77 0,59 0,73 TBC3 7,56 2,66 0,61 0,71 5.2 Exploratory factor analysis 0.5 ≤ KMO≤ 1: KMO coefficient (Kaiser- Meyer-Olkin) is an index used to consider the appropriateness of factor analysis. In this research KMO=0.756 values mean that factor analysis is appropriate.( Table 5.2.1) The Bartlett test has statistical significance (Sig. <0.05): This is a statistical quantity used to consider the hypothesis that variables are not correlated in the overall. This test is statistically significant ( 0.000 <0.05) so the observed variables are correlated with each other in the overall.( Table 5.2.1) With 9 input variables, PCA initially extracts 9 factors (or “components”). Each component has a quality score called an Eigenvalue. Only components with high Eigenvalues are likely to represent a real underlying factor. A common rule of thumb is to select components whose Eigenvalue is at least 1. So our 9 variables seem to measure 3 underlying factors. (Table 5.2.2) Percentage of variance 74.088% > 50%: Shows the percentage variation of observed variables. This means that when the variable is 100%, the value indicates 74.088% the factor analysis explains. Factor loading each item > 0.5 is considered to have practical significance (Table 2) Table 2: Total Variance Explained Component Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative % 1 3,745 41,61 41,61 3,745 41,61 41,61 2,281 25,346 25,346 2 1,599 17,762 59,372 1,599 17,762 59,372 2,229 24,77 50,116 3 1,324 14,717 74,088 1,324 14,717 74,088 2,158 23,972 74,088 4 0,61 6,774 80,862 5 0,513 5,7 86,561 6 0,392 4,359 90,921 7 0,36 4 94,92 8 0,258 2,864 97,784 9 0,199 2,216 100 Extraction Method: Principal Component Analysis We use Rotated Component Matrix as below, results show ATT, SN, TBC are Convergent Validity Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 07 (2018) 70 Table 3: Rotated Component Matrix Rotated Component Matrix a Component 1 2 3 ATT1 ,787 ATT2 ,787 ATT3 ,907 SN1 ,924 SN2 ,844 SN3 ,760 TBC1 ,842 TBC2 ,796 TBC3 ,803 5.3 Testing hypothesises Adjusted R Square, also known as R square correction, it reflects the degree of influence of the independent variables on the dependent variable. Specifically, in this case, 3 independent variables affect 59.4% of the variation of the dependent variable, the remaining 40.6% is due to out-of-model variables and random errors. This value is more than 50%, the study can be used. If d > dU,α, there is no statistical evidence that the error terms are positively autocorrelated( Durbin-Watson Significance Tables) To test for positive autocorrelation at significance α, the test statistic d is compared to lower and upper critical values (dL,α and dU,α): 4-dL>d > dU (4- 1.57>2.112>1.72),α, there is no statistical evidence that the error terms are positively or negative autocorrelated. (Table 3) Table 4: Model summarary Model R R Square Adjusted R Square Std. Error of the Estimate Durbin-Watson 1 ,781 a ,610 ,594 ,580 2,112 a. Predictors: (Constant), TBC, SN, ATT b. Dependent Variable: BH For VIF (variance inflation factor) for each item <2 is not multicollinear and Sig <0.05 so the hypothesises are supported (Table 5.3.2) Hypotheses H1: Consumer UGC attitude affects content creation on the network. (Supported) H2: Social influence has effect on the content creation behavior on the network. (Supported) H3: Behavior control affects the content creation behavior on the network. (Supported) 6. Conclusion and Limitation Theoretically, this study lends support to the theory of planned behavior in explaining intention to generate digital information as UGC. All the factors; attitude, subjective norms and perceived behavioral control, all of them were tested to be positively influenced the intention to users generate content. This indicates that attitude, subjective norms and perceived behavioral control were predictors of intention to use digital coupon. Overall, these factors explained about 59.4% of the variance in intention while the remaining 40.6% may be explained by other factors that were not captured in this model. The attitude was found to be the strongest predictor of intention to use generate content followed by perceived behavioral control and subjective norms. Limitations This study faced a number of limitations. Firstly, data for this study were obtained from a sample including 78 people. If all the large samples size were examined, the result could have been generalized. Secondly, the study focused only on the consumer behavioral intention, but actual usage was not measured. Chuyên mục: Quản trị KD & Marketing - TẠP CHÍ KINH TẾ & QUẢN TRỊ KINH DOANH SỐ 08 (2018) 71 REFERENCES [1]. AC Nielson. (2007). Trust in Advertising. A Global Nielsen Consumer Report, October. [2]. Ajzen, I. (1991). The Theory of planned behavior. Journal of Organizational Behavior and Human Decision Processes, 50, 179–211 [3]. Armitage, C.J. and Corner, M. (2001). Efficacy of the theory of planned behavior: A meta analytic review. British Journal of Social Psychology. 40 (4), 471 - 499. [4]. Bazaarvoice. (2012). Talking to Strangers: Millennials Trust People over Brands. [5]. Brignall, T.W., & van Valey, T. (2005). The impact of Internet communications on social interaction. Sociological Spectrum, 335 - 348. [6]. Comscore. 2012). Produced Video Content And User-Generated Product Videos Exhibit Strong Synergy in Driving Sales Effectiveness. [7]. Delhi school of internet marketing. (2016). 7 User-Generated Content Examples You Can Use Right Away, https://dsim.in/blog/2017/06/05/7-user-generated-content-examples-can-use-right-away/ [8]. Duel. (2017). Worst UGC Fails, June 13, 2017, https://duel.tech/worst-ugc-fails/ [9]. Durbin-Watson Significance, Tables,https://www3.nd.edu/~wevans1/econ30331/Durbin_Watson_tables.pdf [10]. East, R. (2000). Complaining as planned behavior. Psychology & Marketing, 17, 1077–1095. [11]. Hennig-Thurau, Thorsten, Kevin P. Gwinner, Gianfranco Walsh, and Dwayne D. Gremler. (2004). Electronic Word-of-Mouth Via Consumer Opinion Platforms: What Motivates Consumers to Articulate Themselves on theInternet. Journal of Interactive Marketing, 18, 1, 38 – 52 [12]. Kalpathy Ramaiyer Subramanian. (2017). Influence of Social Media in Interpersonal Communication. Internation Journal of Scientific Progress and Research(IJSPR),70 - 75. [13]. Kim, E., Ham, S., Yang, I. S., & Choi, J. G. (2013). The roles of attitude, subjective norm, and perceived behavioral control in the formation of consumers‟ behavioral intentions to read menu labels in the restaurant industry. International Journal of Hospitality Management, 35, 203– 213. doi: 10.1016/j.ijhm.2013.06.008. [14]. Lee, J., Cerreto, F. A., & Lee, J. (2010). Theory of planned behavior and teachers ‟ decisions regarding use of educational technology. Journal of Educational Technology & Society, 13(1), 152–164. [15]. Manning, M. (2009). The effects of subjective norms on behaviour in the theory of planned behaviour: a meta-analysis. The British journal of social psychology / the British Psychological Society, 48(Pt 4), 649–705. doi:10.1348/014466608X393136. [16]. Mohammad Reza Jalilvanda, Sharif Shekarchizadeh Esfahani, Neda Samiei. (2010). Electronic word-of-mouth: challenges and opportunities. Procedia Computer Science, 3 (2011) 42 – 46 [17]. R. E. Goldsmith. (2006). Electronic word-of-mouth, in Khosrow-Pour, M. (Ed.). Encyclopedia of E-Commerce, E-Government and Mobile Commerce, Idea Group Publishing, Hershey, PA, pp. 408-12. [18]. Shirly Taylir & Peter A. Todd. (2001). Understanding information technology usage: A test of competing models. Information Systems Research, 6(2), 144–176. doi:1047-7047/95/0602/0144 [19]. Trong Dat. (2018). Đâu là cơ hội phát triển của các mạng xã hội Made in Việt Nam. Vietnamnet 10/12/2018, https://vietnamnet.vn/vn/cong-nghe/tin-cong-nghe/dau-la-co-hoi-cua-cac-mang-xa-hoi- made-in-viet-nam-493370.html. [20]. Vickery, G., and Wunsch-Vincent, S. (2007). Participative Web And User-Created Content: Web 2.0 Wikis and Social Networking, Paris, France: Organization for Economic Cooperation and Development (OECD). [21]. Ye Wang, Shelly Rodgers. (2011). Chapter 11 Electronic Word of Mouth and Consumer Generated Content: From Concept to Application. Digital Media and Advertising, 212-231 Thông tin tác giả: 1. Zhou Xiao Hong - - Đơn vị công tác: Professor in Nanjing University of Science and Technology 2. Bùi Thị Thúy - - Đơn vị công tác: Student of Nanjing University of Science and Technology - Địa chỉ email: buithithuy.neu@gmail.com Ngày nhận bài: 12/10/2018 Ngày nhận bản sửa: 2/11/2018 Ngày duyệt đăng: 28/12/2018

Các file đính kèm theo tài liệu này:

  • pdftai_sao_nguoi_dung_lai_sang_tao_noi_dung_ung_dung_cua_thuyet.pdf
Tài liệu liên quan