Academic Journal of Science and Technology ISSN: 2771-3032 | Vol. 13, No. 1, 2024 117 The Factors Influencing Consumer Continuous Purchase Intention of Online Micro‐Dramas: Case of Douyin (TIKTOK) Haoyan Cui1, a, *, Nutteera Phakdeephirot1, b 1Rattanakosin International College of Creative Entrepreneurship, Rajamangal University of Technology Rattanakosin, Nakhon Pathom 73170, Thailand a309642873@qq.com, bnutteera.pha@rmutr.ac.th *Corresponding author: Haoyan Cui (Email: 309642873@qq.com) Abstract: The COVID-19 pandemic has significantly impacted social interaction and entertainment, leading to a surge in internet economy and particularly the micro-drama industry on platforms like Douyin. This study investigates the factors influencing consumer continuous purchase intentions for online micro-dramas, focusing on Douyin as a case. Utilizing Expectancy Confirmation Theory (ECT) and Flow Theory, the research identifies key factors such as user satisfaction, perceived usefulness, and flow experience. A survey of 400 valid responses was conducted, and data was analyzed using Statistical Data Analysis, aiming to provide theoretical support and practical suggestions for platform developers, content creators, and policymakers. Keywords: Micro-drama, Douyin, Expectancy Confirmation Theory (ECT), Flow Theory. 1. Introduction The global outbreak of the COVID-19 pandemic at the end of 2019 had a profound impact on various aspects of human society, particularly in areas such as social interaction, work, and entertainment. However, the internet economy experienced unprecedented growth opportunities. As seen during the pandemic control efforts, while offline activities were significantly restricted, many new business models integrating both online and offline channels rapidly developed [1]. By June 2020, the number of online audiovisual users in China had reached 901 million, an increase of 43.8 million compared to March 2020, with a penetration rate among internet users of 95.8% [2]. According to the 54th Statistical Report on the Development of the Internet in China, published by the China Internet Network Information Center , the number of internet users in China has reached nearly 1.1 billion (1.09967 billion), with network video (including short videos and micro-dramas) users totaling 1.068 billion. Among them, micro-drama users reached 576 million, accounting for 52.4% of internet users. The micro-drama industry, with its ability to empower cultural and tourism industries, its incorporation of emerging technologies like AIGC for content creation, and rapid user growth, has now become a new driving force for innovation in the online video industry [3]. As a leader in China’s short video sector, Douyin plays a pivotal role in the development of micro-dramas[4]. However, the micro-drama industry has yet to establish a long-term profitable model due to the complexity of distribution channels and monetization methods. The industry's over-reliance on the "influencer + live commerce" model may lead to misguided development[5]. Additionally, challenges such as the difficulty of maintaining revenue- sharing mechanisms, limited opportunities for brand collaboration due to the small scale of productions, and vague user metrics affecting live commerce data make it hard to cultivate a user payment mindset on short video platforms. Consequently, the commercial transformation of IP-based micro-dramas continues to face difficulties [5]. Beyond profitability challenges, the production side of micro-dramas also suffers from content homogeneity, weak intellectual property awareness, inconsistent content quality, and excessive entertainmentization[7]. Despite these issues, the micro-drama industry shows significant development potential in terms of policy support, market expansion, and business model innovation. At the same time, it faces challenges regarding content quality and monetization models. The core issue is the sustained viewing intentions of consumers, which will determine the future direction of the industry, policy formulation, and business models. This study, using the Douyin platform as an example, investigates the key factors influencing consumers' continuous purchase intentions, particularly focusing on the consumption behavior of short video web dramas on the platform. Based on the Expectancy Confirmation Theory (ECT) and Flow Theory, the study identifies that factors such as user satisfaction, perceived usefulness, flow experience, and expectation confirmation are closely related to continuous purchase intentions. The research collected 400 valid survey responses and analyzed the data using the Statistical Data Analysis , aiming to provide theoretical support and practical suggestions for platform developers, content creators, and policymakers. 2. Literature Review 2.1. Consumer Continuous Purchase Intention The concept of intention was first discussed in the field of psychology. It generally refers to an individual's perception or thoughts about an object, which in turn leads to personal subjective thinking. According to Li Bin [8] and Zhang Qingling [9], there are currently four main theories related to intention. These include the Theory of Reasoned Action 118 proposed by Fishbein and Ajzen [10], the Technology Acceptance Model developed by Davis [11] based on the Theory of Reasoned Action, the Theory of Planned Behavior introduced by Ajzen [8], and the Expectation Confirmation Model put forward by Churchill and Surprenant [8]. 2.2. Expectation Confirmation Theory Expectation Confirmation Theory was first proposed by Oliver to study issues related to product repurchase. However, with the development of information systems, Bhattacherjee, 2001 [12] built a new continuous usage model for the information systems context based on Expectation Confirmation Theory. He demonstrated that variables such as perceived usefulness, satisfaction, and expectation confirmation can significantly influence users’ continuous usage intentions. Subsequently, many scholars have extended this model by incorporating relevant variables and have confirmed that the continuous usage model for information systems has broad applicability and strong explanatory power[13] . When studying the continuous usage model for information systems, many scholars typically combine Expectation Confirmation Theory with other theories to enrich the theoretical framework of the model. They use methods such as structural equation modeling (SEM) to conduct empirical analysis on their respective research subjects, proving the model’s effectiveness in studying users' continuous usage behaviors. Following this approach, Building on this foundation, the present study is grounded in Expectancy Confirmation Theory within the context of information systems and integrates it with Process Theory. The independent variables include Expectancy Confirmation, Perceived Usefulness, and Flow Experience; Satisfaction serves as a mediating variable; and Continued Purchase Intention is designated as the dependent variable.From a theoretical standpoint, this research combines Expectancy Confirmation Theory with Flow Theory to develop a conceptual model applicable to the micro-drama sector of short video platforms. The aim is to elucidate and predict users' intentions for continued engagement with micro- dramas on these platforms. Practically, this study offers theoretical insights that can assist short video platforms in understanding the psychological needs of micro-drama users, thereby enhancing their willingness to continue using such content.Based on the discussion of Expectation Confirmation Theory within the context of information systems, this paper proposes the following hypotheses: H1: Satisfaction is positively correlated with continuous usage intentions, meaning that the higher the satisfaction of consumers watching micro-dramas on Douyin, the stronger their intention to continue using Douyin to watch micro- dramas. H2: Expectation confirmation is positively correlated with satisfaction, meaning that the more consumers feel their expectations are met when using Douyin to watch micro- dramas, the higher their satisfaction with watching micro- dramas on Douyin. H3: Perceived usefulness is positively correlated with satisfaction, meaning that the more useful consumers find Douyin for watching micro-dramas, the higher their satisfaction with watching micro-dramas on Douyin. H4: Expectation confirmation is positively correlated with perceived usefulness, meaning that the more consumers feel their expectations are met when using Douyin to watch micro- dramas, the more useful they find Douyin for watching micro- dramas. H5: Perceived usefulness is positively correlated with continuous usage intentions, meaning that the more useful consumers find Douyin for watching micro-dramas, the stronger their intention to continue using Douyin to watch micro-dramas. 2.3. Flow Theory Flow Theory was first introduced by the renowned psychologist Csikszentmihalyi [14], describing the state in which people become so immersed in an activity that everything else seems insignificant. In recent years, other scholars have confirmed through related studies that flow experience significantly influences internet users' attitudes and behaviors [15]. With the advancement of internet technologies and the extensive research on Expectation Confirmation Theory, the combination of Flow Theory and Expectation Confirmation Theory has been increasingly applied to studies on consumers' continuous usage behaviors in various fields, such as video platforms (Wang Man), automotive information apps (Wang Xiaoyu), and knowledge payment platforms (Li Juanjuan). However, there is a relative gap in research on the micro-drama industry within short video platforms, which have a larger market and more prominent challenges. Therefore, this study attempts to integrate Flow Theory and proposes the following hypotheses: H6: Flow experience is positively correlated with satisfaction, meaning that the stronger the flow experience consumers feel from watching micro-dramas, the higher their satisfaction with watching micro-dramas on Douyin.[16,17,18] H7: Flow experience is positively correlated with continuous usage intentions, meaning that the stronger the flow experience consumers feel from watching micro-dramas, the stronger their intention to continue using Douyin to watch micro-dramas. 2.4. Model development This study will integrate Flow Theory into the Expectation Confirmation Theory model proposed by Bhattacherjee (2001) to develop the research model for this paper (as shown in Figure 1). In this model, confirmation, perceived usefulness, and flow experience serve as independent variables, satisfaction acts as a mediating variable, and continuous purchase intention is the dependent variable. 119 Figure 1. Research Model of Continuous Purchase Intention of Douyin Micro-Dramas Consumers Source: Compiled and drawn from the current study 3. Methodology This study adopts a quantitative research method, which is more suitable for large sample research based on practical considerations. The questionnaire design references existing scales for the five variables, specifically: continuous usage intention references Bhattacherjee (2001), satisfaction references Peter & Vanessa (1996), perceived usefulness references Li Bin (2012) and Wang Juan (2010), expectation confirmation references Bhattacherjee (2001), and flow theory references Novak, Hoffman & Yang (2000), and Chen Lijuan (2006). According to the 54th China Internet Development Status Statistical Report, as of June 2024, the user base of micro-short dramas has reached 576 million people.Furthermore, the questionnaire items were designed based on the micro-drama user profile data .[19] The questionnaire is divided into four parts, totaling 30 items. The first part contains screening questions to exclude respondents who have never used the Douyin platform to watch online micro-dramas, ensuring the efficiency of the returned data. The second part is designed to collect demographic information such as age, gender, occupation, and education level, based on relevant papers and professional reports. The third part collects data on respondents' usage and consumption behaviors. The fourth part employs a five-point Likert scale [20] to measure respondents' perceptions of the five variables: continuous purchase intention, satisfaction, expectation confirmation, perceived usefulness, and flow experience. A score of 1 indicates strong disagreement, and a score of 5 indicates strong agreement, with higher scores representing stronger agreement with the item description, and lower scores indicating less agreement. This study adopted a mixed sampling strategy, combining simple random sampling and convenience sampling in order to ensure the representativeness of the sample while also taking advantage of the convenience and efficiency of convenience sampling. The use of this mixed method is based on the research design principles of Creswell [21], who emphasizes that in some cases, mixed methods can provide more comprehensive data analysis. To efficiently collect data from the target audience, this study allows the researcher to select samples based on convenience and accessibility, thus employing convenience sampling from non-probability sampling methods. As for sample size, the population of micro-drama users is 576 million. Using the formula for calculating simple random sample size (Yamane, 1973)[22]: N = n / (1 + n * e^2) (N = sample size, n = population size = 576 million, e = margin of error, e = 0.05), the calculation is as follows: N = 576,000,000 / (1 + 576,000,000 * 0.05^2) = 399.9. The total number of questionnaires required for this study is calculated to be 399.9; therefore, the total number of questionnaires to be collected is set at 400. Considering the uncontrollable factors that may arise during the questionnaire collection process, the study aims to collect 500 questionnaires. The final research subjects will be users of micro-dramas on the Chinese Douyin platform. The survey questionnaire, titled "Factors Influencing the Continuous Purchase Intent of Douyin Micro-drama Users," will be developed based on existing scales. The questionnaire will be created using the Wenjuanxing app, and users of short drama videos will be individually invited to participate via private messages. The data collection period runs from September 15, 2024, to October 15, 2024, lasting one month. A total of 500 questionnaires were collected during this period, of which 48 were discarded because the respondents selected "I have not watched internet micro-dramas on Douyin." In the remaining 452 questionnaires, 400 were selected using convenience sampling. Ultimately, 400 valid questionnaires were obtained (complete data is shown in Table 1). The data shows that the respondents are predominantly female, accounting for 61.8% (n=247). The majority of respondents are aged 25-34, making up 45% (n=180). Students and freelancers account for 49.3% (n=197) and 30.8% (n=123) of the sample, respectively. In terms of educational background, individuals with junior high school education or below constitute the largest group, representing 50.7% (n=203). Regarding disposable income, the high- income group with monthly incomes above 8,000 yuan constitutes the largest percentage at 42.5% (n=170). The frequency of watching micro-dramas is predominantly once to three times a week, accounting for 45.3% (n=181). Most respondents reported watching for over 10 minutes, representing 49% (n=196). In terms of payment behavior, the majority of respondents indicated that they have not paid for micro-dramas, with 59.3% (n=237) stating they have never made any purchases. Regarding payment methods, buying 120 membership on the short drama platform is the most popular choice, accounting for 47.3% (n=189). Table 1. Description of the overall situation of the sample Variable Classification Frequency Percentage (%) Gender Male 153 38.3 Female 247 61.8 Total 400 Age 18-24 years old 121 30.3 25-34 years old 180 45 35-44 years old 76 19 45-55 years old 23 5.8 Over 40 years old 0 0 Total 400 Occupation white-collar worker 21 5.3 Students 197 49.3 Civil servant or public institution personnel 21 5.3 Freelancers 123 30.8 Other 38 9.5 Total 400 Educational status Middle school and below 203 50.7 Secondary School/Vocational College/Technical Institute 22 5.5 Associate Degree 43 10.8 Bachelor's degree 116 29 Master and above 16 4 Total 400 Monthly Discretionary Income Under 2000 yuan 20 5 2001-4000 yuan 24 6 4001-6000 yuan 125 31.3 6001-8000 yuan 61 15.3 Over 8,000 yuan. 170 42.5 Total 400 The weekly viewing frequency of micro short dramas. 1-3 times (rarely watched) 181 45.3 4-8 times (occasionally watching) 138 34.5 More than 8 times (regularly watching) 81 20.3 Total 400 The duration of each viewing session of micro short dramas. 1 minute or fewer 21 5.3 2-5 minutes 70 17.5 5-8 minutes 70 17.5 8-10 minutes 43 10.8 More than 10 minutes. 196 49 Total 400 Have you ever engaged in any paid activities to view micro short dramas? Yes 163 40.8 no 237 59.3 Total 400 The amount of money spent watching short plays 1-30 yuan 94 23.5 31-60 yuan 0 0 61-100 yuan 46 11.5 Over 100 yuan 2 0.5 I can't remember. 23 5.8 Never engaged in consumption. 235 58.8 Total 400 Watch short dramas and pay for the more inclined to pay Purchase a membership for the micro short drama platform. 189 47.3 View one episode, purchase one episode. 99 24.8 Purchase the complete short drama in a single transaction. 112 28 Total 400 121 4. Data Analysis and Results 4.1. Reliability analysis Reliability, also known as consistency, refers to the degree of consistency in results obtained from repeated measurements on the same sample using the same method. Reliability indicators are often represented by correlation coefficients and can be broadly divided into three categories: internal consistency coefficients (consistency across items), stability coefficients (consistency over time), and equivalence coefficients (consistency across forms). The reliability of a questionnaire generally concerns whether a set of questions (or the entire questionnaire) measures the same concept, in other words, how well the internal consistency among these items is measured (George & Mallery, 2018)[23]. This study employs the most commonly used reliability coefficient method to evaluate the reliability of the questionnaire. It is generally accepted that if the Cronbach's Alpha value is greater than 0.8, it indicates excellent reliability (Tavakol & Dennick, 2011). As shown in Table 2, the Cronbach's Alpha values for all dimensional variables are greater than 0.8. Among them, the reliability of the variables "Satisfaction" and "Flow Theory" even exceeds 0.9. Furthermore, the overall Cronbach's Alpha value for the variables is 0.916, indicating that the reliability of the questionnaire is very good and can be further tested. Table 2. Sample Reliability Analysis Latent Variable Observed Variable Standard Loading Cronbach's α Total Cronbach's α Intention to Continue Use A1 0.690 0.817 0.916 A2 0.690 B1 0.799 0.917 Satisfaction B2 0.796 B3 0.780 B4 0.865 Perceived Usefulness C1 0.753 0.885 C2 0.783 C3 0.793 Expectation Confirmation D1 0.724 0.866 D2 0.767 D3 0.743 Flow Theory E1 0.756 0.927 E2 0.769 E3 0.760 E4 0.761 E5 0.759 E6 0.799 E7 0.772 4.2. Validity analysis Validity refers to the accuracy of the measurement, meaning the extent to which a questionnaire can measure the observed variables it intends to measure in relation to the actual situation. Regarding the construct validity of the questionnaire, this study employs factor analysis as a measurement method. Before conducting factor analysis, it is necessary to test the KMO (Kaiser-Meyer-Olkin) statistic. Kaiser provides the commonly used KMO measurement standards: a value above 0.9 indicates excellent suitability; 0.8 indicates suitability; 0.7 indicates moderate suitability; 0.6 indicates poor suitability; and a value below 0.5 indicates very poor suitability (Kaiser, 1991). Table 3. KMO and Bartlett spheres test analysis KMO Value 0.911 Bartlett’s Sphericity Test Chi-square value 5079.504 Degrees of freedom (df) 171 Significance (sig.) 0.000 122 4.3. Exploratory factor analysis Exploratory factor analysis is used to analyze whether the research items are reasonable and meaningful. It conducts a comprehensive analysis using indicators such as variance explained and factor loading coefficients to verify the factor structure of the data. The variance explained value indicates the level of information extraction, while the factor loading coefficient measures the relationship between the factors and the items (Watkins, 2018)[24]. Table 4. Total variance explanation table Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of Variance Cumulative % Total % of Variance Cumulative % Total % of Variance Cumulative % 1 7.749 40.786 40.786 7.749 40.786 40.786 4.913 25.857 25.857 2 2.953 15.54 56.326 2.953 15.54 56.326 3.144 16.548 42.405 3 1.629 8.573 64.899 1.629 8.573 64.899 2.507 13.195 55.599 4 1.258 6.624 71.522 1.258 6.624 71.522 2.378 12.516 68.115 5 1.041 5.478 77.000 1.041 5.478 77.000 1.688 8.885 77.000 Extraction Method: Principal Component Analysis. Table 5 presents the rotated component matrix, which is used to observe whether each question within each dimension has been properly extracted. This study employs the Varimax rotation method for exploratory factor analysis on all dimensions of the questionnaire. As shown in Table 4-5, the loading coefficients for all dimensions are greater than 0.7. The items for the dimensions of Intention to Continue Use (A1-A3), Satisfaction (B1-B4), Perceived Usefulness (C1- C3), Expectation Confirmation (D1-D3), and Flow Theory (E1-E7) are well extracted, proving the good construct validity of the variables. This indicates that the dimensional division of the questionnaire is highly scientific and reasonable. Table 5. Component matrix after rotation 1 2 3 4 5 A1 0.065 0.262 0.203 0.045 0.849 A2 0.061 0.176 0.239 0.074 0.869 B1 0.258 0.808 0.160 0.114 0.181 B2 0.217 0.826 0.202 0.136 0.089 B3 0.226 0.780 0.222 0.166 0.161 B4 0.217 0.836 0.227 0.175 0.178 C1 0.140 0.207 0.831 0.1 0.167 C2 0.098 0.251 0.863 0.011 0.103 C3 0.084 0.185 0.861 0.055 0.215 D1 0.187 0.157 0.038 0.845 0.014 D2 0.247 0.173 0.039 0.845 0.061 D3 0.277 0.122 0.089 0.827 0.069 E1 0.764 0.253 0.109 0.159 -0.01 E2 0.799 0.182 0.011 0.158 0.067 E3 0.783 0.163 0.098 0.168 0.06 E4 0.812 0.109 0.056 0.125 0.086 E5 0.795 0.115 0.13 0.149 0.077 E6 0.829 0.143 0.112 0.143 -0.009 E7 0.818 0.146 0.021 0.137 0.005 Extraction method: principal component analysis. Rotation method: Caesar's normalized maximum variance method. a rotation converges after 6 iterations. 4.4. Correlation Analysis After the previous data analysis, it was proven that both the questionnaire and the dimensional divisions in this study are valid. The next step involves correlation analysis, which primarily examines the relationships between the variables. Through the correlation test, we can observe the relationships between each variable in this study. The correlation coefficient ranges from -1 to 1, with larger absolute values indicating a stronger correlation between the variables (Senthilnathan, 2019)[25]. 123 Table 6. Correlation analysis table Intention to Continue Use Satisfaction Perceived Usefulness Expectation Confirmation Flow Theory Intention to Continue Use 1 Satisfaction .456** 1 Perceived Usefulness .463** .500** 1 Expectation Confirmation .186** .401** .197** 1 Flow Theory .190** .478** .266** .466** 1 **Significant correlation at the 0.01 level (two-tailed). The detailed results are shown in Table 6, indicating that the influence of multicollinearity is minimal, as all correlation coefficients are below 0.6. The table reveals that the relationships between continuous purchase intention, satisfaction, perceived usefulness, expectation confirmation, and flow theory are significantly positively correlated, which generally aligns with expectations. 4.5. Regression Analysis Regression analysis is a statistical method used to determine the quantitative relationship of dependence between two or more variables. This study employs multiple linear regression analysis to verify the causal relationships among the hypotheses of continuous purchase intention, satisfaction, perceived usefulness, expectation confirmation, and flow theory. 4.5.1. Independent Variables and Continuous Purchase Intention Table 7. Regression analysis of continuous purchase intention R R² Adjusted R² Durbin- Watson F Standardized Coefficient β t p Tolerance VIF Satisfaction 0.532a 0.283 0.278 2.219 52.121 0.321 5.956 0.000 0.622 1.609 Perceived Usefulness 0.314 6.395 0.000 0.749 1.335 Flow Theory -0.047 -0.974 0.331 0.770 1.298 a Predictor Variables: (Constant), E Flow Theory, C Perceived Utility, B Satisfaction b Dependent Variable: A Continued Purchase Intent 4.5.2. Independent variables and satisfaction levels Table 8. Regression analysis of continuous purchase intention R R² Adjusted R² Durbin- Watson F Standardized Coefficient β t p Tolerance VIF Satisfaction 0.638a 0.407 0.403 2.044 90.627 0.386 9.587 0.000 0.923 1.084 Perceived Usefulness 0.192 4.373 0.000 0.777 1.286 Flow Theory 0.286 6.416 0.000 0.752 1.330 a Predictor Variables: (Constant), E Flow Theory, C Perceived Utility, D Expectancy Confirmation b Dependent Variable: B Satisfaction 4.5.3. Independent Variables and Perceived Utility Table 9. Regression analysis of perceived usefulness R R² Adjusted R² Durbin- Watson F Standardized Coefficient β t p Tolerance VIF Perceived Usefulness 0.197a 0.039 0.036 1.867 16.111 0.197 4.014 0.000 1.000 1.000 a Predictor Variable: (Constant), D Expected Confirmation b Dependent Variable: C Perceived Usefulness 124 4.6. Model Assumptions and Findings Table 10. Results of hypothesis testing Research Variable Hypothesis Result Expectation Confirmation Theory H1: Satisfaction is positively correlated with continuous usage intention, meaning that the higher the satisfaction of consumers watching Douyin micro-dramas, the higher their willingness to continuously use the Douyin platform to watch micro- dramas. Established H2: The level of expectation confirmation is positively correlated with satisfaction, meaning that the more consumers' experiences using the Douyin platform to watch micro-dramas meet their expectations, the higher their satisfaction with watching micro-dramas on the Douyin platform. Established H3: Perceived usefulness is positively correlated with satisfaction, meaning that the more useful consumers feel the Douyin platform is for watching micro- dramas, the higher their satisfaction with watching micro-dramas on the Douyin platform. Established H4: The level of expectation confirmation is positively correlated with perceived usefulness, meaning that the more consumers' experiences using the Douyin platform to watch micro-dramas meet their expectations, the more they feel the Douyin platform is useful for watching micro-dramas. Established H5: Perceived usefulness is positively correlated with continuous usage intention, meaning that the more useful consumers feel the Douyin platform is for watching micro-dramas, the higher their willingness to continuously use the Douyin platform to watch micro-dramas. Established Flow Theory H6: The flow experience is positively correlated with satisfaction, meaning that the stronger the flow experience consumers feel from the micro-dramas, the higher their satisfaction with watching micro-dramas on the Douyin platform. Established H7: The flow experience is positively correlated with continuous usage intention, meaning that the stronger the flow experience consumers feel from the micro- dramas, the higher their willingness to continuously use the Douyin platform to watch micro-dramas. Not Established 5. Conclusion and Prospects 5.1. Results and discussion This study is based on Bhattacherjee's (2001) Expectation Confirmation Theory (ECT), combined with Flow Theory, to construct a model. The data was obtained through an online questionnaire survey of users who had watched micro-dramas on the Douyin platform. After data analysis, six out of seven hypotheses were validated, leading to the following conclusions: 1) The micro-drama market has grown exponentially within just three to four years, reaching a scale approaching one trillion yuan. Such rapid development has brought various challenges to the industry, making it both necessary and urgent to analyze the factors influencing consumers' continuous purchase intentions, particularly for Douyin, the benchmark platform in this industry. 2) According to ECT, satisfaction, expectation confirmation, and perceived usefulness are key factors influencing consumers' continuous purchase intentions. The validation of Hypotheses H1-H5 demonstrates the accuracy of Bhattacherjee's (2001) theoretical model in the context of micro-dramas, confirming its applicability beyond the IT environment. Moreover, the positive relationship between expectation confirmation and satisfaction is also supported by the findings of Zhang Mengqi, Chen Caixia, and Li Min (2023) in their research on live-streaming sales of clothing. The positive relationship between perceived usefulness and satisfaction, as well as perceived usefulness and continuous purchase intentions, is similarly validated in studies by Zan Xiangyang (2015), Fan Yanhong (2021), Li Juanjuan (2020), and Wang Man et al. (2021), while the relationship between perceived usefulness and continuous purchase intentions was also confirmed in Shu Wei’s (2021) research on fresh e- commerce. Additionally, referring to Csikszentmihalyi's (1997) description of flow experience, it was anticipated that flow would positively affect both satisfaction and continuous purchase intentions. However, after conducting correlation and regression tests, the data analysis revealed that while flow does positively influence satisfaction (supporting Hypothesis H6), it does not positively influence continuous purchase intentions (disproving Hypothesis H7). This result aligns with Li Juanjuan’s (2020) study on users of knowledge payment platforms, as well as the research of Wu Zhengxiang et al. (2020) and Wang Xiaoyu, who found that flow experience only impacts satisfaction but not continuous purchase intentions. On the contrary, studies by Wang Man et al. (2021) and Li Yingjian (2023) indicated that flow experience does have a positive and significant effect on continuous purchase intentions. The discrepancy in these findings may be attributed to differences in research subjects, as audiences of various platforms may respond differently. To fully understand the underlying patterns, extensive comparative studies would be required. Recommendations for stakeholders include the following: First, for micro-drama creators: they should cater to the interests of young female audiences, particularly with themes like youth dramas and palace dramas, to enhance flow 125 experience. They should focus on originality and resonance with current events, select appropriate actors, and adjust their creative direction based on audience feedback to improve satisfaction and continuous purchase intentions.Second, for Douyin as a platform: the platform should optimize its payment models and intelligent recommendation system to enhance user satisfaction, implement strict content review processes to protect original creators, and provide creative tools to encourage high-quality content production. These efforts will expand the user base and improve user retention and continuous purchase intentions.Third, for the audience: viewers should develop critical viewing skills, resist undesirable content, provide feedback through interaction, and foster healthy entertainment and consumption habits. Additionally, they should improve their information security and rational consumption awareness to avoid excessive or unnecessary purchases.Lastly, for regulatory bodies: they should establish clear and comprehensive review standards to ensure fair and efficient review processes. Collaboration with industry organizations, creators, and platforms to formulate industry norms and strengthen compliance training is essential to maintaining a healthy ecosystem for micro- dramas. 5.2. Limitations of the Study and Recommendations for Future Research This study builds on previous research on consumers' sustained purchase intentions and analyzes the impact of micro-short dramas on consumers' sustained purchase intentions on Douyin platform. However, there are some limitations due to time, energy, and ability constraints. First, data collection is focused on the Douyin platform, and the user group characteristics may differ from those of other micro-short drama platforms, limiting the generalizability of the results. Future research should expand data sources to cover more platforms to enhance the representativeness of the results. Second, the study mainly uses quantitative questionnaire survey methods and does not involve qualitative research, such as interviews or focus groups, which can provide a more comprehensive perspective in the future by combining quantitative and qualitative methods. Finally, with the rapid changes in AIGC technology and government regulations, the focus and business model of Douyin micro-short drama industry may change rapidly, which may affect the timeliness of the research conclusions. Future research should pay attention to industry development trends and adjust research methods and theoretical frameworks. Future research can further explore several directions based on this study. First, there is a phenomenon of monthly income of 6,000 yuan or above but only a junior high school education in the research population, and it is suggested to conduct a special statistical research to determine the prevalence of this situation, which will be of great help in determining the direction of the industry. Second, whether the flow theory has a positive impact on users' sustained purchase intentions under specific conditions still needs further verification, which will make significant contributions to the theoretical research of the flow theory, the expectation confirmation theory, and sustained purchase intentions. Future research can also conduct qualitative research through user interviews and expert surveys to gain a deeper understanding of users' emotional and cognitive factors and conduct cross-platform research on user group differences to further reveal the behavioral characteristics of users on different platforms. References [1] Cui S. (2020). The Internet Empowers and Enhances Intelligence in the Fight Against the Pandemic. People's Daily. Retrieved from http://ip.people.com.cn/GB/n1/2020/0421/c136655- 31681890.html [2] Zhou, Y. (2020). "The 2020 Report on the Development of China's Internet Audio-visual Industry" was released, interpreting the current situation and trends of the industry. China Broadcasting and Television, 2020.11.01. Available from Cnki. [3] CNNIC. (2024). The 54th Statistical Report on Internet Development in China.Retrieved October 14, 2024, from https://www.cnnic.cn/n4/2024/0828/c208-11063.html [4] Yu, Z. H., & Dai, R. (2024). An insight into the development of Douyin micro-dramas: Balancing traffic operation and content quality. International Brand Observer, (10), 50–53. doi:CNKI:SUN:GJPC.0.2024-10-010. [5] Wu C., & Gao H. (2024). Innovative Research on the Profit Model of Micro-Short Dramas in China. Contemporary Television, February 2024. Available from Cnki. [6] Chen H.,Fan R.(2022), Profit Model and Optimization of IP Micro-Short Dramas in the Perspective of Digital Labor, Young Journalist, 2022.09.25 [7] Yu, X., & Liu, Y. (2022). The Creation Model of Emotional Micro-Short Dramas from the Perspective of Consumer Culture - Based on the Analysis of Accounts on Douyin Platform. News Tide, 2022.07.15. [8] Li Bin. (2012). Study on the Influence Factors of Continuous Use Intention of Weibo Users （ Master'sthesis ） https://kns.cnki.net/kcms2/article/abstract?v=JlSlmhnvVuqn5 ADNDZBCm2amJlUH6We8YsdNw0f_0mlRAqkBzw9Apy8 IViUD5DhhuXE9bzB3yghkqNwYsGyxu9NaqXKPgtnrwJld WGrm4Yb1W__WHuDB002_EJKfWY8R0gW_aqWL- 0zSuaO3N_6nLGi_SK5CLPUVu4wEbJyYXeSAOdEenBX- 3fpv8obz- VYfTUHwHtno1a4&uniplatform=NZKPT&language=CHS] [9] Zhang, Q. (2023). Research on influencing factors of consumers' online continuous purchase intention under e- commerce live streaming scenario. Shanxi University of Finance and Economics, June 19, 2023. [10] Fishbein M, Ajzen I. Belief, Attitude, Intention and Behaviour: an introduction to theory and research[J]. Philosophy & Rhetoric, 1975, 41(4):842-844 [11] Davis, F. D. (1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS quarterly, 319-340. [12] Bhattacherjee, A. (2001). Understanding information systems continuance: An expectation-confirmation model. MIS quarterly, 351-370. [13] Qian Ying. Research on the Influence Factors of Continuous Use Behavior of Online Learning Users Based on the Perspective of Social Network Environment and Learning Status Positioning [J]. Modern Intelligence, 2015, 35(3): 50-56. [14] Csikszentmihalyi, H. (1988). Introduction to Part IV in Optimal Experience. Psychological Studies in Consciousness, 23, 421. [15] Hoffman, D. L., & Novak, T. P. (2009). Flow online: Lessons learned and future prospects. Journal of Interactive Marketing, 23(1), 23–34. [16] Wang, M., Lu, M. X., Shi, J., & Yang, H. J. (2021). Research on the influencing factors of user stickiness on video websites based on ECT theory and flow theory: A case study of Bilibili. 126 Journal of Information Engineering, 7(01), 80-92. Retrieved from https://next.cnki.net/middle/abstract?v=QenloEQs_R8M Ypw0W- osBCnnrUf6j3lrPgPgKsnlwnrcvNtIDiwM3ICkCcNecc6UOa ZKINGvJ_K_n04FACdCEeKHKsmLJTcEwdKf1SdMkWZ7 1yZawSL1pg6HTUKqIDI69Nrap4UtdVjMJBSNey2qmhqm Y23AyFoFwFAwOn41Z7FsBmg6X2ee_M7amCHgKJnCum _syLLHZwI=&uniplatform=NZKPT&language=CHS&scenc e=null [17] Wang, X. (2023). Research on the propagation mechanism and optimization strategy of network mini-drama on Kuaishou Platform. Shandong Normal University, 2022.10.15. [18] Li, J. J. (2020). Research on the influencing factors and promotion strategies of users' continuous usage intention on knowledge payment platforms. (Master's thesis). Retrieved from https://link.cnki.net/doi/10.27351/d.cnki.gszhu.2020.002 957 [19] Wei, X. H. (2024). Research on the online marketing strategy of the micro-drama section of DY company. (Master's thesis). Retrieved from https://link.cnki.net/doi/10.27162/d.cnki.gjlin.2024.0056 66 [20] Bertram, D. (2007). Likert scales. Retrieved November, 2(10), 1-10. [21] Creswell, J. W. (2014). Research design: Qualitative, quantitative, and mixed methods approaches (4th ed.). SAGE Publications. [22] Yamane, T. (1967). Statistics: An introductory analysis (2nd ed.). Harper & Row. [23] George, D., & Mallery, P. (2018). Reliability analysis. In IBM SPSS statistics 25 step by step (pp. 249-260): Routledge. [24] Watkins, M. W. (2018). Exploratory factor analysis: A guide to best practice. Journal of black psychology, 44(3), 219-246. [25] Senthilnathan, S. (2019). Usefulness of correlation analysis. Available at SSRN 3416918.