NATIONAL ECONOMICS UNIVERSITY NEU BUSINESS SCHOOL LUU TRAN THUY ANH FACTORS AFFECT INTENTION TO USE EVN MOBILE APPLICATION (EPOINT) AMONG GENERATION Y AND GENERATION Z IN HANOI MASTER OF BUSINESS ADMI[.]
INTRODUCTION
Rationale
According to the statistics of We are social and Hootsuite, the number of mobile connections in Vietnam in 2021 already reached 154 million connections, accounting for more than 157% of the total population The number of Internet users also achieved nearly 69 million users, equivalent to more than 70% Vietnamese population The average time users use the Internet on mobile devices is 3 hours and 8 minutes a day, accounting for 25% of Vietnamese people's day (We are social; Hootsuite, 2021) On the other hand, when using the Internet on mobile devices, users have to access it through mobile applications, so mobile applications become an indispensable part of daily life (Mai & Cho, 2017).
For businesses, according to Appota's report, Vietnam's mobile e-commerce revenue will reach USD 5.6 billion in 2020 with an average growth of 18.6% per year It is expected that in 2021, mobile e-commerce revenue will reach USD 7 billion and by 2023 it is likely to reach USD 10.2 billion (Appota, 2021) This financing comes from mobile apps marketing, the process and analysis of collecting and mining user data, sell apps as well as optimizing shopping experiences On the other hand, the number of applications as of the first quarter of 2020 is 2.56 million and 1.847 million respectively on the two most popular platforms Google Play and Apple App Store, while the number of Vietnamese applications in the list of the most popular applications is still limited, with 1 application in the 25 most important applications of the decade all over the world,
3 of the 10 most downloaded applications in Vietnam (Appota, 2020) Thus, it can be seen that both the great potential and challenges in the current mobile application market for businesses and application developers in Vietnam.
Amid the COVID-19 pandemic, the world is experiencing one of the worst recessions since the second world war (World Bank, 2020) Traditional retailing has had to be reduced due to reasons such as social distance regulations from the governments and be replaced by online retailing (Guthrie, et al., 2021) Besides, COVID-19 has accelerated the transition of physical and traditional retail to online retail even after COVID-19 ends, this trend is expected to be continued to grow (Hassan, et al., 2020) With the increasingly developing technology along with the influence of the Covid-19 pandemic, consumers tend to use mobile devices such as smartphones and tablets for work and entertainment Therefore, developing applications on mobile phones in particular and mobile device support services in general to create an additional channel to reach customers becomes even more important for retailers Understanding consumers' mobile app usage behaviors and the intersection between brick-and-mortar stores and mobile apps plays an important part of developing omnichannel retail as well as omnichannel marketing (Heerde, et al., 2019) In the world as well as in Vietnam, there have been many studies on the factors affecting the behavior of customers using mobile applications, especially the intention to use mobile applications to shop, order food or make online payment Previous studies often focused on the satisfaction or service quality aspects to explain the intention to use e-commerce mobile apps for shopping (Zhang et al., 2012; Polančič et al., 2015; Sanakulov & Karjaluoto, 2015) In the context of services such as e-commerce, perceived value is thought to play a more important role than satisfaction or service quality in explaining consumer behavior (Petrick, 2002; Zeithaml, 1988) Accordingly, studies on intention to use e-commerce mobile apps focus on overall value (Cheah et al., 2015), utility value and recreational value (Kim & Han, 2011), and social value addition (Khoi et al., 2018; Kim et al., 2013; Kim & Han, 2009) Based on the theory of extended technology acceptance model, factors including expected performance, social influences and habits (Lee, et al.,
2019), benefit dynamics (AbdallahAlalwan, 2020) have significant influence on intention to continue using the e-commerce mobile services El-Masri and Tarhini
(2017) showed that trust also affects intention to continue using new services and technologies such as mobile app technology (El-Masri & Tarhini, 2017) For mobile payments, the studies mainly set standards in the mobile payment industry, focus on developing the number of users, and analyze service intentions or trends (Chen, 2008; Abrahão et al., 2016) However, in Vietnam, there is a limited number of studies that specifically evaluate the behavior of using mobile applications to share information, make payments, as well as manage spending and product usage The majority of studies and research on mobile app behaviors of customers in Vietnam tend to focus on single aspect of mobile activities such as shopping, payment and sharing Hanh and partners (2019) investigated the factors impacting on intention to use mobile apps for shopping in Ho Chi Minh City (Hanh, et al., 2019), Quynh and Anh (2021) and Vuong (2021) both studied the factors that affect intention to use mobile payment and mobile payment behaviors of young people in Vietnam (Quynh&Anh, 2021; Vuong, 2021), Trang and Nam (2021) explored and analyzed factors determining the intention to use mobile apps to order food in Ho Chi Minh City (Trang & Nam, 2021) It is obvious that there is no previous research and studies which have been carried on factors determining the intention to adopt a comprehensive mobile app to make both payment and manage shopping and consuming activities or update information such as mobile app by Vietnam electricity corporation (EVN) Meanwhile, EVN is currently the largest and exclusive electricity supplier in Vietnam, having released the EVN mobile app since
2020 to provide customers with comprehensive customer care solutions The application supports customers to update timely information about electricity bills, view electricity bills, schedule power cuts, pay electricity bills online, receive electronic bills, register for electricity services etc In addition, the mobile application also helps customers send help requests to EVN in just a few seconds on the phone completely free of charge.
This application is available on iOS and Android operating systems, as of the end of May 2021, there are 1,056,141 customers in the South and Central regions(EVN, 2021) The Epoint apps plays essentially important role in the business strategy of EVN, especially customer care which is considered as core business’s focus, enable EVN to enhance EVN's position and image in the community With a wide range of customers throughout the country, EVN has always promoted and constantly developed this care work on multiple platforms to bring the best experiences to customers, ensuring that each customer can access and look up information conveniently and receive quick and accurate answers and Epoint is one of the solution to enhance customer service’s performance of the company EPoint has many outstanding advantages, when it can show the temporarily calculated amount for customers, compare electricity consumption with the same period of the previous year, so that customers can assess the current consumption level and make adjustments in their life In addition, with the online payment and automatic debiting feature integrated in the application, customers of EVN can conveniently and quickly pay their monthly electricity bills by bank card or credit card. electronic wallet.Besides, not only is it an electricity monitoring application, with the Epoint application, customers can also participate in accumulating points and focus at the system of EVN's partners That means, Epoint will bring customers more attractive incentives and promotions from hundreds of brands in many fields such as dining, shopping, entertainment, beauty, learning Therefore, it can be said that the adoption of Epoint among customers of EVN plays important roles in the company’s business strategy.
However, in the Northern region, the number of customers using the Epoint application to pay and manage electricity consumption is still limited, specifically this number is much less than the number of customers who have used the Epoint mobile application in the South and Central regions of Vietnam (EVN, 2021).
Therefore, this study aims to explore and analyze the factors affecting the intention of using mobile applications to update information, make payments as well as manage consumers' electricity consumption in the Northern region inVietnam, thereby make contribution to help EVN decide to invest in using mobile applications in business to catch up with the trend of electricity consumption, and at the same time contribute to the development of the electricity retailing industry inVietnam Especially, the study aims to focus on potential customers who belong to generation Y (who was born from 1980-1994) and generation Z (who was born from 1995 to 2012) working and living in Hanoi and have not used Epoint app yet.Because according to the report by Nielsen Vietnam, by 2025, gen Y and gen Z are predicted to be the majority of labor workforce and consumers in Vietnam,accounting for more than 45% of total population (Nielsen Vietnam, 2018).Moreover, potential customers from gen Y and gen Z are considered as tech savvy and active social media users with the tendency to use mobile phone for every activity in their daily life (Nielsen Vietnam, 2018) Therefore, it can be seen clearly that gen Y and gen Z are potential customers for Epoint mobile app to target.
Research objectives
Based on the research rationale above, the study aims to achieve the following objectives:
1 Analyze theories and frameworks on mobile adoption behaviors including theory of reasoned action (TRA), theory of planned behavior (TPB), technology acceptance model (TAM), unified theory of acceptance and use of technology (UTAUT) and theory of perceived risk (TPR).
2 Determine how 7 elements including flexibility, perceived usefulness, perceived ease of use, trust, social influence, hedonic motivation and perceived risk that make influence on intention to adopt EVN mobile application in managing electricity consumption, making payment for electricity bills and updating information from EVN among customers living in the North parts of Vietnam, especially Hanoi
3 Propose some recommendations for EVN to better encourage customers inHanoi to adopt mobile application in managing their electricity consumption and enhance customer service performance.
Research questions
Based on objectives of the study, the study focuses on finding answers for following research questions:
What theories and frameworks explain the user’s adoption of mobile apps?
What are factors that impact on intention to adopt EVN mobile apps among customers in Hanoi?
What are recommendations for EVN to enhance customer’s adoption of mobile app in Hanoi?
Research methodology
Both qualitative and quantitative methods are adopted in this study Qualitative research with secondary data from previous research, reports and academic papers enables to dig deeper into theoretical frameworks related to the study and from that to build the research model Quantitative method with primary data from the survey questionnaire helps to use numerical analysis to confirm research hypotheses on correlations between variables in the research model The survey questionnaire is distributed through online channels with the sample size of 200 participants who are belonging to generation Z, living and working in Hanoi After collection, data is put into SPSS version 26 to process and analyze the findings
1.4.3 Sample size and sampling method
According to Hair et al (2014), the minimum sample size to use EFA is 50, preferably 100 or more The ratio of observations to an analytic variable is 5:1 or 10:1, some researchers suggested that this ratio should be 20:1 (Hair et al., 2014). Number of observations simply means the required number of measured variables in the research model, which are corresponding with questions in the survey For the minimum sample size for regression analysis, Green (1991) gave two cases In the first case, if the purpose of the regression is only to evaluate the general fit of the model such as R_square and F test then the minimum sample size is 50 + 8m (m is the number of independent variables or remaining variables) which are also called the predictors participating in the regression model (Green, 1991) In second case, if the purpose wants to evaluate the factors of each independent variable such as t-test, regression coefficient, p-values then the minimum sample size should be 104 + m (m is the number of independent variables) Green (1991) also specified that m is the number of independent variables we include in the regression analysis, not the number of observations or the number of questions in the survey Harris (1985) suggested that the appropriate sample size to run a multivariate regression should be equal to the number of independent variables plus at least 50 For example, in a regression involving four independent variables, the minimum sample size should be
4 + 50 = 54 Hair et al (2014) suggested that the minimum sample size should be in the ratio 5:1, which means that at least 5 observations for an independent variable.
In the research model of this study, there are total 7 independent variables and
31 observations or measured scales Which means that to conduct EFA analysis, the total sample size needs to be 5*31 = 155 (Hair et al., 2014) For regression, the total sample size should be 50+ 7*8 = 106 (Green, 1991).
Therefore, to ensure enough sample size after removing invalid answers from the survey, the author chooses to take the sample size of 200 participants from generation Z in Hanoi, which satisfies both EFA and regression analysis for this research.
Due to the limitations of time and cost, the sample for the survey questionnaire is chosen based on convenient randomly sampling method
The survey questionnaire is designed based on surveys by previous research and studies on the same topics including research by Venkatesh et al (2012), Kim et al (2016), Linh (2018) and Hung et al (2019) After that, the survey is going to be distributed online through some channels such as Facebook, email, Zalo and Messenger The data then is collected and stored in Google Form, one of the most popular and free online survey tools in Vietnam.
After data collection through survey, the data is going to be classified and removed all invalid answers The data is encoded before being input into SPSS version 26.0 to process.
The process of analysis goes through some steps including frequencies analysis, descriptive statistics analysis, reliability analysis through Cronbach’sAlpha, EFA analysis, regression analysis and one-way ANOVA analysis.
Scope of the research
The study is framed with participants belonging to generation Y and generation Z who are currently living and working in Hanoi These two generations are expected to account for the largest labor workforce as well as main consumers in Vietnam in coming years Customers from these two generations are considered as tech savvy and have tendency to be more familiar with digital transformation in the new era in Vietnam The survey questionnaire to collect data for the study is conducted from 10 June to 30 June 2022.
Thesis structure
To reach research objectives and find out answers for research questions, the study is organized in 4 main chapters as below:
Chapter 1: Research introduction: this chapter presents introduction on rationale of the study, research objectives and question, research methodology, scope of the research and thesis structure.
Chapter 2: Literature review: this chapter analyzes findings and results from previous research on the same topic as well as theoretical framework to build research model and hypotheses for the study.
Chapter 3: Data collection and findings: this chapter introduces in detail the process of choosing sample for the research including sample size and sampling method It also describes the process of data collection and data analysis Besides, this chapter presents the analysis and discussion on research findings.
Chapter 4: Recommendations and conclusion: this chapter provide some suggestions and recommendations for EVN to attract generation Z customers to install and use Epoint app Besides, it also summarizes key points and findings of the study.
LITERATURE REVIEW
Customer’s intention to use the service
Ajzen (1991) stated that behavioral intention is the subjective ability of a person to intend to achieve within a certain time [ CITATION Ice915 \l 1033 ]. According to Tirtiroglu and Elbeck (2008), intention to use is to describe a customer's willingness to use a certain product [ CITATION Erc08 \l 1033 ] Zhao and Othman (2010) defined intention as a course of action that an individual wants to achieve [ CITATION Wen10 \l 1033 ] Ajzen et al (1975) suggested that behavioral intention is a measure of an individual's intention to perform a particular behavior or behavioral intention is positive or negative feelings towards performing a goal behavior [ CITATION Ajz75 \l 1033 ] Intention is the tendency an individual displays to indicate whether they use a new technology or not A person will perform the behavior if they have that intention (Latupeirissa et al.,
2020) Many researchers try to come up with a general concept and measure behavioral intention such as the theory of reasoned action (TRA) which points out two factors including attitude and subjective norm to predict consumer behavior[ CITATION Mar752 \l 1033 ] Davis (1985) proposed a technology acceptance model (TAM) about the relationship and influence of cognitive factors such as ease of use and usefulness on attitude, thereby affecting intention and behavior inInformation technology acceptance of users, intention is considered as a direct premise leading to behavior using technology in the TAM model [ CITATIONFre851 \l 1033 ] Research by Zhang et al (2012) also confirmed that intention to use is a very important concept in the study of consumer behavior and is also the most important factor determining actual consumption behavior [ CITATIONLiy121 \l 1033 ]
Mobile application
Mobile application is defined as software program developed for mobile devices such as smartphones and tablets that allow users to access the content they want right on their mobile device [ CITATION DrM10 \l 1033 ] Most of the mobile devices sold have built-in many mobile applications such as web browser, email, calendar etc However, there are many applications that are not pre-installed that users need to use, when applications are downloaded via distribution platforms, also known as "app store or app market" App store started appearing in 2008 with prominent vendors who provide compatible apps on certain operating system platforms such as Apple App Store for IOS operating system, Google Play for Android and Windows operating system
There are three types of mobile apps including native apps, web apps, and hybrid apps [ CITATION PCl21 \l 1033 ] Native apps are developed for an operating system including iOS and Android, are installed from an app store such as Google Play or Apple's App Store and accessed via the icon on the device’s screen. Native apps have the ability to access device features such as camera, audio recording, contact information, etc Web apps are websites with an interface that looks like native apps, but it is controlled by the browser and usually written in HTML5, operating system-independent and accessed by entering a specific URL into the browser without installation on the device Hybrid apps combine the features of native apps and web apps Like native apps, hybrid apps live in an app store and can take advantage of the device's available features Also, hybrid apps are similar to web apps in that they rely on HTML being displayed in a browser with the message that the browser is embedded in the application Hybrid apps allow for cross-platform development and thus significantly reduce development costs: that is, the same HTML code elements can be reused on different mobile operating systems (Budiu, 2013; Lionbridge, 2012; Skidmore, 2013) In this study mobile apps are defined as native apps or hybrid apps, which are software or programs installed on mobile devices that users can perform certain tasks on.
Theories and frameworks on behaviors to adopt mobile technology
2.3.1 Theory of Reasoned Action (TRA)
The theory of Reasoned Action (TRA), which holds that intentions will determine a person's actual behavior, in which that person's attitudes and subjective norms will affect influence their behavioral tendencies (Fishbein & Ajzen, 1975). Theory of reasoned action (TRA) was developed by Ajzen and Fishbein in 1967 and adjusted and expanded over time The TRA model shows that consumption trends are the best predictors of consumption behavior To pay more attention to the factors that contribute to the buying tendency, consider two factors, which are the attitudes and subjective standards of customers In the TRA model, attitudes are measured by the perception of product attributes Consumers will notice attributes that provide essential benefits and have varying degrees of importance If the weights of those attributes are known, it is possible to predict the outcome of the consumer's choice roughly.
Figure 2:1: TRA model by Fishbein & Ajzen (1975)
Subjective normative factors can be measured through people related to the consumer (such as family, friends, colleagues, etc.); these people like or dislike they buy The degree of impact of subjective standard factors on consumer buying propensity depends on: (1) the degree of support or opposition to the purchase of the consumer and (2) the motivation of the consumer follow the wishes of the influencers The degree of influence of related people on consumer behavioral trends and the motivation of consumers to follow related people are two basic factors for assessing subjective standards The stronger the degree of intimacy of related people with consumers, the greater the influence on their purchasing decisions The greater the consumer's trust in the relevant people, the more their purchasing propensity will be affected The purchase intention of consumers will be affected by these people with varying degrees of strong and weak influence In the rational action theory model, individual consumers' beliefs about a product or brand will influence their attitudes towards behavior, and attitudes towards behavior will affect their propensity to buy Therefore, the attitude will explain the reasons leading to the consumer's buying tendency, and the trend is the best factor to explain the consumer's behavioral trend.
2.3.2 Theory of planned behavior (TPB)
The theory of planned behavior (TPB), this theory was developed by IcekAjzen (1991) from the TRA model when adding the perceived behavioral control factor This factor explains the relationship between a person's beliefs and behavior [ CITATION Ice915 \l 1033 ] Basically, TPB is still used to study and predict human behavior based on the relationship between attitude, subjective norm, intention to perform planned behavior and actual behavior The development of TPB in developing perceived behavioral control factors affecting behavioral intentions, in addition to attitudes and subjective norms.
Figure 2:2: TPB model by Ajzen (1991)
Thus, in TPB, there will be 3 factors affecting intention including (1) attitude towards behavior, (2) subjective norm and (3) perception of behavioral control.
If the subjective norm is an individual's perception of social influence on an individual's behavior, perceived behavioral control is the perceived ability of the individual to perform the behavior Perceived behavioral control is defined as an individual's perception of how easy or difficult a behavior is (Ajzen,
1991) Perceived behavioral control denotes the degree of control over the conduct of the behavior rather than the outcome of the behavior (Ajzen, 2002).
In the TPB model, perceived behavioral control affects both intention and actual behavior The dual effect of perceived behavioral control is supported by perceived behavioral control Lin (1995) and Lin (2007), the authors argued that research on consumer behavior would be incomplete if perceived behavioral control was ignored (Lin 1995; Lin 2007).
The technology acceptance model (TAM), which is developed from the TRA model and agrees that behavior is indeed controlled by the intention to perform the behavior, however, the intention to perform is now influenced by attitude, and perceived usefulness and perceived ease of use are two determinants of a person's attitude (Davis, 1989) Legris et al (2003) described the main purpose of TAM as to provide a basis for identifying the drivers of extrinsic change on intrinsic beliefs, attitudes, and intentions [ CITATION Pau03 \l 1033 ] TAM is based on the TRA described by Fishbien & Ajzen (1975) and the TPB proposed by Ajzen (1991).
To apply the TAM model, it is necessary to consider whether the components considered are suitable for the considered technological characteristics The original TAM model proposed by Davis focused on two factors including perceived usefulness and perceived ease of use According to Davis, perceived usefulness is the degree to which a person believes that using a particular system will improve his or her performance He identified 14 focused elements in 3 groups: work efficiency, productivity and time saving, the importance of the system to one's work Perceived ease of use is attributed to Davis as the degree to which people believe that using the system is not a waste of their time There are 3 groups of factors in the perception of the ease of use feature such as physical effort, mental effort and expectation of personal experience can easily use the system [ CITATION Fre891 \l
1033 ] According to Kaasinen (2005), a specific and important factor affecting the acceptance of mobile phone services is trust [ CITATION Eij05 \l 1033 ] In addition, Keat & Mohan (2004) proposed an additional component describing the trust for the TAM model Trust is a combination of familiarity, company reputation, factual signals, and quality experience [CITATION TKK04 \l 1033 ] Kaasinen
(2005) also incorporated specific components of TAM for mobile services in a new version of TAM specifically for mobile services The author modified the value components (perceived usefulness word) and adds trust components and adaptive feelings Furthermore, another factor “taking to use” is also added before actual usage behavior [ CITATION Eij05 \l 1033 ].
2.3.4 Unified theory of acceptance and use of technology (UTAUT)
The unified theory of acceptance and use of technology (UTAUT) was proposed by Venkatesh et al (2003), the authors said that it is more optimal for the model This is a synthetic model based on previous foundational theories and models, the most important of which are the theory of rational action TRA, the intended behavior TPB and the TAM model The theory holds that four concepts including performance expectancy, effort expectancy, social influence and facilitating conditions are the direct determinants of intention and usage behavior Gender, age, experience and voluntary indirectly influence intention and behavior through the above four concepts This is essentially a theory synthesized based on some previous models and theories such as TRA, TAM,TPB [ CITATION Vis031 \l 1033 ] In fact, the UTAUT theory explains up to70% of the differences in intended use.
Figure 2:4: UTAUT by Venkatesh et al (2003)
Venkatesh et al (2012) continued to expand UTAUT to UTAUT2 by adding 3 factors namely price value, habit and hedonic motivation Venkatesh et al (2012) evaluated that the UTAUT2 model was better than the UTAUT, because the explanatory percentage of the independent variables was higher when considering the impact on intention to use technology in UTAUT2, which compared UTAUT, the expansion in UTAUT2 explains the behavioral intention variables (up to 74% from 56%) and with technology use (up to 52% from 40%) [ CITATION Vis12 \l
Figure 2:5: UTAUT 2 by Venkatesh et al (2012)
UTAUT2 is proposed as a useful model for understanding technology usage by general consumers In this model, the individual is characterized by letting habits influence behavioral intentions Venkatesh, Thong, and Xu (2012) have seen habit as a perceptual construct that reflects the results of previous experiences UTAUT2 models how habits directly and indirectly influence behavior through behavioral intentions The proposed additions to UTAUT2, according to Venkatesh et al.
(2012), reflect major revisions to the factors that explain behavioral intentions and technology use [ CITATION Vis12 \l 1033 ].
2.3.5 Theory of perceived risk (TPR)
The theory of perceived risk (TPR) was developed by Bauer (1960) This theory suggests that a person's behavior is influenced by perceived risks associated with online transactions and perceived risks associated with products or services.
The cognitive component related to online transactions includes the risks that may occur when consumers make transactions on electronic means such as: confidentiality, safety and overall risk perception about transactions The perceived risk component associated with a product or service represents a customer's concern for such things as loss of features, loss of finance, loss of time, loss of opportunity when using the product or service information technology (Bauer, 1960).
The perceived risk theory is widely used in online purchase decision studies, in which two case studies show that this theory is also used in research on ticket purchase decisions, booking tickets (events, trains, planes, hotel reservations) online in general, and buy airline tickets online in particular as studied by Kim, Kim &Shin (2009); Kim, Kim & Leung (2005).
Theoretical framework
In the world, there have been many studies investigating factors related to users' acceptance of different mobile application services, such as e-banking (Chen,
2013), mobile payment (Lu et al., 2011; Zhou, 2013), mobile apps for financial services [ CITATION Hel13 \l 1033 ], mobile app for health service [ CITATION Zha14 \l 1033 ], mobile apps data service (Al-Debei & Al-Lozi, 2014), mobile apps for game (Jiang et al 2015) and mobile apps for learning (Callum et al., 2014) etc. However, there have not been many studies on the factors affecting the intention to install and use of mobile applications in general, specifically, there are some studies below:
The study of Venkatesh et al (2003) is based on the study of 8 models, the theory of TRA, TAM, MM, TPB, C-TAM-TPB, MPCU, IDT and SCT on the factors affecting the intention to use information technology of users, the group of authors synthesizes to form a unified theory called unified theory of acceptance and use of technology (UTAUT) The model of this theory identifies four main independent variables including performance expectancy, effort expectancy, social influence, facilitating conditions, which affect intention and behavior to use technology The survey data is divided into 2 phases Phrase 1 includes group interviews with product development managers, salespeople in the entertainment and telecommunications sectors, investment analysts, accountants Phase 2 includes the sample sample of participants which consist of 645 people, with a total collection time of 6 months Venkatesh and colleagues used PLS model, R-square index, coefficient value to evaluate the corresponding model in 8 theories and the degree of influence of the independent variables on the adoption of technology. From there, the authors synthesized strong influencing factors into the UTAUT model, focusing on 4 factors: performance expectancy, effort expectancy, social influence, facilitating conditions The new theory continues to be tested, the results show that all 4 factors affect the intention and behavior to use technology [ CITATION Vis031 \l 1033 ].
Taylor and partners (2011) used network theory in another study to investigate the use of mobile applications among 143 students in the US The study focuses on evaluating two factors including social network and close similarity to the user's installation of the mobile application Research results showed that social network plays an important role in the decision to download and use a mobile application, especially when the people who have a great influence on users also install the application [ CITATION Dav112 \l 1033 ].
In 2012, Venkatesh and colleagues continued to expand their research into unified theory of acceptance and use of technology 2 (UTAUT 2) by adding 3 independent variables to the model in the UTAUT: hedonic motivation, price value and habit or experience Data is surveyed in 2 phases during 4 months with the participation of 1512 Internet users The research results showed that the independent variables all have an impact on the dependent variable which is behavioral intention, especially the UTAUT2 model when adding 3 independent variables gives better results than the UTAUT model when explain up to 74% behavioral intention (compared to 56% in the old model) and 52% for usage behavior (compared to 40% in previous model) [ CITATION Vis12 \l 1033 ].
Yang's study (2013) conducted a survey with 555 students in the US in 2011. The study used a combination of TPB, TAM and UGT to investigate the mobile application behavioral intention of young people in the US The results of the study showed that usefulness, perceived enjoyment, ease of use and subjective norm are prominent factors predicting attitude towards mobile application usage of the survey group For the factors age, gender, income, education level, the author has shown that there is no significant impact on mobile app behavioral intention [ CITATION Hon131 \l 1033 ].
Research by Hew et al (2015) in Korea, surveying 288 users, using UTAUT2 model The factors used by the author are performance expectancy, effort expectancy, price value, facilitating conditions, habit, social influence, hedonic motivation, gender and education level The authors concluded that all variables except price value and social influence are significantly related to behavioral intention using mobile apps Habit had the strongest influence and gender and education level had no significant influence [ CITATION Jun15 \l 1033 ] Another study by Shen (2015) in Taiwan, with a sample of 476 participants, focused on examining the influence of factors on product type and message framing's moderating to the user's use of the mobile application Hypotheses are built based on TAM, ST and RFT The group of factors includes: source of reputation (App rating scores), level of usefulness (perceived usefulness), type of application (utilitarian app or hedonic app), attitude, mood, regulatory focus framing and perceived risk The results showed that app type and risk perception, source of reputation have an influence on users' attitude towards using the app (She, 2015). Research by Slade et al (2015) in the UK, the study uses a combination of models in UTAUT with 3 extended factors: innovativeness, risk and trust to examine influence on the intention of remote mobile payments SEM analysis was applied with a survey of 268 people who had never purchased products using remote mobile payments (RMP) and those who did not use RMP Initially, the authors proposed 7 hypotheses, the research results supported 5 hypotheses Performance expectancy, social influence, innovativeness significantly positively influence on behavioral intention while perceived risk negatively affects behavioral intention and trust The results also showed that effort expectancy has a negligible influence on behavioral intention and the effect of trust on behavioral intention is not clear [ CITATION Emm15 \l 1033 ] Research by Kim et al (2016) also used TAM to create a model of factors affecting application usage The variables used in this model include perceived informative usefulness, perceived entertaining usefulness, perceived social usefulness, perceived ease of use, attitude toward app usage, user review, and cost-effectiveness The study concluded that perceived information usefulness, perceived usefulness and ease of use all have a significant influence on attitude towards app usage, thus having a significant influence on intention to use mobile application User reviews also significantly influence app usage, but cost effectiveness has no effect on app usage [ CITATION San16 \l 1033 ] Another study by Kim & Koo (2016) in Korea, studied the influence of trust and perceived risk on consumers' participation in online transactions The survey sample was 747 people The authors simultaneously build two models, a unidirectional model and a bidirectional model, to determine the relationship between trust and risk perception. The results of the study showed that trust has a direct and significant effect on transaction intention, while risk perception has a negligible effect [ CITATION Gim16 \l 1033 ].
Research by Thao (2015) used the UTAUT model and regression analysis method to examine the factors affecting the use of mobile banking services atSaigon Thuong Tin Commercial Joint Stock Bank The selected sample is 250 customers at Sacombank The factors used include: effort expectancy, perceived ease of use, social influence, facilitating conditions, perceived trustworthiness and perceived financial costs The results of the study showed that the factors that have a positive influence on the acceptance of using mobile banking include effort expectancy, social influence, facilitation conditions and perceived trustworthiness. The remaining two factors have negligible influence [ CITATION Hoa15 \l 1033 ]. Research by Trung and Ha (2017) was conducted with an observed sample size of
257 consumers in Ho Chi Minh City The authors built a model with many mediating factors such as trust, improvisation, correctness assessment, instant perception, visual attractiveness and direct antecedents such as available products,website’s ease of use which are two antecedents that influence visual attraction The results of the study showed that all 4 factors assessment of correctness, instant perception, visual attractiveness, trust and improvisation have significant and positive effects on the intention to buy impulsively [ CITATION Pha171 \l 1033 ].Another study by Linh (2018) was conducted on the behavior of using mobile applications to share information of customers: A case study of mother and baby products The author built a research model based on the theory of TAM andUTAUT2 with the 6 independent variables including use of social networks, social influence, perception of information, perception of personality personification,habit, motivation for amusement; and frequency of use The study uses the regression analysis method with 349 observations with 6 hypotheses about the positive impact of all 6 factors on the behavioral intention of using mobile applications to share information of customers The results of the study support all 6 hypotheses the author has put forward (Linh, 2018) Research on “factors affecting intention to use mobile applications for online shopping in Ho Chi Minh City” byHung et al (2019) The authors used the SEM linear structural model with the number of observations of 315 sample elements made in 2019 The results of the study pointed out that flexibility, habit, trust, motivation and habit have the strongest positive impact on intention to use mobile applications for shopping while perceived risk has a negative impact on intention to use mobile apps for shopping; perceived ease of use and perceived usefulness had no impact on mobile app behavioral intention to shop [ CITATION Duo19 \l 1033 ].
From previous empirical research and studies related to the acceptance of mobile application and intention to adopt mobile application, it can be seen clearly that UTAUT2 and TAM are most popularly used to build the research model Based on UTAUT2 and TAM, Venkatesh et al (2012), Kim et al (2016), Linh (2018) and Hung et al (2019), flexibility, perceived usefulness, perceived ease of use, trust, social influence, hedonic motivation and perceived risk are key elements that have influence on the intention to use mobile application Therefore, this study mainly focuses on these 7 factors to examine how they impact on generation Z’s intention to adopt Epoint apps
Flexibility is the ability of consumers to use mobile commerce services without being limited in any space and time (Chiu, et al., 2012) As technologies on mobile devices increasingly meet the needs of flexible mobility in consumers' lives, independent use of time and space becomes more and more important for both consumers and service providers (Kalinić & Marinković, 2016) This factor has been proven by many studies to affect the intention to use services such as mobile education (Gunawardana & Ekanayaka, 2009), mobile services (Park, et al., 2013) and payment mobile computing (Thakur & Srivastava, 2014) Hence, the study comes to the below hypothesis:
H1: Flexibility has a positive impact on the intention to use EVN mobile application (EPoint)
The degree to which a person believes that using a particular system will enhance his or her focus on work performance is perceived usefulness (Davis, et al.,
1989) Many empirical studies have demonstrated a direct relationship between perceived usefulness and intention to use mobile phones in the field of e-commerce (Dai & Palvi, 2009; Lu, 1024) Research conducted by Yang et al (2012) shows a positive relationship between perceived usefulness and intention to use mobile phones for shopping purposes (Yang, et al., 2012) In the study of factors affecting purchase intention through mobile applications in Bangkok, a direct relationship of perceived usefulness and intention to use was also found (Punjakunaporn & Techakittiroj, 2015) Therefore, the below hypothesis is proposed for this study:
H2: Perceived usefulness has a positive impact on the intention to use EVN mobile application (EPoint)
Perceived ease of use is defined as the degree to which an individual considers it easy and effortless to use a particular device (Davis, 1989) The relationship between perceived ease of use and intention to use mobile commerce has been studied by many authors This relationship is proven significant in most cases (Wei et al., 2009; Agrebi & Jallais, 2015) Perceived ease of use also affects various technology systems such as mobile commerce services Some limitations of mobile devices, such as small screens and difficult input, can lead to unsatisfied and unwilling consumers to use mobile commerce services, especially those who lack experience (Kalinić & Marinković, 2016) So perceived ease of use is a very important factor for mobile commerce services, regardless of whether the customer is a savvy user of technology or not Therefore, the study proposed following hypothesis:
H3: Perceived ease of use has a positive impact on the intention to use EVN mobile application (EPoint)
Trust is a feeling of certainty about something Customer trust here is a sense of certainty about what e-commerce companies have committed to (McKnight &
Chervany, 2001) Trust is considered an important factor affecting customers' willingness to accept e-commerce services for banks (Akhlaq & Ahmed, 2013). Along with this, Ahmad & Al-zu'bi (2016) examined the effect of trust on the intention to accept the use of e-commerce services at the Bank of Jordan They concluded that this relationship is significant (Ahmad & Al-zu'bi, 2016) Trust is also important in the context of online shopping according to many other studies (Grewal et al., 2004) As a result, customer trust in mobile shopping applications, is believed to play an important role in consumers' e-shopping behavior Therefore, the study comes to the following hypothesis:
H4: Perceived informativeness has a positive impact on the intention to use EVN mobile application (EPoint)
Social influence is the behavior of one person that becomes the guide and direction for the behavior of others (Chong, et al., 2012) In this study, the social influence factor is the level of influence of influential people such as family, friends, colleagues etc who think customers should use mobile apps to shop Social influence as a direct determinant of behavioral intention is shown as subjective norm in TRA theory and TA model; TAM 2; Social factors in MPCU model and IDT model The role of social influence in technology adoption decisions is complex and depends on a range of contingent influences (Venkatesh, et al., 2003). Hence, the following hypothesis is proposed:
H5: Social influence has a positive impact on the intention to use EVN mobile application (EPoint)
Hedonic motivation is the willingness to initiate behaviors that can enhance positive experiences such as comfort and well-being and behaviors that help reduce negative experiences (Bujacz, et al., 2014) Hedonic motivation includes comfort,pleasure, and happiness from using services (Venkatesh, et al., 2003) In the study of information systems, hedonic motivation has a direct influence on the adoption and use of technology (Tarhini et al., 2016) In the consumer context, consumer motivation has also been found to be an important determinant of technology adoption and use Therefore, the study comes to the following hypothesis:
H6: Hedonic motivation has a positive impact on the intention to use EVN mobile application (EPoint)
Perceived risk is the consumer's perception of uncertainty and adverse consequences associated with the purchase of a product or service (Kleijnen, et al.,
RESEARCH FINDINGS AND DISCUSSION
Introduction on EVN
3.1.1 The development history of EVN
Vietnam Electricity Corporation (EVN) was established in 1994 under the decision of the Prime Minister of Vietnam EVN is a state-owned one-member limited liability company, performing the task of supplying electricity for the country's socio-economic development needs EVN currently has 3 power generation corporations, 5 power corporations dealing in electricity and 1 company in charge of electricity transmission EVN affirms its key role in the strategy of developing the electricity industry, ensuring national energy security and quality of electricity supply, meeting the requirements of socio-economic development and the people's daily needs.With the dynamism and creativity in labor, production and business, for many years, EVN has completed the plan assigned by the State, meeting the country's demand for electricity The electricity output of goods supplied for socio-economic development has grown continuously at a high rate, increasing from 85.4 billion kWh in 2010 to 209.77 billion kWh in 2019 The average commercial electricity per person by the end of 2019 is estimated at 2,174.2 kWh/person/year, an increase of 2.21 times compared to 2010 (982.7kWh/person/year) [ CITATION VOV20 \l 1033 ].
According to Decree No 26/2018/ND-CP of the Prime Minister on the Charter of organization and operation of EVN, the main products and services of EVN include:
- Produce, transmit, distribute and trade in electricity; command and operate the system of production, transmission, distribution and distribution of electricity in the national electricity system;
- Import and export of electricity;
- Invest and manage investment capital in power projects;
- Manage, operate, repair, maintain, overhaul, renovate and upgrade electrical, mechanical, control and automation equipment in the production line, transmit and distribute electricity, electrical works and electrical testing;
- Conduct project management consultancy, design survey consultancy, manage investment project formulation consultancy, bidding consultancy, cost estimation, verification and construction supervision consultancy of power source works, road works wires and substations.
EVN is an energy industry group, but it can also be considered a technology corporation in the 4.0 industrial revolution The projects and digital products that EVN has been researching and developing as well as applying new and digital technologies EVN was the first unit to apply e-invoices nationwide in 2015 Then, in 2016, EVN accelerated the process of converting electricity bill collection at home to collecting money through collection points, payment intermediaries, e- wallets and banks By the end of November 2020, the percentage of customers paying electricity bills without cash reached 70.25%; the rate of electricity bills for non-cash payment reached 91.54% At the end of 2018, EVN also officially announced the provision of level 4 "online electricity services" - the highest level in online public service provision according to the Government's regulations, increasing publicity and transparency as well as create every convenience for customers Also in 2018, EVN completed the connection to provide electricity services at the Public Administration Center and online public service portals in 63/63 provinces/cities throughout the country.
According to Decision No 857/QD-TTG on June 6, 2011 of the Prime Minister approving the Charter of organization and operation of the EVN, the organizational structure of management and administration of EVN includes:
- Vice President and Chief Accountant
- Supporting operations including development strategy and general affairs
- Internal audit and financial supervision
The organizational structure of management and administration of EVN can be changed to suit the requirements of production and business, and the provisions of the law.
Below is the organizational structure of EVN:
3.1.4 The characteristics of human resource at EVN
Table 3-1: Characteristics of human resources at EVN in 2019
No Department Number of employees
Bachelor Master PhD Vocational training
From table 3.1, the total number of employees in Vietnam's Electricity sector was 99,815 people, of which female employees were 20,934 people, accounting for 20.8%, a relatively low proportion Human resources in Vietnam's electricity sector under the age of 30 accounted for 12.5%, from 31 - 39 years old accounted for 50.04%, from 40 - 50 years old and over 50 years old accounted for 29%, accounting for 8.46% respectively [ CITATION EVN20 \l 1033 ] Overall, the age structure of employees in the company is aging, affecting labor productivity It is also difficult to arrange a reasonable labor force in the company, especially for heavy jobs such as management and operation of lines, repair of power grid problems Human resources in EVN with graduate degrees including master degree and PhD degree accounted for 19.63%; undergraduate 56.98%, other degrees such as vocational training degree and high school degree 23.39% [ CITATION EVN201
\l 1033 ] The quality of EVN's personnel according to professional qualifications has gradually increased and is stable over the years The distribution of personnel by production and business activities at EVN also has many differences General management and service provision only accounted for 1% of the total number of employees while the electricity distribution and sales division accounted for 75% of the staff Power generation, transmission, as well as consulting and production preparation accounted for 13%, 8% and 3% of the total number of employees atEVN, respectively The structure of labor by industry is still inadequate, many departments have redundant labor force, but cannot be transferred to other departments due to different training expertise The company’s labor productivity compared to the electricity industry of other countries in the region is not high and has not brought into full play its internal strength The main reason for the above shortcomings is that in the past time, the company has focused on management,minimizing the recruitment of new workers at member units in order to improve production and business efficiency and increase labor productivity However, the other side of the problem is reducing the number of young, healthy and qualified workforce to work at the company.
At the end of 2017, the EVN’s Board of Director issued Resolution No 15- NQ/DU on improving the quality of human resources with the goal of training EVN's human resources with high quality, meeting the requirements of EVN’s development in the period of industrial revolution 4.0 Along with that, it is in line with the company’s restructuring roadmap and the Government's requirements on innovation, streamlined organizational structure, effective and efficient operation, building EVN to become one of the most successful companies and leading power corporations in ASEAN.
EVN's 2022 training strategy has been elaborately built, with rich and new content, which is reflected in the request that ERAVCTED provide training topics that not only focus on training to improve human resources for the current competitive electricity market, but also actively train human resources to be able to promptly respond to new contexts of the power industry such as when the power system integrates a large proportion of renewable energy sources or to meet new requirements on energy transition and sustainable development, as these factors will more or less directly or indirectly affect the operation of the competitive electricity market in the future.
In the three years from 2019-2021, EVN continuously recorded positive financial performance despite the impact of the Covid-19 pandemic According to the results of EVN's 2019 electricity production and business cost inspection conducted by the Ministry of Industry and Trade, commercial electricity output in
2019 is 209.77 billion kWh, up 9.05% compared to 2018 Electricity sales revenue in 2019 of EVN reached 388,355.63 billion VND, up 16.63% compared to 2018.The average commercial selling electricity price in 2019 is 1,851.36 VND/kWh, up6.95% compared to 2018 In 2019, EVN's profit is 523.37 billion VND,corresponding to the ratio of profit after tax on equity in 2019 is 0.35%[ CITATION EVN19 \l 1033 ] However, EVN's total sales revenue of commercial electricity in 2020 reached just over 394,892 above the targeted objective of 396,199.38 billion VND, due to a decrease in electricity output sold to customer groups, along with discounts and electricity bills to support the Covid-19 epidemic.
As a result, EVN's electricity production and business activities in 2020 lost 1,307.29 billion VND, but EVN's activities related to electricity production and business gained over 6,049 billion VND; including proceeds from the sale of reactive power; income from financial activities, dividends and distributed profits, and transfer profits As a result, a total of electricity production and business activities and activities related to electricity production and business in 2020 of EVN made a profit of 4,742.24 billion VND Therefore, EVN has preserved and developed the state capital, the total value of consolidated assets by the end of 2020 is 729,452 billion VND (up 1.1% compared to 2019), of which the equity is 240,195 billion VND (an increase of 6% compared to 2019) [ CITATION EVN202 \l 1033 ] According to EVN's semi-annual financial report for 2021, EVN's revenue is more than 211,600 billion VND and pre-tax profit is up to 10,127 billion VND The value of EVN's budget payment in 2021 will reach 22,440 billion VND The production and business results of the parent company EVN and its units in 2021 are all profitable [ CITATION EVN21 \l 1033 ].
The characteristics of sample size
The survey questionnaire received 212 answers, in which 5 answers are not complete therefore only 207 answers are considered as valid and put into data analysis process.
Among 207 respondents, the majority are male, making up for 61.4% while female respondents are more than half of male respondents, accounting for 34.2%.Only 4.2% prefer not to mention about their gender.
The age of participants in the survey are diverse The age groups of 19-25 years old, 26-30 years old and 31-35 years old account for the largest parts with 35.2%, 29.4% and 16.9% respectively More than 10% of participants are from the age of 36 to 40 years old The percentage of participants who are equal or below 18 years old and more than 42 years old is the same, each accounts for 3.8%.
More than half of respondents are employees, contributing for more than 50%.22.2% are freelancers Business owners and other occupations make up for 11% and10.6% respectively Only 5.8% are students.
The participants who own bachelor degree account for more than half of the total with 54% 32.8% answered that they have master degree Other degrees such as college and vocational degrees account for more than 10% Only nearly 3% are owning PhD degree.
Figure 3:5: Education background of participants
The monthly income level of participants is also very different The highest percentage of monthly income ranging from 10.1 to 20 million VND, accounting for nearly 40% More than 22% are earning from 20.1 to 30 million VND per month Approximately 17% earn from 5.1 to 10 million VND on monthly basis. The percentage of high income from 30.1 to 40 million VND per month and more than 40 million VND per month is making up 11% and 8.7% respectively Only 1.4% are gaining equal and below 5 million VND per month.
Figure 3:6: Monthly income of participants
Among 207 respondents, the majority answered that they pay their electricity bill through online non-cash payment via Internet, accounting for more than 88% while only 11% said they they pay their electricity bills by traditional payment by cash at stores.
Figure 3:7: Payment methods for electricity bills of participants
Among online non-cash payment methods, nearly half of participants are using mobile payment apps such as Momo, Zalopay, Moca, VN pay, etc 28.7% and 21.9% admitted that they pay electricity bills through mobile banking apps and Internet banking respectively.
Figure 3:8: Online payment methods for electricity bills
More than 58% of respondents answered that they already installed and used EVN mobile apps which is called Epoint 41% said that they have not installed Epoint apps
Figure 3:9: Number of participants who already installed Epoint apps
Reliability analysis
According to Nunnally (1978), a good scale should have Cronbach's Alpha reliability of 0.7 or higher [ CITATION Jum781 \l 1033 ] Hair et al (2009) also suggested that a scale that ensures unidirectionality and reliability should reach Cronbach's Alpha threshold of 0.7 or higher, however, as a preliminary exploratory study, the threshold is Cronbach's Alpha of 0.6 is acceptable [ CITATION Jos098 \l
Another important indicator is Corrected Item - Total Correlation, this value represents the correlation between the observed variable with the remaining variables in the scale, if this observed variable has a stronger correlation with the variables in the scale group, the higher the Corrected Item – Total Correlation value, the better the observed variable Cristobal et al (2007) said that a good scale is when the observed variables have the Corrected Item - Total Correlation value from 0.3 or more [ CITATION Edu07 \l 1033 ].
From table 3:2, all observed variables have Cronbach’s Alpha bigger than 0.7 and Corrected Item-Total Correlation are all larger than 0.3 Therefore, it can be concluded that all observed variables in this research model meet the requirement on the reliability and can be put into processing for next step.
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted
Descriptive analysis
From Table 3:3, flexibility have tendency to be agreed by the majority of respondents with the mean more than 4.2 Standard deviation of this independent variable group is smaller than 1, which demonstrates that there is no significant difference among the answers of respondents for this factor Generation Z and Y in Vietnam are young and growing up in the digital technology era, which makes them become more tech savvy and prefer the convenience and flexibility that hi-tech can bring to their life.
Table 3-3: Descriptive analysis of Flexibility (FT)
Flexibility (FT) Observed variables Questions Min Max Mean STD
FT1 I don't need to que to use services provided by EVN as before using Epoint app 2 5 4.27 0.867 FT2
I can pay electricity bills and manage other services provided by EVN via Epoint app any time.
I can pay electricity bills and manage other services provided by EVN via Epoint app anywhere if necessary.
Average mean of observed variable group 4.30
Perceived usefulness has mean ranging from 3.8 to 4.2, which illustrating the fact that respondents tend to agree with the statement about the impact of perceived usefulness on their intention to use Epoint app The young customers prefer to use mobile application when the perceive the advantages that mobile app can bring to their daily life The standard deviation smaller than 1 also confirms the answers among participants are not significantly different.
Table 3-4: Descriptive analysis of perceived usefulness (PU)
Perceived usefulness (PU) Observed variables Questions Min Max Mean STD
PU1 Using Epoint app helps me save time and effort 2 5 3.86 0.935 PU2
Using Epoint app helps me to manage electricity bills and related services more efficiently.
PU3 I find it helpful to use Epoint app in managing electricity bills and related services 3 5 4.28 0.735 PU4
I find it more convenient for me to use Epoint app than offline services at EVN's stores and offices.
Using online payment via Epoint app is easier for me to pay electricity bills and other related services.
Average mean of observed variable group 4.05
The mean of perceive ease of use is relatively high compared to the mean of other independent factors This shows that participants all agree that the adoption of mobile application is impacted by its interface and user experience which help customers easily understand and use it without any difficulties Even young generation such as gen Z and gen Y who are familiar with mobile technology, they also agree that the mobile application with user friendly function and easy to use are preferred.
Table 3-5: Descriptive analysis of perceived ease of use (PE)
Perceived ease of use (PE) Observed variables Questions Min Max Mean STD
I find it easy for me to use Epoint app to do what I want to do related to electricity service such as paying electricity bill or looking for information about electricity services.
PE2 I find it easy for me to become more efficient when using Epoint app 3 5 4.31 0.699
I find it easy for me to use Epoint app to pay electricity bills and manage other related services.
PE4 It's easy to keep track of promotions, notifications, order information via Epoint app 2 5 4.21 0.802 PE5 Learning how to use Epoint app is easy for me 2 5 4.24 0.769
Average mean of observed variable group 4.26
The mean of trust is ranging from 3.4 to 4.6, meaning that the participants tend to agree with the statement of the questions related to trust The standard deviation is smaller than 1, which shows that the answers for this factor between participants are not much different Customers only use mobile applications that they put trust on, especially mobile app that is incorporated payment or related to finance,customers even more carefully consider before install and use.
Table 3-6: Descriptive analysis of Trust (TR)
Observed variables Questions Min Max Mean STD
I think personal information will be kept confidential when using Epoint app.
I feel secure when using Epoint app to pay electricity bill and manage other services.
TR3 I feel trustworthy in Epoint app when using it 2 5 3.46 0.933
Average mean of observed variable group 3.87Social influence has the mean relatively lower than other independent factors.The standard deviation smaller than 1 refers to insignificant differences among answers of participants Generation Z and Y in Vietnam tends to be influenced by influencers, key opinion leaders (KOL) or their friends on social media However,recently influencer marketing that takes advantages of influence of influencers, key opinion leaders (KOL) on young generations are overusing, which leads to the suspicion of young customers on the reviews and experience of influencers andKOLs Therefore, in this research, the social influence factors are not agreed at higher level but still over the average.
Table 3-7: Descriptive analysis of Social influence (SI)
Social influence (SI) Observed variables Questions Min Max Mean STD
The reference group (parents, relatives, friends, colleagues ) advised me to use
The media and social networks are giving a lot of advertising information about using
SI3 People around me often use Epoint app 1 5 3.53 0.929
SI4 Using Epoint app makes me to be more updated with people around me 2 5 3.63 0.807
3.58 The mean of hedonic motivation is nearly reaching 4, indicating that the participants agree with the statements about hedonic motivation The standard deviation smaller than 1 also proves that there is no significant difference among participants’ answers Generation Z and Y have tendency to be motivated to use mobile application that makes their life convenient, happy and enjoyable And majority of participants admitted that the enjoyment, convenience and happiness that Epoint brings to their life are one of the strong motivations to encourage them to download and use Epoint.
Table 3-8: Descriptive analysis of Hedonic motivation (HM)
Hedonic motivation Observed variables Questions Min Max Mean STD
HM1 I love using Epoint app rather than I have to go to EVN stores and offices 2 5 3.84 0.745
HM2 Using Epoint app helps me feel comfortable and convenient 3 5 4.07 0.583
HM3 I find it interesting to use Epoint app 3 5 4.07 0.689
Average mean of observed variable group 3.99
The mean of perceived risk is larger than 4, showing that participants tend to agree with statement about perceived risk in the survey There is also no big difference in the answers among respondents Customers nowadays, especially young generation are taking all risks such as privacy, data leak, fraud and hacker into consideration when they decide to adopt new mobile application Especially with mobile application that is incorporated payment function, customers tend to be more careful about possible risks that can cause them loss of money.
Table 3-9: Descriptive analysis of Perceived risks (PR)
Perceived risks Observed variables Questions Min Max Mean STD
PR1 I see a risk of money loss when using
Epoint to pay or make transactions online 4 5 4.41 0.493 PR2
Using Epoint app to make online payments and transactions makes it possible for my money to be stolen.
The possibility of going wrong with the mobile payment system of Epoint app is very high.
In general, using mobile payments on Epoint app is riskier than other types other forms of payment.
Average mean of observed variable group 4.33
Intention to use Epoint app has relatively high mean, which demonstrates a high agreement and high intention to use among participants With standard deviation smaller than 1, no significant difference between answers of participants is found The majority of participants answered that they agreed that they have intention to use Epoint app as well as recommend this app to other people because of its advantages to their life.
Table 3-10: Descriptive analysis of Intention to use Epoint app (IT)
Intention to use Epoint app Observed variables Questions Min Max Mean STD
IT1 I intend to use Epoint app to pay electricity bills and manage other related services 3 5 4.14 0.632 IT2
I will definitely to use Epoint app to pay electricity bills and manage other related services in the future.
IT3 I plan to use Epoint app to get information, vouchers and discounts 3 5 4.43 0.560
IT4 I will recommend other people to use E- point apps 1 5 4.03 0.867
EFA analysis
Exploratory factor analysis includes analysis of 4 important indicators: KMO, Bartlett test, Eigenvalue and Factor loading.
KMO: KMO coefficient is an index used to evaluate the magnitude of the correlation coefficient between two variables with the magnitude of their partial correlation coefficient Kaiser (1974) concluded that the value of KMO must reach 0.5 or higher (0.5 ≤ KMO ≤ 1), the exploratory factor analysis is appropriate, if the KMO is below 0.5, researchers need to consider collecting more data or consider excluding observed variables of little significance [CITATION Kai74 \l 1033 ]
Bartlett test: A very important assumption in EFA is that the observed variables included in the analysis should be correlated with each other Instead of making an assessment based on the rather difficult correlation matrix, Bartlett test is applied This test will consider whether there is a correlation between the variables participating in EFA or not with two hypotheses: o Ho: There is correlation between the observed variables in the population and EFA is appropriate o H1: There is no correlation between the observed variables in the population and EFA is not appropriate
If the sig value of Bartlett test is less than 0.05, Ho is accepted and if sig. value is greater than 0.05, H1 is accepted [ CITATION Jos098 \l 1033 ]
Eigenvalue: Hair et al (2009) suggested that only factors with an eigenvalue of 1 or more are assessed as significant and retained The approach of this method is that the number of factors extracted will explain a certain percentage of variance of the observed variables According to Merenda (1997), the number of factors extracted should achieve a cumulative variance of at least 50% [ CITATION Pet97 \l
1033 ] Meanwhile, Hair et al (2009) said that the number of factors extracted could explain 60% of the total variance as well [ CITATION Jos098 \l 1033 ].
Factor loading: With a minimum sample size of 200, Hair et al (2009) suggested that a loading factor of 0.35 to 0.4 was considered as the minimum condition for the observed variable to be retained Factor loading is at 0.5 or higher,this is the optimal level, the observed variables have good statistical significance[ CITATION Jos098 \l 1033 ].
Results in table 3:11 shows that KMO is larger than 0.9 and sig value of Bartlett test = 0.00 < 0.05, therefore, EFA is suitable for this independent observed variables group.
Table 3-11: KMO and Bartlett's Test For Independent Variables
KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.950
Bartlett's Test of Sphericity Approx Chi-Square 5041.779 df 351
From table 3:12, at the fourth component, the eigenvalue is 1.052 > 1 and the cumulative value is 71.73% > 60%, which illustrates that 4 extracted components are explaining 71.73% the variation of 27 observed variables in EFA.
Table 3-12: Total Variance Explained For Independent Variables
Extraction Sums of Squared Loadings
Rotation Sums of Squared Loadings
Extraction Method: Principal Component Analysis.
Table 3:13 demonstrates that the factor loading of all observed variables are larger than 0.5, meaning that all 27 independent observed variables have good statistical significance and suitable for next step.
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization. a Rotation converged in 7 iterations.
As shown clearly in table 3:14, the KMO = 0.836 > 0.7 and sig value of Bartlett test = 0.00