1. Introduction

Services and industry have been positively affected by the development of science and technology (Nguyen,Nguyen, Dang, & Nguyen, 2016; Werthner & Klein, 1999). That banking services and financial intermediaries operating based on digital platform has been popular in recent years. The digital banking system has brought positive signals to banks with the return over 43% (Statebank, 2019). Also, technological innovation makes changes in management, marketing when applying the digital banking (Mbama & Ezepue, 2018). Therefore, a research on digital banking services is very important because of continuous changes in technology. New electronic services should be studied and applied to enhance the competition among banks.

The digital banking utility has increased the customers’ satisfaction and loyalty. According to Guru, Shanmugam,Alam, and Perera (2003), about 80% of customers will continue using services if banks use the digital banking services. Therefore, a  research to develop the digital banking services will have significant effects in retaining customer loyalty. In fact, there are many researches on developing digital banking services in the world using the theory of reasoned action TRA, Technology AcceptanceModel of Davis (1989).

Digital banks in Vietnam are on the rise quite quickly. In their development strategy, many banks also set the goal of developing digital banking, even some banks consider it as a core business, a key development goal. According to State bank Vietnam's survey, by April 2018, 94% of domestic banks are conducting digital transformation, of which about42% of credit institutions consider digital banking as a business strategy. Over the past 5 years, payment via mobile banking has grown very strongly, reaching 144 /year. Meanwhile, other traditional payment methods grew by only 40%.

Vietnam has had studies on its intention using the digital banking services in Vietnam (Dam & Bui, 2017; Wang &Pho, 2009) using TRA and TAM models with consideration of risk factors and enterprise images. With electric services, customers have concerned information safety during process of using services (Nguyen et al., 2016). However, most of the research have not considered information safety factor yet. Hence, this research will analyse information safety and factors affecting the intention to use digital banking in Vietnam.

2. Literature review

Digital banking is a banking service when information as well as activities are digitized in transactions. Manipulations or transactions are based on digital platform using electric equipments as mobile phones, computers. In other words, customers will have not to go to banks to perform transactions vice versa, banks will have not to meet customers directly to complete the transactions between two sides.

Information safety is defined as control or security ability of customers’ information when using services(Fortes & Rita, 2016; Westin, 1970). Information safety is customers’ personal information which is protected in the system (Smith, Milberg, & Burke, 1996). Customers are always worried that their private information may exploited and used in different purposes which they are not informed, and the information believed to come from supplier information sharing or be stolen on the system (Fortes &Rita, 2016).

Intention using service is considered as customers’ perception or intention on using services in the future(Davis, 1993; Venkatesh, 2000). It is not accidental that the customer leads the intention to use the services of a certain supplier, customers are influenced by many factors in making decision. With TAM model, intention using service are influenced from: (1) Perceived ease of use, (2)Perceived usefulness, (3) Perceived risk, (4) Perceived trust,(5) Convenience; and (6) Attitude toward service. Besides, convenience and enterprise images factors are also shown by some research to influence intention using service of customers.(Chang & Polonsky, 2012; Kang & James, 2004).

3. Research Methodology

3.1. Model and research hypotheses

Base on the previous research on using customer behavior, we proposed a model researching on Informationsafety, Enterprise images, convenience and customer behavior using digital banking services in Vietnam as following figure 1.

Figure 1: Research model

3.2. Research hypotheses

Information safety is concerned by customers in controlling the personal information when conducting online transactions (Fortes & Rita, 2016; Nguyen et al.,2016). That the customers highly appreciate the information safety making the increasing of Perceived usefulness, Perceived ease of use of the product/service(Nguyen et al., 2016; Venkatesh, Morris, Davis, & Davis,2003). At the same time, when feeling safe, customers will try to get used to and realize easily in use (Venkatesh et al.,2003), Perceived risk also decreases (Van Slyke, Shim,Johnson, & Jiang, 2006) and Perceived trust increases(Eastlick, Lotz, & Warrington, 2006). Therefore, this research proposes the following hypotheses:

H1: Information safety positively affects the Perceived usefulness

H2: Information safety positively affects the Perceived ease of use

H3: Information safety positively affects the Perceived risk

H4: Information safety positively affects the Perceivedtrust

Trust is considered as customer’s belief in the product/service they used (Nguyen et al., 2016). Trust makes Perceived risk decrease in online transactions and creates positive attitude toward the product/service (Ha &Stoel, 2009; Pavlou, 2003). Therefore, this research proposes the following hypotheses:

H5: Trust positively affects the Perceived risk

H6: Trust positively affects the attitude towards service

Perceived ease of use is customers’ perception in believing that using technology service does not require much effort and bring freedom, comfort (Davis, 1989;Nguyen et al., 2016). Ease of use helps users save resources, makes them feel useful service (Ha & Stoel,2009) and have positive attitudes toward the service. Therefore, this research proposes the following hypotheses:

H7: Perceived ease of use positively affects the Perceived usefulness

H8: Perceived ease of use positively affects the attitude toward service

Perceived risk is customers’ perception in risks such as abuse of customer privacy information, waste of customers’ time (Glover & Benbasat, 2010; Nguyen et al., 2016). Increase in Perceived risk makes decrease in customers’ positive attitude toward service. Therefore, this research proposes the following hypothesis:

H9: Perceived risk positively affects the attitude toward service

Perceived usefulness is customers’ perception in product/service, the effectiveness in working. The usefulness is expressed by time saving, economic expectation and various utilities (Davis, 1993; Erkan &Evans, 2016). The Perceived usefulness will increase positive attitude toward service and raise intention using service/purchase intention (Fortes & Rita, 2016). Therefore, this research proposes the following hypotheses:

H10: Perceived usefulness positively affects the attitude toward service

H11: Perceived usefulness positively affects the intention using service

Attitude toward service is customer’s assessment on positive or negative meaning in applying service (Ajzen &Fishbein, 1975; Nguyen et al., 2016). The customers’ positive attitude toward service will enhance their intention using service (Davis, 1989; Fortes & Rita, 2016). Hence, this research proposes the following hypothesis:

H12: Attitude toward service positively affects the intention using service

Enterprise image is customers’ awareness or perception in a brand name of a supplier, prestige level of an enterprise and it has positively effects on customers’ intention using service when the enterprise receives goodwill from the customers (Kang & James, 2004). The hypothesis as following:

H13: Enterprise image has positive effects on intention touse service/ purchase intention Convenience is the ability to easily access and use the service anytime, anywhere when having the internet supports (Chen, Hsu, & Lin,2010). The more convenient the service will be, the more likely it will be for customers to use it when traditional service could not be available in many different circumstances. Hence, the hypothesis is:

H14: Convenience has positive effects on intention to use service/ purchase intention

3.3. Research design

Through reference to previous studies, we conducted to build a questionnaire as well as ways to measure the factors in the model. Initially, a pilot will be distributed to customers to review the intelligibility of the questionnaire. The questionnaire was edited through customer contributions. After consulting customers on the content of the questionnaire, the questionnaire continued to be distributed to 50 customers and collected for preliminary analysis. The results of the preliminary analysis based on the evaluation of reliability showed that the questionnaire was reliable and the items reflected the factors for the factors. We then proceeded to distribute the survey on a large scale to ensure the reliability of the questionnaire as well as minimize errors for the analysis in the study.

Table 1: The questionnaire

3.3. Data collection

With the designed questionnaire, we delivered the questionnaires to customers who have used the digital banking service in Vietnam. Form of direct or online coupons by convenience method combined with data collection based on snowball principle. The survey period is from December 2019 to January 2020. The results collected329 valid questionnaires in analysis. The 329 collected questionnaires are appropriate and reliable enough for multivariate analysis in the study (Hair, Black, Babin,Anderson, & Tatham, 2006; Tabachnick & Fidell, 2006).

4. Result

4.1. Descriptive statistics for surveyed subjects

The statistic describes the survey data with 111 men accounting for 33.7% and 218 women accounting for66.3%. In terms of degree level, most of customers are university-educated (271 people account for 82.4%); followed by customers with postgraduate qualification (50people accounted for 15.2%);

high school education accounted for the least proportion with 8 people, accounting for 2.4%. The surveyed customers are mostly office workers (164 people accounting for 49.8%); The group of customers in the group of free and other businesses accounted for the least proportion (the proportion of housewives - pensioners accounted for 4%; other occupations accounted for 9.4%). Income of the main survey subjects is less than 10 million(156 people accounting for 47.4%); The lowest rate of income is over 20 million (11 customers account for 3.3%).

Table 2: Customer characteristic description

4.2.Evaluating the scale reliability and validity

Two criteria to evaluate a factor that is reliable when measuring through observed variables are Cronbach’sAlpha coefficient greater than 0.6 and the item-total correlation greater than 0.3. Observed variables with a total correlation coefficient less than 0.3 will be excluded from the factor and considered as garbage variables. After elimination, this observed variable will not be included and in subsequent analyzes.

The results confirmatory the factor analysis that: Chisquare / df = 1.842 is less than 3, CFI = 0.914; TLI = 0.902;IFI = 0.915 are greater than 0.9, RMSEA = 0.051 is less than 0.08. This shows that the theoretical model is compatible with real data. The factor loading of items are greater than 0.5 (after remove CON4 because the factor loading of CON4 =0.486 is less than 0.5), so it is possible to see the model of convergence validity.

The results of the general reliability analysis and the extracted variance show that the factors are the scales with load factor greater than 0.5, reaching the convergence validity. The composite reliability of the factors above 0.7and the Average Variance Extracted (AVE) are greater than50%. This shows that the factor scales in the formal sample have achieved the necessary reliability (Table 3).

Table 3: The reliability test

IS: Information safety; PEU: Perceived ease of use; PU: Perceived usefulness; TRU: Trust; RIS: Perceived risk; ATT: attitude toward; CON: Convenience; IMG: Corporateimage; INT: Intention to use.

Factor loading greater than 0.5 in each factor are considered to have convergence validity and the square root of the variance greater than the correlation between research concepts are concepts with discriminant validity(see table 4)

Table 4: Discriminant validity

4.3. SEM analysis

The analysis results using structure model show that the model is suitable for market data (Chi-square / df = 1.841 is less than 3, CFI = 0.911; TLI = 0.902, IFI = 0.912 is greater than 0.9, RMSEA) = 0.065 less than 0.08). With the representative factors obtained from analysis of structural models, the author conducted structural analysis to find out the factors affecting the intention to use digital banking services. With the selected significance level of 5%, the pvalue values will be compared with 0.05 to see which factors have an impact and have no impact on the dependent variable. The combined results are presented in the following Figure 2:

Figure 2: The result of SEM

The results of SEM model show that Information safety factor (IS) has a negative effect on Perceived ease of use(PEU) and a positive impact on Perceived risk (RIS). In other words, the hypotheses H3 and H4 are accepted. That Information safety has a negative effect on Perceived ease of use means the higher level of information safety customers feel, the more difficult they will find in the service features. This result shows that the information safety is concerned excessively by the customers when they think that the greater information safety with digital banking services, the more necessary features to design more secure access, making it difficult to use. Customers need better control in their private information when using the utilities of the service. Hence, the secured information is tightened with the utilities to ensure the information security leading to limitations in ease of use (Lee, 2009; Sayar & Wolfe, 2007). The utilities will be limited if information security is put ontop and information safety becomes barriers in developing the utilities of service (Yoon & Barker Steege, 2013). Meanwhile, when customers pay more attention into the information safety, their perceived risk also increases.

That Information safety does not affect Perceived usefulness (PU) and TRU shows the hypotheses H1 and H4are rejected. Information security is considered imperative in online utility services. Customers are well aware that the importance of information safety is not the factor which reflects the usefulness of the service. Although the service has further improved information safety, the customers still feel the Perceived usefulness has not changed (the usefulness of digital banking has been confirmed to customers). At the same time, Information safety does not affect Perceived trust, which indicates that the more customers pay attention to information safety, the less they pay attention or care about trust in digital banking services.

Trust (TRU) has positive effects on Perceived risk, so the hypothesis H5 is accepted. Besides, that TRU does not affect the Attitude towards service (ATT) shows that the hypothesis H6 is rejected. The customers trust in digital banking makes good attitude towards services (Pavlou,2003). Good services bring customers comfort and make them satisfied then have good reviews on the service.

Perceived ease of use has positive effects on Perceivedusefulness of digital banking services, which indicates that the hypothesis H7 is accepted. That PEU does not have impacts on ATT shows the hypothesis H8 is rejected. The Perceived ease of use helps users save resources, or make them feel the useful service (Ha & Stoel, 2009). However, the Perceived ease of use does not have impacts on ATT, indicating that customers have been able to adapt well to electronic applications. Therefore, the access to electronic applications is not a barrier for customers.

In addition, the Perceived usefulness has positive effects on ATT and Perceived Risk has negative effects on ATT. Hence, two hypotheses H8 and H9 are accepted. The more useful customers feel about digital banking, the more they think it is deserved and experienced. The utilities help customers conduct many transactions on online system (Ha& Stoel, 2009; Nguyen et al., 2016). Consequently, the customers have good attitude towards digital banking services. The Perceived risk has negative effects on the attitude towards services, showing that in order to have agood Attitude towards services, besides the Perceivedusefulness factor, customers need to feel less risky in using digital banking services (Glover & Benbasat, 2010).

For the dependent variable is the intention to use the service (INT): The perceived usefulness has a positive impact on the intention to use digital banking services(β_PU=.233). Therefore, the hypothesis H11 is accepted. The attitude towards services also positively influences the intention to use the service (β_ATT= .543); In other words, the hypothesis H12 is accepted. It is the fact that the Perceivedusefulness factors help customers have a good feeling or good attitude towards services, leading to increase the Intention to use services (Fortes & Rita, 2016). The corporate image factor also affects the Intention to use(βIMG= .226), the hypothesis H13 is accepted. At the same time, the image of enterprises (banks) will help increase the intention to use when the image of banks is highly appreciated (Kang & James, 2004). The Convenience factor also have positive effects on the Intention to use (β_CON=.153), the hypothesis H14 is accepted. Finally, the service convenience also brings significant meanings in increasing the customers’ intention to use when they can make transactions anytime, anywhere (Chen et al., 2010).

5. Conclusion

The research results through the structure model show that the factors of Perceived usefulness, Attitude towards services, Convenience and Enterprise image have positive effects on the intention to use digital banking services. Therefore, to raise the intention to use digital banking services, banks must enhance the usefulness of the services through creating many service utilities. The banks need to build an information technology system which ensures24/24-hour transaction to increase good reviews on the convenience of digital banking services. In addition, banks should develop their reputation to create a good image in the customer's awareness.

Meanwhile, the information safety factor has directly impacted on the Perceived ease of use and the Perceive risk of customers in using digital banking services. From the research results, banks should build a strict customer information management system. The customer information should be secured. Confusion in transactions will reduce the perceived security, the Perceived high risk from the customers will have negative effects on the attitude towards services, leading to a decrease in customers’ intention to use digital banking services.

Although the study has achieved the set goals, there are still some limitations regarding the differences between state and non-state banks (Due to the limited number of samples collected, the grouping for analysis was difficult.) Therefore, the proposed study for further studies could collect larger samples and divide them by the types of customers using state and non-state banks.