1. Introduction
Digital transformation has received increasing attention from organizations and businesses that want to remain competitive in the digital world [
1,
2,
3,
4]. Essentially, it refers to “the use of new digital technologies, such as mobile, artificial intelligence, cloud, blockchain, and the Internet of things (IoT) technologies, to enable major business improvements to augment customer experience, streamline operations, or create a new business model” [
5]. Many banks have considered digital transformation essential for improving service quality and increasingly deployed digital services such as internet banking, electronic payment wallets, and mobile banking applications [
6]. These electronic banking (e-banking) services have proved to increase customer value and accelerated the industry’s growth.
Notably, COVID-19 has further encouraged firms to adopt and implement digital transformation [
7]. The pandemic also accelerated the growth of e-banking and digital payment services [
8,
9,
10], while traditional banking has been strongly affected. Specifically, consumers’ health concerns and restrictions such as lockdown and physical distancing have driven them towards e-banking adoption and usage, which is considered safer and more convenient compared to traditional banking services [
11].
Digital transformation is a priority in the development strategy of Vietnam’s banking industry. With a population of over 97.58 million people [
12], of which the proportion of young people is almost 70%, Vietnam is a potential digital banking market [
13]. More than 62% of the population already own smartphones, and 64 million people are Internet users. Following the digital transformation trend, banks in Vietnam have been investing in innovating and digitizing their products and services. Video teller machine (VTM) services appeared in the Vietnam market at the end of 2017 [
14]. A VTM machine is equipped with modern equipment and applies advanced technologies such as biometrics through fingerprint recognition and face recognition. Requests for support during use are handled remotely by a team of consultants. All implementation processes are appropriately designed, and users’ privacy and security are guaranteed. Notably, VTM services operate 24/7 and support customers with complicated transactions (e.g., opening accounts, depositing cash into accounts, and opening savings books) without requiring a transaction point. Several banks have applied VTM solutions under different aliases, e.g., LiveBank (Tien Phong Bank), SmartBank (MB Bank), and R-ATM (Vietin Bank), with the hope of meeting customers’ increasing demand for e-banking services and enhancing their electronic service (e-service) quality.
Santos [
15] refers to e-service quality as “consumers’ overall evaluation and judgment of the excellence and quality of e-service offerings” within the virtual context (p. 235). In general, improving e-service quality is essential to obtaining a competitive advantage [
16], enticing potential customers [
17], and building greater relationships with customers [
18,
19]. In the banking industry, e-service quality appears to be an important driver of customer intention and behavior toward e-banking services [
20,
21,
22]. Previous research has identified factors measuring e-service quality, and there is no consensus regarding its main dimensions [
23,
24]. Furthermore, some studies define service quality as a second-order construct [
25,
26]. Hence, there is insufficient understanding of the direct links between the components of e-service quality and other determinants of consumer behavior toward electronic banking (e-banking). Notably, the issue of e-service quality in Vietnam’s banking industry has been underestimated in the extant literature. This is surprising, given that Vietnam represents a colossal market opportunity for e-banking services [
13].
This study aimed to contribute to the literature by developing and validating a model that explains how key components of e-service quality affect consumer intention to use VTM services, with a specific focus on Vietnam. It particularly concentrates on three important e-service quality dimensions of interface quality, responsiveness, and security, which play an essential role in consumer decisions in e-banking services [
27,
28,
29,
30,
31]. It should be noted that previous research using the technology acceptance model (TAM) [
32] and the theory of planned behavior (TPB) [
33] reports that perceived usefulness (PU), perceived ease of use (PEOU), and attitude are important determinants of consumer intention to use e-banking services [
34,
35,
36,
37]. Hence, this study postulates that the three e-service quality factors (i.e., interface quality, responsiveness, and security) affect PU, PEOU, and attitude, which in turn motivate consumers to use VTM services. In addition, the study seeks to examine the moderating role of time-consciousness, given that the impact of PU and attitude on intention may vary across consumers with different levels of awareness of time and how they spend it [
38]. Essentially, the objectives of this study are threefold:
- (1)
To examine the impact of responsiveness, security, and interface quality on PU, PEOU, and attitude towards using VTM services;
- (2)
To investigate how PU, PEOU, and attitude affect intention to use VTM services;
- (3)
To determine the moderating role of time-consciousness in the impact of PU and attitude on intention to use VTM services.
The rest of this paper is organized as follows. The second section reviews the relevant literature and develops the research hypotheses. Next, the third section describes the research methodology, followed by the fourth section presenting the results. Discussion of the findings and conclusions are demonstrated in the final section.
4. Data Analysis
4.1. Measurement Model, Construct Reliability, and Validity
CFA results demonstrate that the model included 268 degrees of freedom; χ2/df = 1.155 (<3); AGFI = 0.937; GFI = 0.952; CFI = 0.991; TLI = 0.989; NFI = 0.940 (>0.9); and RMSEA = 0.019 (<0.08). The results indicated that the model was suitable for the collected data.
Construct convergent and discriminant validity was assessed using guidelines by Hair et al. [
78]. Accordingly, standardized factor loadings (FLs) needed to be greater than 0.5; composite reliability (CR) needed to be greater than 0.7; the average variance extracted (AVE) needed to be greater than 0.5; and the square root of AVE needed to be greater than the inter-construct correlation between constructs.
Table 2 and
Table 3 show that all the indices came within their requirements. Furthermore, the Cronbach Alpha coefficient values ranged from 0.755 to 0.867, ensuring the construct reliability [
78,
79].
4.2. Common Method Variance
The common method variance (CMV) is a problem that can negatively affect the measure performance since it can threaten reliability and validity and cause measurement errors [
85]. This study followed Podsakoff et al.’s [
85] suggestions for procedural and statistical remedies to minimize CMV occurrence. First, in the data collection process, anonymity and privacy were guaranteed. Secondly, the order of questions was randomized to reduce participants’ ability to predict the structure flow. Finally, Harman’s single factor test and latent common factor were applied to check potential CMV. A Harman’s single-factor test indicated that the single factor explained only 26.532% of the variables’ variance (less than 50%). The latent common factor test results demonstrated that the factor method accounted for 7.29% of the total variance (less than 25%). In addition, when comparing the estimated weight of the measurement model and the latent common factor model, the difference between the estimated weights did not exceed 0.2. Based on these findings and according to Malhotra et al. [
86], it was reasonable to conclude that CMV did not appear to be a pervasive problem in this study.
4.3. Hypothesis Testing
Structural equation modeling (SEM) analysis was applied to test the research hypotheses. The analysis results showed that the model included 178 degrees of freedom; χ
2/df = 1.418 (less than 3); GFI = 0.954; AGFI = 0.938; TLI = 0.976; CFI = 0.980; NFI = 0.936 (greater than 0.9); RMSEA = 0.031 (less than 0.08); and
p-value = 0.000 (less than 0.05). The model explained a significant 63% of consumers’ attitudes towards using VTM services and 30% of their intention to use the services (
Figure 2).
The results also showed that 13 of the 14 proposed research hypotheses were significant, albeit with different significance levels. Notably, the hypothesis that PEOU affects attitude was not supported (p = 0.393 > 0.05).
Table 4 summarizes the hypothesis testing results. The three dimensions of e-service quality significantly enhance PU, PEOU, and attitude towards using VTM services. Specifically, responsiveness was positively related to PU (β = 0.246,
p < 0.001, t = 4.108, PEOU (β = 0.274,
p < 0.001, t = 4.309) and attitude towards using VTM services (β = 0.189,
p < 0.001, t = 3.294). In addition, security positively affected PU (β = 0.214,
p < 0.001, t = 3.285), PEOU (β = 0.295,
p < 0.001, t = 4.261), and attitude towards using VTM services (β = 0.400,
p < 0.001, t = 6.075). Similarly, interface was positively related to PU (β = 0.188,
p < 0.01, t = 2.938), PEOU (β = 0.139,
p < 0.05, t = 2.002) and attitude towards using VTM services (β = 0.154,
p < 0.05, t = 2.538).
As expected, PU positively affected attitude towards using VTM services (β = 0.237, p < 0.001, t = 3.999) and the intention to use (β = 0.373, p < 0.001, t = 4.944). Moreover, PEOU was positively related to PU (β = 0.134, p < 0.05, t = 2.072). However, the effect of PEOU on attitude towards using VTM services was not significant (β = 0.052, p > 0.05, t = 0.854). Finally, attitude was positively related to intention to use VTM services (β = 0.232, p < 0.01, t = 3.059).
4.4. Moderating Effect Hypotheses Testing
The multi-group SEM method was applied to test the moderating role of time consciousness in the effects of PU and attitude on intention. The research sample was divided based on the customer’s assessment of time-consciousness, and the K-mean cluster division method was applied via SPSS 25, resulting in two groups of the samples identified:
Group 1: 214 customers (47.56%) have lower time-consciousness ratings.
Group 2: 236 customers (52.44%) have higher time-consciousness ratings.
The moderation test followed the recommendations of Hair et al. [
78], including an invariance test followed by a structural model analysis.
According to
Table 5, the fit indices of the configurational invariance (CI) and metric invariance (MI) models met the thresholds suggested by Hair et al. [
78]. Furthermore, the
p-value of the χ
2 difference test was 0.755, so there was no difference between the two models. Hence, a structural model was established to evaluate the difference between the two groups of samples.
The results of comparing the difference between the measurement weight model and the structural model show that the
p = 0.000 < 0.05, ∆χ
2 = 50.828, and ∆df = 14. Therefore, the unconstrained model was selected for evaluation [
87]. According to
Table 6, the results showed a difference between the two sample groups. Specifically, the impact of PU on intention was significantly higher for customers with higher time-consciousness (β = 0.451,
p < 0.001) than those with lower time-consciousness (β = 0.230,
p < 0.05). Similarly, the effect of attitude on intention was significantly stronger for customers who were more time-conscious (β = 0.392,
p < 0.001) than those who were less conscious (β = 0.174,
p > 0.05). Thus, it can be concluded that time-consciousness positively moderates the PU–intention and attitude–intention relationships.
5. Discussion, Conclusion, and Implications
5.1. Discussion and Implications
The influences of digital transformation create many opportunities and threats for banks in providing services to customers. Many banks have gradually implemented e-services to increase customer convenience and value. Understanding how customers perceive and respond to these services will help banks develop appropriate strategies for improving the quality and adoption of such services. This study developed and validated a model explaining the impact of different e-service quality dimensions, PEOU and PU, on consumer attitude and intention toward a specific e-banking service, i.e., VTM services. It also highlighted the moderating role of time-consciousness in the model. This study also extended current knowledge about e-banking services in emerging markets by focusing on Vietnam.
5.1.1. Theoretical Implications
First, SEM results showed a positive impact of three e-service quality dimensions, including responsiveness, security, and interface quality, on PEOU, PU, and attitude. These findings enriched the research stream that highlights the role of e-service quality in enhancing consumer intention to adopt e-banking services [
21]. Second, this study revealed the positive relationships between PU, attitude, and intention. These findings align with studies on consumer acceptance of new technology [
32,
36] and further support the application of TAM in the e-banking context.
Third, it is striking that the impact of PEOU on attitude towards using VTM services was insignificant, which contradicts many studies on technology acceptance and use [
32,
88]. One possible explanation for this finding is that some functions used in VTM services, such as fingerprint recognition and face recognition, have not yet operated stably, which might cause inconvenience for some customers. It should be mentioned that the findings also reveal that PEOU positively enhanced PU, which in turn increased the attitude toward using VTM services. That is, PEOU affected attitude indirectly via PU. These interesting findings suggest that more research is necessary to better understand the mechanism through which PEOU motivates consumers’ attitudes toward e-banking services.
Fourth, this study is among the first attempt to verify the moderating role of time-consciousness in the attitude–intention and PU–intention relationships. The findings show that the impact of PU and attitude on intention to use VTM services was more substantial for customers with higher time-consciousness. Such customers often carefully plan to use their time and pay attention to the time efficiency of activities in their lives [
76]. Hence, banking services that can help them save time, such as VTM services, will strengthen the relationship between their perception/attitude and intention to use such services [
38].
Finally, this study’s model validates the direct, indirect, and moderating effects of some key e-quality dimensions and determinants on consumer attitude and intention toward using VTM services. This model explained a significant 30% of the variance in consumer intention. Hence, it can serve as a theoretical framework for future studies into consumer intention and behavior toward e-banking services.
5.1.2. Practical Implications
The findings of this study also provide implications for bank managers and VTM technology providers in improving e-service quality and promoting consumer use of e-banking services. Given this study’s context, these practical implications are particularly relevant to VTM services in emerging markets like Vietnam.
E-service quality dimensions directly impact PU, PEOU, and attitude and indirectly impact intention. Hence, promoting e-service quality dimensions, including responsiveness, security, and interface quality, is essential for banks. It should be noted that among the three dimensions, security has the most substantial impact on PEOU and attitude. Therefore, banks and VTM technology providers should make every effort to increase the security of VTM transactions and customer information. Although VTM integrates recent technologies ensuring consumer security, such as biometrics and fingerprint recognition, transactional data must also be protected. To improve responsiveness and interface quality, banks and VTM technology providers should improve data transmission, increase processing speed, enhance interface design, and provide additional customer support to better address customer problems.
It is essential to enhance the usefulness of VTM services by making the transactions more convenient and efficient. Notably, the VTM transaction system in Vietnam is still at a modest scale, which limits customers in many non-urban areas from accessing this service [
89]. Thus, banks should make VTM services available in more locations to improve accessibility. Although PEOU does not directly impact attitude, it enhances PU, which in turn motivates attitude. Hence it would be beneficial to make it easier for customers to use VTM services. In this regard, instructions provided in the form of manuals at the points of presence or briefings on the interface screen should be utilized. In addition, to let customers use the services without obstacles, the system’s quality needs to be upgraded periodically, including checking to reduce the range of errors that might incur during transactions.
Given the moderating role of time-consciousness, an important target market for VTM services is those who are conscious about their time and how to utilize it. Time-conscious consumers are aware of the scarcity of time as an important resource and value time-convenience. Information and communication programs should therefore highlight the time-saving attribute of VTM services and that such services can be used anytime during the day. Such programs should use different media channels such as TV, magazines, websites, social media, and bank branches and VTM locations. Currently, the VTM services deployed by banks in Vietnam can only deliver two-thirds of the total available banking services [
89]. Therefore, it is necessary to gradually incorporate more services that VTM can perform.
5.2. Conclusion and Future Research
This study examined the impact of electronic service (e-service) quality on customer intention to use VTM services. The findings demonstrate the positive impact of three e-service quality dimensions, including responsiveness, security, and interface quality, on PEOU, PU, and attitude toward using VTM services. In addition, attitude is positively related to intention toward using VTM services, and time-consciousness positively moderates this relationship.
Although this study successfully addressed the research objectives, it has several limitations. First, VTM services are relatively new and have been implemented in only big cities such as Hanoi. Hence, the sample in this study mainly consisted of urban residents. It is suggested that future research should investigate both urban and rural customers. Second, respondents aged 45 and over accounted for only 10% of this study’s sample. Thus, future research should pay more attention to this market segment. In this regard, qualitative research, such as ethnography and life stories, could be used to approach consumers aged 45 and over in rural areas. Third, the cross-sectional survey used in this study is limited in explaining the causal relationships between the variables. Future studies could address this by using longitudinal surveys or experimental methods. Fourth, future studies are advised to test the model validated in the current study in other emerging markets (e.g., China, India, Indonesia, and Malaysia). Fifth, it would be beneficial to extend this research model by including other dimensions of e-service quality (e.g., reliability, competence) and users’ habits, personal need, technological knowledge, and socio-demographic variables such as age, gender, education, income, and religious beliefs. Finally, future research should complement this study by exploring if consumers expect new features and services provided by VTM. This can be done through focus groups, in-depth interviews, or experiments.