Determinants of E-Learning Adoption: Evidence from the Telecommunications Industry in Vietnam

E-learning has received a massive debate over its definition, causing a lack of a common one. Several studies have attempted to define the term. According to Aparicio, et al. (2016), e-learning includes the delivery of content for learning purposes using various electronic means such as a computer network, audio (or videotape), satellite broadcast, interactive television, and so on (Aparicio, et al., 2016). In its general meaning, e-learning refers to any form of electronically enabled learning. In a narrower explanation, it is learning made feasible by using digital technologies or, more specifically, web-based or Internet-enabled learning (Abbad, 2011; Mailizar and Maulina, 2021).

E-learning refers to the attainment and use of knowledge that is predominantly facilitated and distributed electronically (Wentling, et al., 2000; Almulla, 2021; Liao, et al., 2022). Therefore, e-learning is defined as using computer network technology, primarily over an intranet or the internet, to deliver information and instruction to individuals (Welsh, et al., 2003; Amaripuja, et al., 2023). Similarly, Choudhury and Pattnaik (2020) mentioned e-learning as “the transfer of knowledge and skills in a well-designed course content that has established accreditations through electronic media like the Internet, Web 4.0, intranets, and extranets.”

Urdan and Weggen (2000) proposed that e-learning covers a broad range of applications and processes, including computer-based learning, web-based learning, virtual classrooms, and digital collaborations. However, another term – distance learning – was not included in the definition of e-learning and was defined as a learning activity that satisfies three criteria: communication between the trainer and learners is geographically distant; two-way interactive communication; assisted by technology to bridge the space gap (Abuhassna, et al., 2023). The e-learning includes complete courses that are accessible in a just in time manner for supplementing knowledge fulfilling competencies or more fragmented content called “bites” that help learners address immediate issues encountered at work. Learning is and will continue to be a lifelong process that can be accessed anywhere at any time to meet a specific need or want.

The technology acceptance model (TAM) constructed by Davis (1989) is known as one of the most common and widely used models that has been employed to explain the intention to use or adopt e-learning in various contexts. For example, TAM has been presented by numerous studies regarding students’ or employees’ intention to use and continued usage of e-learning systems (Liu, et al., 2010; Tarhini, et al., 2013; Ibrahim, et al., 2017; Alulla, 2021; Ahmed, et al., 2023; Alhur and Alhur, 2023).

For example, Mailizar, et al., (2021) investigated factors that impact behavioral intention of university students on e-learning use during the coronavirus disease 2019 (COVID-19) pandemic in Indonesia using an extended TAM. The model consists of six constructs: system quality, e-learning experience, perceived ease of use (PEOU), perceived usefulness (PU), attitude toward use, and behavioral intention. The findings informed that attitude toward e-learning use was the most prominent construct to predict university students’ behavioral intention to use e-learning during the pandemic. Meanwhile, Jimenez, et al. (2020) also employed TAM to predict and evaluate the adoption of information and communications in e-learning context of European farmers.

Similarly, Kayali and Alaaraj (2020) examined the factors that affect the cloud-based e-learning (CBEL) of students in four Lebanese universities. The findings indicated that user satisfaction is the most important predictor of behavioral intention, followed by relative advantage, social influence, and PEOU. Attitude mediated the effects of social influence and user satisfaction on behavioral intention. Decision makers are recommended to focus on user satisfaction and increase the benefits of CBEL.

In addition, Ajibade and Zaidi (2023) adapted TAM to investigate the extent to which Nigerians are adopting social networking media for e-learning. Intentions to utilize social media for e-learning by students and faculty at Nigerian institutions were shown to be impacted by PEOU and PU.

Furthermore, Stiller and Wager (2023) used the general extended TAM for e-learning and collected data from 113 employees from a medical institution to examine factors influencing the intention to use e-learning in Germany. The PEOU was best explained by the factors computer experience, computer self-efficacy, computer anxiety, and enjoyment.

In TAM, the motivation of an individual to adopt new technology can be explained by three constructs: PEOU, PU, and behavioral intention to use the new technology. Both PEOU and PU impact the adoption of the new technology, particularly the e-learning system.

In the present study, we used TAM as the basic research framework and added three variables, management support (MS), technical support (TS), and course quality (CQ), to explain EA more comprehensively. The proposed hypotheses are presented in Figure 1.

Figure 1.

Proposed research model

(Source: Own elaboration)

In this study, EA is the dependent variable. The measurement scale of EA (three items) was adapted from the study of Liu (2009). Three independent variables, MS (four items), TS (four items), and CQ (six items), were also adopted from Liu (2009).

Two mediators in this research model adapted from TAM are PU (three items) and PEOU (three items). These scales were adopted and adapted from other studies on technological-based products and services, such as Lee (2010). The measurement items are presented in Appendix.

In this study, a self-administered online survey was conducted in 2021 from March to May. The population of this study is the total number of employees working in three largest companies in the Vietnamese telecommunication industry. Thus, the nonprobability and convenient sampling methods were applied. The data collection process is as follows: (1) acquiring permission to collect data from the telecommunication companies; (2) distributing the questionnaire randomly through the company’s email system; and (3) screening to select valid received questionnaires for further analysis.

The questionnaire was randomly distributed to employees within the chosen telecommunication companies through the company mail system. The purpose of the survey was clearly stated in the email. Potential respondents were asked to participate in the study by clicking a hyperlink to the Google Form of the questionnaire. All replies were treated with complete confidentiality. After 2 months, 248 valid questionnaires were returned and employed in the analysis using the Statistical Package for the Social Sciences and Analysis of Moment Structures (AMOS) software.

This study employed structural equation modeling (SEM) to examine the hypotheses. Hypothesized effects were positive and statistically significant, indicating that the three external variables and the two mediating variables were all significant determinants. Test results are shown in Table 4.

As shown in Table 4, seven hypotheses were accepted. Particularly, PU is confirmed to be strongly influenced by MS (β1 = 0.403), followed by PEOU (β5 = 0.317), and CQ (β3 = 0.157). Meanwhile, PEOU is mostly affected by TS (β2 = 0.461) and CQ (β4 = 0.311). In turn, EA is more strongly affected by PU (β6 = 0.484) than PEOU (β7 = 0.199).

The primary objective of this study was to verify the relationship among employees’ PU, PEOU, and EA. In this research, we used the extended TAM, which featured specific variables relating to an enterprise context. In addition, the effect of MS, TS, and CQ on employees’ PU and PEOU was investigated.

Data analysis showed the predictive relevance and validity of the model to evaluate the adoption of e-learning systems among employees. One of the main reasons for this finding may be the ease with which these staff use technology as they work for leading Vietnamese technology enterprises. Moreover, working in this ICT-related field, employees are relatively techproficient and, of course, face less difficulty in mastering the e-learning system. The empirical results were consistent with most studies mentioned in the literature review, such as Alharbi and Drew (2014), Ibrahim, et al. (2017) and Alshehri, et al. (2019). Two dominant determinants, PU and PEOU, positively influenced surveyed employees’ adoption of the e-learning system in corporate training. Our study suggested that the core TAM relationships hold just as well in an enterprise setting as in other fields. The data support all hypothesized relationships. Our findings are in line with the study of Alluma (2021).

In the present study, it could be concluded that all three factors (MS, TS, and CQ) significantly impacted the adoption of e-learning in the Vietnamese telecommunications industry. PU was the most critical factor influencing employees’ adoption of e-learning. Recognizing PU as the factor with the biggest impact on the intention to use e-learning supports existing research (Venkatesh and Davis 2000; Alsabawy, et al., 2016). Although PEOU also recorded a positive effect, it was less impactful. This result concurs with Pituch and Lee (2006), who stated that the effect of PEOU depends on the nature of the task the technology is applied to. Furthermore, our finding is also in line with the studies of Ansong (2017), Castiblanco Jimenez (2020) and Almulla (2021). Hence, a possible explanation for its limited importance on employees’ intention to accept e-learning may be the application of technological training would more likely lead to replacing job tasks, rather than just a higher competence and cognitive capacity for employees. Since e-learning programs could be carried out with fewer expenses on organizing (in comparison with classroom training regarding travel, lodging, meals, and, for some courses, the instructor’s salary) and materials could be reused, in the long run, benefits of some divisions in charge of corporate training would be negatively affected.

The increasing popularity of online learning in the COVID-19 pandemic context has led to the extensive use of digital resources as additional tools for training. The embracing of online instruction could generate additional workload, conflicts, and negative affective responses to technology use for employees. To decrease these negative consequences, it is necessary to accurately identify the predictive factors that can influence the degree of adoption and continuity in the use of technology. Our finding is also supported by several other studies such as Kayali and Alaaraj (2020), Mailizar, et al. (2021), Abuhassna, et al. (2023), and Padalia, et al. (2023).

Furthermore, it is beneficial to identify which factors enable and inhibit employees’ acceptance of technology-driven training programs to take corrective actions to increase acceptability. This study contributed to developing a predictive model for the acceptance of elearning among employees. In line with previous research, we concluded that PU has the biggest predictive power, while PEOU also contributes, but with a lower impact. This finding is similar to the studies of Liao, et al. (2022) and Stiller and Wager (2023). Nonetheless, this factor should not be overlooked since it appears to have a strong predictive power on PU. In addition, our model included external variables, which show a significant correlation with the two aforementioned mediating variables.