COVID-19 in Emerging Countries and Students’ Intention to Use Cloud Classroom: Evidence from Thailand

In many emerging markets, collectivist culture promotes interpersonal relationships which entail sharing both work and personal lives with one other. Nevertheless, the ubiquity of the World Wide Web has provided massive opportunities to teachers and learners around the globe to share knowledge anytime anywhere via online education. It is against this background that this study explores the perceptions of IT students in adopting virtual learning system in higher education institutes in an emerging country context under the COVID-19. We extended Davis’s (1989) Technology Acceptance Model (TAM) and evaluated students’ intention to use cloud classroom. Data were obtained from the five universities IT students in Bangkok, +ailand. Using partial least square structure equation modeling (PLS-SEM), the data of 373 ITstudents were analyzed.+e findings of the research show that all hypotheses were supported, except one that was related to the positive impact of perceived usefulness on students’ intention to use cloud classroom. +e extended TAM model explains 51.6% variance to explain students’ intention to use cloud classroom.+e result of this study has useful implications for educationists and strategists related to the effectiveness and usability of cloud classroom in higher education institutions.


Introduction
e development of the World Wide Web has accelerated the pace of Internet activities. rough Internet activities, more and more people are connecting on the Internet [1]. e universality of the Internet has changed the traditional way of learning [2]. People from different backgrounds and education interact with one another because of shared goals [3]. Kaufmann and Buckner [4] stated that online learning provides opportunities for the advancement of online courses. Harrison et al. [5] emphasized the importance of online learning programs around the world. Technological competencies and instructional design are the crucial elements of online education [6]; therefore, future training programs should give ample focus to these areas [7,8] and understand the needs and characteristics of the people involved in online learning [9,10].
However, for training programs on online learning to be effective, the role of a teacher in traditional classroom learning is authoritative. e teacher determines the content of the course and delivers face-to-face lectures [11][12][13]. e classroom environment is centered on the discussion between teacher and students [3]. However, trends are changing towards technology-based learning methods where teachers use information technology to deliver the lectures [14]. Online courses are being offered to assist the students and improve their learning [15]. Online education is the subset of distance education where teachers adopt a variety of technological applications including web-based learning, digital collaboration, computer-based learning, and virtual classrooms [16]. It is a form of well-structured courses with content for "just-in-time" access to materials and learning [17]. e willingness of higher education institutions to avail themselves the opportunity of cloud classroom continues to grow with the access to high numbers of students [18,19]. Consequently, universities at a point in time may ask the faculty members to consider online teaching partially or fully [20]. Jaschik et al. [21] conducted a survey on faculty attitude towards technology which revealed that only twenty percent (20%) of faculty members used technology to record class lectures. In context of digital startups, Chaiyasoonthorn [22] and Benchrifa et al. [23] also indicated attitude is the main determinant which affects behavioral intention among undergraduate students.
To promote e-learning methodology in higher education institutions, universities must adopt advanced information and communication technologies [24]. Al-Tahitah et al. [25] urged the importance of using social media platforms as a medium of learning during COVID-19 pandemic. ey posited that social media technology has a vital and effective role for students learning during emergency situation like COVID-19 outbreak. erefore, the significant role of cloud classroom cannot be ignored during COVID-19 pandemic. For example, AlAjmi et al. [26] stressed on the role of higher education institution to transform face-to-face education to online learning. In relation to this, Prasetyo et al. [27] indicated that online learning is a viable solution that accelerates the face of students learning. Past studies have focused on the usage of latest technology to facilitate students learning [26,[28][29][30]. e literature remains scarce regarding the use of cloud classroom during COVID-19 pandemic.
erefore, this study focuses on importance dimensions of using cloud classroom and its effectiveness on students' intention to use. e use of cloud computing technology is the most viable option due to its low cost [31]. It promotes e-learning which promotes knowledge sharing, scalability, and reusability of available learning materials around the globe [32][33][34]. Undoubtedly, cloud computing is one of the solutions to solving the problems of e-learning. But the applicability of advanced technology is difficult due to its complex migration process [24,35]. Many factors influence willingness to adopt advanced technology such as the use of the system and facilitating conditions [36]. erefore, it is essential to understand the factors affecting cloud classroom in higher education institutions (HEI). e understanding of respondents' inclination towards the usage of cloud classroom is imperative in the higher education context, as it will improve future learning methodologies in HEIs.
is study explores the perceptions of IT students in adopting virtual learning system in higher education institutes in an emerging country context under the COVID-19. In the context of cloud classroom, facilitating conditions, ease of use, and usefulness are essential [2]. Facilitating conditions help to overcome the perceived barriers during the task [36,37]. Individual confidence regarding the use of advanced technology represents computer self-efficacy [38]. Fathema et al. [39] extended technology acceptance model (TAM) to examine faculty use of learning management systems (LMSs). ey incorporated facilitating conditions and perceived self-efficacy into the extended TAM model. Likewise, many scholars used the technology acceptance model (TAM) to examine students' willingness to advance technology for learning [40][41][42][43]. Many studies have used integrated IS theories and TAM to understand individual behavioral intention towards technology adoption [44,45]. However, a comprehensive integrated framework based on the TAM model for students cloud classroom has yet to be studied. Based on acceptance of information technology literature, we included three additional constructs: facilitating conditions and computer self-efficacy into the technology acceptance model (TAM) to predict students' intention to use cloud classroom in HEIs. Hence, the outcomes of this study provide an understanding regarding students' acceptance of cloud classroom in HEIs.
e first section of this study discussed the importance of cloud classroom in the contemporary world. e second section presented the theoretical framework adopted in this study and the literature of the constructs. In the third section, the methodology adopted in this study was explained. e fourth section elaborated the findings of the study. Finally, we discussed the conclusion and discussions, policy implications, limitations, and future research directions.  [46] and designed to understand people's tendency towards the adoption of technological products [47,48]. TAM was originally derived from the theory of reasoned action (TRA) [49,50], which is widely famous in the domain of social psychology. e TAM was developed in the information management system to assess people's intention to adopt and use new technology or media. e original model of TAM was developed to explain users' behavioral intention to use information management systems through perceived usefulness and perceived ease of use [51]. Later, Venkatesh and Davis [52] extended the original version of TAM by adding many important constructs, such as image, job relevance, subjective norms, experience, and voluntariness, and named it TAM 2. Eight years later, Venkatesh and Bala [53] proposed another extended model named TAM 3 in which experience was used as a moderator defining the relationships among the constructs. For the last three decades, TAM 1, TAM 2, and TAM 3 and its extended versions have been widely used by researchers around the globe to explain behavioral intention. Based on the technology acceptance model, we propose a conceptual model ( Figure 1) that can predict students' intention to use cloud classroom in higher education.

Perceived Usefulness.
In the TAM model, perceived usefulness refers to an individual's believes that using a specific system will accelerate his or her performance [54]. In the TAM model, perceived usefulness is related to task performance, effectiveness, and productivity [51]. It is a major belief construct leading to behavioral intention for the use of specific technology [55][56][57][58]. In the context of the World Wide Web, Moon and Kim [58] confirmed the effectiveness of perceived usefulness on the intention to use  [51,64,65]. Perceived ease of use has the same meaning as the complexity variable in [66] diffusion of innovation theory. Davis and Venkatesh [67] give high value to perceive ease of use because many technology-oriented products were rejected based on users' poor performance due to the poor interface of the systems. Prior researches have shown that site designs include updated information, simple checkout procedures, good layout, transparent navigational structures, effective search engines, and userfriendly interfaces which were important aspects of online shopping [68]. In the context of online learning, Liu et al. [3] explained the significant influence of perceived ease of use to use the online learning community. Saade et al. [63] posited that perceived ease of use develops a favorable perception of students towards the use of learning tools. Past studies found the important role of perceived ease of use on the usefulness of technologies that eventually leads towards the adoption of technology [46,69,70]. Based on the past studies, it can be assumed that teachers' perceived ease of technology use will have a positive effect on the perceived usefulness and adoption of online technology in higher education. Hence, we hypothesized that H2: perceived ease of use has a positive impact on student's intention to use the cloud classroom

Facilitating Conditions (FC).
Ngai et al. [40] define facilitating conditions as "perceived enablers or barriers" that affect an individual's perception of ease or difficulty during the performance of an activity. Facilitating conditions are persons' control beliefs regarding the availability of resources that serve to facilitate the use of a technology [53].
In the context of online teaching, facilitating conditions refer to the availability of hardware, software, technical help, training to faculty, and availability of Internet infrastructure [14]. Prior studies have indicated that facilitating conditions are crucial that influence the use of technology [37,[71][72][73].
Researchers revealed that facilitating conditions have a significant positive influence on intention to use technology [37,74]. e use of technology requires professional development, instructional guidance, and careful management of teaching processes such as course management [14,75]. Adequate technology supports, access to mentors, and a conducive environment enable students to migrate smoothly to this new form of learning and become familiar easily to use them. e successful transition from traditional to technology-based learning requires a change in the pedagogy and acceptance of new skills to deliver online lectures [14,72,[76][77][78]. Based on past studies' evidence related to the effectiveness of facilitating conditions on intention to use technology, we hypothesized that H3: facilitating conditions have a positive impact on perceived usefulness H4: facilitating conditions have a positive impact on perceived ease of use 2.5. Computer Self-Efficacy (CSE). Computer self-efficacy refers to an individual ability to use the computer and perform a task [79]. Deimann and Keller [80] argued that computer self-efficacy accelerates an individual performance of using computer-based technology. A study by Liew et al. [79] on simulation-based learning depicts the influence of computer self-efficacy on task performance. In the context of education, studies revealed the effect of computer self-efficacy on technology acceptance for teaching purposes [81][82][83]. Computer self-efficacy activates intrinsic motivation for cloud classroom [44,81,84]. In the context of current research, computer self-efficacy refers to students' ability and confidence level of using cloud classroom. Based on the above studies, we assume that students' computer self-efficacy serves as an internal motivator that leads to intention to use cloud classroom. Hence, we hypothesized that H5: computer self-efficacy has a positive impact on perceived usefulness H6: computer self-efficacy has a positive impact on perceived ease of use 2.6. Perceived Usefulness and Perceived Ease of Use as a Mediator. Davis [51] posited that both perceived usefulness (PU) and perceived ease of use (PEOU) are the potential mediators of behavioral intention. Davis et al. [64] found the indirect effect PU and PEOU on intention to use technology.

Education Research International
Many researchers have proved that PU and PEOU fully mediate the relationship between external variables and usage intention [52]. In addition, Santhanamery and Ramayah [85] found that PU mediated the relationship between system support and continuance usage intention. Maheshwari [86] studied students' online learning intention and found indirect effect of external factors on online learners' intention. Chen and Aklikokou [87] proved that PU and PEOU mediated the relationships between external factors and e-government adoption. Furthermore, the study of Susanto and Aljoza [88] revealed that direct effect of facilitating conditions was insignificant on acceptance of e-government services, suggested the possibility of indirect effects. Based on this discussion, we have assumed that PU and PEOU will mediate the relationship between facilitating conditions and computer self-efficacy on intention to use cloud classroom. Hence, we hypothesized that H7: PU will mediate the positive relationships between facilitating conditions and intention to use cloud classroom H8: PU will mediate the positive relationships between computer self-efficacy and intention to use cloud classroom H9: PEOU will mediate the positive relationships between facilitating conditions and intention to use cloud classroom H10: PEOU will mediate the positive relationships between computer self-efficacy and intention to use cloud classroom

3.1.
Sampling. e participants of this study are the students from five Bangkok universities. A purposive sampling technique was employed for the collection of students' data. e questionnaire in Table 1 was distributed to past IT students and current ones. e preference for IT students was because they have a better understanding of cloud classroom technology than non-IT students. Participants were informed about the purposive of the study, and we ensured that their demographic details will be kept confidential and will be used for research purposive. e sample size for this study was determined by following the guidelines of researchers [89,90]. ey suggested a ratio of 5 to 10 responses per item. Given the total number of 23 items, a sample size of 230 was appropriate. However, we decided to collect a larger sample to increase data reliability. A total of 700 questionnaires were distributed to the participants of the study. A total of 392 questionnaires were received from the participants of the study with a response rate of 56%. e final analysis was performed on 373 useable data after discarding the incomplete questionnaires and performing a data screening test (outliers' identification).

Instrument.
A pilot test of the questionnaire was conducted on 65 IT students studying at different universities in Bangkok, ailand. Necessary changes related to the language of constructs have been incorporated after consultation with three academic experts and one IT expert. e first part of the questionnaire deals with the demographics of students. e second part of the questionnaire deals with constructs' items. We have employed a five-point Likert scale ranging from (1) strongly disagree to (5) [92] four items scale was used for the measurement of computer self-efficacy. Before formally analyzing the data, it was tested for data bias. We have used Harman single factor test to assess the common method bias. e results of principle axis revealed that a single factor explained less than 50% of the variance in the data which represents that common method bias is not a potential for the collected data.

Students' Profile.
e demographic profile of the participants is shown in Table 2. Out of 373 participants, 207 (55.5%) were and 166 (44.5%) were females. e majority of the participants 246 (66%) belonged to the age group between 21 and 25.
ird-year students were 182 (48.8%) which constitutes the highest percentage. Artificial intelligence was the major of most of the students 108 (29%).

e Measurement Model.
is study has employed partial least square structural equation modeling (PLS-SEM). Partial least square has many advantages such as avoidance of inadmissible solutions and factor indeterminacy, applicability to the theory development, and suitability for prediction [93].
is study has employed a two-step analytical method; first, we analyzed the measurement model and then tested the structural model. e items of the constructs are shown in measurement model, Table 3 and Figure 2. Composite reliability (CR) values more than 0.70 and average variance extracted (AVE) values above 0.50 are considered acceptable for convergent validity [94][95][96]. e values of composite reliability (CR) and average variance extracted (AVE) of this study are greater than the recommended threshold values of 0.70 and 0.50, respectively, thus confirming data robustness. Table 3 showing the values of all constructs composite reliability (CR) ranges from 0.868 to 0.951, and the values of average variance extracted (AVE) ranges from 0.622 to 0.831. Fornell and Larcker [97] criterion was used for the measurement of discriminant validity. e discriminant validity of the constructs was assessed by comparing the square root of average variances extracted with constructs correlations shown in Table 4. e square root of each construct of AVE is greater than the corresponding highest correlation confirming discriminant validity [98].  I feel confident about explaining why software will or will not run on a given computer e structural equation model (Figure 3) includes assessing the significance of path coefficients, analysis of the coefficient of determination (R 2 ), and predictive relevance (Q 2 ) value. Bootstrapping procedure, employing 2000 resampling, was used for the analysis of path coefficients. Path coefficient value ranges between −1 and +1. e estimated path closer to +1 indicates a strong positive relationship while the value closer to −1 indicates a negative relationship between the constructs. e value of R 2 was computed for the endogenous variables (cloud classroom intention) to measure the variances explained in the structural model. e model depicts 51.6% of the variance explained in faculty intention to adopt technology for online teaching (see Table 5). Besides the value of R 2 , we have computed Stone Geisser's Q 2 value as a criterion for the predictive relevance. e value (Q 2 ) above 0 indicates that exogenous variables possess predictive relevance [99]. e result of this study shows that Q 2 is 38.6% which is above 0 which indicates moderate to high predictive relevance (see Table 5).
e results of the structural model are shown in Table 5. e acceptance and rejection of theoretical relationships are  based on the path coefficients and significance values. e results of direct relationships show all hypotheses were accepted. H1 proposed the positive impact of perceived usefulness on students' intention to use cloud classroom system which was supported (β � 0.455, p ≤ 0.001). H2 proposed the positive impact of perceived ease of online teaching on students' intention to use cloud classroom system which was supported (β � 0.366, p ≤ 0.001). H3 proposed positive impact of facilitating conditions on perceived usefulness which was supported (β � 0.504, p ≤ 0.001). H4 proposed positive impact of facilitating conditions on perceived ease of use which was supported (β � 0.392, p ≤ 0.001). H5 proposed positive impact of computer self-efficacy on perceived usefulness which was supported (β � 0.353, p ≤ 0.006). H6 proposed positive impact of computer self-efficacy on perceived ease of use which was supported (β � 0.264, p ≤ 0.006). ese findings are consistent with previous studies [39,52,84,91,100,101].
We have followed Preacher and Hayes [102] method for mediation analysis. A 2000 bootstrapping resampling has been used to test the mediation analysis. e values of t-statistics and p values have been assessed for indirect (mediation) analysis. T value is greater than 1.96 and p ≤ 0.005 corresponds to acceptance of hypotheses. In addition, the mediation was confirmed through absence of "0" value in between confidence interval [102]. e results of mediation analysis presented at Table 6 depict that perceived usefulness mediates the relationships between both facilitation conditions and intention to use cloud classroom (β � 0.229, t � 7.276, p ≤ 0.001) and computer self-efficacy and intention to use cloud classroom (β � 0.161, t � 5.388, p ≤ 0.001). Perceived ease of use also mediates the relationships between both facilitation conditions and intention to use cloud classroom (β � 0.143, t � 5.114, p ≤ 0.001) and computer self-efficacy and intention to use cloud classroom (β � 0.097, t � 3.842, p ≤ 0.001).
All our hypotheses have been accepted and the mediation analysis showed significance. e COVID-19 might have a positive influence on participants of virtual classes given that the social distance requirements need to be maintained to avoid public health crisis. is is a huge shift in an emerging market context where collectivist cultures value person-to-person contacts and sharing.

Discussion.
e research on cloud classrooms is getting momentum due to the outbreak of COVID-19 pandemic. Cloud classrooms are considered the viable medium for the  Education Research International dissemination of knowledge [32]. In the emerging countries like ailand, the importance of cloud classrooms cannot be ignored as it contributes to the development of knowledge. erefore, the current study presented a novel framework that assesses the impact of factors' affecting students' intention to use cloud classrooms as a medium of instruction. e findings of the study revealed the effectiveness of the proposed model that explained more than 51.6% of the variance. e findings of the current study are consistent with previous research where researchers highlighted the importance of perceived usefulness and perceived ease of use of technology in the adoption of technology [27,36]. e results depict that ailand IT students considered the cloud classrooms are effective tools that would help students in learning. Further, the result of the study indicated the importance of computer self-efficacy affecting the perceived usefulness of the technology which is consistent with Wang et al. [81] and Valencia-Vallejo et al.'s [82] studies. ese findings indicate that students feels that they are competent to operate the computer-based technology for learning purpose. Finally, the results revealed that facilitating condition is a vital factor that affects students' intention to use cloud classrooms for their studies. ese findings are consistent with previous researchers where they argued on the importance of facilitating conditions [37,71,74]

Conclusive Remarks.
e progression of e-learning has extended the methods of teaching and learning at all levels of education. Scholars have studied many theoretical frameworks to analyze and understand teachers' and students' tendency towards cloud classroom. is study attempted to extend technology acceptance model (TAM) to predict students' intention to use cloud classroom systems. e extended TAM model included three usability constructs such as computer self-efficacy and facilitating conditions that have a significant role in attracting students for the use of cloud classroom. e findings of the study are encouraging and confirm the effectiveness of TAM in the context of cloud classroom system that matches with previous studies [40,42,103,104]. is study covers a new dimension that focuses on computer self-efficacy on students' intention to use cloud classroom system.
e findings of the current study depict that perceived ease of use, facilitating conditions, and computer self-efficacy have significance in the context of cloud classroom amongst university students. All constructs account 51.6% of the variance in the theoretical model, suggesting the effectiveness of the theoretical model in the context of cloud classroom. Among all studied factors, perceived ease of use has the greatest impact on students' cloud classroom system. Furthermore, the positive impact of facilitating conditions such as professional development and training related to the use of cloud classroom help and motivate students to learn new skills [14,75,105]. e results of mediation analysis depict that PU and PEOU mediated the relationships between two determinants (facilitating conditions and computer self-efficacy) and intention to use cloud classroom. ese findings match with the previous researchers' findings where they found PU and PEOU have mediated effects on intention [52,64,85,87,88].

Policy Implications.
To understand university students' intention to use cloud classroom in higher education institutions, this study highlights the importance of the TAM model and added three additional constructs that are crucial in students' learning systems. Additional constructs that help students to use cloud classroom include facilitating conditions and computer self-efficacy. e findings of the study offer guidance to educationists related to students' use of cloud classroom. First, the educationists and university leaders need to ensure that they have arranged the required facilities and materials that help to promote online teaching. Facilitating conditions vary from the availability of hardware to the availability of technical staff that supports the smooth running of the system [81,92]. Last but not least, the components of the technology acceptance model: perceived ease of use has a significant impact on student's intention to use cloud classroom in higher education institutions. As posited by previous researchers that perceived ease of use internally motivated to perform the task effectively [39], therefore, educationists need to pay attention on attractive and communication friendly interface design. is would help the students to focus on learning and receive maximum content for educational purposes. Poor user-interface designs often create disturbance between the parties involved in the communication. However, this study fails to provide support regarding the effectiveness of perceived usefulness on students' intention to use cloud classroom in higher education institutions. Past studies in the context of students' intention to use cloud classroom applications and YouTube for procedural learning reported the same results [81,106]. is finding suggests that IT students do not exert cognitive efforts on the cloud classroom system.

Limitation and Future Research Directions.
is study has some limitations. e first limitation is related to the methodology of the study. Data of the students have been collected from only five universities of Bangkok. Future researchers can collect data through an online survey technique to include sample sizes from other regions of ailand. is study has only considered IT students' intention to use cloud classroom in higher education institutes with higher homogeneity, therefore generalization cannot be done to all levels in ailand. Future studies can be conducted by included students of other faculties who have used cloud classroom in higher education. ird, this study has adopted a self-reported approach and included subjective measures to answer the questionnaires.

Data Availability
e data used to support the findings of this study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare that they have no conflicts of interest.