From Crisis to Continuity: Exploring Students’ Perspectives on the Future of Online Learning Beyond COVID-19

,


Introduction
With the onset of the Coronavirus (COVID-19) pandemic and the increasing prevalence of information technology, integrating technology into teaching and learning processes has become vital.Therefore, online learning tools have been incorporated among universities across the universe as they provide responsive and flexible learning environments (Dimitriadou & Lanitis, 2023;Lee et al., 2005;Sharma & Chandel, 2013).However, "in general, like any information systems, user acceptance and usage are important primary measures of system success" (Saade et al., 2007, p. 176).Hence, students' acceptance of any information system must be considered.
Electronic learning (e-learning) is a platform used by students and teachers to communicate, access, and exchange information anywhere and at any time (Alonso et al., 2005;Jaswal & Behera, 2023).Accordingly, the e-learning system supports learning and teaching processes either inside or outside any higher education institution's campus (anywhere and anytime), as the information and learning instructions can be delivered to learners via the Internet.Many scholars have provided an extensive range of e-learning definitions as these definitions seek to emphasise the correlation between technology and education, learning or training (Alia, 2017).
A respectable amount of literature has demonstrated numerous factors influencing online learning (Jaber, 2016).Therefore, identifying the factors that encourage learners to utilize e-learning is crucial.To do this, researchers have adapted various technology adoption models and theories, including the Technology Acceptance Model (TAM), Theory of Reasoned Action (TRA), Task Technology Fit (TTF) and Theory of Planned Behaviour (TPB).Nevertheless, "TAM is the most common ground theory in e-learning acceptance literature" (Granić & Marangunić, 2019, p. 2575) and has been used in many studies (Davis et al., 1989).In addition, the adaptation and diffusion of information technology have been studied at two levels: organizational and individual levels (Dasgupta et al., 2002).However, the key focus of this study was on the individual level, as there is an emphasis on individuals' acceptance of technology.In the context of education, attitudes play a significant role in shaping students' perceptions of curriculum, peers, and teachers (Bailey et al., 2022;Liu, 2014).Attitudes are closely associated with the feelings that guide human behaviour, and individuals can develop either positive or negative attitudes towards various subjects or activities (Edo et al., 2023;Genc & Aydin, 2017).
Perceived attitude's influence extends to e-learning systems.Hussein (2017) emphasized its central role in influencing students' intentions to adopt e-learning.Various studies have supported this finding, highlighting attitude as an effective predictor of students' behavioural intentions (Amali et al., 2022;Sharma & Chandel, 2013a).
Attitudes are also particularly relevant in language learning contexts, where learners often show either positive or negative attitudes towards the target language.Positively inclined learners tend to exhibit higher motivation levels, aiding their learning progress, while those with negative attitudes may experience demotivation and potential neglect (Genc & Aydin, 2017;Maruf, 2022).In light of this background, we propose the following hypothesis: H1: Perceived Attitude has a positive and significant association with learners' intention to continue using e-learning systems.
2.2 Perceived Usefulness (PU) Davis (1989, p. 320) defined Perceived Usefulness as "the degree to which a person believes that using a specific system will increase his or her job performance."This concept is not about the general idea of usefulness but is tied to a particular system or technology.It suggests that people evaluate the usefulness of a specific system or tool in achieving a particular purpose.Empirical studies have consistently proved that PU directly influences Perceived Attitude (PA) and, consequently, indirectly affects Behavioral Intention (BI) towards using e-learning websites.
Furthermore, PU emerges as a critical variable affecting BI.It is well-known that PU plays a crucial role in predicting students' intentions to utilise web-based learning systems (Chang & Im, 2014;Humida et al., 2022).
Students are more motivated to accept educational websites when they recognise that these e-learning materials will develop their learning skills and performance.However, it is essential to note that while PU is a prominent influencer of students' intentions to use e-learning systems, some studies, such as those by Saeed and Abdinnour-Helm (2008), Jaber (2016), and M. K. Hsu, Wang, and Chiu (2009), have suggested that PU, while influential, may not be the sole determinant of students' intentions.Therefore, we propose the following two hypotheses: H2: Perceived Usefulness (PU) has a positive influence on students' Perceived Attitudes (PA) toward the use of internet-based learning.
H3: Perceived Usefulness (PU) has a positive direct effect on students' Behavioural Intention (BI) to use the e-learning system.

Perceived Ease of Use (PEOU)
The Perceived Ease of Use is defined as "the degree to which a person believes that using a particular system would be free of effort" (Davis, 1989, p. 320).It suggests the extent to which an individual believes or perceives that using a particular system or technology will involve minimal effort.It is about the subjective perception of ease and convenience in using the system.According to previous research conducted in the past, it has been found that PEOU has both direct and indirect correlations with BI, PU and PA as well.
Several studies have investigated the impact of Perceived Ease of Use (PEOU) on Behavioral Intention (BI) to use e-learning systems.Fan (2023), Hsu et al. (2009), Jaber (2016), and Salloum et al. (2019) have reported a positive effect of PEOU on individuals' intentions to use recent e-learning systems.In contrast, Chesney (2006), Mizher andAlwreikat (2023), andZhou, Xue, andLi (2022) concluded that PEOU has no significant impact on learners' intentions to use e-learning systems.This disparity highlights the complex nature of the relationship between PEOU and BI, which may vary based on context and system characteristics.
PEOU also has been found to significantly affect Perceived Usefulness (PU) in various studies (Al-Adwan et al., 2013;Amali et al., 2022;Dasgupta et al., 2002;Davis et al., 1989;Salloum et al., 2019;Zhou et al., 2022), emphasising its role in shaping perceptions of a system's utility.Additionally, PEOU is associated with Perceived Attitude (PA) towards using e-learning systems, as evidenced by findings in studies such as Davis et al. (1989) and Salloum et al. (2019).In light of this mixed body of research, we propose the following hypotheses: H4: Perceived Ease of Use (BEOU) has a positive influence on Perceived Attitude (PA).
H5: Perceived Ease of Use (BEOU) has a direct positive effect on students' Behavioural Intention (BI).
H6: Perceived Ease of Use (BEOU) has a positive effect on Perceived Usefulness (PU).

Feedback (FD)
Feedback (FD) is defined as "information provided by an agent (e.g., teacher, peer, book, parent, self, experience) regarding aspects of one's performance or understanding" (Hattie & Timperley, 2007, p. 81).Hattie (2008) and Barry (2008) pointed out that FD is a vital factor that influences learning processes.So, in teaching schemes, agent feedback provides learners with information related to the learning process as it will consistently assist students in understanding what they are learning and what they have already learned.Some studies (Petchprasert, 2012) emphasised how FD is closely related to Motivation.Feedback has two side effects on students' motivation in language learning.Receiving feedback can be either in the form of a reward (Deci & Ryan, 1991) or not benefiting from the feedback (Chaudron, 1988).The former exerts a positive impact on enhancing students' learning and performance, whereas the latter's effect stems from students' poor performance on the task.(Petchprasert, 2012).
Even though there are several types of FD, this study investigated the general impact of FD on students' behavioural intentions.
Thus far, only a limited number of studies have explored the effect of FD on PU.As an example, Petchprasert (2012) investigated the effects of two types of FD, process and grade-oriented feedback.His findings demonstrated a positive correlation between process-oriented feedback and PU.Petchprasert's (2012) findings are consistent with previous studies' results (Strijbos et al., 2010;Van der Kleij et al., 2012).Therefore, the following is hypothesised: H7: Feedback is a significant factor in the Perceived Usefulness (PU) of e-learning.

Accessibility (ACC)
Wixom and Todd (2005, p. 90) defined Accessibility (ACC) as "the ease with which information can be accessed or extracted from the system."It refers to the degree of ease or convenience with which users can access or retrieve information from a particular system.It is all about how straightforward it is for users to get the information they need from the system.
Previous studies, such as those by Park (2009), Saoula et al. (2023), Thong et al. (2002), and Wongvilaisakul and Lekcharoen (2015), have expounded on the significance of ACC in relation to Perceived Ease of Use (PEOU).Moreover, several literature reviews, including Revythi and Tselios (2019) and Wongvilaisakul and Lekcharoen (2015), have proven a direct relationship between system accessibility and Behavioral Intention (BI).Given that websites primarily serve the function of providing information, it can be argued that the perceived information accessibility of any website significantly influences its Perceived Usefulness (PU) (Bagdi & Bulsara, 2023;Djamasbi et al., 2006).Based on the above literature review, it is assumed that: H8: Accessibility (ACC) positively correlates with the Perceived Ease of Use (PEOU) of the e-learning system.
H9: Accessibility (ACC) positively correlates with the Perceived Usefulness (PU) of the e-learning system.
H10: Accessibility (ACC) positively correlates with the student's Behavioural Intention (BI) to continue using the e-learning system.

Anxiety (ANX)
Computer anxiety is defined by Chua, Chen, and Wong (1999, p. 610) as "a fear of computers when using one or fearing the possibility of using it when needed".It involves a sense of fear or apprehension related to computers.Individuals experiencing computer anxiety may feel uneasy, nervous, or even scared when interacting with computer technology.It is worth mentioning that computer anxiety differs from negative attitudes toward utilising computers in an e-learning environment.In addition, it entails feelings and personal beliefs about computers and individuals' emotional reactions toward computer usage (Sam et al., 2005).Individuals with higher levels of technology anxiety become unhappy and tense when they use or intend to use the technology.Besides, they tend to avoid using technology, and therefore, their behaviour is affected (Park et al., 2019).Consequently, individuals with high levels of technology anxiety may encounter difficulties in developing a favourable attitude towards utilizing technology, even if they are aware of its advantages (Cebeci et al., 2019).It was demonstrated in prior research that computer anxiety has a significant negative effect on PEOU (Guo et al., 2013;Hu et al., 1999;Tsai et al., 2020) and PU (Chang & Im, 2014;Hu et al., 1999;Igbaria et al., 1996).On the contrary, there is research evidence showing that ANX has no significant influence on computer usage (Compeau et al., 1999).Therefore, according to the prior research, it is postulated that: H11: Anxiety (ANX) is negatively associated with Perceived Usefulness (PU).

Computer Playfulness (CP)
Webster and Martocchio (1992, p. 201) defined Computer Playfulness as "the degree of cognitive spontaneity in microcomputer interaction."It refers to how freely and creatively an individual interacts with a computer.In practical terms, "Computer Playfulness" suggests that some individuals have a greater willingness and ability to explore, experiment, and engage in a creative or spontaneous manner when using microcomputers.Abu-AlSondos et al. ( 2023) and Al-Aulamie et al. (2012) found that CP has a positive influence on perceived usefulness.Likewise, Adetimirin (2015) indicated that the employment of CP has a favourable impact on the Perceived Ease of Use (PEOU).Therefore, this paper hypothesises the following: H13: Computer playfulness (CP) is a significant factor in the Perceived Usefulness (PU) of technology.
H14: Computer playfulness (CP) is a significant determinant of the Perceived Ease of Use (PEOU) of e-learning.

Perceived Enjoyment (PNJ)
Perceived Enjoyment (PNJ) is defined as "the activity of using a specific system is perceived to be enjoyable in its own right, aside from any performance consequences resulting from system use'' (Venkatesh, 2000, p. 351).In other words, individuals find the experience of using the system inherently pleasurable, separate from any functional or performance benefits it might provide.When users perceive a system as enjoyable to use, they are more likely to engage with it, explore its features, and continue using it over time.Many researchers, such as Simonson et al. (1987) and Won et al. (2023), argued that the lack of PNJ might result in a more significant effort to use the system.A considerable amount of prior research has evidenced that PNJ exerts a substantial impact on Perceived Ease of Use (PEOU) (Al-Gahtani, 2016; Kanwal & Rehman, 2017;Zhou et al., 2022) and PU as well (Al-Aulamie et al., 2012;Chang & Im, 2014;Ramírez-Correa et al., 2015).Thus, the following hypotheses were developed: H15: Perceived Enjoyment (PNJ) is a significant factor in the Perceived Usefulness (PU) of an e-learning system.
H16: Perceived Enjoyment (PNJ) is a significant factor in the Perceived Ease of Use (PEOU) of an e-learning system.
Figure 2. The hypothetical model of the study

Research Design
The present study applied a quantitative research methodology by means of administering a bilingual survey questionnaire.This study explained and predicted the students' intention to continue using technology in the Omani context after COVID-19.Besides, it used a structural equation modelling (SEM) research design in order to test the proposed hypothetical model.
To propose the hypothetical research model, the constructs that have been applied in previous research on the Literature of Information Systems (IS) based on the Technology Acceptance model were reviewed.Then, exploratory factor analysis was conducted to reduce the number of constructs and line the theoretical structure of the variables.Later, the correlation between the items and constructs in the structural model was tested by using confirmatory factor analysis.As indicated in Table 1, the survey included 37 modified items used to examine the constructs developed in the hypothetical framework of this study.

Sample and Data Collection
The study's data was collected in public and private universities in Oman during the summer semester of the academic year 2021/2022 by distributing a five-point Likert scale survey among the students using a Google form.A total of 134 students completed the survey.As Table 2 illustrates, the participants of this study were undergraduate students who were studying at public and private universities and colleges in Oman, 49 males (62.6%) and 82 females (37.4%).Their age ranged from 17 to 28 years old.Almost half of the respondents (48.1%) were in the foundation stage compared to the diploma students (23.7%), advanced diploma (9.9%) and bachelor (18.3%).

Research Instrument
A survey was designed and then disseminated among the participants.The survey is composed of two parts.The first part sought to collect participants' demographical information; respondents were asked four questions, including gender, age, academic status and institution.The second part involved the nine variables pertaining to the factors that affect users' intention to continue using online learning.The items in the second section were Behavioural Intention (6 items), Perceived Attitude (4 items), Perceived Ease of Use (6 items), Feedback (3 items), Accessibility (3 items), Anxiety (5 items), Computer Playfulness (3 items) and Perceived Enjoyment (3 items).All of these were measured using 5 Likert-scale items (Strongly agree = 5, and strongly disagree = 1).

Data Analysis
In this study, the data analysis was conducted using SmartPLS 4, a widely recognized software for structural equation modelling.Various statistical techniques were employed, including the calculation of Cronbach's α for internal consistency assessment, the evaluation of composite reliability (CR), average variance extracted (AVE), and factor loadings to assess the measurement model.Discriminant validity was assessed using the Heterotrait-Monotrait ratio of correlations (HTMT).For hypothesis testing, bootstrapping with 5000 resamples was employed to examine the relationships proposed in the theoretical model.This comprehensive approach allowed for the thorough evaluation of the reliability and validity of the constructs and the testing of hypotheses related to students' technology adoption behaviours and attitudes."

Measurement Model
The measurement model of the proposed theoretical model was assessed by scrutinising the convergent validity and discriminant validity.However, Cronbach's α of each construct was calculated prior to evaluating the convergent validity of the measurement model.Cronbach's α, as seen in Table 3, ranged from 0.718 for Anxiety to 0.941 for Behavioural Intention.Nunnally (1975) suggested 0.70 as a standard value of Cronbach's α, and thus, this proves a high internal consistency of the constructs.Fornell and Larcker (1981) recommend that the convergent validity of the measurement model was evaluated by examining composite reliability (CR), average variance extracted (AVE), and the factor loadings.Fornell and Larcker (1981) explain composite reliability (CR) as the collective variance among variables determining a central construct.As specified by Henseler et al. (2014), the value of CR is recommended to reach 0.70 or higher to confirm the high reliability of the measurement.Therefore, as stated in Table 3, all constructs achieved a high value (ranging from 0.847 to 0.954).Henseler et al. (2014) also recommend a standard value of 0.60 or higher for the Average variance extracted (AVE) of each construct to ascertain the convergent validity of the measurement model, and this has been achieved (ranged from 0.604 to 0.839) as demonstrated in Table 3.  Savickas et al. (2002) describe discriminant validity as "the degree to which measures of different constructs are unique."Heterotrait-Monotrait ratio of correlations (HTMT) was conducted to evaluate the discriminant validity.The outputs of HTMT (Table 4) indicate that all the HTMT values are under 0.85 (Henseler et al., 2014), and therefore the results confirmed discriminant validity.This means the constructs are accurately distinct from each other.

Discussion
This study utilized a questionnaire to investigate the impact of external variables, including Feedback, Anxiety, Computer Playfulness, Perceived Enjoyment, and Accessibility, on two critical variables of Technology Acceptance Model (TAM): Perceived Ease of Use (PEOU) and Perceived Usefulness (PU).The primary objective was to understand these factors' influence on students' intentions to adopt e-learning platforms post-COVID-19.Subsequent analysis revealed a significant positive correlation between students' attitudes and their Behavioral Intention (BI), supporting previous research by Amali et al. (2022), Hussein (2017), and Sharma and Chandel (2013a).This finding underscores the idea that a positive attitude towards an e-learning system is a driving force behind a student's intention to continue its use.
The results further underline the significance of PU as a strong determinant, positively impacting both Perceived Ease of Use (PEOU) and Behavioral Intention (BI).These findings resonate with the original Technology Acceptance Model (TAM) proposed by Davis et al. (1989), which suggests that users' acceptance of a learning platform is closely tied to their realization of the system's usefulness and its potential to enhance their performance.The study also showed a significant direct effect of PEOU on Perceived Enjoyment (PA) and Behavioral Intention (BI), aligning with the findings of Salloum et al. (2019).It's evident from these results that users perceiving a digital platform as easy to use or user-friendly are more likely to cultivate a positive attitude and a higher intention to adopt e-learning systems.
Regarding external variables, the study found that Accessibility (ACC) does not have a significant direct effect on Behavioral Intention (BI) to use e-learning systems.This finding diverges from prior studies like Revythi and Tselios (2019) and Wongvilaisakul and Lekcharoen (2015), suggesting that the availability of online learning materials anytime and anywhere does not significantly influence users' behavioural intention.The results also revealed that Feedback (FB), Accessibility (ACC), Anxiety (ANX), Computer Playfulness (CP), and Perceived Enjoyment (PNJ) do not exhibit significant correlations with Perceived Usefulness (PU).These findings contrast with previous research that established positive relationships between FB, CP, and PNJ with PU and negative associations between ANX and PU.This study's outcomes deviate from the existing literature.
Additionally, the study found that Computer Playfulness (CP) does not have a direct impact on Perceived Ease of Use (PEOU), contradicting the findings of Adetimirin (2015).This suggests that users' spontaneous interaction with technology does not necessarily translate to a perception that utilizing an e-learning platform will be straightforward.Despite the disparities observed in this study, there is a consensus among other researchers (Al-Gahtani, 2016;Guo et al., 2013;Hsu, 2019;Kanwal & Rehman, 2017;Park et al., 2019;Park, 2009;Tsai et al., 2020) that quick system access, an enjoyable user experience, and lower levels of technology anxiety contribute to the perception that the e-learning system is easy to use.

Conclusion
The two-year COVID-19 era has made policymakers and educators redefine how education is delivered.With the lifting of COVID-19 restrictions worldwide, neglecting the online education experience and returning to full face-to-face classes is not recommended.This paper helps policymakers and educators to understand the students' behavioural intentions on continuing to use technology in the post-COVID-19 era.The key findings of this paper revealed that students' behavioural intention to continue using technology after COVID-19 is high.The findings also confirmed that students' attitudes have a significant positive impact on their Behavioural Intention.It is important to note that this paper proved that Perceived Usefulness is a strong determinant that positively affects both perceived Attitude and Behavioural Intention.There is also a positive relationship between Accessibility, Perceived Enjoyment and Perceived Ease of Use and a negative association between Anxiety and Perceived Ease of Use.

Implications
The findings of this study have significant practical implications for both e-learning platform developers and educators as they navigate the evolving landscape of online education, especially in the post-COVID-19 era.First, E-learning platform developers should prioritise enhancing the perceived usefulness of their systems.Ensuring that users recognise the value and utility of these platforms is crucial for encouraging their adoption.Furthermore, developers should strive to maintain user-friendly interfaces and easy navigation to improve perceived ease of use.
Our results indicate that both PU and PEOU positively influence attitudes and behavioural intentions.
Our findings also underscore the fundamental role of user attitudes in determining their intentions to continue using e-learning systems.Educators and platform developers should focus on creating engaging, interactive, and enjoyable e-learning experiences to cultivate positive attitudes among students.Encouraging positive perceptions of e-learning platforms can contribute to increased user retention and engagement.
While accessibility (ACC) did not emerge as a significant direct determinant of behavioural intention (BI) in our study, it's essential for developers to provide students with the convenience of accessing educational content from anywhere and at any time.This accessibility remains a fundamental aspect of e-learning systems, ensuring that students can seamlessly integrate learning into their daily routines.
Although our study did not find direct correlations between feedback (FB) and perceived enjoyment (PNJ) with perceived usefulness (PU), these aspects should not be overlooked.Feedback mechanisms play a vital role in improving the user experience, and perceived enjoyment contributes to overall satisfaction.Developers should actively seek and respond to user feedback, integrating features that enhance the overall e-learning experience.

Recommendations
The sudden shift to remote learning during COVID-19 encouraged teachers and students to be familiar with advanced technology tools; however, there was no clear framework that guides teachers and students on how to use the technology tools and for what purpose.This study recommends that policymakers and educators assess the technology used during the pandemic and build a complete framework for how technology might be implemented more efficiently.Students show a high tendency to continue using technology after COVID-19, yet this paper puts forward encapsulating this with a balanced approach that considers Accessibility, Perceived Enjoyment and Perceived Ease of Use.

Limitations
The study provides insights for future directions.First, this study used the quantitative approach in investigating students' behavioural intention to continue using technology in the post-COVID-19 era; thus, future research may triangulate two data sources: quantitative and qualitative.Second, the present study examines students' behavioural intention to use various digital platforms for learning purposes.Therefore, future research may be narrowed to investigate a specific e-learning system.

Model
On the other hand,Teo et al. (2003),Al-Adwan et al. (2013), andAlmaiah et al. (2016) have found that ACC considerably affects both Perceived Usefulness (PU) and PEOU in the context of online communities of learning.

Table 1 .
Scale items

Table 3 .
Reliability analysis and descriptive statistics