Exploring MOOC adoption among aspiring managers – a mediation analysis using the technology readiness and acceptance model (TRAM)

Rapid advancements in information technology (IT) and transformative events like the pandemic have revolutionized sectors globally, including higher education. E-learning involves utilizing technological tools to impart knowledge or skills, with massive open online courses (MOOCs) being the most popular option (Wang, Zhao, Wu, & Goh, 2023). MOOCs involve technology-enabled teaching by academics, used by learners for individual learning. Thus, evolving technology and innovation, the need to stay topical and relevant has heightened the demand for upskilling through MOOCs (Sawant, Thomas, & Kadlag, 2022).

During COVID-19, MOOC adoption surged, with platforms like Coursera reporting 20 million new learners, equal to the enrollments of the previous three years combined (World Economic Forum, 2022). The highest enrollments came from the USA (17.3 million), India (13.6 million) and other Asia-Pacific countries. While the US and China continue to dominate the e-learning market, India is emerging as one of the fastest-growing markets, with the number of users expected to exceed 300 million by 2030 and user penetration to increase to 20% from 15.3% in 2026 (Statista, 2026). The Indian MOOC context is characterized by the digital divide, varying levels of digital literacy and heterogeneity in user capabilities to use technology across rural–urban and various socio-economic clusters. Further, India, as a skills-based economy, places greater emphasis on upskilling and lifelong learning through flexible learning platforms like MOOCs. Online learning in India is mainly self-motivated, which aligns closely with its technological readiness and learning-goal orientation. These factors place India as a unique and relevant context for studying MOOC adoption. Despite the relevance of the Indian context, there is a dearth of robust empirical research studies on MOOC adoption (Virani, Saini, & Sharma, 2023), highlighting the need for further research into the drivers shaping online learning intentions and performance outcomes.

From a service provider’s perspective, understanding the factors influencing MOOC adoption and their impact on task performance is crucial. Recent reports highlight significant growth in online programs, with major platforms such as edX, Udemy and Coursera aggressively expanding their market presence. Despite high enrollments, dropout rates remain elevated (Wang et al., 2023). Thus, understanding consumers’ adoption and continuation journeys is imperative. Since MOOCs are self-driven, learners prioritize quality in delivery, content and return on investment for their education. Evaluating factors contributing to retention and completion requires assessing learners’ self-reported improvements in task performance.

Researchers have investigated factors that facilitate or hinder e-learning adoption (Lee, Hsieh, & Chen, 2013). Various theoretical models, such as task-technology fit (TTF) and technology acceptance model (TAM), explain MOOC adoption but lack comprehensiveness. TTF includes task and technology characteristics but omits users’ beliefs, which are crucial in TAM. Furthermore, theories such as theory of reasoned action (TRA) and TAM assume that positive intention and technology adoption are correlated with enhanced performance, which may not always hold. Self-reporting measures can be subjective and short-duration certifications may not necessarily improve performance.

In addition, researchers argue that existing technology adoption models, such as TAM and TRA, although used extensively, emphasize direct and linear causal relationships between constructs (Davis, Bagozzi, & Warshaw, 1989; Venkatesh, Thong, & Xu, 2016), which do not specify the order or sequence in which the constructs influence adoption behaviors and performance outcomes. In other words, few studies examine the underlying sequence and interplay among mediators between antecedents and outcome variables (Nitzl, Roldan, & Cepeda, 2016; Hair, Hult, Ringle, & Sarstedt, 2017).

To address these gaps and business needs, this study explores the interplay between individual and technological characteristics and their impact on consumers’ MOOC adoption for self-driven skill enhancement and performance. An empirical investigation tackles the following research questions:

The proposed conceptual model was tested and validated on 410 postgraduate management students across India. The results confirm the significant influence of technological readiness and LGO on MOOCs intentions and adoption, offering valuable insights for online learning theory and practice.

MOOCs are technology-enabled learning offerings that deliver boundaryless, structured, high-quality, curated content developed by domain experts. Most MOOCs include concise video lectures and supplementary content, with feedback facilitated through peer review, group collaboration or automated systems. The onus of learning here lies with the user who uses it for skill enhancement and personal development (Akter & Ahmed, 2025).

Against this background, it is important to examine the key motivators that drive learners’ intention and performance outcomes in MOOCs. Prior research indicates that technology-centric factors such as perceived ease of use (PEOU) and perceived usefulness (PU), as well as personality traits, are important determinants of attitude and behavioral intentions (Davis et al., 1989).

Recent research on MOOC adoption (Tseng, Lin, Wang, & Liu, 2022; Virani et al., 2023; Akter & Ahmed, 2025; Huang, Li, & Du, 2025) suggests a shift from technology-centric models toward a more holistic understanding (Gupta, 2020) that incorporates prior technological experiences, cultural background and individual experiences, social and institutional support, self-regulation and performance outcomes. However, MOOC studies in India (Virani et al., 2023; Mohan, Upadhyaya, & Pillai, 2020; Meet, Kala, & Al-Adwan, 2022; Shah and Khanna, 2022) are still largely grounded in the TAM and unified theory of acceptance and use of technology frameworks. A recent work by Singh, Goyal, Joshi, Dhir, and Kumar (2026) emphasizes the importance of individual factors, such as goal orientation, technology readiness (TR) and self-regulation, in driving MOOC intention. Moreover, recent findings highlight a disconnect between intention and actual task performance. Most existing studies are replication models of direct effect with no explanation of the underlying mediation effects.

This study addresses the literature gaps by examining the role of personality and attitude factors in adoption. This combined approach of examining the links between personality traits, attitudes and technology acceptance is crucial for advancing theories and improving practical online education strategies that are better suited to learner diversity and help ensure course completion and lower dropout rates.

The TAM, proposed by Davis et al. (1989), explains adoption in information systems. TAM posits that PU and PEOU shape users’ attitudes toward technology adoption. PU is the belief that a system enhances job performance, while PEOU is the belief that using it is effortless. Initially, TAM included attitude as a mediator, but it was later excluded due to its weak effect (Chatzoglou, Sarigiannidis, Vraimaki, & Diamantidis, 2009).

Later, Lin, Shih, and Sher (2007) integrated TR into the TAM to propose the technology readiness and acceptance model (TRAM), explaining user adoption of technology-enabled innovations.

The TTF theory, proposed by Goodhue and Thompson (1995), asserts that technology acceptance relies on the alignment between tasks and individual characteristics. TTF explores the relationships among task requirements, technology capabilities, individual attributes and their effects on utilization and performance. In e-learning, task characteristics encompass relevant content, affordability, accessibility, evaluation and reporting, while technologies serve as the tools used to carry out these tasks. Individual characteristics, including motivation, TR and LGO, affect how effectively users engage with the technology.

Consumer research posits that personality, attitude and behavior are significant drivers of behavior. This paper examines TR and learning goal orientation (LGO) as critical personality traits that drive e-learning intention.

TR is an overall state of mind, influenced by mental enablers and inhibitors, that encourages the adoption of new technologies (Parasuraman, 2000). TR includes optimism (belief in technology’s benefits), innovativeness (leading in technology use), insecurity (distrust) and discomfort (feeling overwhelmed). These feelings vary across individuals, affecting TR. Self-efficacy, similar to innovativeness, examines an individual’s ability to mobilize motivation and cognitive resources (Chatzoglou et al., 2009).

LGO is “a tendency for individuals to desire to develop the self by acquiring new skills, mastering new situations and improving one’s competence” (VandeWalle & Cummings, 1997). LGO significantly influences individuals’ motivation to learn and master skills, fostering positive behaviors. Prior research has demonstrated that LGO positively affects academic self-efficacy and work performance (Lu, Deng, Yao, & Li, 2022). A student with a high LGO aims to acquire new skills and knowledge to showcase mastery (Rhee, Verma, Plaschka, & Kickul, 2007).

While TTF models include task, technology and individual characteristics, they do not account for users’ perceived beliefs regarding the technology, which is the core construct of TAM. In addition, prior theories of technology acceptance (TRA, TAM) assume that positive intention and technology adoption correlate with enhanced performance. This assumption may compromise predictive accuracy for two reasons: self-reporting measures of technology acceptance are subjective and may differ from objective observations and the intention to adopt technology may not necessarily lead to improvements (Lin et al., 2007; Goodhue & Thompson, 1995; Venkatesh et al., 2016).

To address these gaps, this study proposes integrating TAM with TR, LGO and TTF to enhance understanding of e-learning decisions and subsequent performance improvements. The current study is an endeavor in this direction.

Based on the integrated model, the hypothesized relationships are discussed in the following sub-sections.

User innovativeness refers to a tendency to try out new technologies (Parasuraman, 2000) and to be a technology leader (Lin et al., 2007). Prior literature has validated the relationship between user innovativeness with technology and PEOU. Walczuch, Lemmink, and Streukens (2007) found that high personal innovativeness leads to higher PEOU. Lin et al. (2007) proposed the TRAM as an extension of the TAM, including TR to explain consumers’ adoption of innovations. They found a positive correlation between TR propensities and PEOU.

Studies among Indian consumers found that high personal innovativeness positively influenced PEOU (Roy & Moorthi, 2017). Mukerjee et al. (2019) noted that techno-ready explorers were more likely to adopt mobile self-checkout services when they perceived them as easy to use. User innovativeness and TR have been shown to positively influence PU across various geographies. In Taiwan, Lin et al. (2007) identified TR as a causal antecedent of PU. Roy and Moorthi (2017) evaluated the positive effect of TR on PU among Indian users. Based on this discussion, it is hypothesized that:

Optimism, the belief that technology brings flexibility and efficiency, is essential for technology acceptance. Optimists view technologies as easy to use and valuable. Research by Walczuch et al. (2007) indicated that higher levels of optimism lead to greater PEOU and PU. Optimistic students place a higher value on enrolling in technology-enabled courses (Rhee et al., 2007).

Roy and Moorthi (2017) found that optimism and TR have significant positive effects on PU and PEOU in m-commerce adoption. Lin et al. (2007) reported similar findings for online stock trading systems. Other researchers have also discovered a positive influence of optimism on PEOU and PU (Rahman, Taghizadeh, Ramayah, & Alam, 2017; Park, Ha, & Jeong, 2021). Therefore, it is hypothesized that:

Research indicates that individuals with high levels of insecurity view technology as complex and difficult to use, making it essential to mitigate TR inhibitors such as insecurity (Lin and Chang, 2011). Walczuch et al. (2007) found that heightened personal insecurity diminishes PEOU. Mukerjee et al. (2019) discovered that consumers experiencing high insecurity are distrustful and uneasy with technology, leading to a decrease in their use of mobile self-checkout services. Rhee et al. (2007) found that students’ insecurity regarding online classes negatively impacts enrollment utility. Similarly, Previous researchers (Rahman et al., 2017) corroborated that insecurity affects both PEOU and PU. Therefore, it is hypothesized:

According to Walczuch et al. (2007), individuals with high discomfort experience a lack of control and feel overwhelmed by technology, which negatively influences PEOU but has an insignificant effect on PU. Asian studies corroborate these results in various contexts: electronic medical records, technology acceptance among Bangladeshi micro-entrepreneurs (Rahman et al., 2017) and self-service technology in fashion retail stores (Park et al., 2021). In addition, constructs like computer anxiety also negatively influence PU. Therefore, consumers who are uncomfortable with technology are unlikely to find it easy or valuable in the long run. Thus, it is hypothesized that:

An individual with a high LGO is eager to acquire new skills and demonstrate mastery (Rhee et al., 2007). Studies show that LGO positively affects PEOU and significantly influences intention in web-based systems (Chatzoglou et al., 2009). Thus, users with higher LGO perceive e-learning technologies as easy to use and useful, thereby increasing adoption intentions and positively affecting (TASK-PERF). Consequently, it is hypothesized:

PEOU enhances the PU of technology. Consumers’ perception of ease of use is positively associated with its usefulness (Lin et al., 2007). PEOU is an antecedent of PU, meaning that easier-to-use technology increases PU (Davis et al., 1989; Walczuch et al., 2007). Conversely, complex technology is perceived as useless (Walczuch et al., 2007). Chatzoglou et al. (2009) argued that a system perceived as easy to use is more useful. This positive relationship is confirmed in various contexts (Mukerjee, Deshmukh, & Prasad, 2019; Chiu & Cho, 2021; Mailizar, Burg, & Maulina, 2021). Recent studies in the context of MOOC continue to validate this relationship (Liu, 2021; Virani et al., 2023; Harnadi, Widiantoro, & Prasetya, 2024). This reasoning leads to the following hypothesis:

Consumers are more likely to adopt innovations if they find them easy to use (Lin et al., 2007). Studies indicate that PEOU directly and indirectly impacts users’ intention (INT) to embrace new technology (Davis et al., 1989; Cheng, 2015; Meet et al., 2022). For instance, Chatzoglou et al. (2009) discovered a positive effect of PEOU on INT for web-based training programs. Roy and Moorthi (2017) reported that PEOU positively influenced the adoption of m-commerce services. Mukerjee et al. (2019) observed a positive correlation between Indian customers’ PEOU and their likelihood of utilizing self-checkout services. Similarly, Chiu and Cho (2021) found that PEOU affects the adoption of multimedia messaging services and health apps.

For new or existing technologies, users primarily evaluate PU before forming an intention to use them (Shan and Khanna, 2022). Lin et al. (2007) found PU to be a critical determinant of usage intention. This finding is supported by Chatzoglou et al. (2009) and Cheng (2015). Recent e-learning studies have also validated the relationships between PEOU, PU and intention (Mailizar et al., 2021; Vanitha & Alathur, 2021; Meet et al., 2022; Shah and Khanna, 2022; Harnadi et al., 2024; Huang et al., 2025; Akter & Ahmed, 2025). Therefore, based on the above literature, the following hypotheses are proposed:

According to Goodhue and Thompson (1995), performance impacts are influenced by TTF and utilization. According to Cheng (2024), little attention has been paid to verifying if the intention to adopt e-learning enhances task performance (TASK-PERF). Lin (2012) found that the use of learning management systems and virtual learning systems positively affects performance. An increased intention to adopt MOOCs should enhance performance, as these technologies enable learning, communication and productivity with greater precision and flexibility. Thus, it is hypothesized:

As previously mentioned, LGO positively affects INT, PEOU and PU. The TAM posits that PEOU and PU determine an individual’s intention to use a system. Prior studies have confirmed the mediating role of PEOU and PU on behavioral intention. Humida, Al Mamun, and Keikhosrokiani (2022) found that PU and PEOU mediated the relationship between the predictors and the outcomes of e-learning adoption among Bangladeshi students. Tan et al. (2024) observed that ease of use and usefulness partially mediated the relationship between perceived service quality and students’ attitudes toward hybrid learning in universities. Cheng (2015) suggests that learning goal-oriented employees perceive e-learning systems as easy to use and useful, thereby increasing their acceptance and use. Thus, it is hypothesized:

Combining H10a and H10b, the hypothesis below can be formulated:

When users perceive a MOOC system as easy to use, it enhances their perceived utility, fostering positive intentions. Thus, combining H5a, H6 and H8 leads to the following hypothesis:

Combining H10c and H10d leads to the following hypothesis:

Control variables such as gender, age and work experience were included for the key outcome constructs – intention and task performance, as they may influence MOOC adoption and performance. The results show that none of the control variables had any significant influence on endogenous constructs. Thus, there was no confounding effect of the control variables, indicating the model’s robustness.

The proposed direct and indirect relationships, along with control variables, are presented in Figure 1.

An online survey instrument was designed with five sections to measure nine study constructs. Section 1 assessed respondents’ comfort with technology platforms (e.g. email, social networking, blogs, gaming, browsing) and their TR. Section 2 measured LGO. Section 3 included questions about PEOU, PU and MOOC intention. Section 4 contained self-reporting questions related to task performance (TASK-PERF). The final section captured demographic details (gender, age group, highest qualification, educational background and years of experience). All constructs were measured on a five-point Likert scale, ranging from strongly disagree (1) to strongly agree (5). In addition, user confidence levels with various technology platforms were gauged on a five-point scale from not at all confident (1) to very confident (5).

Standardized scales from the literature were utilized: TR sub-dimensions (optimism, innovativeness, discomfort and insecurity) adapted from Parasuraman (2000), LGO items from Rhee et al. (2007) and items for PEOU, PU and intention from Walczuch et al. (2007) and Chatzoglou et al. (2009). The instrument was pretested with ten management students to ensure clarity and representativeness before the main study.

A purposive sampling technique ensured that respondents had previous experience with essential productivity tools (e.g. data visualization, statistical analysis, spreadsheets and databases). The study, which focused on technological readiness and LGO, gathered responses from Indian management institutes between June and December 2023. Of 2,200 respondents contacted, 206 offline and 294 online responses were received, resulting in a response rate of 22.8%. After screening, 410 usable responses remained. The sample included 286 males (69.8%) and 124 females (30.2%). Age distribution was as follows: 20–22 years (24.1%), 23–24 years (46.3%), 25–28 years (23.9%) and above 28 years (5.6%). In terms of education, 328 (80%) had a bachelor’s degree while 82 (20%) were postgraduates. Backgrounds included engineering (50%), commerce (30.7%), science (11.2%), arts (5.1%) and other courses (2.9%). Regarding work experience, less than 1 year accounted for 28.9%, 1–3 years for 35.9% and over 3 years for 11.2%.

At the time of the study, the authors did not have a formal institutional ethics review board. Participation was voluntary and anonymous and respondents were informed of the study’s objectives. No personally identifiable or sensitive information was collected.

Data analysis utilized partial least squares–structural equation modeling with SmartPLS 4.1.0.9, which is suitable for exploratory research and predictive purposes, particularly with small sample sizes. The sample size met the criterion of being at least ten times the largest number of inner-model paths (Hair et al., 2017).

Before testing the measurement and structural model, the data were examined for common method bias (CMB). As suggested by Chin, Thatcher, Wright and Steel (2013), to assess CMB, a PLS latent marker variable approach was used, whereas a theoretically unrelated construct – in this case, “Confidence with technology platforms” was included as a marker variable in the model. The five technology platforms were email, online music, news and internet TV. These four items were attached to the marker construct, which was connected to all the endogenous constructs. The results indicate that the relationship between the marker variable and the endogenous constructs is not significant. Further, after adding the marker construct, there is no substantial change in the explanatory power, as indicated by R², suggesting that CMB was not a concern and unlikely to confound the study’s results.

The measurement model was evaluated for internal consistency, composite reliability, convergent validity and discriminant validity. Cronbach’s alpha ranged from 0.692 to 0.883, which meets the 0.7 threshold for most constructs. Items with loadings below 0.6 were excluded (Hair et al., 2017). Composite reliability values surpassed 0.7 and the average variance extracted (AVE) was at least 0.5, confirming convergent validity as indicated in Table 1. Discriminant validity (see Table 1) was established using the Fornell and Larcker (1981) criterion, cross-loadings and heterotrait–monotrait ratio (HTMT). The square root of the AVE for each construct was greater than inter-construct correlations and HTMT values ranged from 0.057 to 0.792, remaining below the 0.9 threshold, confirming discriminant validity.

Before validating the structural model, multicollinearity diagnostics were performed. The variance inflation factor values for all constructs ranged from 1.099 to 1.520, which are below the threshold of 3 (Hair et al., 2017), indicating no multicollinearity.

The structural model was estimated using bootstrapping with 5,000 subsamples. Table 2 summarizes the path coefficients (beta values) and their significance (p-values). Of the proposed 16 hypotheses, 10 were significant and accepted, while the remaining 6 (H3a, H3b, H4a, H4b, H5b and H5c) were not significant but indicated the hypothesized direction of influence.

The first eight hypotheses (H1a to H4b) examined the influence of innovativeness, optimism, insecurity and discomfort on PEOU and PU. Positive relationships were identified for innovativeness and optimism, while negative relationships were observed for insecurity and discomfort; however, only the first four were significant.

Furthermore, LGO positively influenced PEOU, PU, INT and TASK-PERF, showing significant effects in H5a and H5d. PEOU positively impacted both PU and INT, supporting H6 and H7. PU’s effect on INT was significant (H8), while INT’s influence on TASK-PERF (H9) was also significant.

The structural model was evaluated using the coefficient of determination (R²), effect sizes (f²), predictive relevance (Q²), path coefficient estimates and overall model fit. The model’s explanatory power for each endogenous construct was assessed using R² values: 0.293 for PEOU, 0.326 for PU, 0.544 for INT and 0.099 for TASK-PERF, all indicating satisfactory explanatory power.

Effect size (f²) was used to evaluate the relative impact of each exogenous construct on the endogenous constructs. According to Cohen (1988), f² values of 0.02, 0.15 and 0.35 signify small, medium and large effects, respectively. PEOU, INNO (f² = 0.079), OPTM (f² = 0.029) and LGO (f² = 0.074) exhibited small effects. PU, PEOU (f² = 0.155) demonstrated a medium effect, while other exogenous constructs showed negligible effects. INT, PU (f² = 0.467) displayed a large effect and PEOU (f² = 0.094) indicated a small effect. TASK-PERF, LGO (f² = 0.043) and INT (f² = 0.030) presented small effects.

Predictive relevance (Q²) was evaluated using cross-validated redundancy with an omission distance of seven. Q² values were 0.266 for PEOU, 0.191 for PU, 0.211 for INT and 0.080 for TASK-PERF, indicating predictive relevance.

The model’s overall fit was assessed using the standardized root mean square residual (SRMR). The SRMR value was 0.057, significantly below the threshold of 0.80, which indicates a good fit.

To test for mediation, the method proposed by Nitzl et al. (2016) was utilized. PEOU and PU were hypothesized to mediate the relationship between LGO and INT, allowing for tests of both parallel and serial mediation (H10a to H10e). In H10a, PEOU significantly mediated the LGO–INT relationship (indirect effect = 0.066, p = 0.000), while the direct effect was not significant (0.053, p = 0.198), indicating partial mediation with a variance accounted for (VAF) of 31.4%. In H10b, PU did not mediate the relationship, as both the indirect and direct effects were insignificant, leading to the acceptance of H10a. PEOU had a stronger indirect effect (0.066) compared with PU (0.037).

For H10c, the examination of PEOU and PU as parallel mediators revealed a significant total indirect effect of 0.103. For H10d, PEOU and PU served as serial mediators in the LGO–INT relationship, demonstrating a serial indirect effect of 0.054, indicating partial mediation with VAFs of 49% for parallel mediation and 25.7% for serial mediation.

Regarding H10e, the total indirect effect of LGO on INT through PEOU and PU was significant (0.157, p = 0.000), while the direct effect remained insignificant (0.053, p = 0.198). This indicates both parallel and serial mediation, with a VAF of 74.8%, showing that PEOU and PU serve as complementary partial mediators in the LGO–INT relationship. Thus, H10e is supported. Refer to Table 3 for mediation results.

In summary, the results confirm that both PEOU and PU mediate the relationship between LGO and INT, with PEOU serving as the stronger mediator. This validates the hypothesized model, demonstrating the robustness of the integrated model combining TAM with TR and LGO in explaining MOOC adoption and performance.

E-learning and MOOCs complement traditional learning. Understanding the intentions behind MOOCs requires a distinct investigation given corporate demands for continuous skill updates. MOOC platforms must evaluate improvements in task performance to enhance users’ intentions to continue using the platform (Cheng, 2024).

This study provides insights into MOOC adoption by integrating TR and LGO with the TAM. Survey data from 410 Indian management students revealed that characteristics such as innovativeness, optimism, insecurity, discomfort and LGO significantly influence PEOU, PU and MOOC intentions. The findings highlight the importance of considering both the positive and negative dimensions of TR in MOOC adoption.

All four TR dimensions—innovativeness, optimism, insecurity and discomfort—were determinants of PEOU and PU. Innovativeness and optimism significantly influenced both PEOU and PU. Interestingly, the relationship between insecurity with both PEOU and PU was found to be insignificant. A plausible explanation is that MOOCs are generally considered low-risk platforms compared to other digital platforms such as e-commerce or digital banking. Further, the respondents comprise management students who are familiar with and comfortable using various digital platforms, including MOOCs. This finding aligns well with prior studies in e-learning (Rahman et al., 2017), which did not identify security as a major concern.

Discomfort did not significantly affect PEOU or PU. This finding aligns with recent studies (Mukerjee et al., 2019). A possible explanation is that young digital natives, who became familiar with e-learning tools during the pandemic, may not view MOOCs as significantly enhancing the teaching and learning process compared to physical classrooms. Path coefficients indicated that the effect of innovativeness on PEOU was greater than that of optimism, whereas the effect of discomfort on PEOU was smaller than that of insecurity.

LGO is a significant determinant of PEOU and task performance. This finding aligns with previous findings suggesting that students with a LGO are intrinsically motivated to learn and improve their skills (Chatzoglou et al., 2009).

The present study makes several theoretical contributions to MOOC literature. First, the study extends TRAM by integrating LGO into the framework. This integration reaffirms the crucial role of PEOU and PU, alongside individual traits, in technology adoption, consistent with previous TAM studies.

Second, the present study also aims to explore multiple and serial mediation effects of PEOU and PU on the relationship between LGO and MOOC intentions. Mediation analysis reveals that PEOU and PU not only directly influence MOOC intentions but also significantly mediate the relationship between LGO and MOOC intentions. This indicates that students’ perceptions of ease and usefulness significantly enhance their intention to adopt MOOC platforms, thereby improving task performance. In other words, intentions must translate into performance outcomes to ensure sustainable adoption.

Finally, the study offers context-specific insights into MOOC adoption, considering heterogeneity in the digital divide and varying levels of TR.

Our study offers several managerial implications. First, the study finds that PU and PEOU are the strongest antecedents of intention. Organizations offering MOOC tools must ensure the platform is easy to understand and use without requiring significant mental effort. MOOC platforms perceived as user-friendly enhance their utility and increase the likelihood of adoption. Training on tool features to support MOOC helps alleviate inhibitions and boost adoption intentions. Including interactive elements such as creative games, heuristic methods, online chat rooms and discussion boards can make learning more engaging and satisfying (Cheng, 2015). MOOC platforms should prioritize user experience design, making the tools intuitive and beneficial for academic and professional growth. Offering introductory sessions and tutorials can help build confidence while reducing anxiety and discomfort associated with new technology.

Second, the direct impact of LGO on task performance highlights the intrinsic motivation of students with high LGO to leverage MOOC tools effectively. By understanding students’ specific needs and characteristics, MOOC providers can design targeted interventions that cater to different segments, ultimately increasing overall adoption and reducing dropout rates.

Third, the mediating role of PEOU and PU between LGO and intention offers MOOC platform providers insights into alternative pathways to drive adoption, including ease of navigation and interface design, as well as interventions to improve productivity and efficiency.

The social implications of this study are multifaceted. By identifying how TR and learning orientation drive MOOC adoption, the study underscores the potential of MOOCs to bridge educational inequities and democratize access to quality learning opportunities for diverse populations worldwide. Moreover, the findings highlight how enhancing learner confidence, innovation and motivation in digital environments can enhance digital inclusion and workforce employability.

This work represents a first step in understanding the significance of TR and LGO in shaping MOOC intention and satisfaction.

Even though this study provides valuable insights, it has several limitations. First, it is a cross-sectional study involving students from various management institutes in India. Although the sample is diverse in socio-economic strata, educational backgrounds, learning cultures and preferences, it limits the generalizability of the findings to other countries and contexts. Future research should include a more diverse sample and provide a comprehensive understanding of e-learning adoption across various cultural and educational contexts.

Second, data collection relied on self-administered surveys, which may introduce potential biases and inaccuracies. Future research should use qualitative methods, such as personal interviews and focus groups, to investigate characteristics like self-efficacy, compatibility, facilitating conditions and information quality. These methods can reveal underlying motivations and barriers, thereby enriching quantitative findings.

Third, the study did not examine moderating variables that can significantly influence adoption. Future research should explore the effects of variables such as age, gender and prior technology experience. Understanding these interactions can provide deeper insights into the nuances of MOOC adoption.

In addition, examining the impact of specific MOOC features and content types on user satisfaction and performance can yield actionable insights for MOOC platform developers. Future studies could also investigate the long-term effects of MOOC adoption on academic and professional outcomes, offering a longitudinal perspective on the benefits and challenges of MOOCs.

Author contribution: Himanshu Joshi: conceptualization (Equal), data curation (Equal), formal analysis (Equal), methodology (Equal), writing – original draft (Equal), writing – review and editing (Equal).

Neena Sondhi: data curation (Equal), formal analysis (Equal), methodology (Equal), writing – original draft (Equal), writing – review and editing (Equal).