Current Issue

Presence in a metaverse learning environment is described as the sense of "being there" within a virtual space, which is crucial for immersive learning experiences[11]. Research in virtual learning contexts highlights that when learners feel more present in such environments, they are more likely to report higher satisfaction due to enhanced engagement and interaction with the content[12][13]. Similarly, studies have found that satisfaction in educational settings is heavily influenced by the quality of the immersive experience, suggesting a direct correlation between the sense of presence and overall satisfaction[14][15]. Thus, this study suggests the following hypotheses.

H1. Presence positively affects Satisfaction.

Interactivity in educational technology is characterized by the dynamic interaction between learners and digital content, which facilitates active learning and engagement[11]. It has been consistently shown that higher levels of interactivity can lead to increased learner satisfaction, as these interactions make learning experiences more personalized and responsive[16][17]. This positive association is supported by the premise that more engaging and interactive environments contribute significantly to positive educational outcomes, including learner satisfaction[18]. Thus, this study suggests the following hypotheses.

H2. Interactivity positively affects Satisfaction.

Cyber sickness, characterized by symptoms such as dizziness, nausea, and eye strain, typically arises from immersive digital environments[19]. Research indicates that experiencing cyber sickness can significantly detract from user satisfaction, as discomfort directly undermines the usability and overall experience of learning technologies[20][21]. This inverse relationship is further emphasized in studies linking the severity of cyber sickness symptoms to reduced satisfaction and hindered learning outcomes[22][23]. Thus, this study suggests the following hypotheses.

H3. Cyber sickness negatively affects satisfaction.

Loading time refers to the duration it takes for digital content and interfaces to become fully operational and accessible to the user[24]. Studies have shown that prolonged loading times can lead to frustration and a decrease in user satisfaction, as delays disrupt the learning flow and reduce the efficiency of the educational experience[25][26]. The negative impact of loading time on satisfaction is significant, suggesting that smoother, quicker access to learning resources is crucial for maintaining learner engagement and satisfaction. Thus, this study suggests the following hypothesis.

H4. Loading time negatively affects satisfaction.

For this study, the questionnaire design included two distinct parts, utilizing a five-point Likert scale ranging from "strongly disagree" to "strongly agree" to measure responses. The first part collected demographic information, including age, gender, education level, and prior experience with metaverse platforms. The second part assessed users' perceptions of the main constructs related to the metaverse learning environment. The questionnaire items were initially developed in Korean to ensure clarity and relevance to the local context, drawing from existing validated scales and literature[11][19][24][27][28]. Table 1 organizes the questionnaire items into five key constructs. For instance, the presence construct includes items such as the learner's sense of being in a metaverse environment and recognizing its difference from the real world.

The target sample for this study comprised students from Tech university of Korea currently enrolled in either the 'Future VR Lab' course or the 'Meeting with a Robotics Celebrity' lecture. These subjects were selected as they represent a technologically engaged demographic likely to have familiarity with metaverse environments, enhancing the relevance and applicability of the findings to educational technologies.

The survey was conducted from April 20, 2022, to May 13, 2022, using both online and offline methods to ensure broad participation. Data collection was performed through an online survey distributed via Naver Form, a platform chosen for its accessibility and ease of use, which is crucial for maximizing response rates. The voluntary nature of participation was emphasized, and anonymity was strictly maintained to protect respondent privacy and encourage honesty in responses.

A total of 151 responses were initially collected. After data preprocessing to ensure quality and accuracy, 55 responses were removed due to incompleteness or random answering patterns. Consequently, 96 responses were retained for the final analysis. This filtering process was essential to maintain the integrity and reliability of the data used in the study. The sample size of 96 for this study is considered adequate for the analysis, given the complexity of the model which includes 5 latent variables and 14 observed variables. According to guidelines Wolf, Harrington[29], a minimum sample size of 10 times the largest number of structural paths directed at a single latent variable in the model is suggested. This rule ensures sufficient statistical power for the analysis and valid results.

Table 2 categorizes 96 respondents by gender and academic grade, showing that 63.5% were male and 42.7% were seniors, reflecting a diverse yet academically mature sample.

The measurement model assessment in this study utilized factor analysis and reliability tests, confirming the validity and reliability of the constructs. Each construct's reliability was verified through Cronbach's Alpha, Composite Reliability(CR), and Average Variance Extracted(AVE) scores, all detailed in Table 3. Presence, interactivity, cyber sickness, loading time, and satisfaction demonstrated high internal consistency, with Cronbach's Alpha values well above the acceptable threshold of 0.7, indicating substantial reliability[32]. Furthermore, the factor loadings for all items were notably strong, significantly exceeding the recommended threshold of 0.7, suggesting that the items were well-suited to measure the intended constructs[33]. The AVE values for each construct were above 0.5, confirming adequate convergent validity, which measures the extent to which items of a specific construct converge in comparison to measures of different constructs.

Table 4 provides the correlation matrix along with discriminant assessment, where the diagonal elements represent the square root of AVE and confirm that all constructs share more variance with their indicators than with other constructs, thereby establishing discriminant validity[33].

SEM analysis was performed to evaluate and verify the proposed relationships between the constructs of this study. A bootstrap resampling technique, utilizing 5000 resamples, was employed to determine the significance of the hypotheses in the research model. The outcomes of this analysis are detailed in Table 5.

This study contributes to the existing literature on metaverse learning by exploring the nuanced relationships between various factors and satisfaction. Previous studies have primarily focused on isolated aspects of learner satisfaction in virtual environments, such as presence or interactivity[11][17]. However, they often did not fully explore the combined effects of cyber sickness and loading times alongside these factors. Our comprehensive approach reveals the interdependencies and collective impact of these variables on satisfaction within the metaverse. The empirical findings underscore the critical role of presence and interactivity in enhancing learner satisfaction. This research bridges that gap by providing a holistic view of how these elements interact and influence learner satisfaction within the metaverse.

A significant theoretical advancement of this study is the understanding of how cyber sickness negatively impacts satisfaction, an area less covered in prior research. While previous studies have acknowledged the discomfort caused by cyber sickness[20][21], they did not fully quantify its direct impact on learning satisfaction compared to other variables like presence and interactivity. The findings of this study suggest that mitigating cyber sickness is just as crucial as enhancing presence and interactivity for improving learner satisfaction.

Additionally, the non-significant impact of loading time on satisfaction contradicts previous assumptions[25][26], indicating that users may prioritize content quality over speed, which shifts the focus for future technology enhancements. This revelation prompts a shift in how developers might prioritize elements of metaverse design, suggesting a potential reevaluation of resource allocation towards enhancing interactive and immersive qualities rather than merely reducing loading times. By presenting these nuanced insights, this study not only guides future empirical and theoretical work but also assists practitioners in designing more effective and satisfying metaverse learning environments.

The practical implications of this research are profound for various stakeholders in the realm of educational technology, particularly in metaverse learning environments. For educational institutes, the findings suggest that investing in technologies that enhance presence and interactivity should be a priority, as these factors are strongly associated with increased learner satisfaction. Providing students with immersive and interactive learning experiences can lead to higher engagement and potentially better educational outcomes.

For metaverse developers, the study underscores the importance of addressing cyber sickness by optimizing system performance and enhancing user interface design to minimize discomfort. Such improvements can significantly enhance user satisfaction, retention, and overall success in metaverse applications. Additionally, the insignificant role of loading time suggests developers might focus more on content richness and interactive features rather than solely on reducing load times.

Practitioners, including educators and curriculum designers, should consider these factors when developing or selecting metaverse platforms for educational purposes. By focusing on enhancing the quality of presence and interactivity while minimizing cyber sickness, they can create more effective and satisfying learning environments for students.

These recommendations can guide the development of more effective and user-friendly educational technologies, ultimately enhancing learning experiences and outcomes in the increasingly popular digital learning spaces.

This study, while comprehensive, has limitations that pave the way for future research. The cross-sectional nature of the data limits our ability to draw conclusions about the causality of relationships between presence, interactivity, cyber sickness, loading time, and satisfaction. Longitudinal studies could provide deeper insights into how these factors impact satisfaction over time. Additionally, the study was confined to a single metaverse platform, which might limit the generalizability of the findings. Future research should explore these relationships across various platforms and with diverse user groups to validate and extend the applicability of the results, potentially exploring the role of demographic variables in moderating these effects. Moreover, this study did not incorporate variables such as information security, which is often emphasized in online platforms[35]. Future research should comprehensively include variables like information security and individual sensitivity to security to enhance the explanatory power of the model. Finally, the analysis being limited to a single metaverse platform may restrict the generalizability of the findings. Recognizing the potential variability across different platforms, future research should examine these dynamics within a broader range of virtual learning environments to ascertain the consistency of the observed effects.