Next Article in Journal
Learning a Convolutional Autoencoder for Nighttime Image Dehazing
Previous Article in Journal
Measurement of Text Similarity: A Survey
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Information
Volume 11
Issue 9
10.3390/info11090423
information-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

From Traditional to VR-Based Online Education Platforms: A Model of the Mechanism Influencing User Migration

by
Jing Chen
,
Chang Liu
,
Ronghua Chang
,
Pengfei Gui
and
Sanggyun Na
*
School of Business Administration, Wonkwang University, 460 Iksandae-ro, Iksan, Jeonbuk 54538, Korea
*
Author to whom correspondence should be addressed.
Information 2020, 11(9), 423; https://doi.org/10.3390/info11090423
Submission received: 13 July 2020 / Revised: 16 August 2020 / Accepted: 19 August 2020 / Published: 31 August 2020
(This article belongs to the Special Issue The Evolutions of Blended Learning: New Forms of Mixed Learning)
Browse Figures
Versions Notes

Abstract

:
VR technology can help create optimal virtual learning spaces. Such spaces offer new visual experiences that break through the limitations of time and space and greatly stimulate people’s imagination and creativity in learning. Currently, the bandwidth required for such spaces limits the large-scale application of virtual reality (VR) technology for this purpose. With the large-scale deployment and application of high-speed networks, however, online education platforms based on VR technology will be better able to meet the diversified and personalized learning needs of learners. To promote the development and popularization of new online education platforms based on VR, the factors influencing the migration of learners from traditional online education platforms to new platforms need to be understood more clearly. A model based on the theory of negative, positive, and anchoring effects can explain learners’ migration behavior in this connection. To this end, a structural equation model based on the PLS variance algorithm was used to analyze data obtained through offline and online questionnaires. It was found that in terms of “negative effects”, the afunction and loyalty associated with traditional online education platforms reduced learners’ willingness to migrate to new platforms based on VR technology. In terms of “positive effects”, the novel interactivity and personalization brought by the new platform increased the willingness of users of traditional platforms to migrate to new platforms. In terms of “anchoring effects”, the system quality and relationship quality of learners’ use of traditional online education platforms, as well as the transfer costs associated with the new platform, generated learners’ risk perception about platform migration. In addition, risk perception not only negatively affects learners’ migration to the new platforms, but also strengthens their cognition of the system quality and relationship quality of the traditional platforms, while reducing their interactive awareness of those platforms. Therefore, by adjusting the psychological component of virtual learning, the online education platforms based on VR technology can create high-quality platforms migrating from traditional platforms.

1. Introduction

With the emphasis on lifelong learning becoming an important trend, online education, thus far, has been treated as a kind of web-based mode of distance education [1]. The number of users of online education platforms has grown rapidly in recent years, because these platforms are not restricted by considerations of space and time, and they offer flexible and convenient ways of teaching as well as rich teaching resources [2]. These advantages, for some users, make online education platforms an attractive alternative to what they see as the insufficiency and flaws of traditional offline education [3,4]. The number of online education platform users in China is expected to reach 296 million by 2020 [5].
With the rapid development of science and technology, cloud computing, big data, virtual reality (VR), and other technologies have had an increasing impact on the wider population [6]. Accordingly, one of the most urgent problems in the field of education centers on how to provide an informed analysis of students’ educational needs and learning conditions, in order to develop the best learning plan for students via personalized intelligent educational platforms [7]. VR, an important branch of computer science, is a research development that has been used in the simulation and extension of the theories, methods, technologies, and applications of intelligent systems [8]. VR is based on the computer technologies at the core of modern high-tech applications; it generates interactive 3D environments to provide immersive sensory experiences, allowing for unprecedented simulations of real experience, which are marked by strong interactivity [9]. Although VR has been used for recreational purposes, it has also been applied to the economic, health-care-related, educational, and national-defense-related issues [10]. In the educational field in particular, VR has a wide range of applications, and has become increasingly influential [11,12].
Representing the innovative potentials of Internet technology at the present stage, VR systems have already garnered attention internationally, for popular- science initiatives as well as in other contexts [13]. VR plays a positive role in enhancing user participation and improving user experience. These advantages can be seen in the construction of a virtual science museum in Japan by the science organization JST [14]. This museum uses multimedia technology and VR to create scene simulations via 3D imaging. The museum’s dinosaur exhibit is a popular product enabled by these technologies [15]; it simulates the earth’s history and the evolution of the environment, helping students better understand life on earth as well as the evolutionary histories of dinosaur species. Likewise, the US’s Weather Channel, on television, uses augmented reality (AR) technology to explore issues of ice safety and show how climate change affects ice [16]. In this way, AR raises public awareness of weather phenomena and their impact on people’s everyday lives. By extension, it can be argued that the empowerment of VR (and AR) technologies in the field of education will bring unprecedented innovation and new opportunities for learning [17,18].
However, an important research topic at this stage, when online education platforms based on VR technology are beginning to be built, concerns how to induce learners to migrate from traditional online education platforms to VR-enabled platforms, so as to quickly realize the network effect of the new generation of platforms [19].The migration cost to learners moving from existing platforms to platforms based on VR technology needs to be studied, especially when such migration makes it necessary for them to switch between different platform service providers [20]. According to the idea of status quo bias, risk perception makes people unwilling to change the status quo even when facing better choices, so risk perception may also reduce learners’ willingness to migrate [21]. More generally, the factors that might affect learners’ use of online education platform based on VR technology are complex, and include the shortage of traditional platforms, the advantages of new platforms, and cognitive risk perceptions [22]. All these factors may affect the promotion and popularization of new online education platforms based on VR technology.
At present, however, research on the applications of VR technology in education is still at an early stage of development, focusing on issues at a macro-qualitative level [23]. To date, there has been no research on the mechanisms explaining learners’ migration from traditional online education platforms to VR-based platforms. This gap may hinder research on and the development of new platforms, while also restricting the application, promotion, and popularization of platforms using VR technology. Accordingly, the present study integrates predictors of migration behavior into a unified framework based on the “negative-positive-anchor” theory, so as to help researchers and managers understand the factors influencing migration behavior [24,25]. The paper addresses the following two research questions: Which factors lead to the transfer of learners from traditional online education platforms to platforms based on VR technology? How do these factors affect learners’ willingness to transfer? The results of the study may not only help widen (and deepen) individual participation in online learning based on VR technology, but also enrich and expand the existing theories of migration behavior vis-à-vis online education platforms. At the same time, the study has important reference value for VR technology developers and managers, particularly when it comes to applying and popularizing VR technology in the education field.
Thus, the remainder of this paper is organized as follows: The theoretical background is presented in Section 2. The research hypotheses and research model is proposed in Section 3. The research design is described in Section 4. The data analysis and results are shown in Section 5. The conclusions are illustrated in Section 6. Finally, the suggestions for future applications are presented in Section 7.

2. Theoretical Background

2.1. Migration Behavior and “Negative-Positive-Anchor” Theory

The migration behavior of individuals has long been studied by researchers. In the narrow sense, migration refers to the flow of populations [26]; in a broader sense, however, the concept of migration can be applied to consumers switching from one service platform to another [27]. User migration may cause companies to lose established customers or gain new customers, so it is an important topic for marketing scholars. Previous research has focused on the factors influencing consumers’ willingness to migrate, including satisfaction, the quality of their relationship with current providers, inertia, and transfer costs [28,29].
In this context, the “negative-positive-anchor” theory proposed by Moon (1995) productively explains migration behavior [30]. The “negative” factors in the theory refer to the factors that force individuals to leave the original site (or current platform). The “positive” factors refer to the factors that attract individuals to the destination (or alternative platform). The “anchoring force” factor is generally related to the individual and his or her surrounding culture, which will affect not only the migration behavior, but also affect the “push” and “pull” of negative and positive factors, respectively. This theory is widely used in the field of social science. For example, Meyer et al. (1991) studied the transfer behavior of people in the use of instant messaging platforms [31]. Their study found that boredom and a sense of discontent with the existing platform pushed users to transfer, while an attractive alternative platform pulled users. Meanwhile, individuals’ subjective norms affected migration, with the force and inertia of anchoring factors not only hindering user transfers, but also weakening the tension between the push and pull factors. Li et al. (2014) emphasized the similarity between transfers and switching behavior [32]. In other words, when learners shift among various online education service providers, their changes are essentially a kind of transfer behavior, such that the “negative-positive-anchor” theory can be used as a theoretical framework to understand the migration patterns involved. This theory is also applicable to the behavior of learners migrating from traditional online education platforms to online education platforms based on VR technology.

2.2. VR Technology and Its Potential Applications in the Field of Education

As the VR technology embedded into the field of education, it will bring potentially radical transformations of traditional modes of teaching, as has been discussed by a number of researchers [33]. Some researchers describe how VR technology, the Internet of things, cloud computing, and other technologies can be applied in the domain of teaching. Educational institutions can use VR technology to access information, as a form of collaboration and communication with teachers, or as a tool for transferring educational contents, as well as a means of online teaching. A communication platform allows participants to exchange information about specific domains and interact and learn cooperatively. However, some researchers point out that the difficulties in applying VR technology [34,35]. Other researchers have pointed out that AR technology as the teaching modes should be changed to promote the concept of “teachers are the best” and “courses are the king,” in order to realize the breakthrough potential of the new educational ecosystem with the help of VR technology [36].
In 3D model fields, VR technology is used in engineering schools to aid both the lecturers and students. They offer students the opportunity to visualize the engineering concepts they learn in the classroom. In order to achieve a better understanding of circuit issues presented in formal lectures [37], including the remote physics experiments [38], simulation control testing [39] and virtual laboratories [40] are also widely used the VR technology. The application of VR in training, both surgical [41] and laboratory [42], are useful references used in professional instruction. The above research intended essentially to highlight the new possibilities that VR could bring to engineering education and later in their professional activity. Both of the learner and lectures are benefited from the VR technology.
Immersion, presence, and interactivity are regarded as the core characteristics of VR technologies [43]. As for the psychological component of virtual learning, each of theories offers a different perspective on the learning goals, motivational process, learning performance, transfer of knowledge process, the role of emotions, and implications for the teaching methods. Specifically, Robinson and Hullinger (2008) reported that online learners’ emotion has a significant impact on their learning engagement [44]. Milliganet al. (2013) revealed that learners’ motivation, online learning experience and self-confidence are the main factors driving learning engagement by Immersion [45].
The existing literature demonstrates that online education platforms based on VR technology have good prospects for application [46]. However, how to prompt users to migrate from traditional online education platforms to VR-based platforms is a necessary condition for the application and sustained development of VR technology in the field of education [47]. At present, there is no literature on the migration behavior of learners moving between these two kinds of platforms, or on the mechanism that leads to the formation of an intention to migrate. Therefore, it is necessary to fully understand the migration process of learners based on VR technology.

3. Research Hypotheses and Research Model

3.1. Negative Effects

In this context, negative effects derive from the negative factors associated with using traditional online education platforms. Such factors generate a negative experience for users, and hence “push” them away from the original platform. According to a report, many users believe that live-streaming courses on online education platforms have shortcomings such as poor image flow, a poor learning atmosphere, and delayed feedback [48]. However, after a long period of learning with the help of these platforms, users become familiar with the environment they help create and have to consider the risks of entering a new environment. At the same time, these shortcomings lead to the loyalty of traditional online education platforms. In this study, these problems are summarized as the afunction and loyalty [49]. That is, the afunction reflects the quality problem of traditional education platform, while the loyalty reflects the non-quality emotional problem. The concept of system quality is derived from the information system success model, which is generally used to evaluate information system technical indicators, such as reliability and ease of use. As a kind of information system, the quality of online education platform is closely related to the user experience. Previous studies have shown that the negative performance of the system quality with afunction in the user experience will act as a driving factor for them to move to the new platform. Therefore, the afunction will decrease learners’ dependence and give rise to a risk-perception assessment with respect to the new platform. Accordingly, the following hypothesis is proposed in this study:
Hypothesis 1 (H1a). 
Afunction negatively affect learners’ perceived risks with respect to traditional online education platforms.
Loyalty index as one way of emotional commitment reflects an individual’s desire to maintain relationships with traditional online education platforms [42]. In this study, the loyalty index refers to learners’ emotional attachment to, identification with, and participation in traditional online education platforms. Studies have shown that emotional commitment is a key factor affecting users’ continued use of online services; in other words, without loyalty, users are prone to migration. If learners have a solid loyalty to the traditional online education platforms, they may consider a perceived risk and maintain a relationship with and continue to use established platforms. Therefore, the following hypothesis is proposed in this study:
Hypothesis 1 (H1b). 
Loyalty index negatively affects learners’ perceived risks with respect to traditional online education platforms.

3.2. Positive Effects

In this context, positive effects derive from the positive factors associated with VR-based online education platforms; these factors make users eager to use the new platforms, and “pull” users toward the VR-based technology. In this study, tension factors relevant to positive effects were linked to deprivation. Deprivation can be defined in terms of people’s feelings about the lack of something and their belief that they should have it [50]. The perception of deprivation is the result of a comparison of possible circumstances with existing circumstances. Hart et al. (2016) found that the sense of relative deprivation created by the new platforms (that is, the interactivity, personalization, and economic possibilities represented by those platforms) would prompt people to migrate to them [51]. Accordingly, when the online education platforms based on VR technology are made available, and the learners who have not yet migrated to them observe the rich new functions they afford, this can produce a kind of envy-based risk perception, which causes the feeling that traditional online education platforms are lacking. This sense of lack could, in turn, prompt the shift from traditional to new platforms.
The present study discusses this kind of relative deprivation, or sense of lack, created by VR-based online education platforms, focusing on the aspects of interactivity and personalization. Interactivity is defined as the functional value that users feel they should have but do not actually experience [52]. VR technology and the Internet, fused with other technologies such as cloud-computing applications, bring fundamentally new experiences for learners, such as all-round, three-dimensional interactive and personalized image management [53]. These new experiences will make learners using traditional platforms feel envy because of their lack of such features. Therefore, the following hypothesis is proposed in this study:
Hypothesis 2 (H2a). 
The interactivity enabled by VR-based online education platforms has a positive impact on learners’ willingness to migrate.
Personalized teaching, through VR-based platforms, will link up with users’ learning styles and psychological needs. The application of VR technology in online education platforms will thus bring about “edutainment” [54]. For example, the learning resources used on VR-based platforms can contain gamification modules. These modules may make traditional platforms seem boring by comparison with the more personalized new platforms. Therefore, the following hypothesis is proposed in this study:
Hypothesis 2 (H2b). 
The personalization enabled by VR-based platforms has a positive impact on learners’ willingness to migrate.

3.3. Anchoring Effects

In this context, anchoring effects come from factors that either directly impede migration behavior or weaken push and pull effects. These factors are independent of the learning platform itself, and are mainly reflected in individual behavior patterns [55]. In this study, risk perception was selected as an anchoring force of this kind. Risk perception reflects the individual’s decision about existing behavior patterns vis-à-vis new environments. Previous studies have shown that risk perception is an important factor in explaining an individual’s current behavior, including his or her migration behavior.
The status quo bias theory holds that risk perception involves risk assessments in the face of better choices [56]. Studies have shown that risk perception can lead people to choose to maintain the status quo in the face of possible migration. When using an online education platform, the familiar teaching modes and feedback mechanisms associated with that platform make learners feel comfortable, whereas switching to a new platform may lead to an increase of learning costs due to users’ inability to adapt to the new environment. All of this may lead, in turn, to inertia vis-à-vis platform migration. Therefore, this study takes risk perception as an anchoring force and proposes the following hypothesis:
Hypothesis 3 (H3). 
Risk perception has a negative impact on learners’ willingness to migrate from traditional online education platforms to VR-based platforms.
In this study, these problems are summarized in terms of system quality and relationship quality [57]. System quality reflects the functional qualities or capabilities of the platform, while relationship quality reflects the non-functional qualities or aspects of the platform. The concept of system quality is derived from the information-system success model, which is generally used to evaluate information systems in terms of technical indicators, such as reliability and ease of use. Given that online education platforms constitute a kind of information system, the quality of the platforms is closely related to user experience. Previous studies have shown that negative performance with respect to system quality, in users’ experience, will act as a driving factor, pushing them to move to a new platform [58]. Accordingly, the following hypotheses are proposed in this study:
Hypothesis 4 (H4a). 
System quality positively affects learners’ risk perception with respect to traditional online education platforms.
Hypothesis 4 (H4b). 
Relationship quality positively affects learners’ risk perception with respect to traditional online education platforms.
In this study, transfer costs refer to the costs incurred when learners migrate from traditional online education platforms to online education platforms based on VR technology, including process costs, economic costs, and relationship costs [59]. Process cost refers to the assessment cost and learning cost caused by the learner switching platforms, including the time and energy spent learning to use the new platform. Economic cost refers to the financial cost caused by switching platforms, including the expense of purchasing new equipment [60]. Relationship cost refers to the loss of personal relationship after learners switch platforms, such as the loss of the original teacher–student relationship and classmate relationships made possible by the traditional platform. A large number of studies have shown that when the transfer cost is high, people will have a strong cognitive risk perception due to external constraints [61]. Accordingly, the following hypothesis is proposed in this study:
Hypothesis 4 (H4c). 
Transfer costs positively affect learners’ risk perception vis-à-vis traditional online education platforms.
Risk perception may also affect the push of traditional online education platforms [62]. On the one hand, according to the status quo bias theory, one of the causes of risk perception is people’s enjoyment and attachment to the current mode of behavior. At this point, people will be satisfied with the status quo [63], and such satisfaction will create a higher cognition of the status quo. This higher perception may not be due to the actual situation, but the user’s own “beautification” of present circumstances. Therefore, for learners using traditional online education platforms, inertia will lead them to form a higher evaluation of the system quality and relationship quality of the platform they are currently using. On the other hand, according to the theory of cognitive dissonance, with the formation of inertia, people will take a negative view of migration behavior to avoid falling into contradiction. Learners with strong inertia are subconsciously reluctant to change, so they will highly evaluate the system quality and relationship quality of traditional online education platforms, which can provide subjective rationality for their inertia. Accordingly, the following hypotheses are proposed in this study:
Hypothesis 5 (H5a). 
Risk perception positively affects the perceived system quality of traditional online education platforms.
Hypothesis 5 (H5b). 
Risk perception positively affects the relationship quality experienced by learners with respect to traditional online education platforms.
Similarly, risk perception may affect the pull of online education platforms based on VR technology. Risk perception makes people tend to reject migration when evaluating risk subjectively [64]. At the same time, in order to avoid cognitive dissonance, people with strong risk perception will have a low evaluation of the quality of the new system. Hence, for learners who have not used VR-based platforms, risk-perception assessments will cause them subconsciously to weaken the advantages of the new platform [65]. In turn, faced with the functional richness and emotional pleasure enabled by VR-based platforms, learners with strong risk-perception evaluation are not likely to have a corresponding sense of lack. Accordingly, the following hypotheses are proposed in this study:
Hypothesis 6 (H6a). 
Risk perception negatively affects the perceived interactivity of VR-based online education platforms.
Hypothesis 6 (H6b). 
Risk perception negatively affects learners’ perception of the personalization enabled by VR-based online education platforms.
In addition, because previous online-learning-behavior studies have found that learners’ demographic attributes can affect individual behaviors, this study added age, gender, occupation, and education level as control variables to the research model, in order to improve the effectiveness of statistical tests [66].
Based on the research assumptions just outlined, this paper proposes the research model shown in Figure 1.

4. Research Design

4.1. Measurement of Variables

In this study, questionnaires (Appendix A) were used to collect data. All the measurement items in the model consisted of independent variables and dependent variables [67]. The 7-point Likert scale was used to measure all potential variables, with 1 indicating strong disapproval and 7 indicating strong approval. Before the formal survey, a preliminary survey was conducted with 46 undergraduates to ensure that the reliability and validity of the scale met the requirements; the scale was adjusted according to the results of the preliminary survey to make it more suitable for the Chinese context. The final questionnaire used for data collection consisted of two parts: a measurement scale for 9 latent variables, and measurement questions for 4 demographic variables. More specifically, there were 4 measurement items related to emotional commitment, and 8 items related to transfer cost, used to measure the process cost, economic cost, and relational cost. In addition, 4 items were used for loyalty index. In terms of measurement questions, 6 were used for system quality, 4 for relationship quality, 3 for interactivity, 4 for risk perception and 4 for intention to migrate.

4.2. Data Collection

The researchers conducted formal data collection in the form of an offline paper questionnaire plus an online questionnaire from October to December 2019. They issued invitations to participate in the questionnaires through a WeChat group and supermarkets. After careful screening of the questionnaire, the researchers excluded not meet the basic quality requirements for responses (for example, all the answers were 4 and the time taken to complete the questionnaire was less than 30 s). Ultimately, 500 valid questionnaires were obtained. Among them, 232 were male (46.40%) and 268 were female (53.60%). The majority of respondents had a bachelor’s degree or a postgraduate degree (87.00%). They were mainly aged between 18 and 28 years old and engaged in occupations that included students and freelancers.

5. Data Analysis and Results

5.1. Measurement Model

In this study, Similar to Fang et al., 2018, a two-step method was used to test the model [68]. In order to ensure the validity of the measurements, before analyzing the structural model, “confirmatory factor analysis” was used to evaluate the adequacy of the measurement model of each latent variable in the model, including reliability, convergent validity, and discriminant validity [69]. In this study, Cronbach’s coefficients and combined reliability scores were used to evaluate the internal consistency reliability of the measurement model. Cronbach’s coefficient is a conservative estimate of measurement reliability, while combination reliability tends to overestimate the intrinsic consistency reliability of measurement, resulting in relatively high reliability estimations. As can be seen from Table 1, Cronbach’s coefficient and the combination reliability score of each latent variable are both greater than the threshold value of 0.70 but not higher than 0.95, indicating that the measurement of each latent variable has good internal consistency.
The convergent validity and discriminant validity of the scale were further tested by confirmatory factor analysis [70]. The convergent validity reflects the degree of correlation between items measuring the same latent variable, while the discriminant validity reflects the difference between different latent variables. The convergence validity was evaluated by means of average variance extracted (AVE) and factor loading significance. As can be seen from Table 1, the AVE value of each latent variable is greater than 0.50. The results of factor load and cross load analysis show that the loads of most measurement items on their theoretical potential variables are greater than 0.70 and therefore statistically significant. Only three of the measurement items had factor loads slightly less than 0.70. After deleting these items that measured less than 0.70 neither increases the combination reliability nor causes the content validity of the latent variables to change, we retain these measurements. To sum up, the scale has good convergent validity.
Factor load and crossing load analysis also showed that the measurement problem of the load on the latent variables is significantly higher than in the other cross on the latent variable load. The analysis also showed that each value of the latent variables of the AVE square root value is greater than the latent variables’ phase-relationship value. These results indicate that measures of the latent variables used in the study have enough discriminant validity. In recent years, research has shown that the simple comparison of factor loading, like the comparison of AVE square root values, is a defective method of measuring correlation coefficients between latent variables. As a result, this study also uses a heterogeneous method of evaluating discriminant validity—namely, Henseler et al.’s (2015) homogeneity-related ratio (i.e., the heterotrait-monotrait ratio of correlations, or HTMT) [71]. HTMT reflects the ratio of (1) the mean value of the correlation of measurement indexes of different latent variables to (2) the mean value of the correlation of measurement indexes of the same latent variable. According to the suggestion of Henseler et al., if the HTMT value is higher than 0.90, it indicates that the measurement lacks discriminant validity. The results reported in Table 2 show that the HTMT of the measurement index (for risk perception and relation quality) proposed in this study is 0.812, less than the threshold value of 0.90, which once again verifies that the measurement of each latent variable has sufficient discriminant validity.
Considering that the data collected in this study were obtained through a questionnaire-based survey method, there may be Common Method Variance (CMV) among variables [72]. The presence of CMV amplifies the correlation between measurements, which can skew the results. In this study, the Harman single factor method was used to evaluate potential CMV problems. After the exploratory factor analysis based on principal component analysis was carried out for all the measurement indexes, and 9 factors were extracted, the first factor obtained without rotation only explained a variance of 22.64%. Furthermore, the correlation coefficient between potential variables reported in Table 1 is also low. These results suggest that the CMV of this study is not obvious, and that further testing in this regard is warranted.

5.2. Structural Model

Before providing a structural model parameter estimation, this study first evaluated the multivariate assumptions, including normality, linearity, multicollinearity, and homovarianc [73]. The results of the Doornik–Hansen multivariate normality test showed that the measured variables in this study did not meet the requirements of multivariate normal distribution (p < 0.001). Considering the complexity of the model proposed in this study and the small sample size, a PLS algorithm based on variance (PLS-ABV) was chosen instead of a structural equation model based on covariance [74]. Compared with the structural equation model based on covariance, the PLS algorithm has lower requirements for the measurement scale, sample size, model complexity, and data normality, and has been widely used in online-user behavior research in the fields of marketing management and management information systems. Specifically, the matrix PLS algorithm in R software is used to estimate the parameters of the structural model. The PLS algorithm is used with other software (such as SmartPLS, GraphPLS, WarpPLS, etc.) to treat the original data as the input of the algorithm; the matrix PLS algorithm operates in accordance with the covariance matrix of the input data model parameter estimation, so as to create higher computational efficiency. The algorithm can thus satisfy the requirements of calculations based on mass data. Simulation results show that the computational results of the matrix PLS are exactly the same as those of commercial software such as Smart PLS, but the former is significantly faster [75]. Figure 2 reports the estimated results of the model.
As shown in Figure 2, the model proposed in this study explains 65.7% of the sample users’ intentions regarding online education platform migration. According to the threshold judgment standard of R2 proposed by Hair et al. (2011), the explanatory power of the model in this study is above the medium level. Specifically, inertia, as an anchoring effect of migration intention formation, had the greatest negative effect on migration intention (β = −0.401, p < 0.001). In turn, emotional commitment, transfer costs, and user habits significantly contribute to inertia. The further path coefficient difference analysis results of the push effect (system quality and relationship quality) show that the influence of system quality on migration intention is significantly less than that of interactivity and personalization (where system βafunction-interactivity = −0.254, t = −3.055, p < 0.001; βafunction-personalization = −0.335, t =−4.407, p < 0.001), but there is no significant difference in the effect of relationship quality (βafunction-relationship quality = 0.222, t = 0.135, p = 0.402). There was no significant difference in the effect of interactivity and personalization on the formation of intention regarding migration (for example, βinteractivity-personalization = −0.098, t = −0.724, p = 0.208). All the proposed research hypotheses were verified except the hypothesis H6b (inertia negatively affects personalization). The reason for the failure of H6b may be that existing online education platforms focus on learner-oriented functional modules, and the use of such platforms basically fails to meet the individual’s entertainment and emotional needs. Therefore, inertia will not significantly reduce learners’ cognition of personalization.

5.3. Analysis of Mediating Effects

In order to test whether cognitive risk perception affects users’ intentions with respect to migration through such mediating variables as system quality and relationship quality (negative effects) and interactivity and personalization (positive effects), this study tested potential mediating effects. Specifically, the self-sampling method (Bootstrap) proposed by Preacher and Hayes was used to calculate the mediating effects [76]. The results in Table 3 show that, on the whole, the negative effects and positive effect partially mediate the anchoring effect with respect to migration behavior. The mediating effects accounted for 3.1% of the total effect of inertia on migration intentions. Among these effects, the mediating effect of individuation is not significant.

6. Conclusions

The application of VR technology to online education platforms can improve the efficiency of online learning through individual portraits, visual recognition, voice recognition, and other innovations, so as to meet the diversified and personalized learning needs of learners and in the process bring unprecedented changes to the field of education. An important topic for research, in this connection, is how to encourage learners to move from the existing online education platforms to VR-based platforms. The present paper proposes a model of the mechanism influencing learner migration; the model is based on the “negative-positive-anchor” theory. After testing this model with survey data from 500 learners using traditional online education platforms, the following conclusions can be drawn.
First, the afunction and loyalty of traditional online education platforms, constituting two major negative factors, reduce learners’ willingness to migrate to VR-based online education platforms. That is to say, when learners have a poor evaluation of the system performance of traditional online education platforms (for example, with respect to the ease of use of the system), it is easier for them to migrate to VR-based online education platforms. This may be because the online education platform with higher system quality is more conducive to the efficient learning of learners. On the other hand, the results also show that learners pay attention not only to the functional quality of the platform, but also to its non-functional quality, particularly when it comes to building interpersonal relationships. When learners feel that the connectivity that an online education platform provides with teachers and other learners is weak (for example, they lack interaction with teachers and classmates and cannot communicate effectively with each other in a timely manner), learners are also more likely to migrate to a new platform.
Second, the improved interactivity and personalized perception enabled by VR-based online education platforms in comparison with traditional online education platforms increases learners’ willingness to transfer, in accordance with the positive effects of migration. The rich functionality of VR-based platforms (manifested, for example, in a more intelligent system and more personalized learning support) makes learners’ experience more enjoyable, with the ramification process potentially also making learning experiences more effective. The more functionality the new platforms have, the more likely it is that learners will perceive their traditional platforms as relatively scarce, and the feeling of lack will lead them to migrate to a new platform. Overall, then, the rich and novel functions provided by VR-based platforms are important influencing factors with respect to learners’ migration behavior.
Third, the risk-perception evaluation involved in the anchoring effect vis-à-vis the formation of migration intention negatively affects learners’ migration intention; it also positively effects their cognition of the system quality and relationship quality of traditional online education platforms, and reduces their cognition of the interactivity of the original platforms. In addition, the emotional commitment and usage habits of learners relative to the original platforms, as well as the transfer cost caused by migration to new, VR-based platforms, are the preconditions of cognitive risk perception. This shows that when the learner has a strong sense of attachment to the original platform, they will make a higher assessment of risk (i.e., not be willing to move to a new platform). At the same time, this kind of emotion weakens not only the push caused by insufficient system quality and relationship quality, but also the pull from the increased interactivity of the new platform. Similarly, learners’ existing usage habits and perceived high transfer costs will reinforce their cognitive risk perception as well as emotional commitment, thus negatively affecting the push of the original platform and the pull of the new platform. It is worth noting that the results of this study show that while inertia can reduce learners’ cognition of the richness of VR-based platforms at the functional level. It will not affect their cognition of the richness of those platforms at the emotional level. This indicates that the existing online education platforms have not satisfied learners’ emotional needs, while the comparative advantages of the new platform will greatly enhance learners’ willingness to migrate.

7. Suggestions for Future Applications

Based on the conclusions just outlined, we also offer several suggestions about how to better promote the application and popularization of VR technology in the field of online education.
One is to build high-quality VR-based platforms. “High quality” refers not only to the high quality of the platform in terms of learning functions, but also to the high quality of the platform architecture. High-quality platform architecture is conducive to the construction of interpersonal relationships between participants, thus realizing learners’ interpersonal communication needs at different levels, including emotional orientation and utilitarian orientation. VR technology can greatly expand the social scene and attention range of individuals, and help them realize the connection between virtual interpersonal relationships and real interpersonal relationships, so that virtual space and the real world can be seamlessly integrated. Therefore, the targeted use of VR technology to build holographic and immersive interpersonal interaction functions can help users build interpersonal relationships through VR-based platforms.
Second, through the combination of VR technology with big data, cloud computing, the Internet of things, blockchains, and other technologies, VR-based platforms can provide learners with rich functional experiences and pleasant emotional experiences, so as to create a more personalized sense of interaction than learners experience on the traditional platforms. VR techniques integrated with the Internet of things, cloud computing technology, wearable interactive terminals, virtual scene generators, facial emotion AI technology, real-time data recorders, and electronic teaching instruments, tools, and technologies that place teachers and learners in the same virtual space—all of this affords teachers real-time control of learning environments and helps teachers monitor physical and mental status changes among their students during the teaching process. Furthermore, by recording and analyzing teaching activities, the weaknesses of each learner can be identified, and then AI technology can be used to achieve targeted after-class review and guidance. At the same time, it is also convenient for teachers to know in a timely fashion the effect of the teaching process on learners, so as to focus on issues that present challenges to learning. Through these methods, learners can have interesting emotional experiences when using these new functions; these experiences can prompt users of traditional platforms to have a sense of lack, thus promoting their migration behavior.
The third suggestion is to reduce the cost of the transfer to VR-based platforms as much as possible, so as to weaken the impact of individual risk perception and thus promote the migration of learners to the new platforms. Transfer costs include process costs, economic costs, and relationship costs. In terms of process costs, the new platform can be designed to be easier to use, so as to reduce the learning time and energy that learners have to invest to adapt to the new platform. In terms of economic costs, innovative business models can be employed to avoid a single mode of paid use, allowing more learners to use the new platforms and experience their comparative advantages, thus increasing their perception of the new platforms’ capacity for personalization and interactivity. In terms of relationship costs, data sharing between platforms, strategic cooperation vis-à-vis platform acquisition, and mutual recommendations among friends can be used to reduce the loss of interpersonal relationships that learners might otherwise experience in the process of platform migration.

Author Contributions

Data curation, J.C. and C.L.; investigation, J.C and R.C.; methodology, J.C.; project administration, S.N.; resources, P.G.; writing—original draft, J.C. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Conflicts of Interest

The authors declare no conflict of interest.

Appendix A. Questionnaire

1Gender: Male Female
2Age: What is your age?
3Education: What is the highest degree or level of school you have completed? If currently enrolled, highest degree received.
1Associate college degree or below
2Bachelor’s degree
3Master’s degree or above
4VR Experience: How much time do you spend on the Web?
1Never
1A few times a month or less
2A few times a week
3About once a day
4Several times each day
5Please rate the usefulness of VR-based online education platforms you are learning or you’ve just completed
1 (Not at all useful) 2 3 4 (Neutral) 5 6 7 (Very useful)
6Please rate the interestingness of VR-based online education platforms you are learning or you’ve just completed
1 (Not at all interesting) 2 3 4 (Neutral) 5 6 7 (Very interesting)
To what extent do you agree or disagree with the following statements about VR-based online education platforms that you are learning or have just completed?
Emotional commitment (Strongly Disagree/Agree, 1–7 Scale)
Q1
The VR-based online course teacher/instructor creates a good atmosphere facilitates social interaction.
Q2
The VR-based online course teacher/instructor encourages communications between learners and teachers.
Q3
I communicate with other learners during VR-based online course learning.
Q4
I exchange and share opinions with other learners during VR-based online course learning.
Loyalty index (Strongly Disagree/Agree, 1–7 Scale)
Q1
When I learn VR-based online course, I feel like a competent person.
Q2
When I learn VR-based online course, I feel very capable and effective.
Q3
I am satisfied with my performance at VR-based online course.
Q4
I think I am good at VR-based online course learning.
Interactivity (Strongly Disagree/Agree, 1–7 Scale)
Q1
When I learn VR-based online course, I can recognize other learners.
Q2
When I learn VR-based online course, I feel loved and cared about.
Q3
When I learn VR-based online course, I feel a lot of closeness and intimacy.
Relationship (Strongly Disagree/Agree, 1–7 Scale)
Q1
When I learn VR-based online course, I feel free to be who I am.
Q2
When I learn VR-based online course, I have a say to what happens and can voice my opinions.
Q3
I have some choice in what I want to learn in VR-based online course.
Q4
I feel that I learn VR-based online course because I want to.
System quality (Strongly Disagree/Agree, 1–7 Scale)
Q1
Learning VR-based online course makes me forget my immediate surroundings.
Q2
Learning VR-based online course makes me forget the reality of the outside world.
Q3
Learning VR-based online course makes me forget the reality of the outside world.
Q4
Learning VR-based online course makes me forget the knowledge in the class.
Q5
Learning VR-based online course makes me like to study in the virtual platform.
Learning willingness (Strongly Disagree/Agree, 1–7 Scale)
Q1
I feel a strong sense of belonging to the VR-based online education platforms.
Q2
I feel strong ties to the VR-based online education platforms.
Q3
I feel a strong sense of identification with the VR-based online education platforms.
Risk perception (Strongly Disagree/Agree, 1–7 Scale)
Q1
It is impossible that I will not continue VR-based online education platforms in the future.
Q2
It is likely that I will continue VR-based online education platforms in the next few months.
Q3
If possible, I am willing to engage in VR-based online education platforms in the next few months.
Q4
If I could, I am willing to contribute to VR-based online education platforms conversation and discussions in the next few months.
Transfer cost
Q1
In general, I am glad to be a member of the VR-based online education platforms in the next few months.
Q2
It is impossible that I will mind the cost transfer from traditional learning platforms to the VR-based online education platforms.
Q3
It is impossible that I will continue renew the cost to VR-based online education platforms after finishing the probation period.
Q4
If possible, I am willing to accept the cost beyond the traditional cost in VR-based online education platforms.
Q5
If I could, I am willing to contribute to VR-based online education platforms conversation within few months.
Q6
It is impossible that I will consider the cost in the VR-based online education platforms.
Q7
If I could, I am willing to give up the traditional course and transfer to VR-based online education platforms.
Q8
If I could, I am willing to study in the VR-based online education platforms forever.

References

  1. Hughes, J.; Herrington, M.; McDonald, T.; Rhodes, A. E-portfolios and personalized learning: Research in practice with two dyslexic learners in UK higher education. Dyslexia 2011, 17, 48–64. [Google Scholar] [CrossRef] [PubMed]
  2. Iandoli, L.; Quinto, I.; De Liddo, A.; Shum, S.B. On online collaboration and construction of shared knowledge: Assessing mediation capability in computer supported argument visualization tools. J. Assoc. Inf. Sci. Technol. 2016, 67, 1052–1067. [Google Scholar] [CrossRef] [Green Version]
  3. Serrano, C.M.; Wesselink, P.R.; Vervoorn, J.M. First experiences with patient-centered training in virtual reality. J. Dent. Educ. 2020, 84, 607–614. [Google Scholar] [CrossRef] [PubMed]
  4. Violante, M.G.; Vezzetti, E. Virtual interactive e-learning application: An evaluation of the student satisfaction. Comput. Appl. Eng. Educ. 2015, 23, 72–91. [Google Scholar] [CrossRef] [Green Version]
  5. Vogel, J.; Bowers, C.; Meehan, C.; Hoeft, R.; Bradley, K. Virtual reality for life skills education: Program evaluation. Deaf. Educ. Int. 2004, 6, 39–50. [Google Scholar] [CrossRef]
  6. McMillan, K.; Flood, K.; Glaeser, R. Virtual reality, augmented reality, mixed reality, and the marine conservation movement. Aquat. Conserv. Mar. Freshw. Ecosyst. 2017, 27, 162–168. [Google Scholar] [CrossRef] [Green Version]
  7. Kim-Berman, H.; Karl, E.; Sherbel, J.; Sytek, L.; Ramaswamy, V. Validity and User Experience in an Augmented Reality Virtual Tooth Identification Test. J. Dent. Educ. 2019, 83, 1345–1352. [Google Scholar] [CrossRef]
  8. Vanderplank, R. A Virtual Learning Environment seen as a System of Instrumented Activities Video and Informal Language Learning. Model. Next Gener. Digit. Learn. Environ. 2010, 1–62. [Google Scholar] [CrossRef]
  9. Parboosingh, I.J.; Reed, V.A.; Palmer, J.C.; Bernstein, H.H. Enhancing practice improvement by facilitating practitioner interactivity: New roles for providers of continuing medical education. J. Contin. Educ. Health Prof. 2011, 31, 122–127. [Google Scholar] [CrossRef]
  10. Pantelidis, V.S. Virtual reality and engineering education. Comput. Appl. Eng. Educ. 1997, 5, 3–12. [Google Scholar] [CrossRef]
  11. Reznek, M.A.; Rawn, C.L.; Krummel, T.M. Evaluation of the Educational Effectiveness of a Virtual Reality Intravenous Insertion Simulator. Acad. Emerg. Med. 2002, 9, 1319–1325. [Google Scholar] [CrossRef] [PubMed]
  12. Schott, C.; Marshall, S. Virtual reality and situated experiential education: A conceptualization and exploratory trial. J. Comput. Assist. Learn. 2018, 34, 843–852. [Google Scholar] [CrossRef]
  13. Moro, C.; Štromberga, Z.; Raikos, A.; Stirling, A. The effectiveness of virtual and augmented reality in health sciences and medical anatomy. Anat. Sci. Educ. 2017, 10, 549–559. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  14. Chen, L.-W.; Tsai, J.-P.; Kao, Y.-C.; Wu, Y.-X. Investigating the learning performances between sequence- and context-based teaching designs for virtual reality (VR)-based machine tool operation training. Comput. Appl. Eng. Educ. 2019, 27, 1043–1063. [Google Scholar] [CrossRef]
  15. Peng, Y.; Su, C.; Heidrich, W. 21-2: Exploring 3D Interactive Performance Animation for VR/AR Applications Using Low-cost Motion Capture. SID Symp. Dig. Tech. Pap. 2016, 47, 251–254. [Google Scholar] [CrossRef]
  16. Bower, M.; Lee, M.J.W.; Dalgarno, B. Collaborative learning across physical and virtual worlds: Factors supporting and constraining learners in a blended reality environment. Br. J. Educ. Technol. 2017, 48, 407–430. [Google Scholar] [CrossRef]
  17. Akbulut, A.; Catal, C.; Yıldız, B. On the effectiveness of virtual reality in the education of software engineering. Comput. Appl. Eng. Educ. 2018, 26, 918–927. [Google Scholar] [CrossRef] [Green Version]
  18. Park, S.C.; Ryoo, S.Y. An empirical investigation of end-users’ switching toward cloud computing: A two factor theory perspective. Comput. Hum. Behav. 2013, 29, 160–170. [Google Scholar] [CrossRef]
  19. Karel, K.; Paul, A.K.; Wim, J. The Sociability of Computer-Supported Collaborative Learning Environment. J. Educ. Technol. Soc. 2002, 5, 8–22. [Google Scholar]
  20. Lee, E.S. A theory of migration. Demography 1966, 3, 47–57. [Google Scholar] [CrossRef]
  21. Samuelson, W.; Zeckhauser, R. Status quo bias in decision making. J. Risk Uncertain. 1988, 1, 7–59. [Google Scholar] [CrossRef]
  22. Sjöberg, L.; Engelberg, E. Risk Perception and Movies: A Study of Availability as a Factor in Risk Perception. Risk Anal. 2010, 30, 95–106. [Google Scholar] [CrossRef] [PubMed]
  23. Nordström, P.; Franks, P.W.; Gustafson, Y.; Nordström, A. Cognitive function in young men and the later risk of fractures. J. Bone Miner. Res. 2012, 27, 2291–2297. [Google Scholar] [CrossRef] [PubMed]
  24. Henseler, J.; Ringle, C.M.; Sarstedt, M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. J. Acad. Mark. Sci. 2015, 43, 115–135. [Google Scholar] [CrossRef] [Green Version]
  25. Ghomi, E.J.; Rahmani, A.M.; Qader, N.N. Applying queue theory for modeling of cloud computing: A systematic review. Concurr. Comput. Pract. Exp. 2019, 31, e5186. [Google Scholar] [CrossRef]
  26. Fang, J.; Zhao, Z.; Wen, C.; Wang, R. Design and performance attributes driving mobile travel application engagement. Int. J. Inf. Manag. 2017, 37, 269–283. [Google Scholar] [CrossRef]
  27. Manouselis, N.; Vuorikari, R.; Van Assche, F. Collaborative recommendation of e-learning resources: An experimental investigation. J. Comput. Assist. Learn. 2010, 26, 227–242. [Google Scholar] [CrossRef]
  28. Balalaie, A.; Heydarnoori, A.; Jamshidi, P.; Tamburri, D.A.; Lynn, T. Microservices migration patterns. Softw. Pract. Exp. 2018, 48, 2019–2042. [Google Scholar] [CrossRef]
  29. Sun, Y.; Liu, D.; Chen, S.; Wu, X.; Shen, X.-L.; Zhang, X. Understanding users’ switching behavior of mobile instant messaging applications: An empirical study from the perspective of push-pull-mooring framework. Comput. Hum. Behav. 2017, 75, 727–738. [Google Scholar] [CrossRef]
  30. Hsieh, J.-K.; Hsieh, Y.-C.; Chiu, H.-C.; Feng, Y.-C. Post-adoption switching behavior for online service substitutes: A perspective of the push–pull–mooring framework. Comput. Hum. Behav. 2012, 28, 1912–1920. [Google Scholar] [CrossRef]
  31. Meyer, J.P.; Allen, N.J. A three-component conceptualization of organizational commitment. Hum. Resour. Manag. Rev. 1991, 1, 61–89. [Google Scholar] [CrossRef]
  32. Lin, T.-C.; Huang, S.-L. Understanding the Determinants of Consumers’ Switching Intentions in a Standards War. Int. J. Electron. Commer. 2014, 19, 163–189. [Google Scholar] [CrossRef]
  33. Gropen, J.; Clark-Chiarelli, N.; Hoisington, C.; Ehrlich, S.B. The Importance of Executive Function in Early Science Education. Child Dev. Perspect. 2011, 5, 298–304. [Google Scholar] [CrossRef]
  34. Guzman, J.L.; Dormido, S.; Berenguel, M. Interactivity in education: An experience in the automatic control field. Comput. Appl. Eng. Educ. 2013, 21, 360–371. [Google Scholar] [CrossRef]
  35. Shin, Y.-S. Virtual reality simulations in Web-based science education. Comput. Appl. Eng. Educ. 2002, 10, 18–25. [Google Scholar] [CrossRef]
  36. Foley, J.M.; Kaiser, L.M.R. Learning Transfer and Its Intentionality in Adult and Continuing Education. New Dir. Adult Contin. Educ. 2013, 2013, 5–15. [Google Scholar] [CrossRef]
  37. Hu, X.; Ng, J.; Lee, J.H. VR creation experience in cultural heritage education: A preliminary exploration. Proc. Assoc. Inf. Sci. Technol. 2019, 56, 422–426. [Google Scholar] [CrossRef]
  38. Bell, J.T.; Fogler, H.S. The Application of Virtual Reality to Chemical Engineering and Education; American Institute of Chemical Engineers: New York, NY, USA, 1998. [Google Scholar]
  39. Cashman, P.R.; Kitney, C.J.I.; Ying, C.L. Segmentation and Deformable Modelling Techniques for a Virtual Reality Surgical Simulator in Hepatic Oncology. In Proceedings of the 3th Conferencia Ibérica de Sistemas y Tecnologias de Información, Ourense, Spain, 12–15 October 2009; pp. 90–98. [Google Scholar]
  40. Gibbon, G.J. Combining Theory and Experimentation to Develop Inductive Learning Skills in an Electric Circuits Course. In Proceedings of the ICEE 2008 Conference, Pécs and Budapest, Hungary, 28–31 July 2008; pp. 20–23. Available online: https://www.ineer.org/Events/ICEE2008/full_papers/full_paper50.pdf (accessed on 31 August 2020).
  41. Ozvoldova, M.; Cernansky, P.; Schuer, F. Internet remote physics experiments in a student laboratory. Innovations 2006, 25, 297–304. [Google Scholar]
  42. Perez, J.F.R.; Barea, L.; Boquete, M.; Hidalgo, A.; Dapena, I. Cataract Surgery Simulator for Medical Education & Finite Element/3D Human Eye Model; Springer: Berlin/Heidelberg, Germany, 2008; pp. 1802–1804. [Google Scholar]
  43. Sampaio, A.Z.; Henriques, P.G. Visual simulation of Civil Engineering activities: Didactic virtual models. Innovations 2008, 297–304. [Google Scholar]
  44. Schofield, D. Mass Effect: A Chemical Engineering Education Application of Virtual Reality Simulator Technology. J. Online Learn. Teach. 2012, 8, 63. [Google Scholar]
  45. Yang, H.; Fu, X. Research on Innovation and Entrepreneurship Education for College Students of Electronic and Information Majors. Int. Conf. Inf. Technol. Med. Educ. 2017. [Google Scholar] [CrossRef]
  46. Kolmanič, S.; Kohek, Š.; Žalik, B. From computer edit geometric design education to virtual worlds: Advantages and pitfalls of Computer-Based Education. Comput. Appl. Eng. Educ. 2018, 26, 1614–1625. [Google Scholar] [CrossRef]
  47. Zhang, H.; Lu, Y.; Gupta, S.; Zhao, L.; Chen, A.; Huang, H. Understanding the antecedents of customer loyalty in the Chinese mobile service industry: A push–pull–mooring framework. Int. J. Mob. Commun. 2014, 12, 551–577. [Google Scholar] [CrossRef]
  48. Alhabeeb, M.J. On consumer trust and product loyalty. Int. J. Consum. Stud. 2007, 31, 609–612. [Google Scholar] [CrossRef]
  49. Mio, C.; Zaro, E.S.; Fasan, M. Are loyalty shares an effective antidote against short-termism? Empirical evidence from Italy. Bus. Strategy Environ. 2020, 29, 1785–1796. [Google Scholar] [CrossRef]
  50. Rostami, M.; Ahmadian, L.; Jahani, Y.; Niknafs, A. The effect of patient satisfaction with academic hospitals on their loyalty. Int. J. Health Plan. Manag. 2019, 34, e726–e735. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  51. Waheed, M.; Kaur, K.; Kumar, S. What role does knowledge quality play in online students’ satisfaction, learning and loyalty? An empirical investigation in an eLearning context. J. Comput. Assist. Learn. 2016, 32, 561–575. [Google Scholar] [CrossRef]
  52. Hart, S.A. Precision Education Initiative: Moving Toward Personalized Education. Mindbrainand Educ. 2016, 10, 209–211. [Google Scholar] [CrossRef]
  53. Encarnacao, L.M.; Fechter, J.; Grunert, T.; Straßer, W. A Platform for User-Tailored Interaction Development in 2D, 3D, and VR. Comput. Graph. Forum 1996, 15, 433–442. [Google Scholar] [CrossRef]
  54. Trelease, R.B.; Nieder, G.L.; Dørup, J.; Hansen, M.S. Going virtual with quicktime VR: New methods and standardized tools for interactive dynamic visualization of anatomical structures. Anat. Rec. 2000, 261, 64–77. [Google Scholar] [CrossRef]
  55. Ma, M.; Fallavollita, P.; Seelbach, I.; Von Der Heide, A.M.; Euler, E.; Waschke, J.; Navab, N. Personalized augmented reality for anatomy education. Clin. Anat. 2016, 29, 446–453. [Google Scholar] [CrossRef] [PubMed]
  56. Esch, F.-R.; Schmitt, B.H.; Redler, J.; Langner, T. The brand anchoring effect: A judgment bias resulting from brand awareness and temporary accessibility. Psychol. Mark. 2009, 26, 383–395. [Google Scholar] [CrossRef]
  57. O’Neill, E.; Brereton, F.; Shahumyan, H.; Clinch, J.P. The Impact of Perceived Flood Exposure on Flood-Risk Perception: The Role of Distance. Risk Anal. 2016, 36, 2158–2186. [Google Scholar] [CrossRef] [PubMed]
  58. Lovat, T.J. Synergies and Balance between Values Education and Quality Teaching. Educ. Philos. Theory 2010, 42, 489–500. [Google Scholar] [CrossRef]
  59. MacDonald, M.J. Quality academy: Quality management education in a multifacility health care organization. J. Contin. Educ. Health Prof. 1995, 15, 175–176. [Google Scholar] [CrossRef]
  60. Lo, S.-F.; Wang, Y.-T.; Wu, L.-Y.; Hsu, M.-Y.; Chang, S.-C.; Hayter, M. A cost–effectiveness analysis of a multimedia learning education program for stoma patients. J. Clin. Nurs. 2010, 19, 1844–1854. [Google Scholar] [CrossRef]
  61. Bradford, A.B.; Adler-Baeder, F.; Ketring, S.A.; Bub, K.L.; Pittman, J.F.; Smith, T.A. Relationship Quality and Depressed Affect Among a Diverse Sample of Relationally Unstable Relationship Education Participants. Fam. Relat. 2014, 63, 219–231. [Google Scholar] [CrossRef]
  62. Yizengaw, T. Implementation of Cost Sharing in the Ethiopian Higher Education Landscape: Critical Assessment and the Way Forward. High. Educ. Q. 2007, 61, 171–196. [Google Scholar] [CrossRef]
  63. Nam, S.-J.; Hwang, H. Consumers’ Needs for Public Education and Corporate Participation Regarding Child Internet Addiction: Based on the Risk Perception Attitude Framework. J. Consum. Aff. 2019, 53, 1220–1233. [Google Scholar] [CrossRef]
  64. Burnham, T.A.; Frels, J.K.; Mahajan, V. Consumer Switching Costs: A Typology, Antecedents, and Consequences. J. Acad. Mark. Sci. 2003, 31, 109–126. [Google Scholar] [CrossRef]
  65. Miller, J.A.; Crapo, J.S.; Bradford, K.; Higginbotham, B.J. Relationship Beliefs Patterns Among Relationship Education Participants at Different Venues. Fam. Relat. 2019, 68, 390–404. [Google Scholar] [CrossRef]
  66. Brower, A.M.; Humphreys, D.; Karoff, R.; Kallio, S. Designing quality into direct-assessment competency-based education. J. Competency Based Educ. 2017, 2, e01043. [Google Scholar] [CrossRef]
  67. Graf, S.; Liu, T.C. Analysis of learners’ navigational behaviour and their learning styles in an online course. J. Comput. Assist. Learn. 2010, 26, 116–131. [Google Scholar] [CrossRef]
  68. Fang, J.M.; Shi, Z.H.; Liu, L. The influencing mechanism of learners migration behavior on online education platform based on 5G technology: A new model for learner support. Res. Modern Distance Educ. 2019, 31, 22–31. [Google Scholar]
  69. Kang, M.; Im, T. Factors of learner–instructor interaction which predict perceived learning outcomes in online learning environment. J. Comput. Assist. Learn. 2013, 29, 292–301. [Google Scholar] [CrossRef]
  70. Li, N.; Marsh, V.; Rienties, B. Modelling and Managing Learner Satisfaction: Use of Learner Feedback to Enhance Blended and Online Learning Experience. Decis. Sci. J. Innov. Educ. 2016, 14, 216–242. [Google Scholar] [CrossRef]
  71. Wu, Y. Factors impacting students’ online learning experience in a learner-centred course. J. Comput. Assist. Learn. 2016, 32, 416–429. [Google Scholar] [CrossRef]
  72. Pardo, L.; Martín, N. Homogeneity/Heterogeneity Hypotheses for Standardized Mortality Ratios Based on Minimum Power-divergence Estimators. Biom. J. 2009, 51, 819–836. [Google Scholar] [CrossRef]
  73. Teo, T. Considering common method variance in educational technology research. Br. J. Educ. Technol. 2011, 42, E94–E96. [Google Scholar] [CrossRef]
  74. Hair, J.F.; Ringle, C.M.; Sarstedt, M. PLS-SEM: Indeed a Silver Bullet. J. Mark. Theory Pract. 2011, 19, 139–152. [Google Scholar] [CrossRef]
  75. Fang, J.; Tang, L.; Yang, J.; Peng, M. Social interaction in MOOCs: The mediating effects of immersive experience and psychological needs satisfaction. Telemat. Inform. 2019, 39, 75–91. [Google Scholar] [CrossRef]
  76. Preacher, K.J.; Hayes, A.F. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behav. Res. Methods Instrum. Comput. 2004, 36, 717–731. [Google Scholar] [CrossRef] [PubMed] [Green Version]
Information 11 00423 g001 550
Figure 1. Research model.
Figure 1. Research model.
Information 11 00423 g001
Information 11 00423 g002 550
Figure 2. Coefficients of pathway for structural models. Note: about the p value: * p < 0.05; ** p < 0.01; *** p < 0.001.
Figure 2. Coefficients of pathway for structural models. Note: about the p value: * p < 0.05; ** p < 0.01; *** p < 0.001.
Information 11 00423 g002
Table
Table 1. Reliability of a variable. Correlation coefficient and AVE matrix.
Table 1. Reliability of a variable. Correlation coefficient and AVE matrix.
AlphCRSystem QualityRelation QualityTransfer CostRisk PerceptionAfunctionLoyaltyInteractivityPersonalizationLearner’s Willingness
system quality0.8740.9420.872
relation quality0.8790.8750.5140.854
transfer cost0.8540.8910.5920.5380.982
risk perception0.8710.92567,840.5010.7240.8750.864
afunction0.8690.8970.5430.2110.6210.7550.821
loyalty0.8720.9120.5480.0140.5870.7850.5610.798
interactivity0.9540.875−0.4210.098−0.0870.597−0.012−0.0410.867
personalization0.8650.8960.087−0.2570.0970.2140.081−0.0120.8920.901
learner’s willingness0.7510.824−0.471−0.256−0.421−0.512−0.259−0.4250.6580.6240.878
Table
Table 2. Heterotrait-monotrait ratio of correlations.
Table 2. Heterotrait-monotrait ratio of correlations.
System QualityRelation QualityTransfer CostRisk PerceptionAfunctionLoyaltyInteractivityPersonalizationLearner’s Willingness
system quality1
relation quality0.5351
transfer cost0.6040.6221
risk perception0.5180.8120.7851
afunction0.5340.2540.7010.7861
loyalty0.5570.0170.6330.7740.5071
interactivity−0.4150.095−0.1020.612−0.113−0.1051
personalization0.109−0.2040.0920.3350.098−0.1360.7541
learner’s willingness−0.052−0.198−0.337−0.609−0.35−0.3270.6125891
Table
Table 3. Results of the mediating effect.
Table 3. Results of the mediating effect.
Mediating Effect PathwayCoefficentBootstrap Standard ErrorBootstrap (95% CI) (Confidence Interval)Mediating Effect
risk perception → afuntion → learner’s willingness−0.0210.008[−0.035, −0.001]Significant
risk perception → loyality index → learner’s willingness−0.0350.017[−0.052, −0.012]Significant
risk perception → Intractivity → learner’s willingness−0.0170.015[−0.043, −0.002]Significant
risk perception → personalization → learner’s willingness0.0210.019[−0.045,0.024]Not Significant
total mediating effecct−0.0310.033[−0.014, −0.015]Significant
Note: Standard error estimates based on 500 bootstrap self-sampling were provided with a 95% confidence interval of sample percentage.

Share and Cite

MDPI and ACS Style

Chen, J.; Liu, C.; Chang, R.; Gui, P.; Na, S. From Traditional to VR-Based Online Education Platforms: A Model of the Mechanism Influencing User Migration. Information 2020, 11, 423. https://doi.org/10.3390/info11090423

AMA Style

Chen J, Liu C, Chang R, Gui P, Na S. From Traditional to VR-Based Online Education Platforms: A Model of the Mechanism Influencing User Migration. Information. 2020; 11(9):423. https://doi.org/10.3390/info11090423

Chicago/Turabian Style

Chen, Jing, Chang Liu, Ronghua Chang, Pengfei Gui, and Sanggyun Na. 2020. "From Traditional to VR-Based Online Education Platforms: A Model of the Mechanism Influencing User Migration" Information 11, no. 9: 423. https://doi.org/10.3390/info11090423

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop