The learning styles of an individual pertain to his or her ability to effectively understand and absorb information. ⁸ In terms of student-centric learning, it is essential to accommodate all students with different learning preferences. Many teachers believe that teaching according to individual style can help improve learners’ performance. ²⁴ Mirza and Khurshid ⁸ identified the learning modalities of health professional students and suggested that student motivation is positively enhanced when students recognized their learning styles. The study by Good et al. ²⁴ also suggested that VARK (Visual, Aural, Read/write, Kinesthetic) is a significant teaching tool and also helps enhance performance.

Mirza and Khurshid ⁸ have stated that VARK (Visual, Auditory, Reader/Writer and Kinesthetic) is the most accepted teaching model that categorizes learners with respect to their sensory characteristics. It is one of the earliest and most popular learner style tools developed by Fleming and Mills. ¹⁹ According to these researchers, the success of the model stems from its authenticity, usability, and the array of learning resources available to complement it.

The term e-learning or electronic learning originated in the mid-1990s when the Internet began to gather momentum. ²⁰ The two core elements of e-learning include computer-based learning as well as web-based learning. ⁹

Researchers have developed a number of models to guide e-learning research and e-learning effectiveness. By studying and synthesizing previous models of e-learning research from management, information systems, education, sociology and psychology, Johnson and Brown ¹⁰ concluded that e-learning outcomes are influenced by multiple processes and inputs. They proposed a model to summarize the literature consisting of five inputs: organizational context, technology, design/pedagogy, instructions and trainee or learner traits. The study by Johnson and Brown focuses more on technology as an input and influencing factor in the learning process. The sub focuses under technology are reliability, usefulness, ease of use and media richness ( Figure 1).

Mobile learning is an integral part of e-learning. As cited by Basak et al., ⁹ Behera stated that mobile learning or m-learning is correlated with mobile computing and e-learning. According to Sanchez-Prieto et al., ¹⁸ mobile learning is a learning environment that is closely related to e-learning, belonging to a separate typology, where the process of teaching, as well as learning, would have a digital dimension ( Figure 2).

Virtual reality (VR) is a computer-generated rendering of a 3D world or image that can be communicated in a relatively natural or tactile manner by a human wearing special electronic devices (headgear with an incorporated display inside or accessories equipped with sensors). VR provides 3D digital worlds featuring sophisticated ways of interaction that can motivate students to understand more completely. Its truly interactive learning environment has the power to improve the learning process and skill acquisition. ¹² Previous research has shown that the use of VR technologies will help to motivate the learning process by allowing students to experience ‘natural phenomena’ while being safe from the real-world consequences. ¹² In other words, the incorporation of VR allows new educational opportunities to emerge.

Teachers frequently encounter issues concerning student engagement with teaching materials. ²² Higher levels of engagement with learning activities can be achieved if the virtual world is more interactive. For instance, debates are typically effective at involving learners in subjects needing critical thinking, ²¹ but they are less suitable for learning factual science subjects like physics or chemistry. VR may be particularly useful for those subjects where a spatial arrangement is important or there are dynamic changes.

VR can enhance engagement and improve retention, as students get to learn through experience, which is necessary for STEM subjects. VR also influences the students’ spatial ability for real-time visualization, which are crucial skills in the engineering field. Students can explore “the use of advanced holographic technologies to bring virtual 3D building components to life” [ 11, para 3]. The implications of VR require investment in resources and technical equipment. Hence there is a need for educational institutions to start with “cheaper alternatives and small mobile devices first” [ 11, para 5].

VR and learner motivation

Many professions, including medical science, engineering, architecture, product development, and geology, have used VR to study and visualize abstract concepts. ²⁵ According to previous research, utilizing VR technology as a teaching tool enhances learners' grasp on the concept, test scores as well as learning motivation while lowering training costs and experimental risks. Employing VR in instruction can potentially increase student learning motivation and positively enhance student performance. ²⁶ However, there is a paucity of studies on the impact of mobile VR on student motivation in STEM courses.

Mobile VR

Although VR has been available for a long time, the associated technology required to access it has been prohibitively expensive, heavy and high battery/power consumption. VR apps have been able to proliferate into the general market because of mobile VR headsets, which are essentially eyewear that can support a smartphone. ¹³

With a number of available ways to experiment with VR content, headsets, video and apps, this tool has the potential to become yet another method in an educator's repertoire for assisting learners in understanding critical theories and mastering a range of concepts. As smartphones with integrated video capability became widely available, the influence of what could be shared and generated in the classroom has improved significantly. VR technology is essentially the first phase leading to the advancement of interactive learning, ¹ and mobile VR adds accessibility and cost-efficiency to that advancement.

Based on the above review of literature, the following hypotheses are proposed for this study: H1:

Visual learners have the highest level of engagement in a mobile VR environment.

The literature review shows that VR is swiftly becoming a household word, thanks to the recent boom of VR-compatible devices. Despite having a significant impact on student engagement and performance, mobile VR is greatly neglected in education. The proposed framework considers factors such as learning styles of learners and technology used to teach in order to measure the motivation and engagement of students belonging to the K-12 group ( Figure 3).

A minimum cluster of 30 students from the same age group under the K-12 curriculum should be selected for the pilot study to be statistically significant. It is very important that the students be selected from a definite age group, preferably high schoolers (grade 9-12), and course, preferably STEM curriculum, as different age groups may produce different results. In addition, the study includes surveys in multiple levels; it’s desired that they be filled in with proper understanding of the questions being asked. High schoolers are young adults that fit the mentioned criteria. During the sample selection, diverse VARK learner types should be included in an even ratio to avoid a biased outcome and a healthy ratio of male to female should be considered to avoid skewing the results. This could be ensured with the “VARK LS Questionnaire” stage mentioned in Figure 3.

The most appropriate device for this study is Google Cardboard, as it is affordable, easy-to-build and has numerous vendors. It also supports both Android and iOS-based mobile devices. The VR devices (Google Cardboard) are to be acquired and distributed by the research team whereas the mobile devices can be the students’ own devices. A list of students’ existing mobile phones can be produced before initiating the test to ensure a smooth execution.

Although most young learners are familiar with VR and can rapidly pick up technology usage or its trends, the teachers or instructors may need to get a better understanding of the possible uses and integration of the existing mobile VR software in the curriculum. Coming up with suitable educational content may require some training or professional help. This can be ensured by allowing time to the educators to explore the technology first and look though the manuals and tutorials available online. Then they can communicate their concerns or areas of difficulty and training/troubleshooting with the help of Google cardboard community can be arranged. For activity logs to be tracked, students will be taught a topic (in line with their study curriculum) using mobile VR technology which will be accompanied by a simple quiz to gauge how well the students understood the topic and how well they retained that information. This process should be repeated several times over the period of one semester to investigate the patterns over a decent time span. The design of the content should be appropriate for the age group, relevant to the topic, and have the ability to be supported by existing mobile VR software such as Expeditions, InCell VR, Titans of space, and Google Daydream. These VR software are some of the most well-established and widely used platforms for educational purposes with a variety of existing materials.

A mixed-method for data collection will be conducted to find the degree of engagement and motivation the learners learning in the mobile VR-assisted e-learning context.

A questionnaire method will be utilized as the first method to collect learner motivation and engagement-related data. •

Measuring motivation: the questionnaire developed by Vallerand et al. ¹⁵ known as EME [(Échelle de Motivation en Éducation (Measure of Motivation towards Education)], which comprises seven subscales measuring “three different kinds of intrinsic motivation and three different kinds of extrinsic motivation”. ¹⁶

A system log or activity log tracking will be used as the second method to collect data regarding engagement. Some of the system logs parameters proposed to be collected in this study are shown in Table 1.

The data collected through activity logs and questionnaires will be mapped to each individual learner correctly. Further visualization, analysis and investigation will be performed using Smart PLS, Python or R language to summarize the main characteristics of the data collected. Various methods of clustering, regression and correlation heatmaps will be applied to have an improved understanding of the relationships of the variables, detect patterns, spot anomalies, and test the hypothesis and other assumptions. Principal component analysis (PCA) will be explored to understand which attributes contribute most to the variance of each model. The Panda and Seaborn libraries in Python are very powerful in terms of data visualization and assessing correlation.