Impact of Metaverse-Based Speaking Activities for Second Language Learning*

Article information

J Eng Teach Movie Media. 2023;24(4):57-72
Publication date (electronic) : 2023 November 30
doi :
1Associate professor, Department of General Education, Seoul Women’s University, 621 Hwarangro, Nowon-Gu, Seoul, 01797, Korea
Corresponding author, Hea-Suk Kim, Associate professor, Department of General Education, Seoul Women’s University, 621 Hwarangro, Nowon-Gu, Seoul, 01797, Korea (E-mail:
*This work was supported by a research grant from Seoul Women’s University (2023-0204).
Received 2023 October 15; Revised 2023 November 6; Accepted 2023 November 25.


This study aimed to investigate whether metaverse-based speaking activities had an impact on enhancing speaking proficiency. 45 students who enrolled in College English during the 2022 spring semester were divided into high and low proficiency groups based on their mock TOEIC scores before the beginning of the semester. Students engaged in speaking activities via a metaverse environment twice a week. All students underwent speaking assessments, which included reading a text aloud, describing a picture, responding to questions, and expressing an opinion, both before and after the experiment. Additionally, a survey was conducted to gauge learner perceptions of metaverse-based speaking activities. The collected data were analyzed using paired sample t-tests and ANCOVA to examine differences within and between groups. The results indicated significant improvements in both groups from pre- to post-experiment assessments. However, it was observed that there existed no discernible disparity in speaking performance by varying degrees of English proficiency. This implies that English proficiency did not yield a statistically significant influence on the overall outcomes. In terms of learner perceptions, both groups exhibited positive attitudes, with no significant differences between them. In conclusion, this study suggests diverse approaches for incorporating metaverse-based speaking activities into L2 courses.

Keywords: tertiary


As the COVID-19 pandemic has extended the period of non-face-to-face interactions, our way of life has evolved to embrace the concept of “untact,”1 wherein numerous offline activities have adapted to align with this trend. Simultaneously, the content and Information Technology (IT) industries have undergone innovative transitions, accelerating due to this shift. Advancements in IT are instigating substantial shifts in our established societal norms across diverse domains, notably in the spheres of economics, culture, society, education, and beyond. In the realm of education, the utilization of information and computer technology is of importance. One of the most critical skills for educators and practitioners is the ability to integrate modern technology into their teaching practices (Hussin, 2018). Furthermore, with the advent of smartphones, social media, and artificial intelligence (AI), educators continually face the challenge of identifying which of these cutting-edge technologies is best suited for developing their instructional learning method.

Recently, a new form of reality, known as Metaverse, has been rapidly emerging by combining the advantages of Virtual Reality (VR) and Augmented Reality (AR). Metaverse is a portmanteau of meta, which means virtual, and universe, signifying the world or cosmos, representing an extension of virtual reality. In the metaverse, people can partake in a range of social activities, encompassing discussions on various subjects, collaborative project work, gaming, and the acquisition of knowledge by experiencing and solving problems (Jovanović & Milosavljević, 2022).

Hwang and Chien (2022) asserted that due to the characteristics inherent in the metaverse, it is anticipated that metaverse-based education distinguishes itself from traditional VR or AR education methods. Within the metaverse context, the objective of language acquisition extends beyond a learning activity. Instead, it seeks to empower L2 (second language) learners to have another life, creating an environment where English becomes their predominant medium for work, education, social interaction, and entertainment. Moreover, one of the differences between conventional online learning and metaverse-based learning is the immediate entry into a virtual world using one’s avatar, facilitating learning in a highly realistic and immersive classroom or situations that closely resemble the real world. In this regard, the metaverse has been recognized as being the next generation of social connection (Hwang & Chien, 2022) and can offer substantial educational benefits (Bae et al., 2022; Hwang, 2022; Jeong et al., 2021).

Currently, research related to the metaverse is underway in various fields, including language education, generating considerable interest (Hidayati et al., 2023; Hussin, 2018; Hwang, 2022; Jeong et al., 2021; H.-S. Kim et al., 2023; S. Kim, 2021). According to Kim (2021), students find the infinite possibilities of the virtual world more appealing than the physical world’s limitations. She highlights the significant educational advantages of metaverse learning. Conversely, certain studies have yielded mixed findings, wherein learners have encountered adverse effects in terms of their interest and motivation for learning within the metaverse (Kim et al., 2023).

Despite the noteworthy interest in metaverse, there is a substantial research gap that requires attention in order to ascertain its effectiveness. While the COVID-19 pandemic has abated, it remains a period during which students must maintain physical distance from one another, making in-person interactions challenging. As a result, there has been a surge of interest in developing metaverse-based methods that can effectively and engagingly support English language learning.

This study endeavors to examine the influence of employing a metaverse environment, which provides immersive experiential learning and encourages active peer communication, as noted by Lee (2017). This examination is particularly focused on diverse proficiency levels (Kim et al., 2023). That is, the present study aims to evaluate the impact of metaverse-based speaking activities on English speaking skills by contrasting lower and higher proficiency levels. To achieve the objectives of this research, the specific research questions are as follows:

  • 1. Do metaverse-based speaking activities improve speaking skills?

  • 2. Do metaverse-based speaking activities impact improvement based on the student’s proficiency level?

  • 3. Does student’s proficiency level affect their perception of metaverse-based activities?


According to Warschauer (1996), the use of technology can help learners develop a positive perception of English learning and enhance their English skills. In recent times, significant strides in technology have revolutionized the way English as a Foreign Language (EFL) learners acquire and improve their language skills. These technological advancements have initiated a period in which learners are no longer confined by the constraints of time and place. In the domain of education, students are increasingly inclined to depart from traditional pedagogical methods (Kim & Hyun, 2021). A noteworthy development in this context is the rise of English language learning programs that employ metaverse-based platforms. Metaverse platforms, which offer high accessibility and a sense of presence, can stimulate learners’ motivation and curiosity in educational environments. They have the potential to enhance learning outcomes by ultimately boosting learners’ engagement and increasing interest and attention (Jeong et al., 2021).

Studies involving the application of augmented reality and virtual environments in education have asserted that these approaches are effective in heightening students’ interest and immersion in the learning process, aligning well with the preferences of Generation Z learners who place a premium on visual elements and experiential learning (Amponsah et al., 2022). The term “MZ generation” refers to a neologism that encompasses both the Millennial (born between 1981 and 1995) and Generation Z (born between 1996 and 2005) generations. The MZ generation is characterized by their adeptness in using digital and IT devices, as well as having individualistic and present-oriented tendencies. Therefore, educational settings should adapt to cater to these generational differences. The VR platforms allow for the easy creation of various spaces such as classroom group spaces and lounges to cater to educational purposes. Instructors can design spaces tailored to the objectives of their lessons, share materials, and discussion topics, making learner interactions more comfortable and effective (Cheon, 2021). Furthermore, from the learners’ perspective, these platforms encourage active participation and engagement, which can lead to increased efficiency and effectiveness in achieving educational goals, ultimately contributing to learners’ knowledge and experience for the future.

Kaplan-Rakowski and Gruber (2023) studied intermediate English learners in eight public speaking sessions over three months, yielding 160 research observations. One group used high-immersion VR, in which participants wore VR headsets and addressed a virtual audience. Another group used Zoom to present to a live learning environment. When compared to Zoom, those practicing in VR had much reduced anxiety levels. This demonstrated VR’s favorable impact on improving foreign language speaking. That is, language learners are more likely to be willing to speak, and the anonymity of VR is likely to reduce anxiety levels. The results are supportive of Hwang’s (2022) study, indicating that the utilization of the metaverse platform can be both motivating and entertaining.

Bae et al. (2022) found that metaverse-based language learning holds significant appeal for EFL students. Having the advantages of gamification and edutainment, the metaverse has the potential to serve as an efficient and beneficial educational platform (Aydin, 2022; Hwang & Chien, 2022). Metaverse boasts features such as lifelike virtual reality implementation and gamification effects, facilitating novel forms of social interaction (Bae et al., 2022). According to Kim (2014), game-based learning can transform learners into gamers, increasing interest in learning and participation in learning activities. It also promotes cooperative learning through friendly competition and facilitates interaction. In this regard, educational games that simultaneously satisfy the elements of learning and motivation have significant potential in the EFL environment (Jin, 2021; Kim, 2014). Moreover, users can create avatars and interact with peers within the same virtual space, amplifying the sense of realism. This is particularly viewed as providing opportunities for learners to actively engage and take initiative in their educational experiences, fostering their interest and immersion through virtualization (Yoon, 2022).

According to a study by Hwang (2022), it explored the effectiveness of metaverse integration in online education, employing the ifland application for Korean university students learning English as a foreign language. The study found that students showed active participation in interactive learning process through avatars representing themselves. He emphasized the platform’s efficacy, attributing it to the incorporation of gamification and edutainment, which lead to substantial educational advantages.

Several platforms have embraced the metaverse concept, including prominent ones such as, Animal Crossing, Decentraland, Roblox, Minecraft, Second Life, Fortnite, and others. In the context of domestic platforms, Cheon (2021) highlighted the emergence of platforms like Zepeto. Zepeto offers a variety of virtual spaces, encompassing classrooms, meeting rooms, and cafes, while integrating essential metaverse elements that facilitate the use of avatars. When introducing the metaverse into the educational environment, a critical consideration is to evaluate the user-friendliness of the new learning environment and app interface, as well as the capacity of the virtual space to engage learners. In view of these criteria, Zepeto could be the most fitting and widely adopted platform (Hidayati et al., 2023).

On the other hand, research has substantiated the efficacy of utilizing the Zepeto platform within the metaverse. Jang (2021) have opted for metaverse as a tool to enhance Korean speaking abilities and have introduced an instructional framework. They also evaluated learners’ levels of satisfaction, interest, self-confidence, and the usefulness of the learning approach. The research findings indicated that a majority of the participants showed favorable perceptions, and the efficacy of language learning was substantiated.

Hidayati et al. (2023) conducted a study to explore how Zepeto impacted speaking skills when connecting with foreign friends. It employed a quasi-experimental design with two classes: one as an experimental group and the other as a control group. The study included pre-tests, treatment, and post-tests with 60 students. The experimental group utilized the Zepeto as a learning tool, while the control group received traditional instruction. The results of the statistical analysis indicated a significant difference between the two groups. In terms of the difficulties encountered when using Zepeto, participants highlighted initial challenges with connecting to foreign users. However, they managed to find solutions and successfully made foreign friends. Occasionally, they felt confusion when communicating with these friends but tackled it by using dictionaries to understand unfamiliar words. Despite these hurdles, they were expected to meet people from different countries.

On the contrary, language learning endeavors linked to the metaverse have yielded unsatisfactory results. Kim et al. (2023) explored the impact of different interactions (face-to-face and Metaverse) on speaking skills. 110 participants were divided into three groups: face-to-face chatting, text-chatting on metaverse, and voice-chatting on metaverse. They completed speaking tasks before and after the experiments. The post-test results did not show any significant difference in language improvement between the traditional and metaverse groups. This suggests that the use of the metaverse did not seem to have a discernible impact on improving students’ language skills. In other words, despite students interacting through the metaverse platform, it did not appear to be beneficial. It can be inferred that the learning outcomes were not influenced by the choice between the metaverse platform and conventional methods. In addition, most students had a negative attitude towards the metaverse learning platform, showing dissatisfaction with confidence, motivation, self-directed learning, and various language skills. They expressed concerns about the platform’s effectiveness in addressing their language learning needs. While some might consider using it in the future, a majority indicated they would not.

Despite the evident benefits of the metaverse in foreign language learning, challenges persist in its utilization as a language learning platform. Some limitations encompass weakened social connections and communication, along with the inability to foster real-world relationships and potential maladaptation to the real world (Kye et al., 2021). Recent research exploring the application of the metaverse has revealed favorable learner perceptions. Nevertheless, contrasting findings have emerged regarding its impact on speaking skills, and there is a marked scarcity of studies involving metaverse. In this regard, there is a prominent need for more comprehensive investigation in the field of metaverse-based research, with a particular emphasis on the utilization of established metaverse platforms like Zepeto. Consequently, the objective of this study is to assess whether the metaverse platform, representing a novel concept within the evolving post-COVID educational environment, can address the limitations associated with conventional pedagogical methods and demonstrate efficacy in foreign language learning.


1. Participants

The study looked into how speaking practice in the metaverse can improve speaking performance. The study involved 53 university students who had enrolled in compulsory English courses. These students were categorized into two groups based on their mock TOEIC scores before the beginning of the semester: one group with low proficiency (M = 297.50) and the other group with high proficiency (M = 693.04). There were originally 25 students in the class with low English proficiency, and 28 students in the class with high English proficiency. However, the data was analyzed with students who participated in both two tests and survey, and the final number was 22 and 23, respectively. The students in the two groups came from various academic backgrounds. Their majors included fields like Contemporary Art, Ceramics & Fiber Arts, Mathematics, Applied Food System, Digital Media Design and Applications, Information Security, and Software Convergence, etc. The participating students were in their first year of college and fall within the age range of 19 to 21.

Table 1 presents a comparative overview of responses from two distinct groups: individuals with lower proficiency in English (n = 22) and those with higher proficiency (n = 23). It delved into various facets of their language learning experiences and interactions with metaverse.

Demographic Information Between Lower Proficiency vs. Higher Proficiency Groups

Firstly, concerning their experiences living in English-speaking countries, only one participant in the lower proficiency group had lived in Australia, while none had lived in other countries. In contrast, the higher proficiency group displayed a more significant degree of international exposure. Specifically, five individuals had lived in countries such as the USA, Canada, France, and the Philippines, while the remaining 18 participants had no prior experience of living abroad.

When it comes to their main purposes for learning English, participants in the lower proficiency group had varied goals. Specifically, 11 participants were driven by the prospect of enhancing their employability, while six aimed for gaining grades. Additionally, three participants had a general aspiration to boost their English skills, and two expressed an interest in English and culture. In contrast, the higher proficiency group demonstrated a more homogeneous set of purposes. The 10 participants were motivated by employability and eight participants shared a keen interest in English and culture as their primary driving force. Four participants prioritized achieving better grades. Interestingly, only one individual in this group expressed a specific desire to enhance English skill.

Regarding their experiences with the metaverse, the majority of students in both groups had limited exposure. When it comes to the specific areas of English they aimed to improve through the metaverse, both groups expressed a strong preference for enhancing their speaking skills. In contrast, reading and writing garnered relatively less attention.

In terms of their areas of vulnerability in English, participants in the lower proficiency group identified two key areas of challenge. Specifically, seven participants found listening to be a significant challenge, while 11 participants cited speaking as their primary area of difficulty. Conversely, participants in the higher proficiency group predominantly expressed vulnerability in speaking, with a substantial 16 participants recognizing it as their most challenging aspect of the language. Lastly, when asked about the areas they wished to improve the most in English, both groups predominantly emphasized speaking, highlighting its significance in their language learning goals.

2. Teaching Procedures and Instruments

After the advent of the COVID-19 pandemic, while classroom-based teaching was possible, students wearing masks were hesitant about face-to-face conversations. Hence, utilizing metaverse for communication seemed like a more natural and convenient mode of interaction.

Prior to the beginning of the semester, each participant completed a series of mock TOEIC tests to assess their proficiency levels. Additionally, they undertook a pre-speaking assessment in the first week of the spring semester in spring in 2022. The questions were selected from the mock TOEIC speaking test2 which included four tasks: reading a text aloud, describing a picture, responding to questions, and expressing an opinion. They answered each question and they recorded their answers and saved their recording files. Later, students were asked to upload their sound files on e-class assignment board.

In the second week, students were instructed to install the Zepeto app, and they were given a demonstration of its basic usage before trying it out. All students were required to use it in their weekly classes. While initially unfamiliar with using metaverse, the students gradually became more accustomed to it, although there were occasional delays in using the app. They conducted metaverse-based speaking activities for approximately 10 weeks, from week 3 to week 13, excluding the midterm exam period.

Students received classes twice a week, each lasting 75 minutes, with approximately 15 minutes dedicated to speaking practice in every session. The course was predominantly focused on listening and speaking skills. The speacking exercises included activities such as topic-related questioning and answering (i.e., “How much time do you think a college student should spend on social networking sites each day?”) and expressing opinions (i.e., “Are there more advantages or disadvantages if college students use social media?”). During the speaking practice sessions, students participated in metaverse-based speaking activities, either individually or as part of a group. The speaking prompts were derived from the course textbook, encompassing various topics such as music, movies, shopping, health, games, social media, food, and travel. Structured questions and given tasks were provided to students, enabling them to engage in communicating and interacting.

In week 14, participants completed a post-test using the same set of questions as the pre-test (Kim et al., 2023). The pre- and post-tests were explained as an assessment to see if there was any change in one’s speaking skills from before to after the semester, with the findings not impacting grades. They were encouraged to take the test in a relaxed manner. Finally, they were directed to respond to a survey questionnaire consisting of ten multiple-choice items pertaining to their experiences of using metaverse and to offer their insights on the strengths and weaknesses of metaverse-based speaking activities.

3. Metaverse: Zepeto

Zepeto, a prominent social networking application originating from South Korea, boasted nearly 300 million global users, as reported by Naver (2022). In Zepeto, users have the ability to personalize their distinct avatars by taking selfies and dressing them up. They can use these digital avatars to interact with others from a distance through actions, voice calls, and capturing moments. Furthermore, users have the opportunity to generate income by creating and selling AR fashion items. The study integrated the metaverse platform, Zepeto, into the curriculum for the participating students (Cheon, 2021; Hidayati et al., 2023). Students were directed to access the virtual space by following the link provided by the instructor in every class and participate in speaking activities within that virtual environment.

4. Analysis

The data included three main components: pre-speaking tests, post-speaking tests, and post-questionnaires. The analysis was conducted using SPSS 25.0 software. Paired sample t-tests were employed to examine how the interaction methods influenced the performance of the four speaking tasks within each group when comparing the pre- and post-tests. To further understand the comparative effects of these interactions between the two groups, an analysis of covariance (ANCOVA) was carried out. This analysis allowed for a thorough investigation into the impact of the different interaction methods depending on participants’ speaking abilities.

Table 2 illustrates a scoring system for evaluating participants in four communication tasks. It assessed aspects such as pronunciation, intonation, stress, grammar, vocabulary, cohesion, content relevance, and completeness. The tasks were scored on a scale from 0 to 3 or 0 to 5, aligning with TOEIC criteria, and this scoring approach allowed for a thorough evaluation of participants’ language and communication abilities across various dimensions. In order to ascertain the inter-rater reliability, the Cronbach’s alpha coefficient was employed. This examination focused on the Cronbach’s alpha values between the researcher and a fellow academic with 16 years of teaching experience. The inter-rater reliability for both the pre- and post-test scores was found to be between 0.93 to 0.95, which seen as an acceptable level.

ETS TOEIC Speaking Evaluation Criteria

The study examined students’ perceptions regarding the utilization of metaverse for speaking practice, employing descriptive statistics to draw comparisons between two proficiency groups. A total of 10 close-ended items were incorporated in the analysis to gather insights into their perceptions and experiences with this technology-driven approach to language learning and speaking practice. A survey employing a six-point Likert scale (ranging from 1 = strongly disagree to 6 = strongly agree) was employed. Furthermore, there were two open-ended questions inquiring about the benefits and drawbacks of metaverse usage. Student responses were scrutinized, categorized into essential themes, and their frequency and proportions were calculated.


1. Changes in Speaking Performance

The current research aimed to investigate how engaging with speaking practice on metaverse impacts speaking abilities. The first research question focused on comparing the average scores of participants before and after using the metaverse. Paired sample t-tests were employed to compare the test scores. Table 3 displays the results of the four speaking tasks’ pre- and post-tests for the lower proficiency group.

Result of Paired Sample t-Tests: Lower Level Group

The test scores evaluated their performance in both pre-tests and post-tests, with the goal of examining the impact of the metaverse on their communication skills (see Table 3). In the “Read a text aloud” task, participants began with an average score of 1.43 in the pre-test, and following an intervention or practice, their scores significantly improved to an average of 2.02 in the post-test. The t-value of -11.06 (p < .00) underscored the statistical significance of this improvement, indicating a substantial enhancement in their performance. Regarding the second task, “Describe a picture”, participants initially scored an average of 0.93 in the pre-test, which notably increased to 1.59 in the post-test after the intervention. The t-value of -8.63 (p < .00) confirmed the statistical significance of this improvement, indicating a significant positive impact on their performance. The “Respond to questions” task displayed a similar pattern, where participants began with an average score of 0.77 in the pre-test. However, following the treatment, their performance substantially improved, with an average score of 1.44 in the post-test. (t = -10.96, p < .00). The result confirmed the substantial positive impact of the intervention on their ability to respond to questions effectively. In the “Express an opinion” task, participants initially scored an average of 0.93 in the pre-test. Remarkably, their scores witnessed a significant increase to an average of 2.02 in the post-test. The results proved the statistical significance of this improvement (t = -7.30, p < .00).

In summary, the outcomes observed across all four communication tasks consistently indicated that the speaking practice on metaverse had a profoundly significant and positive impact on the lower-level students’ speaking performance. These results strongly indicate substantial enhancements in both language and communication skills as a direct result of the intervention.

The presented Table 4 compares higher proficiency groups’ performance in pre- and post-tests. Participants in the “Read a text aloud” task scored an average of 2.41 in the pre-test, but after the intervention or practice, their scores improved dramatically to an average of 2.78. This improvement was supported by a statistically significant t-value of -5.73 (p < .00). Similarly, in the “Describe a picture” task, the participants started with an average score of 1.96 in the pre-test, which increased to 2.30 in the post-test following the intervention. The significant improvement was highlighted by a t-value of -3.81 (p < .00). The “Respond to questions” task followed a similar pattern, with individuals scoring 2.21 on average in the pre-test and 2.52 on average in the post-test. The t-value of -3.89 (p < .00) confirmed the statistical significance. Finally, in the pre-test, they scored an average of 2.89 on the “Express an opinion” task. Laster, their scores climbed significantly to an average of 3.54. This improvement was supported by a substantial t-value of -4.59, signifying statistical significance. Overall, the findings demonstrate that metaverse-based speaking practice had a significant and advantageous impact on the performance of participants with higher proficiency in all four tasks.

Result of Paired Sample t-Tests: High Level Group

The study’s findings suggest that metaverse-based speaking activities have a positive impact on improving learners’ speaking skills, irrespective of their proficiency level in English. This observation aligns with previous research (Hidayati et al., 2023; Hwang, 2022; Jang, 2021), further confirming that the utilization of the metaverse can be an effective tool in assisting learners in enhancing their communication abilities. These results highlight the potential of metaverse-based language learning as a valuable pedagogical approach. Regardless of whether learners have a high or low level of proficiency in English, engaging in speaking activities within the metaverse appears to provide them with opportunities for meaningful language practice and improvement. This outcome is particularly significant in the context of language education, as it demonstrates that metaverse platforms can cater to a diverse range of learners with varying language abilities.

2. Influence of Proficiency Levels on Metaverse-Based Speaking Activities

To address the second research question, whether there were significant differences between the two groups, ANCOVA was conducted, regarding the achievements in the post-tests. The findings dealt with the comparison of mean scores between two proficiency levels.

Table 5 presents descriptive statistics including mean (M), standard deviation (SD), adjusted mean (Adjusted-M), and standard error (SE) for each group’s performance. In “Read a text aloud,” the lower group had an average score of 2.02 (SD = .39), and their adjusted mean was 2.38 (SE = .080). In contrast, the higher group in the same task scored 2.78 (SD = .33), an adjusted-M of 2.44 (SE = .077). For the “Describe a picture” task, the lower group had an M of 1.59 (SD = .53), an adjusted-M of 1.99 (SE = .102). The higher group in this task had an M of 2.30 (SD = .56), an adjusted-M of 1.92 (SE = .089). In the “Respond to questions” task, the lower group achieved an M of 1.44 (SD = .47), an adjusted-M of 1.92 (SE = .096). Meanwhile, the higher group scored 2.52 (SD = .39), an adjusted-M of 2.06 (SE = .093). Finally, for the “Express an opinion” task, the lower group’s mean score was 2.02 (SD = .92), with an adjusted-M of 2.64 (SE = .183). The higher group had an M of 3.54 (SD = .71), an adjusted-M of 2.96 (SE = .159).

Descriptive Statistics of the Post-Test Scores Between Groups

To examine potential variations in speaking performance between the two proficiency levels, one-way ANCOVAs were employed, with the pre-test serving as a covariate for the post-tests of each speaking task. The findings are displayed in Table 6, demonstrating that there were no statistically significant differences in the four tasks among the groups: reading a text aloud (F = .263, p = .611), describing a picture (F = .156, p = .694), responding to questions (F = .690, p = .411), and expressing an opinion (F = 1.404, p = .243).

Results of ANCOVA

Therefore, language activities based on the metaverse may be perceived as not significantly affecting learners’ proficiency levels. This result implies that irrespective of their proficiency levels, learners could enhance their performance in English speaking within metaverse-based activities. Unlike the study of Kim et al. (2023) findings, which highlighted potential communication issues for learners who involved in activities on metaverse, both groups demonstrated noticeable improvement in their communication abilities.

3. Students’ Perceptions Towards Metaverse-Based Language Learning

1) Analysis of Closed-Ended Survey Regarding Perceptions of Metaverse

A post-survey was used to examine students’ perceptions on speaking activities using metaverse. There were 10 items asking about the perceptions towards using metaverse for English speaking practice. Table 7 shows the mean scores for each level and identifies whether there were any significant differences between the two proficiency levels.

Results of Post-Survey: Low and High Proficiency Levels

In the study’s findings, several key statements were evaluated to gauge participants’ perceptions and experiences when using metaverse for learning and enhancing their English skills in two distinct groups (see Table 7). The first statement, “Using metaverse improves my learning skills,” received an average rating of 3.77 in the lower group, while the higher group rated it slightly higher at 3.87. Similarly, the statement “Using metaverse improves my English skills” had an average rating of 3.95 in the lower group and 4.09 in the higher group. Notably, “Using metaverse results in a learning benefit” and “Using metaverse is helpful for learning” both had closely aligned average scores between the two groups, demonstrating little differentiation in their responses. Furthermore, the survey assessed participants’ perceptions of ease of interaction with people through the metaverse. The lower group rated it at 4.05, while the higher group gave it a lower rating of 3.61. That is, the students in the higher proficiency group tended to find it relatively easier to engage in face-to-face conversations. This implies that speaking in person is more comfortable and straightforward for them. In terms of cognitive load reduction, “Using metaverse reduces cognitive load” had an average rating of 4.45 in the lower group and 4.09 in the higher group. This can be interpreted that students with lower proficiency levels might experience a lower degree of anxiety while engaging in conversations within the metaverse. Interestingly, the statements regarding the enjoyment and interest in using metaverse for learning English, as well as the desire to continue using it, displayed varying scores, with the lower group generally having slightly higher scores compared to the higher group. In sum, the differences in averages and the results of t-tests between the two groups were not statistically significant across all the survey items.

According to the findings, when it came to using metaverse to learn and enhance their English skills, lower and higher proficiency level students had similar attitudes. Among the survey items, both groups exhibited a strong preference, especially regarding the perceived benefits of metaverse for learning, its ability to reduce cognitive load, and the engaging and enjoyable nature of metaverse-based speaking activities. The findings of this study align with those of the earlier research conducted by Jang (2021), where a significant number of participants exhibited positive perceptions toward metaverse, thus reinforcing the effectiveness of language learning.

2) Analysis of Open-Ended Survey: Benefits and Limitations of Metaverse-Based Speaking Activity

An open-ended survey was conducted after finishing a semester-long experiment to better investigate students’ thoughts and ideas on metaverse-based speaking activities. Students were encouraged to freely describe the benefits and drawbacks of utilizing metaverse in English classrooms. Their answers were classified and analyzed. Tables 8-11 give an examination of the responses from both groups. Out of the lower proficiency group, two students did not submit their responses, making the total number of responses 20, whereas all students from the higher proficiency group responded, resulting in a total of 23 responses.

Benefits Towards Metaverse-Based Speaking Activity: Lower Proficiency Group

Drawbacks Towards Metaverse-Based Speaking Activity: Lower Proficiency Group

Benefits Towards Metaverse-Based Speaking Activity: Higher Proficiency Group

Drawbacks Towards Metaverse-Based Speaking Activity: Higher Proficiency Group

Table 8 illustrates feedback from participants from the lower proficiency group regarding their experiences with metaverse, categorized into four themes. Participants in this category appreciated that metaverse-based speaking reduced the burden of direct communication, which could boost their confidence. It made speaking in English less awkward, especially when interacting with unfamiliar peers. This category received 50% of the responses, indicating its significance.

Next, the students highlighted the convenience of being able to communicate with individuals who were seated far away without having to physically move. They valued the ease of speaking without changing their seating arrangement, and this category received 30% of the responses. Some participants found metaverse to be an enjoyable and interesting way to make new acquaintances in the virtual world. They likened studying English on metaverse to playing games and believed it captured people’s interest effectively. This theme constituted 10% of the responses. Lastly, as for the miscellaneous comments, some could meet many of their classmates and experiencing fewer limitations in speaking activities (10%).

Therefore, these responses offer valuable insights into the varied experiences and perceptions of students utilizing the metaverse for language learning. It sheds light on the aspects that contributed to lower anxiety and increased confidence among the students.

Table 9 summarizes feedback from lower proficiency participants regarding issues they encountered while using the metaverse-based Zepeto app, categorized into four themes. The lower level group reported difficulty receiving errors when attempting to join metaverse activities, and they found it challenging to stay engaged when their internet connection was unstable. Network issues accounted for 40% of the responses, highlighting the significance of these concerns. Next, 30% mentioned that it took a while for audio to load, and there were instances of overlapping sounds. Additionally, some participants expressed a preference for hearing only their conversation with others, highlighting issues with the audio settings or functionality. 15% of the participants described issues related to system errors, particularly poor connection quality and unpleasant experiences resulting from errors. Lastly, regarding miscellaneous comments, some participants expressed a preference for face-to-face communication over using the Zepeto app, while others mentioned feeling less concentrated when interacting with others face-to-face.

Together, these categories offer significant insights into the obstacles and issues encountered by participants while utilizing the metaverse platform. It highlights the importance of addressing network and audio issues, as well as improving system performance, to enhance the overall user experience.

Table 10 shows the benefits of their experiences with the metaverse environment, categorized into five themes for higher-level students. The most prominent theme, accounting for 39.1% of responses, focused on how metaverse reduced the perceived burden and enhanced confidence in speaking. Participants found that speaking through metaverse was less burdensome compared to face-to-face interactions. They expressed that using metaverse for exchanging opinions allowed them to speak more comfortably and confidently, as it provided them with the opportunity to think about expressions at their own pace.

Approximately 26.1% of the responses highlighted the theme of more changes to communicate. The higher level group appreciated that metaverse allowed them to converse freely with unfamiliar peers, providing opportunities to speak in English that might not have been available in face-to-face interactions. This category emphasized the accessibility of communication within the virtual environment. Some participants (17.4%) noted that metaverse had a positive impact on improving their English language skills. They mentioned that the metaverse environment facilitated practice and improvement in their conversation skills, as they had to respond immediately when listening to questions in English. Two participants found the experience of using the metaverse to be enjoyable and interesting. They relished the opportunity to meet classmates in the virtual world and engage more in conversations in various dialects of English. This theme highlighted the aspect of enjoyment and novelty in the learning process. Lastly, two participants expressed convenience in communication and the feasibility of socializing with others even in a virtual world. As a result, they emphasize the metaverse platform’s capacity to alleviate communication challenges, promote meaningful dialogues, facilitate language practice and enhancement, and offer an enjoyable and convenient mode of interaction.

Table 11 summarizes feedback from higher proficiency participants regarding challenges they encountered while using the metaverse environment, categorized into five themes. 34.8% of responses primarily focused on audio-related problems. Participants mentioned difficulties in hearing responses and challenges in communicating clearly. These issues highlighted the importance of addressing audio quality and clarity within the platform. Likewise, 34.8% responses revolved around system-related errors. Participants reported extended login times and issues accessing rooms due to lengthy loading times. Network-related challenges accounted for 17.4% of the responses. They described poor network connections that hindered their ability to hear the other person’s voice and resulted in communication disruptions. Inexperience with the platform was highlighted. Two of them expressed that they were not accustomed to using it and found it negative when the platform took time to get used to. Finally, one student expressed a sense of humiliation when initiating conversations.

According to the results, it is clear that the use of English in the metaverse platform had a positive influence on students’ learning. This approach offers students valuable opportunities to assume various roles in a realistic setting, providing them with immersive and engaging language learning experiences. Moreover, the anonymity factor may contribute to reducing burden, increase confidence, meeting new friends, and creating a more relaxed learning environment for students to practice English. Additionally, students found the metaverse platform to be more enjoyable and interesting for learning English.

However, it is important to acknowledge that not all students may benefit from this learning method, as some may not find it suitable for their preferences. Moreover, in metaverse-based education, it is vital to ensure that students have a solid understanding of the operation of individual devices and applications before commencement. Additionally, conducting pre-assessments of the system errors, audio problems, and network issues can proactively minimize any potential inefficiencies. These technical challenges, as identified in prior research (Kim et al., 2023), are perceived as drawbacks. This underscores the significance of instructors possessing the experience and knowledge to address such metaverse-related issues.


Although the COVID-19 pandemic has diminished, the need for students to adhere to physical distancing measures remains, posing challenges to traditional in-person interactions. Consequently, there has been a notable surge of interest in developing metaverse-based methods to provide effective and engaging support for English language learning. This study endeavored to explore how a metaverse-based learning environment affects English speaking skills across various proficiency levels.

The first research question aimed to investigate the impact of metaverse-based speaking activities on students’ speaking skills. A comparison of students’ pre-test and post-test speaking scores for both groups revealed improvements in the four speaking tasks: reading a text aloud, describing a picture, responding to questions, and expressing an opinion. All the students demonstrated significant improvements in their speaking performance following metaverse-based speaking practice. These findings align with prior research, suggesting that the use of the metaverse platform contributes to the enhancement of English-speaking skills (Hidayati et al., 2023; Hwang, 2022; Jang, 2021). These results underscore the potential of metaverse-based language learning as a valuable pedagogical approach. Participating in speaking activities within the metaverse appears to provide valuable opportunities for meaningful language practice and improvement.

Concerning the second research question, ANCOVAs were utilized to assess differences in speaking performance between two proficiency levels. The analysis of the data revealed that both groups exhibited enhanced speaking scores, with no significant distinctions observed between the groups. It suggests that regardless of a student’s current proficiency level in L2, the use of the metaverse as an educational tool has a positive impact on their learning experience. It can further booster their participation in the learning process, ultimately resulting in speaking skill enhancements. This outcome holds particular significance in the field of language education, showcasing the metaverse’s capacity to accommodate a diverse range of learners with varying language skills. It implies that educators should contemplate the integration of such platforms into their language teaching approaches to cater to learners of different proficiency levels.

Lastly, the study evaluated participants’ perceptions and experiences in two proficiency groups when using metaverse for learning and practice their English skills. The findings showed that, overall, both groups had similar perceptions and experiences. There were only modest differences between the two groups and these differences were not statistically significant. Participants generally had favorable perceptions and experiences when they performed metaverse-based speaking practice in the metaverse platform.

Concerning the advantages of utilizing the metaverse, feedback from the lower proficiency group regarding its use for language learning can be categorized into four themes: reducing the burden of direct communication and boosting confidence; making English-speaking less awkward, especially with unfamiliar peers; the convenience of communicating with individuals without altering their seating arrangements; enjoyable and akin to playing games, which makes language learning engaging. In terms of the drawbacks of their experiences, the students encountered issues such as network problems, delays and overlapping sounds in audio and system errors.

When it comes to the benefits described by higher-level students using the metaverse platform, similar to the lowerlevel group, several responses were observed. Their comments encompassed reduced communication burden and increased confidence, the facilitation of meaningful conversations, language skill enhancement, enjoyment of the experience, and convenience of communication. These results emphasize the favorable influence of the metaverse on communication and language learning. Nevertheless, they also provided feedback on challenges, such as audio issues, system errors, network-related obstacles, unfamiliarity with the platform, and mixed sentiments about initiating conversations.

In summary, this approach enabled students to assume various roles in authentic settings, offering them immersive and captivating language learning experiences. Anonymity played a role in reducing stress, boosting confidence, fostering new connections, and creating a more relaxed atmosphere for English speaking practice (Kaplan-Rakowski & Gruber, 2023). Students also expressed enjoyment and more engagement with the metaverse platform for learning English, supporting earlier research findings (Aydin, 2022; Hwang, 2022). However, when integrating the metaverse into the learning environment, it is important to consider technical challenges, as highlighted in previous research (Kim et al., 2023). This underscores the importance of instructors being prepared to address potential technical issues unique to the metaverse context.

In light of the study’s results, it is affirmed that incorporating the metaverse into English classes has the capacity to instill learners with engagement and interest. Particularly noteworthy is the observation that sustained motivation and the subsequent facilitation of self-directed learning can be achieved when learners engage in diverse practical tasks within a metaverse environment designed to emulate real-world scenarios.

The study identified educational implications, yet it had limitations. The small number of participants implies that the findings cannot be generalized. Furthermore, as the study aimed to establish the effectiveness of using Zepeto within the metaverse, different outcomes might be obtained with different types of the metaverse platforms.

Based on the current study’s findings, the following future studies are recommended. Conducting metaverse experiments in a classroom setting may not entirely encompass the features of the metaverse in language education. Therefore, within an online classroom environment, it could offer more useful insights. In addition, it may be valuable to investigate how metaverse-based learning activities affect not only L2 speaking but also other language domains. Lastly, there is a necessity for research to examine how metaverse-based activities influence learners’ global competencies and digital literacy across a range of metaverse-based language tasks.



Untact is a Korean term created by combining contact with the prefix un, indicating the absence of face-to-face interaction and physical touch.


Hackers TOEIC Speaking (Hackers Language Research Institute Publisher, 2011)


Amponsah S., van Wyk M. M., Kolugu M. K.. 2022;Academic experiences of “Zoom-fatigue” as a virtual streaming phenomenon during the COVID-19 pandemic. International Journal of Web-based Learning and Teaching Technologies 17(6):1–16.
Aydin S.. 2022;The metaverse in foreign language learning: A theoretical framework. Innovational Research in ELT 3(2):12–19.
Bae Y., Seok H., Jun Y., Lee J.. 2022;Development research of English speaking program using metaverse. Journal of Learner-Centered Curriculum and Instruction 22(24):777–800.
Cheon S.-M.. 2021;A study on students perceptions of blended learning in English education. The Journal of Linguistics Science Academic Journal 99:145–167.
Hackers Language Research Institute Publisher. (2011). Hackers TOEIC speaking. Hackers Language Research Institute.
Hidayati A. N., Muin A., Rahmawati E.. 2023;The effectiveness of Zepeto game to improve students speaking skill. Journal Bahasa dan Sastra 15(1):44–49.
Hussin A. A.. 2018;Education 4.0 made simple: Ideas for teaching. International Journal of Education & Literacy Studies 6(3):92–98.
Hwang G. J., Chien S. Y.. 2022;Definition, roles, and potential research issues of the metaverse in education: An artificial intelligence perspective. Computers and Education: Artificial Intelligence 3:1–6.
Hwang Y.. 2022;Preliminary investigation on student perspectives and satisfaction with distance education in the metaverse world: Focusing on the use of ifland app. The Korea Contents Association 22(3):121–133.
Jang J.. 2021;A study on a Korean speaking class based on metaverse: Using Gather. town. Journal of Korean Language Education 32(4):279–301.
Jeong Y. S., Lim T., Ryu J.. 2021;The effects of spatial mobility on metaverse based online class on learning presence and interest development in higher education. Journal of Korean Association for Educational Information and Media 27(3):1167–1188.
Jin S.. 2021;The effects of digital virtual reality game-based language learning on English language learners’ development of vocabulary and cultural knowledge and affective attitudes. Journal of English Teaching through Movies and Media 22(3):78–94.
Jovanovi&#x00107; A., Milosavljevi&#x00107; A.. 2022;VoRtex metaverse platform for gamified collaborative learning. Electronics 11(3):317–337.
Kaplan-Rakowski R., Gruber A.. 2023;The impact of high-immersion virtual reality on foreign language anxiety. Smart Learning Environments 10:46.
Kim H.-S., Kim N.-Y., Cha Y.. 2023;Exploring the potential of Metaverse as a future learning platform for enhancing EFL learners’ English proficiency. Korean Journal of English Language and Linguistics 23:220–236.
Kim S.. 2014;Effects of a gamified learning environment on learning experiences: A case study on a general English course using relative evaluation policy. Multimedia-Assisted Language Learning 17(3):68–94.
Kim Y. A., Hyun I. S.. 2021;A study of global competence-English communication convergence class. Korean Journal of General Education 15(1):135–148.
Kim Y.-J.. 2021;A study on the convergence of types in game and non-game metaverse contents. The Korean Journal of Animation 17(3):80–99.
Kye B., Han N., Kim E., Park Y., Jo S.. 2021;Educational applications of metaverse: possibilities and limitations. Journal of Educational Evaluation for Health Professions 18:32.
Lee D. H.. 2017;A development plan for a screen English learning system through 3D augmented reality. STEM Journal 18(3):99–117.
Naver. (2022, March 4). Asia’s No. 1 metaverse platform ZEPETO reaches 300 million global users. Naver.
Yoon T.. 2022;Comparison study of pre-service primary English teachers on online learning platforms. The Jungang Journal of English Language and Literature 64(2):195–216.
Warschauer M.. 1996;Comparing face-to-face and electronic communication in the second language classroom. CALICO Journal 13(2):7–26.

Article information Continued


Demographic Information Between Lower Proficiency vs. Higher Proficiency Groups

Lower level (n = 22) Higher level (n = 23)
Experience of living in an English-speaking country Yes 1 5
No 21 18
Main purpose of learning English Interest in English and culture 2 8
Grades 6 4
Employability 11 10
Desire to improve English 3 1
Experience of using metaverse Yes 2 4
No 20 19
Area you want to learn English using metaverse Listening 3 4
Speaking 17 16
Reading 1 2
Writing 1 1
Area in which you feel most vulnerable in English Listening 7 3
Speaking 11 16
Reading 2 2
Writing 2 2
Area in which you want to improve the most in English Listening 5 2
Speaking 14 19
Reading 2 1
Writing 1 1


ETS TOEIC Speaking Evaluation Criteria

Question Task Evaluation criteria Score
1 Read a text aloud Pronunciation, intonation and stress 0-3
2 Describe a picture all of the above, plus grammar, vocabulary, cohesion 0-3
3-5 Respond to questions all of the above, plus relevance of content, completeness of content 0-3
6 Express an opinion all of the above 0-5


Result of Paired Sample t-Tests: Lower Level Group

Task Test M SD df t p
Read a text aloud Pre-test 1.43 .39 21 -11.06 .00**
Post-test 2.02 .39
Describe a picture Pre-test .93 .44 21 -8.63 .00**
Post-test 1.59 .53
Respond to questions Pre-test .77 .55 21 -10.96 .00**
Post-test 1.44 .47
Express an opinion Pre-test .93 .94 21 -7.30 .00**
Post-test 2.02 .92

p < .01


Result of Paired Sample t-Tests: High Level Group

Task Test M SD df t p
Read a text aloud Pre-test 2.41 .33 22 -5.73 .00**
Post-test 2.78 .33
Describe a picture Pre-test 1.96 .56 22 -3.81 .00**
Post-test 2.30 .56
Respond to questions Pre-test 2.21 .39 22 -3.89 .00**
Post-test 2.52 .39
Express an opinion Pre-test 2.89 .88 22 -4.59 .00**
Post-test 3.54 .71

p < .01


Descriptive Statistics of the Post-Test Scores Between Groups

Task Group M SD Adjusted-M SE
Read a text aloud Low 2.02 .39 2.38 .080
High 2.78 .33 2.44 .077
Describe a picture Low 1.59 .53 1.99 .102
High 2.30 .56 1.92 .089
Respond to questions Low 1.44 .47 1.92 .096
High 2.52 .39 2.06 .093
Express an opinion Low 2.02 .92 2.64 .183
High 3.54 .71 2.96 .159


Results of ANCOVA

Task Source Type III sum of squares df MS F Sig.
Read a text aloud Pre-test 2.707 1 2.707 38.608 .000
Group .018 1 .018 .263 .611
Describe a picture Pre-test 6.416 1 6.416 42.967 .000
Group .023 1 .023 .156 .694
Respond to questions Pre-test 4.175 1 4.175 45.954 .000
Group .063 1 .063 .690 .411
Express an opinion Pre-test 13.453 1 13.453 37.073 .000
Group .510 1 .510 1.404 .243


Results of Post-Survey: Low and High Proficiency Levels

M SD t p
1. Using metaverse improves my learning skills. Group 1 3.77 .81 -.41 .68
Group 2 3.87 .76
2. Using metaverse improves my English skills. Group 1 3.95 .84 -.49 .62
Group 2 4.09 .95
3. Using metaverse results in a learning benefit. Group 1 4.05 1.17 .01 1.00
Group 2 4.04 .77
4. Using metaverse is helpful for learning English. Group 1 3.91 1.06 -.33 .75
Group 2 4.00 .80
5. It is easy to interact with people through metaverse. Group 1 4.05 1.17 1.24 .22
Group 2 3.61 1.20
6. Using metaverse reduces cognitive load. Group 1 4.45 1.01 1.15 .26
Group 2 4.09 1.12
7. It is fun to use metaverse to learn English. Group 1 3.91 1.02 -1.17 .25
Group 2 4.22 .74
8. It is interesting to use metaverse to learn English. Group 1 4.41 1.05 .69 .49
Group 2 4.22 .80
9. Communicating with others through metaverse makes me comfortable. Group 1 4.59 1.01 1.56 .13
Group 2 4.17 .78
10. I want to continue to use metaverse to learn English. Group 1 3.86 1.28 .46 .65
Group 2 3.70 1.15

Note. Group 1 = Lower level; Group 2 = Higher level


Benefits Towards Metaverse-Based Speaking Activity: Lower Proficiency Group

Rank Responses n %
1 Reduce burden & increase confidence 10 50
- Not having to communicate directly to individuals reduces the load.
- I suppose one perk of adopting Zepeto was that it made conversing in English less awkward, especially with friends I didn’t know.
2 More communication possible 6 30
- It was convenient to be able to communicate with those who sit far away.
- I was able to speak without moving my seat.
3 Fun & interesting 2 10
- It was enjoyable (interesting, fun) to make new acquaintances in the new virtual world.
- I believe I was able to study English in the manner of playing games on Zepeto.
4 Others 2 10
- I got to meet a lot of my classmates.
- I had fewer limits on speaking activities.
Total 20 100


Drawbacks Towards Metaverse-Based Speaking Activity: Lower Proficiency Group

Rank Responses n %
1 Network issues 8 40
- It was a little difficult to receive errors when joining the metaverse activity.
- It was difficult to engage when the internet connection was unstable.
2 Audio issues 6 30
- It takes a while to load and there is overlapping sound.
- I only need to hear my talk with the other person, but I can hear others.
3 System errors 3 15
- The connection is poor.
- It was unpleasant because there were some errors cased.
4 Others 3 15
- I preferred face-to-face communication over using metaverse.
- I believe I am less concentrated when interacting with others face-to-face.
Total 20 100


Benefits Towards Metaverse-Based Speaking Activity: Higher Proficiency Group

Rank Responses n %
1 Reduce burden & improve confidence 9 39.1
- Speaking through metaverse was less burdensome.
- It appears that using metaverse to exchange opinions is less burden than talking face to face and spending a bit more time thinking about expressions.
- I could get greater confidence in speaking in a more relaxed manner.
2 More communication possible 6 26.1
- You can talk freely with friends you don’t know.
- I have more opportunities to speak in English compared to face-to-face conversation.
3 Improvement in speaking skills 4 17.4
- My conversation skills have increased.
- Because I had to respond immediately when I heard it, I was able to practice speaking English a lot.
4 Fun & interesting 2 8.7
- I enjoyed the metaverse experience.
- Without a doubt, I think it was a lot of fun to meet my classmates in the virtual world and chat in different dialects of English.
5 Others 2 8.7
- It was convenient because it was easy to communicate with others.
- I believe it is feasible to hang out with people even when they are in virtual world.
Total 23 100


Drawbacks Towards Metaverse-Based Speaking Activity: Higher Proficiency Group

Rank Responses n %
1 Audio issues 8 34.8
- It was difficult to hear the responses.
- I faced challenges communicating clearly because of audio problems.
1 System errors 8 34.8
- It took too long to log in.
- There were some issues when I couldn’t access the room due to the lengthy loading time.
3 Network issues 4 17.4
- The network connection was poor, so I couldn’t hear the other person’s voice well, and the connection was cut off, resulting in poor communication.
- Access delays due to school Wi-Fi issues.
4 Inexperience in use 2 8.7
- I wasn’t used to using it.
- I consider it a drawback if it requires a significant adjustment period.
5 Others 1 4.3
- It is humiliating to initiate the conversation.
Total 23 100